├── other ├── des.py ├── yolov8n.pt ├── framework.jpg ├── yolov8.py ├── gpt4_utils.ipynb ├── bert_utils.ipynb └── blip2_utils.ipynb ├── Client_Split ├── Unsupervised │ └── des.py └── Weakly_Supervised │ ├── des.py │ ├── ST │ ├── des.py │ ├── test_gts │ │ ├── client1.npy │ │ ├── client2.npy │ │ ├── client3.npy │ │ ├── client4.npy │ │ ├── client5.npy │ │ ├── client6.npy │ │ └── client7.npy │ ├── test_clients │ │ ├── client2.list │ │ ├── client3.list │ │ ├── client5.list │ │ ├── client7.list │ │ ├── client6.list │ │ ├── client4.list │ │ └── client1.list │ └── new_split │ │ ├── client2.list │ │ ├── client3.list │ │ ├── client5.list │ │ ├── client7.list │ │ ├── client4.list │ │ ├── client6.list │ │ ├── client0.list │ │ └── client1.list │ └── XD │ ├── des.py │ ├── test_gts │ ├── client1.npy │ ├── client2.npy │ ├── client3.npy │ ├── client4.npy │ ├── client5.npy │ └── client6.npy │ └── test_clients │ └── clients │ └── client4.list ├── Code ├── log.py ├── cluster_utils.py ├── modules.py ├── test.py ├── loss.py ├── model.py ├── fusion.py ├── layers.py ├── distill.py ├── environment.yml ├── train.py └── main.py └── README.md /other/des.py: -------------------------------------------------------------------------------- 1 | print("framework") 2 | -------------------------------------------------------------------------------- /Client_Split/Unsupervised/des.py: -------------------------------------------------------------------------------- 1 | print("unsupervised client split." 2 | -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/des.py: -------------------------------------------------------------------------------- 1 | print('weakly supervised client split.') 2 | -------------------------------------------------------------------------------- /other/yolov8n.pt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/other/yolov8n.pt -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/des.py: -------------------------------------------------------------------------------- 1 | print("ShanghaiTech dataset client split.") 2 | -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/XD/des.py: -------------------------------------------------------------------------------- 1 | print("XD-Violence dataset client split.") 2 | -------------------------------------------------------------------------------- /other/framework.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/other/framework.jpg -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_gts/client1.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/ST/test_gts/client1.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_gts/client2.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/ST/test_gts/client2.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_gts/client3.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/ST/test_gts/client3.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_gts/client4.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/ST/test_gts/client4.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_gts/client5.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/ST/test_gts/client5.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_gts/client6.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/ST/test_gts/client6.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_gts/client7.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/ST/test_gts/client7.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/XD/test_gts/client1.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/XD/test_gts/client1.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/XD/test_gts/client2.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/XD/test_gts/client2.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/XD/test_gts/client3.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/XD/test_gts/client3.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/XD/test_gts/client4.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/XD/test_gts/client4.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/XD/test_gts/client5.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/XD/test_gts/client5.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/XD/test_gts/client6.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Eurekaer/FedVAD/HEAD/Client_Split/Weakly_Supervised/XD/test_gts/client6.npy -------------------------------------------------------------------------------- /other/yolov8.py: -------------------------------------------------------------------------------- 1 | from ultralytics import YOLO 2 | 3 | # Create a new YOLO model from scratch 4 | model = YOLO('yolov8n.yaml') 5 | 6 | # Load a pretrained YOLO model (recommended for training) 7 | model = YOLO('yolov8n.pt') 8 | 9 | # Train the model using the 'coco128.yaml' dataset for 3 epochs 10 | results = model.train(data='coco128.yaml', epochs=3) 11 | 12 | # Evaluate the model's performance on the validation set 13 | results = model.val() 14 | 15 | with open('video filename txt','r') as file: 16 | data = file.readlines() 17 | for d in data: 18 | d = d.strip() 19 | print(d) 20 | results = model(d) 21 | # break 22 | 23 | -------------------------------------------------------------------------------- /Code/log.py: -------------------------------------------------------------------------------- 1 | import logging 2 | 3 | 4 | def get_logger(filename, verbosity=1, name=None): 5 | level_dict = {0: logging.DEBUG, 1: logging.INFO, 2: logging.WARNING} 6 | # formatter = logging.Formatter( 7 | # "[%(filename)s][%(levelname)s] %(message)s" 8 | # ) 9 | formatter = logging.Formatter( 10 | "[%(asctime)s][%(filename)s][line:%(lineno)d][%(levelname)s] %(message)s" 11 | ) 12 | logger = logging.getLogger(name) 13 | logger.setLevel(level_dict[verbosity]) 14 | 15 | fh = logging.FileHandler(filename, "a") 16 | fh.setFormatter(formatter) 17 | logger.addHandler(fh) 18 | 19 | sh = logging.StreamHandler() 20 | sh.setFormatter(formatter) 21 | logger.addHandler(sh) 22 | 23 | return logger -------------------------------------------------------------------------------- /Code/cluster_utils.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import numpy as np 3 | from sklearn.cluster import KMeans 4 | def Dist(a, b): 5 | return torch.norm(a - b, dim=1) 6 | 7 | def cluster_assignment_update(params_matrix, bert_features, cluster_centers_w, cluster_centers_o, alpha, beta): 8 | num_clients = params_matrix.shape[0] 9 | num_clusters = cluster_centers_w.shape[0] 10 | assignment = torch.zeros(num_clients, num_clusters) 11 | 12 | for i in range(num_clients): 13 | model_weights = torch.tensor(params_matrix[i]).float() 14 | combined_distribution = torch.tensor(bert_features[i]).float() 15 | 16 | distances = alpha * Dist(model_weights, cluster_centers_w) + beta * Dist(combined_distribution, cluster_centers_o) 17 | cluster_index = torch.argmin(distances) 18 | assignment[i, cluster_index] = 1 19 | 20 | return assignment 21 | 22 | def update_cluster_centers(r_ik, params_matrix, bert_features): 23 | r_sum = torch.sum(r_ik, dim=0, keepdim=True) # Shape (1, num_clusters) 24 | r_sum += 1e-8 * (r_sum == 0).float() # Prevent division by zero 25 | new_cluster_centers_w = torch.mm(r_ik.t(), torch.tensor(params_matrix).float()) / r_sum.t() 26 | new_cluster_centers_o = torch.mm(r_ik.t(), torch.tensor(bert_features).float()) / r_sum.t() 27 | return new_cluster_centers_w, new_cluster_centers_o 28 | -------------------------------------------------------------------------------- /Code/modules.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import torch.nn.init as torch_init 3 | import torch.nn as nn 4 | 5 | from layers import * 6 | 7 | 8 | class XEncoder(nn.Module): 9 | def __init__(self, d_model, hid_dim, out_dim, n_heads, win_size, dropout, gamma, bias, norm=None): 10 | super(XEncoder, self).__init__() 11 | self.n_heads = n_heads 12 | self.win_size = win_size 13 | self.self_attn = TCA(d_model, hid_dim, hid_dim, n_heads, norm) 14 | self.linear1 = nn.Conv1d(d_model, d_model // 2, kernel_size=1) 15 | self.linear2 = nn.Conv1d(d_model // 2, out_dim, kernel_size=1) 16 | self.dropout1 = nn.Dropout(dropout) 17 | self.dropout2 = nn.Dropout(dropout) 18 | self.norm = nn.LayerNorm(d_model) 19 | self.loc_adj = DistanceAdj(gamma, bias) 20 | 21 | def forward(self, x, seq_len): 22 | adj = self.loc_adj(x.shape[0], x.shape[1]) 23 | mask = self.get_mask(self.win_size, x.shape[1], seq_len) 24 | x = x + self.self_attn(x, mask, adj) 25 | x = self.norm(x).permute(0, 2, 1) 26 | x = self.dropout1(F.gelu(self.linear1(x))) 27 | x_e = self.dropout2(F.gelu(self.linear2(x))) 28 | 29 | return x_e, x 30 | 31 | def get_mask(self, window_size, temporal_scale, seq_len): 32 | m = torch.zeros((temporal_scale, temporal_scale)) 33 | w_len = window_size 34 | for j in range(temporal_scale): 35 | for k in range(w_len): 36 | m[j, min(max(j - w_len // 2 + k, 0), temporal_scale - 1)] = 1. 37 | 38 | m = m.repeat(self.n_heads, len(seq_len), 1, 1).cuda() 39 | 40 | return m 41 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # FedVAD: Enhancing Federated Video Anomaly Detection with GPT-Driven Semantic Distillation 2 | ## Introduction 3 | This repo is the official implementation of "FedVAD: Enhancing Federated Video Anomaly Detection with GPT-Driven Semantic Distillation". 4 | ![image](https://github.com/Eurekaer/FedVAD/blob/main/other/framework.jpg) 5 | ## Requirements 6 | The code requires ```python>=3.8``` and the following packages: 7 | ``` 8 | torch==1.8.0 9 | torchvision==0.9.0 10 | numpy==1.21.2 11 | scikit-learn==1.0.1 12 | scipy==1.7.2 13 | pandas==1.3.4 14 | tqdm==4.63.0 15 | xlwt==2.5 16 | ``` 17 | The environment with required packages can be created directly by running the following command: 18 | ``` 19 | conda env create -f environment.yml 20 | ``` 21 | 22 | ## Datasets 23 | For the **UCF-Crime** and **XD-Violence** datasets, we use off-the-shelf features extracted by [Wu et al](https://github.com/Roc-Ng). For the **ShanghaiTech** dataset, we used this [repo](https://github.com/v-iashin/video_features) to extract I3D features (highly recommended:+1:). 24 | | Dataset | Origin Video | I3D Features | 25 | | -------- | -------- | -------- | 26 | |   UCF-Crime |   [homepage](https://www.crcv.ucf.edu/projects/real-world/) | [download link](https://stuxidianeducn-my.sharepoint.com/:f:/g/personal/pengwu_stu_xidian_edu_cn/EvYcZ5rQZClGs_no2g-B0jcB4ynsonVQIreHIojNnUmPyA?e=xNrGxc) | 27 | |  XD-Violence |   [homepage](https://roc-ng.github.io/XD-Violence/) | [download link](https://roc-ng.github.io/XD-Violence/) | 28 | | ShanghaiTech |   [homepage](https://svip-lab.github.io/dataset/campus_dataset.html) | [download link](https://drive.google.com/file/d/1kIv502RxQnMer-8HB7zrU_GU7CNPNNDv/view?usp=drive_link) | 29 | |    UBnormal |   [homepage](https://github.com/lilygeorgescu/ubnormal#download) | [download link](https://drive.google.com/file/d/1KbfdyasribAMbbKoBU1iywAhtoAt9QI0/view]) | 30 | 31 | ## Acknowledgement 32 | Our codebase mainly refers to [STG-NF](https://github.com/orhir/STG-NF) and [PEL4VAD](https://github.com/yujiangpu20/PEL4VAD). We greatly appreciate their excellent contribution with nicely organized code! 33 | -------------------------------------------------------------------------------- /other/gpt4_utils.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": null, 6 | "metadata": { 7 | "colab": { 8 | "base_uri": "https://localhost:8080/" 9 | }, 10 | "id": "k2Nt4nqfrUvi", 11 | "outputId": "1c086cfb-73b3-4259-bb9c-124f0740adc0" 12 | }, 13 | "outputs": [], 14 | "source": [ 15 | "from openai import OpenAI\n", 16 | "\n", 17 | "client = OpenAI(\n", 18 | " # defaults to os.environ.get(\"OPENAI_API_KEY\")\n", 19 | " api_key=\"OPENAI_API_KEY\",\n", 20 | ")\n", 21 | "statement = 'video descrime.'\n", 22 | "chat_completion = client.chat.completions.create(\n", 23 | " messages=[\n", 24 | " {\n", 25 | " \"role\": \"user\",\n", 26 | " \"content\": statement + \"Please analyze the statement step by step, list the possible explanations, and categotize them as one of [category list], then determine which category is most likely.\",\n", 27 | " }\n", 28 | " ],\n", 29 | " model=\"gpt-4\",\n", 30 | ")\n", 31 | "response = chat_completion.choices[0].message.content\n", 32 | "print(response)" 33 | ] 34 | }, 35 | { 36 | "cell_type": "code", 37 | "execution_count": null, 38 | "metadata": { 39 | "id": "T3oiIT641kvw" 40 | }, 41 | "outputs": [], 42 | "source": [] 43 | } 44 | ], 45 | "metadata": { 46 | "colab": { 47 | "provenance": [] 48 | }, 49 | "kernelspec": { 50 | "display_name": "Python 3 (ipykernel)", 51 | "language": "python", 52 | "name": "python3" 53 | }, 54 | "language_info": { 55 | "codemirror_mode": { 56 | "name": "ipython", 57 | "version": 3 58 | }, 59 | "file_extension": ".py", 60 | "mimetype": "text/x-python", 61 | "name": "python", 62 | "nbconvert_exporter": "python", 63 | "pygments_lexer": "ipython3", 64 | "version": "3.8.13" 65 | }, 66 | "vscode": { 67 | "interpreter": { 68 | "hash": "d4d1e4263499bec80672ea0156c357c1ee493ec2b1c70f0acce89fc37c4a6abe" 69 | } 70 | } 71 | }, 72 | "nbformat": 4, 73 | "nbformat_minor": 0 74 | } 75 | -------------------------------------------------------------------------------- /Code/test.py: -------------------------------------------------------------------------------- 1 | from sklearn.metrics import auc, roc_curve, confusion_matrix, precision_recall_curve 2 | import numpy as np 3 | import torch 4 | 5 | 6 | def cal_false_alarm(gt, preds, threshold=0.5): 7 | preds = list(preds.cpu().detach().numpy()) 8 | gt = list(gt.cpu().detach().numpy()) 9 | 10 | preds = np.repeat(preds, 16) 11 | preds[preds < threshold] = 0 12 | preds[preds >= threshold] = 1 13 | tn, fp, fn, tp = confusion_matrix(gt, preds, labels=[0, 1]).ravel() 14 | 15 | far = fp / (fp + tn) 16 | 17 | return far 18 | 19 | 20 | def test_func(dataloader, model, gt, dataset): 21 | with torch.no_grad(): 22 | model.eval() 23 | pred = torch.zeros(0).cuda() 24 | abnormal_preds = torch.zeros(0).cuda() 25 | abnormal_labels = torch.zeros(0).cuda() 26 | normal_preds = torch.zeros(0).cuda() 27 | normal_labels = torch.zeros(0).cuda() 28 | gt_tmp = torch.tensor(gt.copy()).cuda() 29 | 30 | for i, (v_input, label) in enumerate(dataloader): 31 | v_input = v_input.float().cuda(non_blocking=True) 32 | seq_len = torch.sum(torch.max(torch.abs(v_input), dim=2)[0] > 0, 1) 33 | 34 | logits, _ = model(v_input, seq_len) 35 | logits = torch.mean(logits, 0) 36 | pred = torch.cat((pred, logits)) 37 | labels = gt_tmp[: seq_len[0] * 16] 38 | 39 | abnormal_labels = abnormal_labels.double() # 将abnormal_labels转换为double类型 40 | normal_labels = normal_labels.double() 41 | labels = labels.double() # 将labels转换为double类型 42 | if torch.sum(labels) == 0: 43 | normal_labels = torch.cat((normal_labels, labels)) 44 | normal_preds = torch.cat((normal_preds, logits)) 45 | else: 46 | abnormal_labels = torch.cat((abnormal_labels, labels)) 47 | abnormal_preds = torch.cat((abnormal_preds, logits)) 48 | gt_tmp = gt_tmp[seq_len[0] * 16:] 49 | 50 | pred = list(pred.cpu().detach().numpy()) 51 | n_far = cal_false_alarm(normal_labels, normal_preds) 52 | fpr, tpr, _ = roc_curve(list(gt), np.repeat(pred, 16)) 53 | roc_auc = auc(fpr, tpr) 54 | pre, rec, _ = precision_recall_curve(list(gt), np.repeat(pred, 16)) 55 | pr_auc = auc(rec, pre) 56 | 57 | if dataset == 'ucf-crime': 58 | return roc_auc, n_far 59 | elif dataset == 'xd-violence': 60 | return pr_auc, n_far 61 | elif dataset == 'shanghaiTech': 62 | return roc_auc, n_far 63 | else: 64 | raise RuntimeError('Invalid dataset.') 65 | -------------------------------------------------------------------------------- /Code/loss.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import torch.nn as nn 3 | import torch.nn.functional as F 4 | import numpy as np 5 | from utils import gen_label 6 | 7 | 8 | def CLAS2(logits, label, seq_len, criterion): 9 | logits = logits.squeeze() 10 | ins_logits = torch.zeros(0).cuda() # tensor([]) 11 | for i in range(logits.shape[0]): 12 | if label[i] == 0: 13 | tmp, _ = torch.topk(logits[i][:seq_len[i]], k=1, largest=True) 14 | else: 15 | tmp, _ = torch.topk(logits[i][:seq_len[i]], k=int(seq_len[i]//16+1), largest=True) 16 | tmp = torch.mean(tmp).view(1) 17 | ins_logits = torch.cat((ins_logits, tmp)) 18 | 19 | clsloss = criterion(ins_logits, label) 20 | return clsloss 21 | 22 | 23 | def KLV_loss(preds, label, criterion): 24 | preds = F.softmax(preds, dim=1) 25 | preds = torch.log(preds) 26 | if torch.isnan(preds).any(): 27 | loss = 0 28 | else: 29 | # preds = F.log_softmax(preds, dim=1) 30 | target = F.softmax(label * 10, dim=1) 31 | loss = criterion(preds, target) 32 | 33 | return loss 34 | 35 | 36 | def temporal_smooth(arr): 37 | arr2 = torch.zeros_like(arr) 38 | arr2[:-1] = arr[1:] 39 | arr2[-1] = arr[-1] 40 | loss = torch.sum((arr2-arr)**2) 41 | return loss 42 | 43 | 44 | def temporal_sparsity(arr): 45 | loss = torch.sum(arr) 46 | # loss = torch.mean(torch.norm(arr, dim=0)) 47 | return loss 48 | 49 | 50 | def Smooth(logits, seq_len, lamda=8e-5): 51 | smooth_mse = [] 52 | for i in range(logits.shape[0]): 53 | tmp_logits = logits[i][:seq_len[i]-1] 54 | sm_mse = temporal_smooth(tmp_logits) 55 | smooth_mse.append(sm_mse) 56 | smooth_mse = sum(smooth_mse) / len(smooth_mse) 57 | 58 | return smooth_mse * lamda 59 | 60 | 61 | def Sparsity(logits, seq_len, lamda=8e-5): 62 | spar_mse = [] 63 | for i in range(logits.shape[0]): 64 | tmp_logits = logits[i][:seq_len[i]] 65 | sp_mse = temporal_sparsity(tmp_logits) 66 | spar_mse.append(sp_mse) 67 | spar_mse = sum(spar_mse) / len(spar_mse) 68 | 69 | return spar_mse * lamda 70 | 71 | 72 | def Smooth_Sparsity(logits, seq_len, lamda=8e-5): 73 | smooth_mse = [] 74 | spar_mse = [] 75 | for i in range(logits.shape[0]): 76 | tmp_logits = logits[i][:seq_len[i]] 77 | sm_mse = temporal_smooth(tmp_logits) 78 | sp_mse = temporal_sparsity(tmp_logits) 79 | smooth_mse.append(sm_mse) 80 | spar_mse.append(sp_mse) 81 | smooth_mse = sum(smooth_mse) / len(smooth_mse) 82 | spar_mse = sum(spar_mse) / len(spar_mse) 83 | 84 | return (smooth_mse + spar_mse) * lamda -------------------------------------------------------------------------------- /Code/model.py: -------------------------------------------------------------------------------- 1 | 2 | import torch 3 | from modules import * 4 | import torch.nn.init as torch_init 5 | 6 | 7 | def weight_init(m): 8 | classname = m.__class__.__name__ 9 | if classname.find('Conv') != -1 or classname.find('Linear') != -1: 10 | torch_init.xavier_uniform_(m.weight) 11 | # m.bias.data.fill_(0.1) 12 | 13 | class XModel(nn.Module): 14 | def __init__(self, cfg): 15 | super(XModel, self).__init__() 16 | self.t = cfg.t_step 17 | self.self_attention = XEncoder( 18 | d_model=cfg.feat_dim, 19 | hid_dim=cfg.hid_dim, 20 | out_dim=cfg.out_dim, 21 | n_heads=cfg.head_num, 22 | win_size=cfg.win_size, 23 | dropout=cfg.dropout, 24 | gamma=cfg.gamma, 25 | bias=cfg.bias, 26 | norm=cfg.norm, 27 | ) 28 | self.classifier = nn.Conv1d(cfg.out_dim, 1, self.t, padding=0) 29 | self.logit_scale = nn.Parameter(torch.ones([]) * np.log(1 / cfg.temp)) 30 | self.apply(weight_init) 31 | 32 | def forward(self, x, seq_len): 33 | x_e, x_v = self.self_attention(x, seq_len) 34 | logits = F.pad(x_e, (self.t - 1, 0)) 35 | logits = self.classifier(logits) 36 | 37 | logits = logits.permute(0, 2, 1) 38 | logits = torch.sigmoid(logits) 39 | 40 | return logits, x_v 41 | 42 | 43 | 44 | # model test 45 | if __name__ == '__main__': 46 | import torch 47 | import numpy as np 48 | # 模型配置 49 | class Config: 50 | def __init__(self): 51 | self.t_step = 10 # 替换为你的配置 52 | self.feat_dim = 1024 # 替换为你的配置 53 | self.hid_dim = 512 # 替换为你的配置 54 | self.out_dim = 256 # 替换为你的配置 55 | self.head_num = 4 # 替换为你的配置 56 | self.win_size = 3 # 替换为你的配置 57 | self.dropout = 0.1 # 替换为你的配置 58 | self.gamma = 2.0 # 替换为你的配置 59 | self.bias = True # 替换为你的配置 60 | self.norm = 'layer' # 替换为你的配置 61 | self.temp = 1.0 # 替换为你的配置 62 | 63 | # 创建模型实例 64 | cfg = Config() 65 | model = XModel(cfg) 66 | 67 | 68 | device = torch.device("cuda") 69 | model = model.to(device) 70 | # 生成随机数据 71 | batch_size = 1 72 | seq_len = 98 73 | input_shape = (batch_size, seq_len, cfg.feat_dim) 74 | x = torch.rand(input_shape).to(device) 75 | seq_len_tensor = torch.tensor([seq_len]).to(device) 76 | 77 | # 将模型置于评估模式 78 | model.eval() 79 | 80 | # 前向传播 81 | with torch.no_grad(): 82 | logits, x_v = model(x, seq_len_tensor) 83 | 84 | # 打印输出的 logits 和 x_v 的形状 85 | print("Logits Shape:", logits.shape) 86 | print("x_v Shape:", x_v.shape) 87 | -------------------------------------------------------------------------------- /Code/fusion.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import torch.nn.functional as F 3 | from torch import nn 4 | class FeatureFusionMHSA(nn.Module): 5 | def __init__(self, feature_dim, num_heads): 6 | super(FeatureFusionMHSA, self).__init__() 7 | self.num_heads = num_heads 8 | self.feature_dim = feature_dim 9 | self.head_dim = feature_dim // num_heads 10 | 11 | # 确保特征维度可以被头数整除 12 | assert feature_dim % num_heads == 0, "feature_dim must be divisible by num_heads" 13 | 14 | # 定义多头自注意力中的线性层 15 | self.query = nn.Linear(feature_dim, feature_dim) 16 | self.key = nn.Linear(feature_dim, feature_dim) 17 | self.value = nn.Linear(feature_dim, feature_dim) 18 | 19 | # 定义输出层 20 | self.fc_out = nn.Linear(feature_dim, feature_dim) 21 | 22 | # 定义层归一化 23 | self.layer_norm = nn.LayerNorm(feature_dim) 24 | 25 | def forward(self, value, query): 26 | batch_size, feature_dim, seq_length = query.size() 27 | value = value.transpose(1, 2) 28 | query = query.transpose(1, 2) 29 | # 将输入进行归一化 30 | value_ln = self.layer_norm(value) 31 | query_ln = self.layer_norm(query) 32 | 33 | # 将输入分割成多头 34 | Q = self.query(query_ln).view(batch_size, seq_length, self.num_heads, self.head_dim) 35 | K = self.key(value_ln).view(batch_size, seq_length, self.num_heads, self.head_dim) 36 | V = self.value(value_ln).view(batch_size, seq_length, self.num_heads, self.head_dim) 37 | 38 | # 转置以得到适合点积的形状 39 | Q = Q.transpose(1, 2) # (batch_size, num_heads, seq_length, head_dim) 40 | K = K.transpose(1, 2) # (batch_size, num_heads, seq_length, head_dim) 41 | V = V.transpose(1, 2) # (batch_size, num_heads, seq_length, head_dim) 42 | 43 | # 计算注意力得分 44 | attention = torch.matmul(Q, K.transpose(-2, -1)) / torch.sqrt(torch.tensor(self.head_dim, dtype=torch.float32)) 45 | attention = torch.softmax(attention, dim=-1) 46 | 47 | # 通过注意力得分加权值向量 48 | out = torch.matmul(attention, V) # (batch_size, num_heads, seq_length, head_dim) 49 | out = out.transpose(1, 2).contiguous().view(batch_size, seq_length, feature_dim) # (batch_size, seq_length, feature_dim) 50 | 51 | # 应用最后的线性层 52 | out = self.fc_out(out) 53 | out = out.transpose(1, 2) 54 | return out 55 | 56 | # 初始化视频特征和文本特征 57 | # video_features = torch.randn(batch_size, feature_dim, seq_length) 58 | # text_features = torch.randn(batch_size, feature_dim, seq_length) 59 | 60 | # # 初始化融合模型 61 | # feature_fusion_model = FeatureFusionMHSA(feature_dim, num_heads) 62 | 63 | # # 特征融合 64 | # fused_features = feature_fusion_model(video_features, text_features) 65 | 66 | # # 验证融合后的特征维度 67 | # print(fused_features.shape) # 输出融合后的特征维度以验证 -------------------------------------------------------------------------------- /Code/layers.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import torch.nn as nn 3 | from torch import FloatTensor 4 | from torch.nn.parameter import Parameter 5 | from scipy.spatial.distance import pdist, squareform 6 | import torch.nn.functional as F 7 | import numpy as np 8 | import math 9 | 10 | 11 | class DistanceAdj(nn.Module): 12 | def __init__(self, sigma, bias): 13 | super(DistanceAdj, self).__init__() 14 | # self.sigma = sigma 15 | # self.bias = bias 16 | self.w = nn.Parameter(torch.FloatTensor(1)).cuda() 17 | self.b = nn.Parameter(torch.FloatTensor(1)).cuda() 18 | self.w.data.fill_(sigma) 19 | self.b.data.fill_(bias) 20 | 21 | def forward(self, batch_size, seq_len): 22 | arith = np.arange(seq_len).reshape(-1, 1) 23 | dist = pdist(arith, metric='cityblock').astype(np.float32) 24 | dist = torch.from_numpy(squareform(dist)).cuda() 25 | # dist = torch.exp(-self.sigma * dist ** 2) 26 | # self.w = self.w.cuda() 27 | # self.b = self.b.cuda() 28 | # self.w = torch.nn.Parameter(self.w.cuda()) 29 | # self.b = torch.nn.Parameter(self.b.cuda()) 30 | 31 | dist = torch.exp(-torch.abs(self.w * dist ** 2 - self.b)) 32 | dist = torch.unsqueeze(dist, 0).repeat(batch_size, 1, 1) 33 | 34 | return dist 35 | 36 | 37 | class TCA(nn.Module): 38 | def __init__(self, d_model, dim_k, dim_v, n_heads, norm=None): 39 | super(TCA, self).__init__() 40 | self.dim_v = dim_v 41 | self.dim_k = dim_k # same as dim_q 42 | self.n_heads = n_heads 43 | self.norm = norm 44 | 45 | self.q = nn.Linear(d_model, dim_k) 46 | self.k = nn.Linear(d_model, dim_k) 47 | self.v = nn.Linear(d_model, dim_v) 48 | self.o = nn.Linear(dim_v, d_model) 49 | 50 | self.norm_fact = 1 / math.sqrt(dim_k) 51 | self.alpha = nn.Parameter(torch.tensor(0.)) 52 | self.act = nn.Softmax(dim=-1) 53 | 54 | def forward(self, x, mask, adj=None): 55 | Q = self.q(x).view(-1, x.shape[0], x.shape[1], self.dim_k // self.n_heads) 56 | K = self.k(x).view(-1, x.shape[0], x.shape[1], self.dim_k // self.n_heads) 57 | V = self.v(x).view(-1, x.shape[0], x.shape[1], self.dim_v // self.n_heads) 58 | 59 | if adj is not None: 60 | g_map = torch.matmul(Q, K.permute(0, 1, 3, 2)) * self.norm_fact + adj 61 | else: 62 | g_map = torch.matmul(Q, K.permute(0, 1, 3, 2)) * self.norm_fact 63 | l_map = g_map.clone() 64 | l_map = l_map.masked_fill_(mask.data.eq(0), -1e9) 65 | 66 | g_map = self.act(g_map) 67 | l_map = self.act(l_map) 68 | glb = torch.matmul(g_map, V).view(x.shape[0], x.shape[1], -1) 69 | lcl = torch.matmul(l_map, V).view(x.shape[0], x.shape[1], -1) 70 | 71 | alpha = torch.sigmoid(self.alpha) 72 | tmp = alpha * glb + (1 - alpha) * lcl 73 | if self.norm: 74 | tmp = torch.sqrt(F.relu(tmp)) - torch.sqrt(F.relu(-tmp)) # power norm 75 | tmp = F.normalize(tmp) # l2 norm 76 | tmp = self.o(tmp).view(-1, x.shape[1], x.shape[2]) 77 | return tmp 78 | -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_clients/client2.list: -------------------------------------------------------------------------------- 1 | test/02_0161_0.npy 2 | test/02_0161_1.npy 3 | test/02_0161_2.npy 4 | test/02_0161_3.npy 5 | test/02_0161_4.npy 6 | test/02_0161_5.npy 7 | test/02_0161_6.npy 8 | test/02_0161_7.npy 9 | test/02_0161_8.npy 10 | test/02_0161_9.npy 11 | test/03_0031_0.npy 12 | test/03_0031_1.npy 13 | test/03_0031_2.npy 14 | test/03_0031_3.npy 15 | test/03_0031_4.npy 16 | test/03_0031_5.npy 17 | test/03_0031_6.npy 18 | test/03_0031_7.npy 19 | test/03_0031_8.npy 20 | test/03_0031_9.npy 21 | test/03_0032_0.npy 22 | test/03_0032_1.npy 23 | test/03_0032_2.npy 24 | test/03_0032_3.npy 25 | test/03_0032_4.npy 26 | test/03_0032_5.npy 27 | test/03_0032_6.npy 28 | test/03_0032_7.npy 29 | test/03_0032_8.npy 30 | test/03_0032_9.npy 31 | test/03_0033_0.npy 32 | test/03_0033_1.npy 33 | test/03_0033_2.npy 34 | test/03_0033_3.npy 35 | test/03_0033_4.npy 36 | test/03_0033_5.npy 37 | test/03_0033_6.npy 38 | test/03_0033_7.npy 39 | test/03_0033_8.npy 40 | test/03_0033_9.npy 41 | test/03_0059_0.npy 42 | test/03_0059_1.npy 43 | test/03_0059_2.npy 44 | test/03_0059_3.npy 45 | test/03_0059_4.npy 46 | test/03_0059_5.npy 47 | test/03_0059_6.npy 48 | test/03_0059_7.npy 49 | test/03_0059_8.npy 50 | test/03_0059_9.npy 51 | test/02_001_0.npy 52 | test/02_001_1.npy 53 | test/02_001_2.npy 54 | test/02_001_3.npy 55 | test/02_001_4.npy 56 | test/02_001_5.npy 57 | test/02_001_6.npy 58 | test/02_001_7.npy 59 | test/02_001_8.npy 60 | test/02_001_9.npy 61 | test/02_003_0.npy 62 | test/02_003_1.npy 63 | test/02_003_2.npy 64 | test/02_003_3.npy 65 | test/02_003_4.npy 66 | test/02_003_5.npy 67 | test/02_003_6.npy 68 | test/02_003_7.npy 69 | test/02_003_8.npy 70 | test/02_003_9.npy 71 | test/02_007_0.npy 72 | test/02_007_1.npy 73 | test/02_007_2.npy 74 | test/02_007_3.npy 75 | test/02_007_4.npy 76 | test/02_007_5.npy 77 | test/02_007_6.npy 78 | test/02_007_7.npy 79 | test/02_007_8.npy 80 | test/02_007_9.npy 81 | test/02_008_0.npy 82 | test/02_008_1.npy 83 | test/02_008_2.npy 84 | test/02_008_3.npy 85 | test/02_008_4.npy 86 | test/02_008_5.npy 87 | test/02_008_6.npy 88 | test/02_008_7.npy 89 | test/02_008_8.npy 90 | test/02_008_9.npy 91 | test/02_009_0.npy 92 | test/02_009_1.npy 93 | test/02_009_2.npy 94 | test/02_009_3.npy 95 | test/02_009_4.npy 96 | test/02_009_5.npy 97 | test/02_009_6.npy 98 | test/02_009_7.npy 99 | test/02_009_8.npy 100 | test/02_009_9.npy 101 | test/02_012_0.npy 102 | test/02_012_1.npy 103 | test/02_012_2.npy 104 | test/02_012_3.npy 105 | test/02_012_4.npy 106 | test/02_012_5.npy 107 | test/02_012_6.npy 108 | test/02_012_7.npy 109 | test/02_012_8.npy 110 | test/02_012_9.npy 111 | test/02_015_0.npy 112 | test/02_015_1.npy 113 | test/02_015_2.npy 114 | test/02_015_3.npy 115 | test/02_015_4.npy 116 | test/02_015_5.npy 117 | test/02_015_6.npy 118 | test/02_015_7.npy 119 | test/02_015_8.npy 120 | test/02_015_9.npy 121 | test/03_002_0.npy 122 | test/03_002_1.npy 123 | test/03_002_2.npy 124 | test/03_002_3.npy 125 | test/03_002_4.npy 126 | test/03_002_5.npy 127 | test/03_002_6.npy 128 | test/03_002_7.npy 129 | test/03_002_8.npy 130 | test/03_002_9.npy 131 | test/03_005_0.npy 132 | test/03_005_1.npy 133 | test/03_005_2.npy 134 | test/03_005_3.npy 135 | test/03_005_4.npy 136 | test/03_005_5.npy 137 | test/03_005_6.npy 138 | test/03_005_7.npy 139 | test/03_005_8.npy 140 | test/03_005_9.npy -------------------------------------------------------------------------------- /Code/distill.py: -------------------------------------------------------------------------------- 1 | import os 2 | import json 3 | import os 4 | import csv 5 | 6 | from transformers import BertModel, BertTokenizer 7 | import torch 8 | from scipy.spatial.distance import cosine 9 | 10 | # Initialize the tokenizer and model from the pre-trained BERT base model 11 | tokenizer = BertTokenizer.from_pretrained('bert-base-uncased') 12 | model = BertModel.from_pretrained('bert-base-uncased') 13 | model.eval() # Ensure the model is in evaluation mode 14 | 15 | # Define a function to get the BERT embeddings for a given text 16 | def get_bert_embedding(text): 17 | # Tokenize the text and get the required input tensors 18 | inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=512) 19 | with torch.no_grad(): # No need to track gradients for embeddings 20 | # Get the outputs from the BERT model 21 | outputs = model(**inputs) 22 | # Use the embedding of the [CLS] token (first token) as the sentence representation 23 | return outputs.last_hidden_state[:, 0, :].squeeze().detach() 24 | 25 | def get_public_train_list(tag): 26 | total_object_number = {} 27 | public_train_list = [] 28 | with open('Object target distribution json file sent to each client on the server side.','r') as coc: 29 | data = json.load(coc) 30 | print(data) 31 | for i in tag: 32 | numbers = data['client' + str(int(i) + 1)] 33 | for n in numbers: 34 | if total_object_number.get(n): 35 | total_object_number[n] += int(numbers[n]) 36 | else: 37 | total_object_number[n] = int(numbers[n]) 38 | print(total_object_number) 39 | # 按值对字典进行排序 40 | sorted_items = sorted(total_object_number.items(), key=lambda x: x[1], reverse=True) 41 | print(sorted_items) 42 | # 计算前70%的元素数量 43 | num_items = int(len(sorted_items) * 0.7) 44 | 45 | # 创建新的字典，只包含前70%的元素 46 | sorted_dict = dict(sorted_items[:num_items]) 47 | print(sorted_dict) 48 | long_string = ' '.join(sorted_dict.keys()) 49 | 50 | # 打印结果 51 | print(long_string) 52 | embedding1 = get_bert_embedding(long_string) 53 | print(embedding1.shape) 54 | print(type(embedding1)) 55 | embedding1_np = embedding1.numpy() 56 | import numpy as np 57 | path = 'Path to the object target distribution feature folder corresponding to each video file on the server side.' 58 | names = os.listdir(path) 59 | for name in names: 60 | embedding2 = np.load(path + name) 61 | embedding2_np = embedding2 62 | # Calculate the cosine similarity between the embeddings 63 | cosine_similarity = 1 - cosine(embedding1_np, embedding2_np) 64 | print(cosine_similarity) 65 | if float(cosine_similarity) > 0.70: 66 | public_train_list.append(name) 67 | return public_train_list 68 | 69 | 70 | def get_public_aug_train_list(tags): 71 | public_train_list = get_public_train_list(tags) 72 | tmp_path = 'Temporary storage of public dataset file paths.' 73 | with open(tmp_path,'w') as t: 74 | for train in public_train_list: 75 | name = train.split('.npy')[0] 76 | mid = name.split('_')[1] 77 | if len(mid) == 3: 78 | flags = ' 0.0' 79 | else: 80 | flags = ' 1.0' 81 | for i in range(0,10): 82 | n_name = 'train/' + name + '_' + str(i) + '.npy' + flags 83 | t.writelines(n_name + '\n') 84 | return tmp_path 85 | 86 | 87 | get_public_aug_train_list([1,2,3]) 88 | -------------------------------------------------------------------------------- /Code/environment.yml: -------------------------------------------------------------------------------- 1 | channels: 2 | - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch 3 | - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free 4 | - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge 5 | - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main 6 | - defaults 7 | dependencies: 8 | - _libgcc_mutex=0.1=main 9 | - _openmp_mutex=4.5=1_gnu 10 | - blas=1.0=mkl 11 | - bzip2=1.0.8=h7b6447c_0 12 | - ca-certificates=2022.3.29=h06a4308_1 13 | - certifi=2021.10.8=py38h06a4308_2 14 | - cudatoolkit=11.1.1=h6406543_8 15 | - freetype=2.11.0=h70c0345_0 16 | - giflib=5.2.1=h7b6447c_0 17 | - gmp=6.2.1=h2531618_2 18 | - gnutls=3.6.15=he1e5248_0 19 | - intel-openmp=2021.4.0=h06a4308_3561 20 | - jpeg=9b=0 21 | - lame=3.100=h7b6447c_0 22 | - lcms2=2.12=h3be6417_0 23 | - ld_impl_linux-64=2.35.1=h7274673_9 24 | - libffi=3.3=he6710b0_2 25 | - libgcc-ng=9.3.0=h5101ec6_17 26 | - libgomp=9.3.0=h5101ec6_17 27 | - libidn2=2.3.2=h7f8727e_0 28 | - libopus=1.3.1=h7b6447c_0 29 | - libpng=1.6.37=hbc83047_0 30 | - libstdcxx-ng=9.3.0=hd4cf53a_17 31 | - libtasn1=4.16.0=h27cfd23_0 32 | - libtiff=4.2.0=h85742a9_0 33 | - libunistring=0.9.10=h27cfd23_0 34 | - libuv=1.40.0=h7b6447c_0 35 | - libvpx=1.7.0=h439df22_0 36 | - libwebp=1.2.0=h89dd481_0 37 | - libwebp-base=1.2.0=h27cfd23_0 38 | - lz4-c=1.9.3=h295c915_1 39 | - mkl=2021.4.0=h06a4308_640 40 | - mkl-service=2.4.0=py38h7f8727e_0 41 | - mkl_fft=1.3.1=py38hd3c417c_0 42 | - mkl_random=1.2.2=py38h51133e4_0 43 | - ncurses=6.3=h7f8727e_2 44 | - nettle=3.7.3=hbbd107a_1 45 | - ninja=1.10.2=py38hd09550d_3 46 | - numpy-base=1.21.2=py38h79a1101_0 47 | - olefile=0.46=pyhd3eb1b0_0 48 | - openh264=2.1.0=hd408876_0 49 | - openssl=1.1.1n=h7f8727e_0 50 | - pip=21.2.4=py38h06a4308_0 51 | - portaudio=19.6.0=h7b6447c_4 52 | - pyaudio=0.2.11=py38h7b6447c_2 53 | - python=3.8.12=h12debd9_0 54 | - pytorch=1.8.0=py3.8_cuda11.1_cudnn8.0.5_0 55 | - readline=8.1=h27cfd23_0 56 | - setuptools=58.0.4=py38h06a4308_0 57 | - six=1.16.0=pyhd3eb1b0_0 58 | - sqlite=3.36.0=hc218d9a_0 59 | - tk=8.6.11=h1ccaba5_0 60 | - torchaudio=0.8.0=py38 61 | - torchvision=0.9.0=py38_cu111 62 | - typing_extensions=3.10.0.2=pyh06a4308_0 63 | - wheel=0.37.0=pyhd3eb1b0_1 64 | - x264=1!157.20191217=h7b6447c_0 65 | - xz=5.2.5=h7b6447c_0 66 | - zlib=1.2.11=h7b6447c_3 67 | - zstd=1.4.9=haebb681_0 68 | - pip: 69 | - av==9.2.0 70 | - charset-normalizer==2.0.12 71 | - click==8.1.2 72 | - clip==0.2.0 73 | - cycler==0.11.0 74 | - decorator==4.4.2 75 | - docker-pycreds==0.4.0 76 | - einops==0.6.0 77 | - ffmpeg==1.4 78 | - fonttools==4.28.1 79 | - ftfy==6.1.1 80 | - fvcore==0.1.5.post20220414 81 | - gensim==4.1.2 82 | - gitdb==4.0.9 83 | - gitpython==3.1.27 84 | - h5py==3.7.0 85 | - idna==3.3 86 | - imageio==2.16.1 87 | - imageio-ffmpeg==0.4.5 88 | - imutils==0.5.4 89 | - iopath==0.1.9 90 | - joblib==1.1.0 91 | - kiwisolver==1.3.2 92 | - matplotlib==3.5.0 93 | - moviepy==1.0.3 94 | - numpy==1.21.4 95 | - opencv-contrib-python==4.1.2.30 96 | - opencv-python==4.1.2.30 97 | - packaging==21.3 98 | - pandas==1.3.4 99 | - pathtools==0.1.2 100 | - pillow==8.4.0 101 | - portalocker==2.4.0 102 | - proglog==0.1.9 103 | - promise==2.3 104 | - protobuf==3.20.0 105 | - psutil==5.9.0 106 | - pyparsing==3.0.6 107 | - python-dateutil==2.8.2 108 | - pytz==2021.3 109 | - pyyaml==6.0 110 | - regex==2022.9.13 111 | - requests==2.27.1 112 | - scikit-learn==1.0.1 113 | - scipy==1.7.2 114 | - seaborn==0.12.0 115 | - sentry-sdk==1.5.10 116 | - setproctitle==1.2.3 117 | - setuptools-scm==6.3.2 118 | - shortuuid==1.0.8 119 | - simplejson==3.17.6 120 | - smart-open==5.2.1 121 | - smmap==5.0.0 122 | - tabulate==0.8.9 123 | - termcolor==1.1.0 124 | - thop==0.0.31-2005241907 125 | - threadpoolctl==3.0.0 126 | - tomli==1.2.2 127 | - tqdm==4.63.0 128 | - urllib3==1.26.9 129 | - wandb==0.12.14 130 | - wcwidth==0.2.5 131 | - xlwt==1.3.0 132 | - yacs==0.1.8 133 | -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_clients/client3.list: -------------------------------------------------------------------------------- 1 | test/04_0004_0.npy 2 | test/04_0004_1.npy 3 | test/04_0004_2.npy 4 | test/04_0004_3.npy 5 | test/04_0004_4.npy 6 | test/04_0004_5.npy 7 | test/04_0004_6.npy 8 | test/04_0004_7.npy 9 | test/04_0004_8.npy 10 | test/04_0004_9.npy 11 | test/04_0011_0.npy 12 | test/04_0011_1.npy 13 | test/04_0011_2.npy 14 | test/04_0011_3.npy 15 | test/04_0011_4.npy 16 | test/04_0011_5.npy 17 | test/04_0011_6.npy 18 | test/04_0011_7.npy 19 | test/04_0011_8.npy 20 | test/04_0011_9.npy 21 | test/04_0012_0.npy 22 | test/04_0012_1.npy 23 | test/04_0012_2.npy 24 | test/04_0012_3.npy 25 | test/04_0012_4.npy 26 | test/04_0012_5.npy 27 | test/04_0012_6.npy 28 | test/04_0012_7.npy 29 | test/04_0012_8.npy 30 | test/04_0012_9.npy 31 | test/04_0046_0.npy 32 | test/04_0046_1.npy 33 | test/04_0046_2.npy 34 | test/04_0046_3.npy 35 | test/04_0046_4.npy 36 | test/04_0046_5.npy 37 | test/04_0046_6.npy 38 | test/04_0046_7.npy 39 | test/04_0046_8.npy 40 | test/04_0046_9.npy 41 | test/07_0006_0.npy 42 | test/07_0006_1.npy 43 | test/07_0006_2.npy 44 | test/07_0006_3.npy 45 | test/07_0006_4.npy 46 | test/07_0006_5.npy 47 | test/07_0006_6.npy 48 | test/07_0006_7.npy 49 | test/07_0006_8.npy 50 | test/07_0006_9.npy 51 | test/07_0008_0.npy 52 | test/07_0008_1.npy 53 | test/07_0008_2.npy 54 | test/07_0008_3.npy 55 | test/07_0008_4.npy 56 | test/07_0008_5.npy 57 | test/07_0008_6.npy 58 | test/07_0008_7.npy 59 | test/07_0008_8.npy 60 | test/07_0008_9.npy 61 | test/07_0048_0.npy 62 | test/07_0048_1.npy 63 | test/07_0048_2.npy 64 | test/07_0048_3.npy 65 | test/07_0048_4.npy 66 | test/07_0048_5.npy 67 | test/07_0048_6.npy 68 | test/07_0048_7.npy 69 | test/07_0048_8.npy 70 | test/07_0048_9.npy 71 | test/04_002_0.npy 72 | test/04_002_1.npy 73 | test/04_002_2.npy 74 | test/04_002_3.npy 75 | test/04_002_4.npy 76 | test/04_002_5.npy 77 | test/04_002_6.npy 78 | test/04_002_7.npy 79 | test/04_002_8.npy 80 | test/04_002_9.npy 81 | test/04_005_0.npy 82 | test/04_005_1.npy 83 | test/04_005_2.npy 84 | test/04_005_3.npy 85 | test/04_005_4.npy 86 | test/04_005_5.npy 87 | test/04_005_6.npy 88 | test/04_005_7.npy 89 | test/04_005_8.npy 90 | test/04_005_9.npy 91 | test/04_007_0.npy 92 | test/04_007_1.npy 93 | test/04_007_2.npy 94 | test/04_007_3.npy 95 | test/04_007_4.npy 96 | test/04_007_5.npy 97 | test/04_007_6.npy 98 | test/04_007_7.npy 99 | test/04_007_8.npy 100 | test/04_007_9.npy 101 | test/04_008_0.npy 102 | test/04_008_1.npy 103 | test/04_008_2.npy 104 | test/04_008_3.npy 105 | test/04_008_4.npy 106 | test/04_008_5.npy 107 | test/04_008_6.npy 108 | test/04_008_7.npy 109 | test/04_008_8.npy 110 | test/04_008_9.npy 111 | test/04_011_0.npy 112 | test/04_011_1.npy 113 | test/04_011_2.npy 114 | test/04_011_3.npy 115 | test/04_011_4.npy 116 | test/04_011_5.npy 117 | test/04_011_6.npy 118 | test/04_011_7.npy 119 | test/04_011_8.npy 120 | test/04_011_9.npy 121 | test/04_014_0.npy 122 | test/04_014_1.npy 123 | test/04_014_2.npy 124 | test/04_014_3.npy 125 | test/04_014_4.npy 126 | test/04_014_5.npy 127 | test/04_014_6.npy 128 | test/04_014_7.npy 129 | test/04_014_8.npy 130 | test/04_014_9.npy 131 | test/04_016_0.npy 132 | test/04_016_1.npy 133 | test/04_016_2.npy 134 | test/04_016_3.npy 135 | test/04_016_4.npy 136 | test/04_016_5.npy 137 | test/04_016_6.npy 138 | test/04_016_7.npy 139 | test/04_016_8.npy 140 | test/04_016_9.npy 141 | test/04_017_0.npy 142 | test/04_017_1.npy 143 | test/04_017_2.npy 144 | test/04_017_3.npy 145 | test/04_017_4.npy 146 | test/04_017_5.npy 147 | test/04_017_6.npy 148 | test/04_017_7.npy 149 | test/04_017_8.npy 150 | test/04_017_9.npy 151 | test/04_018_0.npy 152 | test/04_018_1.npy 153 | test/04_018_2.npy 154 | test/04_018_3.npy 155 | test/04_018_4.npy 156 | test/04_018_5.npy 157 | test/04_018_6.npy 158 | test/04_018_7.npy 159 | test/04_018_8.npy 160 | test/04_018_9.npy 161 | test/07_002_0.npy 162 | test/07_002_1.npy 163 | test/07_002_2.npy 164 | test/07_002_3.npy 165 | test/07_002_4.npy 166 | test/07_002_5.npy 167 | test/07_002_6.npy 168 | test/07_002_7.npy 169 | test/07_002_8.npy 170 | test/07_002_9.npy 171 | test/07_005_0.npy 172 | test/07_005_1.npy 173 | test/07_005_2.npy 174 | test/07_005_3.npy 175 | test/07_005_4.npy 176 | test/07_005_5.npy 177 | test/07_005_6.npy 178 | test/07_005_7.npy 179 | test/07_005_8.npy 180 | test/07_005_9.npy 181 | test/07_008_0.npy 182 | test/07_008_1.npy 183 | test/07_008_2.npy 184 | test/07_008_3.npy 185 | test/07_008_4.npy 186 | test/07_008_5.npy 187 | test/07_008_6.npy 188 | test/07_008_7.npy 189 | test/07_008_8.npy 190 | test/07_008_9.npy 191 | -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/new_split/client2.list: -------------------------------------------------------------------------------- 1 | train/02_002_0.npy 0.0 2 | train/02_002_1.npy 0.0 3 | train/02_002_2.npy 0.0 4 | train/02_002_3.npy 0.0 5 | train/02_002_4.npy 0.0 6 | train/02_002_5.npy 0.0 7 | train/02_002_6.npy 0.0 8 | train/02_002_7.npy 0.0 9 | train/02_002_8.npy 0.0 10 | train/02_002_9.npy 0.0 11 | train/02_006_0.npy 0.0 12 | train/02_006_1.npy 0.0 13 | train/02_006_2.npy 0.0 14 | train/02_006_3.npy 0.0 15 | train/02_006_4.npy 0.0 16 | train/02_006_5.npy 0.0 17 | train/02_006_6.npy 0.0 18 | train/02_006_7.npy 0.0 19 | train/02_006_8.npy 0.0 20 | train/02_006_9.npy 0.0 21 | train/02_010_0.npy 0.0 22 | train/02_010_1.npy 0.0 23 | train/02_010_2.npy 0.0 24 | train/02_010_3.npy 0.0 25 | train/02_010_4.npy 0.0 26 | train/02_010_5.npy 0.0 27 | train/02_010_6.npy 0.0 28 | train/02_010_7.npy 0.0 29 | train/02_010_8.npy 0.0 30 | train/02_010_9.npy 0.0 31 | train/02_011_0.npy 0.0 32 | train/02_011_1.npy 0.0 33 | train/02_011_2.npy 0.0 34 | train/02_011_3.npy 0.0 35 | train/02_011_4.npy 0.0 36 | train/02_011_5.npy 0.0 37 | train/02_011_6.npy 0.0 38 | train/02_011_7.npy 0.0 39 | train/02_011_8.npy 0.0 40 | train/02_011_9.npy 0.0 41 | train/02_0128_0.npy 1.0 42 | train/02_0128_1.npy 1.0 43 | train/02_0128_2.npy 1.0 44 | train/02_0128_3.npy 1.0 45 | train/02_0128_4.npy 1.0 46 | train/02_0128_5.npy 1.0 47 | train/02_0128_6.npy 1.0 48 | train/02_0128_7.npy 1.0 49 | train/02_0128_8.npy 1.0 50 | train/02_0128_9.npy 1.0 51 | train/02_013_0.npy 0.0 52 | train/02_013_1.npy 0.0 53 | train/02_013_2.npy 0.0 54 | train/02_013_3.npy 0.0 55 | train/02_013_4.npy 0.0 56 | train/02_013_5.npy 0.0 57 | train/02_013_6.npy 0.0 58 | train/02_013_7.npy 0.0 59 | train/02_013_8.npy 0.0 60 | train/02_013_9.npy 0.0 61 | train/02_014_0.npy 0.0 62 | train/02_014_1.npy 0.0 63 | train/02_014_2.npy 0.0 64 | train/02_014_3.npy 0.0 65 | train/02_014_4.npy 0.0 66 | train/02_014_5.npy 0.0 67 | train/02_014_6.npy 0.0 68 | train/02_014_7.npy 0.0 69 | train/02_014_8.npy 0.0 70 | train/02_014_9.npy 0.0 71 | train/02_0164_0.npy 1.0 72 | train/02_0164_1.npy 1.0 73 | train/02_0164_2.npy 1.0 74 | train/02_0164_3.npy 1.0 75 | train/02_0164_4.npy 1.0 76 | train/02_0164_5.npy 1.0 77 | train/02_0164_6.npy 1.0 78 | train/02_0164_7.npy 1.0 79 | train/02_0164_8.npy 1.0 80 | train/02_0164_9.npy 1.0 81 | train/03_001_0.npy 0.0 82 | train/03_001_1.npy 0.0 83 | train/03_001_2.npy 0.0 84 | train/03_001_3.npy 0.0 85 | train/03_001_4.npy 0.0 86 | train/03_001_5.npy 0.0 87 | train/03_001_6.npy 0.0 88 | train/03_001_7.npy 0.0 89 | train/03_001_8.npy 0.0 90 | train/03_001_9.npy 0.0 91 | train/03_003_0.npy 0.0 92 | train/03_003_1.npy 0.0 93 | train/03_003_2.npy 0.0 94 | train/03_003_3.npy 0.0 95 | train/03_003_4.npy 0.0 96 | train/03_003_5.npy 0.0 97 | train/03_003_6.npy 0.0 98 | train/03_003_7.npy 0.0 99 | train/03_003_8.npy 0.0 100 | train/03_003_9.npy 0.0 101 | train/03_0035_0.npy 1.0 102 | train/03_0035_1.npy 1.0 103 | train/03_0035_2.npy 1.0 104 | train/03_0035_3.npy 1.0 105 | train/03_0035_4.npy 1.0 106 | train/03_0035_5.npy 1.0 107 | train/03_0035_6.npy 1.0 108 | train/03_0035_7.npy 1.0 109 | train/03_0035_8.npy 1.0 110 | train/03_0035_9.npy 1.0 111 | train/03_0039_0.npy 1.0 112 | train/03_0039_1.npy 1.0 113 | train/03_0039_2.npy 1.0 114 | train/03_0039_3.npy 1.0 115 | train/03_0039_4.npy 1.0 116 | train/03_0039_5.npy 1.0 117 | train/03_0039_6.npy 1.0 118 | train/03_0039_7.npy 1.0 119 | train/03_0039_8.npy 1.0 120 | train/03_0039_9.npy 1.0 121 | train/03_004_0.npy 0.0 122 | train/03_004_1.npy 0.0 123 | train/03_004_2.npy 0.0 124 | train/03_004_3.npy 0.0 125 | train/03_004_4.npy 0.0 126 | train/03_004_5.npy 0.0 127 | train/03_004_6.npy 0.0 128 | train/03_004_7.npy 0.0 129 | train/03_004_8.npy 0.0 130 | train/03_004_9.npy 0.0 131 | train/03_0041_0.npy 1.0 132 | train/03_0041_1.npy 1.0 133 | train/03_0041_2.npy 1.0 134 | train/03_0041_3.npy 1.0 135 | train/03_0041_4.npy 1.0 136 | train/03_0041_5.npy 1.0 137 | train/03_0041_6.npy 1.0 138 | train/03_0041_7.npy 1.0 139 | train/03_0041_8.npy 1.0 140 | train/03_0041_9.npy 1.0 141 | train/03_0060_0.npy 1.0 142 | train/03_0060_1.npy 1.0 143 | train/03_0060_2.npy 1.0 144 | train/03_0060_3.npy 1.0 145 | train/03_0060_4.npy 1.0 146 | train/03_0060_5.npy 1.0 147 | train/03_0060_6.npy 1.0 148 | train/03_0060_7.npy 1.0 149 | train/03_0060_8.npy 1.0 150 | train/03_0060_9.npy 1.0 151 | train/03_0061_0.npy 1.0 152 | train/03_0061_1.npy 1.0 153 | train/03_0061_2.npy 1.0 154 | train/03_0061_3.npy 1.0 155 | train/03_0061_4.npy 1.0 156 | train/03_0061_5.npy 1.0 157 | train/03_0061_6.npy 1.0 158 | train/03_0061_7.npy 1.0 159 | train/03_0061_8.npy 1.0 160 | train/03_0061_9.npy 1.0 -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_clients/client5.list: -------------------------------------------------------------------------------- 1 | test/06_0144_0.npy 2 | test/06_0144_1.npy 3 | test/06_0144_2.npy 4 | test/06_0144_3.npy 5 | test/06_0144_4.npy 6 | test/06_0144_5.npy 7 | test/06_0144_6.npy 8 | test/06_0144_7.npy 9 | test/06_0144_8.npy 10 | test/06_0144_9.npy 11 | test/06_0153_0.npy 12 | test/06_0153_1.npy 13 | test/06_0153_2.npy 14 | test/06_0153_3.npy 15 | test/06_0153_4.npy 16 | test/06_0153_5.npy 17 | test/06_0153_6.npy 18 | test/06_0153_7.npy 19 | test/06_0153_8.npy 20 | test/06_0153_9.npy 21 | test/06_003_0.npy 22 | test/06_003_1.npy 23 | test/06_003_2.npy 24 | test/06_003_3.npy 25 | test/06_003_4.npy 26 | test/06_003_5.npy 27 | test/06_003_6.npy 28 | test/06_003_7.npy 29 | test/06_003_8.npy 30 | test/06_003_9.npy 31 | test/06_006_0.npy 32 | test/06_006_1.npy 33 | test/06_006_2.npy 34 | test/06_006_3.npy 35 | test/06_006_4.npy 36 | test/06_006_5.npy 37 | test/06_006_6.npy 38 | test/06_006_7.npy 39 | test/06_006_8.npy 40 | test/06_006_9.npy 41 | test/06_009_0.npy 42 | test/06_009_1.npy 43 | test/06_009_2.npy 44 | test/06_009_3.npy 45 | test/06_009_4.npy 46 | test/06_009_5.npy 47 | test/06_009_6.npy 48 | test/06_009_7.npy 49 | test/06_009_8.npy 50 | test/06_009_9.npy 51 | test/06_011_0.npy 52 | test/06_011_1.npy 53 | test/06_011_2.npy 54 | test/06_011_3.npy 55 | test/06_011_4.npy 56 | test/06_011_5.npy 57 | test/06_011_6.npy 58 | test/06_011_7.npy 59 | test/06_011_8.npy 60 | test/06_011_9.npy 61 | test/06_015_0.npy 62 | test/06_015_1.npy 63 | test/06_015_2.npy 64 | test/06_015_3.npy 65 | test/06_015_4.npy 66 | test/06_015_5.npy 67 | test/06_015_6.npy 68 | test/06_015_7.npy 69 | test/06_015_8.npy 70 | test/06_015_9.npy 71 | test/06_017_0.npy 72 | test/06_017_1.npy 73 | test/06_017_2.npy 74 | test/06_017_3.npy 75 | test/06_017_4.npy 76 | test/06_017_5.npy 77 | test/06_017_6.npy 78 | test/06_017_7.npy 79 | test/06_017_8.npy 80 | test/06_017_9.npy 81 | test/06_018_0.npy 82 | test/06_018_1.npy 83 | test/06_018_2.npy 84 | test/06_018_3.npy 85 | test/06_018_4.npy 86 | test/06_018_5.npy 87 | test/06_018_6.npy 88 | test/06_018_7.npy 89 | test/06_018_8.npy 90 | test/06_018_9.npy 91 | test/06_020_0.npy 92 | test/06_020_1.npy 93 | test/06_020_2.npy 94 | test/06_020_3.npy 95 | test/06_020_4.npy 96 | test/06_020_5.npy 97 | test/06_020_6.npy 98 | test/06_020_7.npy 99 | test/06_020_8.npy 100 | test/06_020_9.npy 101 | test/06_021_0.npy 102 | test/06_021_1.npy 103 | test/06_021_2.npy 104 | test/06_021_3.npy 105 | test/06_021_4.npy 106 | test/06_021_5.npy 107 | test/06_021_6.npy 108 | test/06_021_7.npy 109 | test/06_021_8.npy 110 | test/06_021_9.npy 111 | test/06_022_0.npy 112 | test/06_022_1.npy 113 | test/06_022_2.npy 114 | test/06_022_3.npy 115 | test/06_022_4.npy 116 | test/06_022_5.npy 117 | test/06_022_6.npy 118 | test/06_022_7.npy 119 | test/06_022_8.npy 120 | test/06_022_9.npy 121 | test/06_026_0.npy 122 | test/06_026_1.npy 123 | test/06_026_2.npy 124 | test/06_026_3.npy 125 | test/06_026_4.npy 126 | test/06_026_5.npy 127 | test/06_026_6.npy 128 | test/06_026_7.npy 129 | test/06_026_8.npy 130 | test/06_026_9.npy 131 | test/06_028_0.npy 132 | test/06_028_1.npy 133 | test/06_028_2.npy 134 | test/06_028_3.npy 135 | test/06_028_4.npy 136 | test/06_028_5.npy 137 | test/06_028_6.npy 138 | test/06_028_7.npy 139 | test/06_028_8.npy 140 | test/06_028_9.npy 141 | test/06_029_0.npy 142 | test/06_029_1.npy 143 | test/06_029_2.npy 144 | test/06_029_3.npy 145 | test/06_029_4.npy 146 | test/06_029_5.npy 147 | test/06_029_6.npy 148 | test/06_029_7.npy 149 | test/06_029_8.npy 150 | test/06_029_9.npy 151 | test/06_030_0.npy 152 | test/06_030_1.npy 153 | test/06_030_2.npy 154 | test/06_030_3.npy 155 | test/06_030_4.npy 156 | test/06_030_5.npy 157 | test/06_030_6.npy 158 | test/06_030_7.npy 159 | test/06_030_8.npy 160 | test/06_030_9.npy 161 | test/09_002_0.npy 162 | test/09_002_1.npy 163 | test/09_002_2.npy 164 | test/09_002_3.npy 165 | test/09_002_4.npy 166 | test/09_002_5.npy 167 | test/09_002_6.npy 168 | test/09_002_7.npy 169 | test/09_002_8.npy 170 | test/09_002_9.npy 171 | test/09_003_0.npy 172 | test/09_003_1.npy 173 | test/09_003_2.npy 174 | test/09_003_3.npy 175 | test/09_003_4.npy 176 | test/09_003_5.npy 177 | test/09_003_6.npy 178 | test/09_003_7.npy 179 | test/09_003_8.npy 180 | test/09_003_9.npy 181 | test/09_005_0.npy 182 | test/09_005_1.npy 183 | test/09_005_2.npy 184 | test/09_005_3.npy 185 | test/09_005_4.npy 186 | test/09_005_5.npy 187 | test/09_005_6.npy 188 | test/09_005_7.npy 189 | test/09_005_8.npy 190 | test/09_005_9.npy 191 | test/09_007_0.npy 192 | test/09_007_1.npy 193 | test/09_007_2.npy 194 | test/09_007_3.npy 195 | test/09_007_4.npy 196 | test/09_007_5.npy 197 | test/09_007_6.npy 198 | test/09_007_7.npy 199 | test/09_007_8.npy 200 | test/09_007_9.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_clients/client7.list: -------------------------------------------------------------------------------- 1 | test/10_0038_0.npy 2 | test/10_0038_1.npy 3 | test/10_0038_2.npy 4 | test/10_0038_3.npy 5 | test/10_0038_4.npy 6 | test/10_0038_5.npy 7 | test/10_0038_6.npy 8 | test/10_0038_7.npy 9 | test/10_0038_8.npy 10 | test/10_0038_9.npy 11 | test/10_0042_0.npy 12 | test/10_0042_1.npy 13 | test/10_0042_2.npy 14 | test/10_0042_3.npy 15 | test/10_0042_4.npy 16 | test/10_0042_5.npy 17 | test/10_0042_6.npy 18 | test/10_0042_7.npy 19 | test/10_0042_8.npy 20 | test/10_0042_9.npy 21 | test/12_0142_0.npy 22 | test/12_0142_1.npy 23 | test/12_0142_2.npy 24 | test/12_0142_3.npy 25 | test/12_0142_4.npy 26 | test/12_0142_5.npy 27 | test/12_0142_6.npy 28 | test/12_0142_7.npy 29 | test/12_0142_8.npy 30 | test/12_0142_9.npy 31 | test/12_0151_0.npy 32 | test/12_0151_1.npy 33 | test/12_0151_2.npy 34 | test/12_0151_3.npy 35 | test/12_0151_4.npy 36 | test/12_0151_5.npy 37 | test/12_0151_6.npy 38 | test/12_0151_7.npy 39 | test/12_0151_8.npy 40 | test/12_0151_9.npy 41 | test/12_0173_0.npy 42 | test/12_0173_1.npy 43 | test/12_0173_2.npy 44 | test/12_0173_3.npy 45 | test/12_0173_4.npy 46 | test/12_0173_5.npy 47 | test/12_0173_6.npy 48 | test/12_0173_7.npy 49 | test/12_0173_8.npy 50 | test/12_0173_9.npy 51 | test/12_0174_0.npy 52 | test/12_0174_1.npy 53 | test/12_0174_2.npy 54 | test/12_0174_3.npy 55 | test/12_0174_4.npy 56 | test/12_0174_5.npy 57 | test/12_0174_6.npy 58 | test/12_0174_7.npy 59 | test/12_0174_8.npy 60 | test/12_0174_9.npy 61 | test/10_004_0.npy 62 | test/10_004_1.npy 63 | test/10_004_2.npy 64 | test/10_004_3.npy 65 | test/10_004_4.npy 66 | test/10_004_5.npy 67 | test/10_004_6.npy 68 | test/10_004_7.npy 69 | test/10_004_8.npy 70 | test/10_004_9.npy 71 | test/10_005_0.npy 72 | test/10_005_1.npy 73 | test/10_005_2.npy 74 | test/10_005_3.npy 75 | test/10_005_4.npy 76 | test/10_005_5.npy 77 | test/10_005_6.npy 78 | test/10_005_7.npy 79 | test/10_005_8.npy 80 | test/10_005_9.npy 81 | test/10_008_0.npy 82 | test/10_008_1.npy 83 | test/10_008_2.npy 84 | test/10_008_3.npy 85 | test/10_008_4.npy 86 | test/10_008_5.npy 87 | test/10_008_6.npy 88 | test/10_008_7.npy 89 | test/10_008_8.npy 90 | test/10_008_9.npy 91 | test/10_009_0.npy 92 | test/10_009_1.npy 93 | test/10_009_2.npy 94 | test/10_009_3.npy 95 | test/10_009_4.npy 96 | test/10_009_5.npy 97 | test/10_009_6.npy 98 | test/10_009_7.npy 99 | test/10_009_8.npy 100 | test/10_009_9.npy 101 | test/10_011_0.npy 102 | test/10_011_1.npy 103 | test/10_011_2.npy 104 | test/10_011_3.npy 105 | test/10_011_4.npy 106 | test/10_011_5.npy 107 | test/10_011_6.npy 108 | test/10_011_7.npy 109 | test/10_011_8.npy 110 | test/10_011_9.npy 111 | test/12_001_0.npy 112 | test/12_001_1.npy 113 | test/12_001_2.npy 114 | test/12_001_3.npy 115 | test/12_001_4.npy 116 | test/12_001_5.npy 117 | test/12_001_6.npy 118 | test/12_001_7.npy 119 | test/12_001_8.npy 120 | test/12_001_9.npy 121 | test/12_004_0.npy 122 | test/12_004_1.npy 123 | test/12_004_2.npy 124 | test/12_004_3.npy 125 | test/12_004_4.npy 126 | test/12_004_5.npy 127 | test/12_004_6.npy 128 | test/12_004_7.npy 129 | test/12_004_8.npy 130 | test/12_004_9.npy 131 | test/12_007_0.npy 132 | test/12_007_1.npy 133 | test/12_007_2.npy 134 | test/12_007_3.npy 135 | test/12_007_4.npy 136 | test/12_007_5.npy 137 | test/12_007_6.npy 138 | test/12_007_7.npy 139 | test/12_007_8.npy 140 | test/12_007_9.npy 141 | test/12_008_0.npy 142 | test/12_008_1.npy 143 | test/12_008_2.npy 144 | test/12_008_3.npy 145 | test/12_008_4.npy 146 | test/12_008_5.npy 147 | test/12_008_6.npy 148 | test/12_008_7.npy 149 | test/12_008_8.npy 150 | test/12_008_9.npy 151 | test/12_011_0.npy 152 | test/12_011_1.npy 153 | test/12_011_2.npy 154 | test/12_011_3.npy 155 | test/12_011_4.npy 156 | test/12_011_5.npy 157 | test/12_011_6.npy 158 | test/12_011_7.npy 159 | test/12_011_8.npy 160 | test/12_011_9.npy 161 | test/12_012_0.npy 162 | test/12_012_1.npy 163 | test/12_012_2.npy 164 | test/12_012_3.npy 165 | test/12_012_4.npy 166 | test/12_012_5.npy 167 | test/12_012_6.npy 168 | test/12_012_7.npy 169 | test/12_012_8.npy 170 | test/12_012_9.npy 171 | test/12_015_0.npy 172 | test/12_015_1.npy 173 | test/12_015_2.npy 174 | test/12_015_3.npy 175 | test/12_015_4.npy 176 | test/12_015_5.npy 177 | test/12_015_6.npy 178 | test/12_015_7.npy 179 | test/12_015_8.npy 180 | test/12_015_9.npy 181 | test/13_001_0.npy 182 | test/13_001_1.npy 183 | test/13_001_2.npy 184 | test/13_001_3.npy 185 | test/13_001_4.npy 186 | test/13_001_5.npy 187 | test/13_001_6.npy 188 | test/13_001_7.npy 189 | test/13_001_8.npy 190 | test/13_001_9.npy 191 | test/13_005_0.npy 192 | test/13_005_1.npy 193 | test/13_005_2.npy 194 | test/13_005_3.npy 195 | test/13_005_4.npy 196 | test/13_005_5.npy 197 | test/13_005_6.npy 198 | test/13_005_7.npy 199 | test/13_005_8.npy 200 | test/13_005_9.npy 201 | test/13_006_0.npy 202 | test/13_006_1.npy 203 | test/13_006_2.npy 204 | test/13_006_3.npy 205 | test/13_006_4.npy 206 | test/13_006_5.npy 207 | test/13_006_6.npy 208 | test/13_006_7.npy 209 | test/13_006_8.npy 210 | test/13_006_9.npy 211 | -------------------------------------------------------------------------------- /Code/train.py: -------------------------------------------------------------------------------- 1 | import torch 2 | from loss import * 3 | from utils import * 4 | from bert_model import BERT_Arch 5 | from transformers import AutoModel, BertTokenizerFast 6 | import pickle as pk 7 | from fusion import * 8 | 9 | def train_func(dataloader, model, text_model_weights, optimizer, criterion, criterion2, dl_loss_function,client_idx, lamda=0): 10 | t_loss = [] 11 | s_loss = [] 12 | 13 | with torch.set_grad_enabled(True): 14 | model.train() 15 | for i, (v_input, t_input, label, multi_label, t_id) in enumerate(dataloader): 16 | # for i, (v_input, t_input, label, multi_label) in enumerate(dataloader): 17 | seq_len = torch.sum(torch.max(torch.abs(v_input), dim=2)[0] > 0, 1) 18 | v_input = v_input[:, :torch.max(seq_len), :] 19 | v_input = v_input.float().cuda(non_blocking=True) 20 | t_input = t_input.float().cuda(non_blocking=True) 21 | label = label.float().cuda(non_blocking=True) 22 | multi_label = multi_label.cuda(non_blocking=True) 23 | 24 | logits, v_feat = model(v_input, seq_len) 25 | 26 | # Prompt-Enhanced Learning 27 | logit_scale = model.logit_scale.exp() 28 | video_feat, token_feat, video_labels = get_cas(v_feat, t_input, logits, multi_label) 29 | 30 | v2t_logits, v2v_logits = create_logits(video_feat, token_feat, logit_scale) 31 | ground_truth = torch.tensor(gen_label(video_labels), dtype=v_feat.dtype).cuda() 32 | 33 | 34 | loss2 = KLV_loss(v2t_logits, ground_truth, criterion2) 35 | 36 | loss1 = CLAS2(logits, label, seq_len, criterion) 37 | 38 | loss = loss1 + lamda * loss2 39 | 40 | optimizer.zero_grad() 41 | loss.backward() 42 | optimizer.step() 43 | 44 | t_loss.append(loss1) 45 | s_loss.append(loss2) 46 | 47 | return sum(t_loss) / len(t_loss), sum(s_loss) / len(s_loss) 48 | 49 | 50 | def public_train_func(dataloader, global_model,visual_model, optimizer, criterion, criterion2,dl_loss_function, lamda=0): 51 | t_loss = [] 52 | s_loss = [] 53 | mse_criterion = nn.MSELoss() 54 | with torch.set_grad_enabled(True): 55 | global_model.train().cuda() 56 | visual_model.eval().cuda() 57 | feature_dim = 512 58 | num_heads = 8 59 | feature_fusion_model = FeatureFusionMHSA(feature_dim, num_heads).to("cuda") 60 | for i, (v_input, t_input, label, multi_label, t_id) in enumerate(dataloader): 61 | seq_len = torch.sum(torch.max(torch.abs(v_input), dim=2)[0] > 0, 1) 62 | v_input = v_input[:, :torch.max(seq_len), :] 63 | v_input = v_input.float().cuda(non_blocking=True) 64 | t_input = t_input.float().cuda(non_blocking=True) 65 | label = label.float().cuda(non_blocking=True) 66 | multi_label = multi_label.cuda(non_blocking=True) 67 | 68 | logits, v_feat = global_model(v_input, seq_len) 69 | logits2, v_feat2 = visual_model(v_input, seq_len) 70 | 71 | # t_feature = get_xd_text_feature(t_id) 72 | t_feature = get_sh_text_feature(t_id) 73 | t_feature_expanded = t_feature.unsqueeze(2).cuda(non_blocking=True) # shape: [128, 512, 1] 74 | t_feature_expanded = t_feature_expanded.expand(-1, -1, v_feat.size(-1)) 75 | 76 | 77 | fused_feature = feature_fusion_model(v_feat,t_feature_expanded) 78 | 79 | mse_loss = mse_criterion(v_feat,fused_feature) 80 | 81 | 82 | # Prompt-Enhanced Learning 83 | logit_scale = global_model.logit_scale.exp() 84 | video_feat, token_feat, video_labels = get_cas(v_feat, t_input, logits, multi_label) 85 | 86 | v2t_logits, v2v_logits = create_logits(video_feat, token_feat, logit_scale) 87 | ground_truth = torch.tensor(gen_label(video_labels), dtype=v_feat.dtype).cuda() 88 | 89 | 90 | loss2 = KLV_loss(v2t_logits, ground_truth, criterion2) 91 | 92 | loss1 = CLAS2(logits, label, seq_len, criterion) 93 | 94 | loss = loss1 + lamda * loss2 + 0.7 * mse_loss 95 | 96 | optimizer.zero_grad() 97 | loss.backward() 98 | optimizer.step() 99 | 100 | t_loss.append(loss1) 101 | s_loss.append(loss2) 102 | 103 | return global_model.state_dict() 104 | 105 | def get_text_id_based_name(name): 106 | import csv 107 | data_dict = {} 108 | csv_file = name # 替换为你的 CSV 文件路径 109 | with open(csv_file, 'r') as file: 110 | csv_reader = csv.DictReader(file) 111 | for row in csv_reader: 112 | video_name = row['name'] 113 | sentence = row['id'] 114 | data_dict[sentence] = video_name 115 | return data_dict 116 | def get_sh_text_feature(t_id): 117 | fe = np.zeros((len(t_id), 512)) 118 | linear_layer = create_linear_layer(768, 512) 119 | 120 | with open('all sh_text_feature path','rb') as tf: 121 | infos = pk.load(tf) 122 | ids = get_text_id_based_name('Path to the id of the text corresponding to the video in the public dataset SH.') 123 | for i, id in enumerate(t_id): 124 | name = ids[str(id)] 125 | 126 | feature = infos[name].reshape(768) 127 | f = linear_layer(feature).reshape(1, 512) 128 | fe[i] = f.cpu().detach().numpy() 129 | fe_tensor = torch.from_numpy(fe).float() 130 | 131 | return fe_tensor 132 | 133 | import torch.nn as nn 134 | 135 | def create_linear_layer(input_dim, output_dim): 136 | return nn.Linear(input_dim, output_dim) 137 | 138 | # 使用函数 139 | get_sh_text_feature([10]) 140 | 141 | -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/new_split/client3.list: -------------------------------------------------------------------------------- 1 | train/04_0003_0.npy 1.0 2 | train/04_0003_1.npy 1.0 3 | train/04_0003_2.npy 1.0 4 | train/04_0003_3.npy 1.0 5 | train/04_0003_4.npy 1.0 6 | train/04_0003_5.npy 1.0 7 | train/04_0003_6.npy 1.0 8 | train/04_0003_7.npy 1.0 9 | train/04_0003_8.npy 1.0 10 | train/04_0003_9.npy 1.0 11 | train/04_0010_0.npy 1.0 12 | train/04_0010_1.npy 1.0 13 | train/04_0010_2.npy 1.0 14 | train/04_0010_3.npy 1.0 15 | train/04_0010_4.npy 1.0 16 | train/04_0010_5.npy 1.0 17 | train/04_0010_6.npy 1.0 18 | train/04_0010_7.npy 1.0 19 | train/04_0010_8.npy 1.0 20 | train/04_0010_9.npy 1.0 21 | train/04_0013_0.npy 1.0 22 | train/04_0013_1.npy 1.0 23 | train/04_0013_2.npy 1.0 24 | train/04_0013_3.npy 1.0 25 | train/04_0013_4.npy 1.0 26 | train/04_0013_5.npy 1.0 27 | train/04_0013_6.npy 1.0 28 | train/04_0013_7.npy 1.0 29 | train/04_0013_8.npy 1.0 30 | train/04_0013_9.npy 1.0 31 | train/04_003_0.npy 0.0 32 | train/04_003_1.npy 0.0 33 | train/04_003_2.npy 0.0 34 | train/04_003_3.npy 0.0 35 | train/04_003_4.npy 0.0 36 | train/04_003_5.npy 0.0 37 | train/04_003_6.npy 0.0 38 | train/04_003_7.npy 0.0 39 | train/04_003_8.npy 0.0 40 | train/04_003_9.npy 0.0 41 | train/04_006_0.npy 0.0 42 | train/04_006_1.npy 0.0 43 | train/04_006_2.npy 0.0 44 | train/04_006_3.npy 0.0 45 | train/04_006_4.npy 0.0 46 | train/04_006_5.npy 0.0 47 | train/04_006_6.npy 0.0 48 | train/04_006_7.npy 0.0 49 | train/04_006_8.npy 0.0 50 | train/04_006_9.npy 0.0 51 | train/04_009_0.npy 0.0 52 | train/04_009_1.npy 0.0 53 | train/04_009_2.npy 0.0 54 | train/04_009_3.npy 0.0 55 | train/04_009_4.npy 0.0 56 | train/04_009_5.npy 0.0 57 | train/04_009_6.npy 0.0 58 | train/04_009_7.npy 0.0 59 | train/04_009_8.npy 0.0 60 | train/04_009_9.npy 0.0 61 | train/04_010_0.npy 0.0 62 | train/04_010_1.npy 0.0 63 | train/04_010_2.npy 0.0 64 | train/04_010_3.npy 0.0 65 | train/04_010_4.npy 0.0 66 | train/04_010_5.npy 0.0 67 | train/04_010_6.npy 0.0 68 | train/04_010_7.npy 0.0 69 | train/04_010_8.npy 0.0 70 | train/04_010_9.npy 0.0 71 | train/04_012_0.npy 0.0 72 | train/04_012_1.npy 0.0 73 | train/04_012_2.npy 0.0 74 | train/04_012_3.npy 0.0 75 | train/04_012_4.npy 0.0 76 | train/04_012_5.npy 0.0 77 | train/04_012_6.npy 0.0 78 | train/04_012_7.npy 0.0 79 | train/04_012_8.npy 0.0 80 | train/04_012_9.npy 0.0 81 | train/04_013_0.npy 0.0 82 | train/04_013_1.npy 0.0 83 | train/04_013_2.npy 0.0 84 | train/04_013_3.npy 0.0 85 | train/04_013_4.npy 0.0 86 | train/04_013_5.npy 0.0 87 | train/04_013_6.npy 0.0 88 | train/04_013_7.npy 0.0 89 | train/04_013_8.npy 0.0 90 | train/04_013_9.npy 0.0 91 | train/04_015_0.npy 0.0 92 | train/04_015_1.npy 0.0 93 | train/04_015_2.npy 0.0 94 | train/04_015_3.npy 0.0 95 | train/04_015_4.npy 0.0 96 | train/04_015_5.npy 0.0 97 | train/04_015_6.npy 0.0 98 | train/04_015_7.npy 0.0 99 | train/04_015_8.npy 0.0 100 | train/04_015_9.npy 0.0 101 | train/04_019_0.npy 0.0 102 | train/04_019_1.npy 0.0 103 | train/04_019_2.npy 0.0 104 | train/04_019_3.npy 0.0 105 | train/04_019_4.npy 0.0 106 | train/04_019_5.npy 0.0 107 | train/04_019_6.npy 0.0 108 | train/04_019_7.npy 0.0 109 | train/04_019_8.npy 0.0 110 | train/04_019_9.npy 0.0 111 | train/04_020_0.npy 0.0 112 | train/04_020_1.npy 0.0 113 | train/04_020_2.npy 0.0 114 | train/04_020_3.npy 0.0 115 | train/04_020_4.npy 0.0 116 | train/04_020_5.npy 0.0 117 | train/04_020_6.npy 0.0 118 | train/04_020_7.npy 0.0 119 | train/04_020_8.npy 0.0 120 | train/04_020_9.npy 0.0 121 | train/07_0007_0.npy 1.0 122 | train/07_0007_1.npy 1.0 123 | train/07_0007_2.npy 1.0 124 | train/07_0007_3.npy 1.0 125 | train/07_0007_4.npy 1.0 126 | train/07_0007_5.npy 1.0 127 | train/07_0007_6.npy 1.0 128 | train/07_0007_7.npy 1.0 129 | train/07_0007_8.npy 1.0 130 | train/07_0007_9.npy 1.0 131 | train/07_0009_0.npy 1.0 132 | train/07_0009_1.npy 1.0 133 | train/07_0009_2.npy 1.0 134 | train/07_0009_3.npy 1.0 135 | train/07_0009_4.npy 1.0 136 | train/07_0009_5.npy 1.0 137 | train/07_0009_6.npy 1.0 138 | train/07_0009_7.npy 1.0 139 | train/07_0009_8.npy 1.0 140 | train/07_0009_9.npy 1.0 141 | train/07_001_0.npy 0.0 142 | train/07_001_1.npy 0.0 143 | train/07_001_2.npy 0.0 144 | train/07_001_3.npy 0.0 145 | train/07_001_4.npy 0.0 146 | train/07_001_5.npy 0.0 147 | train/07_001_6.npy 0.0 148 | train/07_001_7.npy 0.0 149 | train/07_001_8.npy 0.0 150 | train/07_001_9.npy 0.0 151 | train/07_003_0.npy 0.0 152 | train/07_003_1.npy 0.0 153 | train/07_003_2.npy 0.0 154 | train/07_003_3.npy 0.0 155 | train/07_003_4.npy 0.0 156 | train/07_003_5.npy 0.0 157 | train/07_003_6.npy 0.0 158 | train/07_003_7.npy 0.0 159 | train/07_003_8.npy 0.0 160 | train/07_003_9.npy 0.0 161 | train/07_004_0.npy 0.0 162 | train/07_004_1.npy 0.0 163 | train/07_004_2.npy 0.0 164 | train/07_004_3.npy 0.0 165 | train/07_004_4.npy 0.0 166 | train/07_004_5.npy 0.0 167 | train/07_004_6.npy 0.0 168 | train/07_004_7.npy 0.0 169 | train/07_004_8.npy 0.0 170 | train/07_004_9.npy 0.0 171 | train/07_0047_0.npy 1.0 172 | train/07_0047_1.npy 1.0 173 | train/07_0047_2.npy 1.0 174 | train/07_0047_3.npy 1.0 175 | train/07_0047_4.npy 1.0 176 | train/07_0047_5.npy 1.0 177 | train/07_0047_6.npy 1.0 178 | train/07_0047_7.npy 1.0 179 | train/07_0047_8.npy 1.0 180 | train/07_0047_9.npy 1.0 181 | train/07_0049_0.npy 1.0 182 | train/07_0049_1.npy 1.0 183 | train/07_0049_2.npy 1.0 184 | train/07_0049_3.npy 1.0 185 | train/07_0049_4.npy 1.0 186 | train/07_0049_5.npy 1.0 187 | train/07_0049_6.npy 1.0 188 | train/07_0049_7.npy 1.0 189 | train/07_0049_8.npy 1.0 190 | train/07_0049_9.npy 1.0 191 | train/07_006_0.npy 0.0 192 | train/07_006_1.npy 0.0 193 | train/07_006_2.npy 0.0 194 | train/07_006_3.npy 0.0 195 | train/07_006_4.npy 0.0 196 | train/07_006_5.npy 0.0 197 | train/07_006_6.npy 0.0 198 | train/07_006_7.npy 0.0 199 | train/07_006_8.npy 0.0 200 | train/07_006_9.npy 0.0 201 | train/07_007_0.npy 0.0 202 | train/07_007_1.npy 0.0 203 | train/07_007_2.npy 0.0 204 | train/07_007_3.npy 0.0 205 | train/07_007_4.npy 0.0 206 | train/07_007_5.npy 0.0 207 | train/07_007_6.npy 0.0 208 | train/07_007_7.npy 0.0 209 | train/07_007_8.npy 0.0 210 | train/07_007_9.npy 0.0 -------------------------------------------------------------------------------- /other/bert_utils.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": null, 6 | "metadata": {}, 7 | "outputs": [], 8 | "source": [ 9 | "import numpy as np\n", 10 | "import pandas as pd\n", 11 | "import torch\n", 12 | "import torch.nn as nn\n", 13 | "from sklearn.model_selection import train_test_split\n", 14 | "from sklearn.metrics import classification_report\n", 15 | "from transformers import AutoModel, BertTokenizerFast\n", 16 | "\n", 17 | "# specify GPU\n", 18 | "device = torch.device(\"cuda\")\n", 19 | "# TRANSFORMERS_OFFLINE=1\n", 20 | "bert = AutoModel.from_pretrained('bert-base-uncased')\n", 21 | "# Load the BERT tokenizer\n", 22 | "tokenizer = BertTokenizerFast.from_pretrained('bert-base-uncased')\n" 23 | ] 24 | }, 25 | { 26 | "cell_type": "code", 27 | "execution_count": null, 28 | "metadata": {}, 29 | "outputs": [], 30 | "source": [ 31 | "inputs = tokenizer([\"hello world\"], return_tensors=\"pt\",padding=True) # \"pt\"表示\"pytorch\"\n", 32 | "outputs = bert(**inputs)\n", 33 | "print(inputs)\n", 34 | "print(outputs[0].shape)\n", 35 | "print(outputs[1].shape)" 36 | ] 37 | }, 38 | { 39 | "cell_type": "code", 40 | "execution_count": null, 41 | "metadata": {}, 42 | "outputs": [], 43 | "source": [ 44 | "import csv\n", 45 | "\n", 46 | "# 创建一个空字典来存储数据\n", 47 | "data_dict = {}\n", 48 | "# 使用with语句打开csv文件\n", 49 | "with open('', 'r', encoding='utf-8') as csvfile:\n", 50 | " reader = csv.DictReader(csvfile)\n", 51 | " for row in reader:\n", 52 | " name = row['name']\n", 53 | " sentence = row['sentence']\n", 54 | " data_dict[name] = sentence\n", 55 | "print(data_dict['RoadAccidents061_x264'])" 56 | ] 57 | }, 58 | { 59 | "cell_type": "code", 60 | "execution_count": null, 61 | "metadata": {}, 62 | "outputs": [], 63 | "source": [ 64 | "import pickle as pk\n", 65 | "\n", 66 | "feature_list = {}\n", 67 | "\n", 68 | "for dd in data_dict:\n", 69 | " name = dd\n", 70 | " sentence = data_dict[name]\n", 71 | " print(name)\n", 72 | " print(sentence)\n", 73 | " print(type(sentence))\n", 74 | " inputs = tokenizer(sentence, return_tensors=\"pt\",padding=True) # \"pt\"表示\"pytorch\"\n", 75 | " outputs = bert(**inputs)\n", 76 | " print(inputs)\n", 77 | " print(outputs[0].shape)\n", 78 | " print(outputs[1].shape) \n", 79 | " feature = outputs[1]\n", 80 | " feature_list[name] = feature\n", 81 | " break" 82 | ] 83 | }, 84 | { 85 | "cell_type": "code", 86 | "execution_count": null, 87 | "metadata": {}, 88 | "outputs": [], 89 | "source": [ 90 | "import pickle as pk\n", 91 | "\n", 92 | "feature_list = {}\n", 93 | "\n", 94 | "for dd in data_dict:\n", 95 | " name = dd\n", 96 | " sentence = data_dict[name]\n", 97 | " print(name)\n", 98 | " print(sentence)\n", 99 | "\n", 100 | " inputs = tokenizer(sentence, return_tensors=\"pt\",padding=True) # \"pt\"表示\"pytorch\"\n", 101 | " outputs = bert(**inputs)\n", 102 | " print(inputs)\n", 103 | " print(outputs[0].shape)\n", 104 | " print(outputs[1].shape) \n", 105 | "\n", 106 | " feature = outputs[1]\n", 107 | " feature_list[name] = feature\n", 108 | " # break\n", 109 | "\n", 110 | "print(feature_list)\n", 111 | "with open('','wb') as feature:\n", 112 | " pk.dump(feature_list,feature)" 113 | ] 114 | }, 115 | { 116 | "cell_type": "code", 117 | "execution_count": null, 118 | "metadata": {}, 119 | "outputs": [], 120 | "source": [ 121 | "import csv\n", 122 | "\n", 123 | "# 创建一个空字典来存储数据\n", 124 | "data_dict = {}\n", 125 | "print(\"===\")\n", 126 | "# 使用with语句打开csv文件\n", 127 | "with open('', 'r', encoding='utf-8') as csvfile:\n", 128 | " reader = csv.DictReader(csvfile)\n", 129 | " for row in reader:\n", 130 | " name = row['Name']\n", 131 | " name = name[:-1] + '_' + name[-1]\n", 132 | " sentence = row['Description']\n", 133 | " data_dict[name] = sentence\n", 134 | " # break\n", 135 | "# print(data_dict['RoadAccidents061_x264'])\n" 136 | ] 137 | }, 138 | { 139 | "cell_type": "code", 140 | "execution_count": null, 141 | "metadata": {}, 142 | "outputs": [], 143 | "source": [ 144 | "print(data_dict)" 145 | ] 146 | }, 147 | { 148 | "cell_type": "code", 149 | "execution_count": null, 150 | "metadata": {}, 151 | "outputs": [], 152 | "source": [ 153 | "import pickle as pk\n", 154 | "\n", 155 | "\n", 156 | "feature_list = {}\n", 157 | "\n", 158 | "for dd in data_dict:\n", 159 | " name = dd\n", 160 | " sentence = data_dict[name]\n", 161 | " MAX_LENGTH = 512\n", 162 | " print(\"sentence lengtg = {}\".format(len(sentence)))\n", 163 | " inputs = tokenizer(sentence, return_tensors=\"pt\", padding=True, truncation=True, max_length=MAX_LENGTH)\n", 164 | " feature = outputs[1]\n", 165 | " feature_list[name] = feature\n", 166 | " # break\n", 167 | "\n", 168 | "with open('','wb') as feature:\n", 169 | " pk.dump(feature_list,feature)" 170 | ] 171 | }, 172 | { 173 | "cell_type": "code", 174 | "execution_count": null, 175 | "metadata": {}, 176 | "outputs": [], 177 | "source": [] 178 | } 179 | ], 180 | "metadata": { 181 | "kernelspec": { 182 | "display_name": "STG-NF", 183 | "language": "python", 184 | "name": "python3" 185 | }, 186 | "language_info": { 187 | "codemirror_mode": { 188 | "name": "ipython", 189 | "version": 3 190 | }, 191 | "file_extension": ".py", 192 | "mimetype": "text/x-python", 193 | "name": "python", 194 | "nbconvert_exporter": "python", 195 | "pygments_lexer": "ipython3", 196 | "version": "3.8.13" 197 | }, 198 | "orig_nbformat": 4 199 | }, 200 | "nbformat": 4, 201 | "nbformat_minor": 2 202 | } 203 | -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/new_split/client5.list: -------------------------------------------------------------------------------- 1 | train/06_002_0.npy 0.0 2 | train/06_002_1.npy 0.0 3 | train/06_002_2.npy 0.0 4 | train/06_002_3.npy 0.0 5 | train/06_002_4.npy 0.0 6 | train/06_002_5.npy 0.0 7 | train/06_002_6.npy 0.0 8 | train/06_002_7.npy 0.0 9 | train/06_002_8.npy 0.0 10 | train/06_002_9.npy 0.0 11 | train/06_004_0.npy 0.0 12 | train/06_004_1.npy 0.0 13 | train/06_004_2.npy 0.0 14 | train/06_004_3.npy 0.0 15 | train/06_004_4.npy 0.0 16 | train/06_004_5.npy 0.0 17 | train/06_004_6.npy 0.0 18 | train/06_004_7.npy 0.0 19 | train/06_004_8.npy 0.0 20 | train/06_004_9.npy 0.0 21 | train/06_005_0.npy 0.0 22 | train/06_005_1.npy 0.0 23 | train/06_005_2.npy 0.0 24 | train/06_005_3.npy 0.0 25 | train/06_005_4.npy 0.0 26 | train/06_005_5.npy 0.0 27 | train/06_005_6.npy 0.0 28 | train/06_005_7.npy 0.0 29 | train/06_005_8.npy 0.0 30 | train/06_005_9.npy 0.0 31 | train/06_007_0.npy 0.0 32 | train/06_007_1.npy 0.0 33 | train/06_007_2.npy 0.0 34 | train/06_007_3.npy 0.0 35 | train/06_007_4.npy 0.0 36 | train/06_007_5.npy 0.0 37 | train/06_007_6.npy 0.0 38 | train/06_007_7.npy 0.0 39 | train/06_007_8.npy 0.0 40 | train/06_007_9.npy 0.0 41 | train/06_008_0.npy 0.0 42 | train/06_008_1.npy 0.0 43 | train/06_008_2.npy 0.0 44 | train/06_008_3.npy 0.0 45 | train/06_008_4.npy 0.0 46 | train/06_008_5.npy 0.0 47 | train/06_008_6.npy 0.0 48 | train/06_008_7.npy 0.0 49 | train/06_008_8.npy 0.0 50 | train/06_008_9.npy 0.0 51 | train/06_010_0.npy 0.0 52 | train/06_010_1.npy 0.0 53 | train/06_010_2.npy 0.0 54 | train/06_010_3.npy 0.0 55 | train/06_010_4.npy 0.0 56 | train/06_010_5.npy 0.0 57 | train/06_010_6.npy 0.0 58 | train/06_010_7.npy 0.0 59 | train/06_010_8.npy 0.0 60 | train/06_010_9.npy 0.0 61 | train/06_012_0.npy 0.0 62 | train/06_012_1.npy 0.0 63 | train/06_012_2.npy 0.0 64 | train/06_012_3.npy 0.0 65 | train/06_012_4.npy 0.0 66 | train/06_012_5.npy 0.0 67 | train/06_012_6.npy 0.0 68 | train/06_012_7.npy 0.0 69 | train/06_012_8.npy 0.0 70 | train/06_012_9.npy 0.0 71 | train/06_014_0.npy 0.0 72 | train/06_014_1.npy 0.0 73 | train/06_014_2.npy 0.0 74 | train/06_014_3.npy 0.0 75 | train/06_014_4.npy 0.0 76 | train/06_014_5.npy 0.0 77 | train/06_014_6.npy 0.0 78 | train/06_014_7.npy 0.0 79 | train/06_014_8.npy 0.0 80 | train/06_014_9.npy 0.0 81 | train/06_0145_0.npy 1.0 82 | train/06_0145_1.npy 1.0 83 | train/06_0145_2.npy 1.0 84 | train/06_0145_3.npy 1.0 85 | train/06_0145_4.npy 1.0 86 | train/06_0145_5.npy 1.0 87 | train/06_0145_6.npy 1.0 88 | train/06_0145_7.npy 1.0 89 | train/06_0145_8.npy 1.0 90 | train/06_0145_9.npy 1.0 91 | train/06_0147_0.npy 1.0 92 | train/06_0147_1.npy 1.0 93 | train/06_0147_2.npy 1.0 94 | train/06_0147_3.npy 1.0 95 | train/06_0147_4.npy 1.0 96 | train/06_0147_5.npy 1.0 97 | train/06_0147_6.npy 1.0 98 | train/06_0147_7.npy 1.0 99 | train/06_0147_8.npy 1.0 100 | train/06_0147_9.npy 1.0 101 | train/06_0150_0.npy 1.0 102 | train/06_0150_1.npy 1.0 103 | train/06_0150_2.npy 1.0 104 | train/06_0150_3.npy 1.0 105 | train/06_0150_4.npy 1.0 106 | train/06_0150_5.npy 1.0 107 | train/06_0150_6.npy 1.0 108 | train/06_0150_7.npy 1.0 109 | train/06_0150_8.npy 1.0 110 | train/06_0150_9.npy 1.0 111 | train/06_0155_0.npy 1.0 112 | train/06_0155_1.npy 1.0 113 | train/06_0155_2.npy 1.0 114 | train/06_0155_3.npy 1.0 115 | train/06_0155_4.npy 1.0 116 | train/06_0155_5.npy 1.0 117 | train/06_0155_6.npy 1.0 118 | train/06_0155_7.npy 1.0 119 | train/06_0155_8.npy 1.0 120 | train/06_0155_9.npy 1.0 121 | train/06_019_0.npy 0.0 122 | train/06_019_1.npy 0.0 123 | train/06_019_2.npy 0.0 124 | train/06_019_3.npy 0.0 125 | train/06_019_4.npy 0.0 126 | train/06_019_5.npy 0.0 127 | train/06_019_6.npy 0.0 128 | train/06_019_7.npy 0.0 129 | train/06_019_8.npy 0.0 130 | train/06_019_9.npy 0.0 131 | train/06_023_0.npy 0.0 132 | train/06_023_1.npy 0.0 133 | train/06_023_2.npy 0.0 134 | train/06_023_3.npy 0.0 135 | train/06_023_4.npy 0.0 136 | train/06_023_5.npy 0.0 137 | train/06_023_6.npy 0.0 138 | train/06_023_7.npy 0.0 139 | train/06_023_8.npy 0.0 140 | train/06_023_9.npy 0.0 141 | train/06_024_0.npy 0.0 142 | train/06_024_1.npy 0.0 143 | train/06_024_2.npy 0.0 144 | train/06_024_3.npy 0.0 145 | train/06_024_4.npy 0.0 146 | train/06_024_5.npy 0.0 147 | train/06_024_6.npy 0.0 148 | train/06_024_7.npy 0.0 149 | train/06_024_8.npy 0.0 150 | train/06_024_9.npy 0.0 151 | train/06_027_0.npy 0.0 152 | train/06_027_1.npy 0.0 153 | train/06_027_2.npy 0.0 154 | train/06_027_3.npy 0.0 155 | train/06_027_4.npy 0.0 156 | train/06_027_5.npy 0.0 157 | train/06_027_6.npy 0.0 158 | train/06_027_7.npy 0.0 159 | train/06_027_8.npy 0.0 160 | train/06_027_9.npy 0.0 161 | train/09_001_0.npy 0.0 162 | train/09_001_1.npy 0.0 163 | train/09_001_2.npy 0.0 164 | train/09_001_3.npy 0.0 165 | train/09_001_4.npy 0.0 166 | train/09_001_5.npy 0.0 167 | train/09_001_6.npy 0.0 168 | train/09_001_7.npy 0.0 169 | train/09_001_8.npy 0.0 170 | train/09_001_9.npy 0.0 171 | train/09_004_0.npy 0.0 172 | train/09_004_1.npy 0.0 173 | train/09_004_2.npy 0.0 174 | train/09_004_3.npy 0.0 175 | train/09_004_4.npy 0.0 176 | train/09_004_5.npy 0.0 177 | train/09_004_6.npy 0.0 178 | train/09_004_7.npy 0.0 179 | train/09_004_8.npy 0.0 180 | train/09_004_9.npy 0.0 181 | train/09_0057_0.npy 1.0 182 | train/09_0057_1.npy 1.0 183 | train/09_0057_2.npy 1.0 184 | train/09_0057_3.npy 1.0 185 | train/09_0057_4.npy 1.0 186 | train/09_0057_5.npy 1.0 187 | train/09_0057_6.npy 1.0 188 | train/09_0057_7.npy 1.0 189 | train/09_0057_8.npy 1.0 190 | train/09_0057_9.npy 1.0 191 | train/09_006_0.npy 0.0 192 | train/09_006_1.npy 0.0 193 | train/09_006_2.npy 0.0 194 | train/09_006_3.npy 0.0 195 | train/09_006_4.npy 0.0 196 | train/09_006_5.npy 0.0 197 | train/09_006_6.npy 0.0 198 | train/09_006_7.npy 0.0 199 | train/09_006_8.npy 0.0 200 | train/09_006_9.npy 0.0 201 | train/09_008_0.npy 0.0 202 | train/09_008_1.npy 0.0 203 | train/09_008_2.npy 0.0 204 | train/09_008_3.npy 0.0 205 | train/09_008_4.npy 0.0 206 | train/09_008_5.npy 0.0 207 | train/09_008_6.npy 0.0 208 | train/09_008_7.npy 0.0 209 | train/09_008_8.npy 0.0 210 | train/09_008_9.npy 0.0 211 | train/09_010_0.npy 0.0 212 | train/09_010_1.npy 0.0 213 | train/09_010_2.npy 0.0 214 | train/09_010_3.npy 0.0 215 | train/09_010_4.npy 0.0 216 | train/09_010_5.npy 0.0 217 | train/09_010_6.npy 0.0 218 | train/09_010_7.npy 0.0 219 | train/09_010_8.npy 0.0 220 | train/09_010_9.npy 0.0 221 | -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/new_split/client7.list: -------------------------------------------------------------------------------- 1 | train/10_001_0.npy 0.0 2 | train/10_001_1.npy 0.0 3 | train/10_001_2.npy 0.0 4 | train/10_001_3.npy 0.0 5 | train/10_001_4.npy 0.0 6 | train/10_001_5.npy 0.0 7 | train/10_001_6.npy 0.0 8 | train/10_001_7.npy 0.0 9 | train/10_001_8.npy 0.0 10 | train/10_001_9.npy 0.0 11 | train/10_002_0.npy 0.0 12 | train/10_002_1.npy 0.0 13 | train/10_002_2.npy 0.0 14 | train/10_002_3.npy 0.0 15 | train/10_002_4.npy 0.0 16 | train/10_002_5.npy 0.0 17 | train/10_002_6.npy 0.0 18 | train/10_002_7.npy 0.0 19 | train/10_002_8.npy 0.0 20 | train/10_002_9.npy 0.0 21 | train/10_006_0.npy 0.0 22 | train/10_006_1.npy 0.0 23 | train/10_006_2.npy 0.0 24 | train/10_006_3.npy 0.0 25 | train/10_006_4.npy 0.0 26 | train/10_006_5.npy 0.0 27 | train/10_006_6.npy 0.0 28 | train/10_006_7.npy 0.0 29 | train/10_006_8.npy 0.0 30 | train/10_006_9.npy 0.0 31 | train/10_007_0.npy 0.0 32 | train/10_007_1.npy 0.0 33 | train/10_007_2.npy 0.0 34 | train/10_007_3.npy 0.0 35 | train/10_007_4.npy 0.0 36 | train/10_007_5.npy 0.0 37 | train/10_007_6.npy 0.0 38 | train/10_007_7.npy 0.0 39 | train/10_007_8.npy 0.0 40 | train/10_007_9.npy 0.0 41 | train/10_0074_0.npy 1.0 42 | train/10_0074_1.npy 1.0 43 | train/10_0074_2.npy 1.0 44 | train/10_0074_3.npy 1.0 45 | train/10_0074_4.npy 1.0 46 | train/10_0074_5.npy 1.0 47 | train/10_0074_6.npy 1.0 48 | train/10_0074_7.npy 1.0 49 | train/10_0074_8.npy 1.0 50 | train/10_0074_9.npy 1.0 51 | train/10_0075_0.npy 1.0 52 | train/10_0075_1.npy 1.0 53 | train/10_0075_2.npy 1.0 54 | train/10_0075_3.npy 1.0 55 | train/10_0075_4.npy 1.0 56 | train/10_0075_5.npy 1.0 57 | train/10_0075_6.npy 1.0 58 | train/10_0075_7.npy 1.0 59 | train/10_0075_8.npy 1.0 60 | train/10_0075_9.npy 1.0 61 | train/10_010_0.npy 0.0 62 | train/10_010_1.npy 0.0 63 | train/10_010_2.npy 0.0 64 | train/10_010_3.npy 0.0 65 | train/10_010_4.npy 0.0 66 | train/10_010_5.npy 0.0 67 | train/10_010_6.npy 0.0 68 | train/10_010_7.npy 0.0 69 | train/10_010_8.npy 0.0 70 | train/10_010_9.npy 0.0 71 | train/12_002_0.npy 0.0 72 | train/12_002_1.npy 0.0 73 | train/12_002_2.npy 0.0 74 | train/12_002_3.npy 0.0 75 | train/12_002_4.npy 0.0 76 | train/12_002_5.npy 0.0 77 | train/12_002_6.npy 0.0 78 | train/12_002_7.npy 0.0 79 | train/12_002_8.npy 0.0 80 | train/12_002_9.npy 0.0 81 | train/12_003_0.npy 0.0 82 | train/12_003_1.npy 0.0 83 | train/12_003_2.npy 0.0 84 | train/12_003_3.npy 0.0 85 | train/12_003_4.npy 0.0 86 | train/12_003_5.npy 0.0 87 | train/12_003_6.npy 0.0 88 | train/12_003_7.npy 0.0 89 | train/12_003_8.npy 0.0 90 | train/12_003_9.npy 0.0 91 | train/12_009_0.npy 0.0 92 | train/12_009_1.npy 0.0 93 | train/12_009_2.npy 0.0 94 | train/12_009_3.npy 0.0 95 | train/12_009_4.npy 0.0 96 | train/12_009_5.npy 0.0 97 | train/12_009_6.npy 0.0 98 | train/12_009_7.npy 0.0 99 | train/12_009_8.npy 0.0 100 | train/12_009_9.npy 0.0 101 | train/12_010_0.npy 0.0 102 | train/12_010_1.npy 0.0 103 | train/12_010_2.npy 0.0 104 | train/12_010_3.npy 0.0 105 | train/12_010_4.npy 0.0 106 | train/12_010_5.npy 0.0 107 | train/12_010_6.npy 0.0 108 | train/12_010_7.npy 0.0 109 | train/12_010_8.npy 0.0 110 | train/12_010_9.npy 0.0 111 | train/12_013_0.npy 0.0 112 | train/12_013_1.npy 0.0 113 | train/12_013_2.npy 0.0 114 | train/12_013_3.npy 0.0 115 | train/12_013_4.npy 0.0 116 | train/12_013_5.npy 0.0 117 | train/12_013_6.npy 0.0 118 | train/12_013_7.npy 0.0 119 | train/12_013_8.npy 0.0 120 | train/12_013_9.npy 0.0 121 | train/12_014_0.npy 0.0 122 | train/12_014_1.npy 0.0 123 | train/12_014_2.npy 0.0 124 | train/12_014_3.npy 0.0 125 | train/12_014_4.npy 0.0 126 | train/12_014_5.npy 0.0 127 | train/12_014_6.npy 0.0 128 | train/12_014_7.npy 0.0 129 | train/12_014_8.npy 0.0 130 | train/12_014_9.npy 0.0 131 | train/12_0143_0.npy 1.0 132 | train/12_0143_1.npy 1.0 133 | train/12_0143_2.npy 1.0 134 | train/12_0143_3.npy 1.0 135 | train/12_0143_4.npy 1.0 136 | train/12_0143_5.npy 1.0 137 | train/12_0143_6.npy 1.0 138 | train/12_0143_7.npy 1.0 139 | train/12_0143_8.npy 1.0 140 | train/12_0143_9.npy 1.0 141 | train/12_0148_0.npy 1.0 142 | train/12_0148_1.npy 1.0 143 | train/12_0148_2.npy 1.0 144 | train/12_0148_3.npy 1.0 145 | train/12_0148_4.npy 1.0 146 | train/12_0148_5.npy 1.0 147 | train/12_0148_6.npy 1.0 148 | train/12_0148_7.npy 1.0 149 | train/12_0148_8.npy 1.0 150 | train/12_0148_9.npy 1.0 151 | train/12_0152_0.npy 1.0 152 | train/12_0152_1.npy 1.0 153 | train/12_0152_2.npy 1.0 154 | train/12_0152_3.npy 1.0 155 | train/12_0152_4.npy 1.0 156 | train/12_0152_5.npy 1.0 157 | train/12_0152_6.npy 1.0 158 | train/12_0152_7.npy 1.0 159 | train/12_0152_8.npy 1.0 160 | train/12_0152_9.npy 1.0 161 | train/12_0154_0.npy 1.0 162 | train/12_0154_1.npy 1.0 163 | train/12_0154_2.npy 1.0 164 | train/12_0154_3.npy 1.0 165 | train/12_0154_4.npy 1.0 166 | train/12_0154_5.npy 1.0 167 | train/12_0154_6.npy 1.0 168 | train/12_0154_7.npy 1.0 169 | train/12_0154_8.npy 1.0 170 | train/12_0154_9.npy 1.0 171 | train/12_0175_0.npy 1.0 172 | train/12_0175_1.npy 1.0 173 | train/12_0175_2.npy 1.0 174 | train/12_0175_3.npy 1.0 175 | train/12_0175_4.npy 1.0 176 | train/12_0175_5.npy 1.0 177 | train/12_0175_6.npy 1.0 178 | train/12_0175_7.npy 1.0 179 | train/12_0175_8.npy 1.0 180 | train/12_0175_9.npy 1.0 181 | train/13_002_0.npy 0.0 182 | train/13_002_1.npy 0.0 183 | train/13_002_2.npy 0.0 184 | train/13_002_3.npy 0.0 185 | train/13_002_4.npy 0.0 186 | train/13_002_5.npy 0.0 187 | train/13_002_6.npy 0.0 188 | train/13_002_7.npy 0.0 189 | train/13_002_8.npy 0.0 190 | train/13_002_9.npy 0.0 191 | train/13_003_0.npy 0.0 192 | train/13_003_1.npy 0.0 193 | train/13_003_2.npy 0.0 194 | train/13_003_3.npy 0.0 195 | train/13_003_4.npy 0.0 196 | train/13_003_5.npy 0.0 197 | train/13_003_6.npy 0.0 198 | train/13_003_7.npy 0.0 199 | train/13_003_8.npy 0.0 200 | train/13_003_9.npy 0.0 201 | train/13_004_0.npy 0.0 202 | train/13_004_1.npy 0.0 203 | train/13_004_2.npy 0.0 204 | train/13_004_3.npy 0.0 205 | train/13_004_4.npy 0.0 206 | train/13_004_5.npy 0.0 207 | train/13_004_6.npy 0.0 208 | train/13_004_7.npy 0.0 209 | train/13_004_8.npy 0.0 210 | train/13_004_9.npy 0.0 211 | train/13_007_0.npy 0.0 212 | train/13_007_1.npy 0.0 213 | train/13_007_2.npy 0.0 214 | train/13_007_3.npy 0.0 215 | train/13_007_4.npy 0.0 216 | train/13_007_5.npy 0.0 217 | train/13_007_6.npy 0.0 218 | train/13_007_7.npy 0.0 219 | train/13_007_8.npy 0.0 220 | train/13_007_9.npy 0.0 221 | -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_clients/client6.list: -------------------------------------------------------------------------------- 1 | test/08_0058_0.npy 2 | test/08_0058_1.npy 3 | test/08_0058_2.npy 4 | test/08_0058_3.npy 5 | test/08_0058_4.npy 6 | test/08_0058_5.npy 7 | test/08_0058_6.npy 8 | test/08_0058_7.npy 9 | test/08_0058_8.npy 10 | test/08_0058_9.npy 11 | test/08_0077_0.npy 12 | test/08_0077_1.npy 13 | test/08_0077_2.npy 14 | test/08_0077_3.npy 15 | test/08_0077_4.npy 16 | test/08_0077_5.npy 17 | test/08_0077_6.npy 18 | test/08_0077_7.npy 19 | test/08_0077_8.npy 20 | test/08_0077_9.npy 21 | test/08_0080_0.npy 22 | test/08_0080_1.npy 23 | test/08_0080_2.npy 24 | test/08_0080_3.npy 25 | test/08_0080_4.npy 26 | test/08_0080_5.npy 27 | test/08_0080_6.npy 28 | test/08_0080_7.npy 29 | test/08_0080_8.npy 30 | test/08_0080_9.npy 31 | test/08_0157_0.npy 32 | test/08_0157_1.npy 33 | test/08_0157_2.npy 34 | test/08_0157_3.npy 35 | test/08_0157_4.npy 36 | test/08_0157_5.npy 37 | test/08_0157_6.npy 38 | test/08_0157_7.npy 39 | test/08_0157_8.npy 40 | test/08_0157_9.npy 41 | test/08_0158_0.npy 42 | test/08_0158_1.npy 43 | test/08_0158_2.npy 44 | test/08_0158_3.npy 45 | test/08_0158_4.npy 46 | test/08_0158_5.npy 47 | test/08_0158_6.npy 48 | test/08_0158_7.npy 49 | test/08_0158_8.npy 50 | test/08_0158_9.npy 51 | test/08_001_0.npy 52 | test/08_001_1.npy 53 | test/08_001_2.npy 54 | test/08_001_3.npy 55 | test/08_001_4.npy 56 | test/08_001_5.npy 57 | test/08_001_6.npy 58 | test/08_001_7.npy 59 | test/08_001_8.npy 60 | test/08_001_9.npy 61 | test/08_002_0.npy 62 | test/08_002_1.npy 63 | test/08_002_2.npy 64 | test/08_002_3.npy 65 | test/08_002_4.npy 66 | test/08_002_5.npy 67 | test/08_002_6.npy 68 | test/08_002_7.npy 69 | test/08_002_8.npy 70 | test/08_002_9.npy 71 | test/08_004_0.npy 72 | test/08_004_1.npy 73 | test/08_004_2.npy 74 | test/08_004_3.npy 75 | test/08_004_4.npy 76 | test/08_004_5.npy 77 | test/08_004_6.npy 78 | test/08_004_7.npy 79 | test/08_004_8.npy 80 | test/08_004_9.npy 81 | test/08_006_0.npy 82 | test/08_006_1.npy 83 | test/08_006_2.npy 84 | test/08_006_3.npy 85 | test/08_006_4.npy 86 | test/08_006_5.npy 87 | test/08_006_6.npy 88 | test/08_006_7.npy 89 | test/08_006_8.npy 90 | test/08_006_9.npy 91 | test/08_007_0.npy 92 | test/08_007_1.npy 93 | test/08_007_2.npy 94 | test/08_007_3.npy 95 | test/08_007_4.npy 96 | test/08_007_5.npy 97 | test/08_007_6.npy 98 | test/08_007_7.npy 99 | test/08_007_8.npy 100 | test/08_007_9.npy 101 | test/08_008_0.npy 102 | test/08_008_1.npy 103 | test/08_008_2.npy 104 | test/08_008_3.npy 105 | test/08_008_4.npy 106 | test/08_008_5.npy 107 | test/08_008_6.npy 108 | test/08_008_7.npy 109 | test/08_008_8.npy 110 | test/08_008_9.npy 111 | test/08_010_0.npy 112 | test/08_010_1.npy 113 | test/08_010_2.npy 114 | test/08_010_3.npy 115 | test/08_010_4.npy 116 | test/08_010_5.npy 117 | test/08_010_6.npy 118 | test/08_010_7.npy 119 | test/08_010_8.npy 120 | test/08_010_9.npy 121 | test/08_011_0.npy 122 | test/08_011_1.npy 123 | test/08_011_2.npy 124 | test/08_011_3.npy 125 | test/08_011_4.npy 126 | test/08_011_5.npy 127 | test/08_011_6.npy 128 | test/08_011_7.npy 129 | test/08_011_8.npy 130 | test/08_011_9.npy 131 | test/08_016_0.npy 132 | test/08_016_1.npy 133 | test/08_016_2.npy 134 | test/08_016_3.npy 135 | test/08_016_4.npy 136 | test/08_016_5.npy 137 | test/08_016_6.npy 138 | test/08_016_7.npy 139 | test/08_016_8.npy 140 | test/08_016_9.npy 141 | test/08_021_0.npy 142 | test/08_021_1.npy 143 | test/08_021_2.npy 144 | test/08_021_3.npy 145 | test/08_021_4.npy 146 | test/08_021_5.npy 147 | test/08_021_6.npy 148 | test/08_021_7.npy 149 | test/08_021_8.npy 150 | test/08_021_9.npy 151 | test/08_023_0.npy 152 | test/08_023_1.npy 153 | test/08_023_2.npy 154 | test/08_023_3.npy 155 | test/08_023_4.npy 156 | test/08_023_5.npy 157 | test/08_023_6.npy 158 | test/08_023_7.npy 159 | test/08_023_8.npy 160 | test/08_023_9.npy 161 | test/08_024_0.npy 162 | test/08_024_1.npy 163 | test/08_024_2.npy 164 | test/08_024_3.npy 165 | test/08_024_4.npy 166 | test/08_024_5.npy 167 | test/08_024_6.npy 168 | test/08_024_7.npy 169 | test/08_024_8.npy 170 | test/08_024_9.npy 171 | test/08_025_0.npy 172 | test/08_025_1.npy 173 | test/08_025_2.npy 174 | test/08_025_3.npy 175 | test/08_025_4.npy 176 | test/08_025_5.npy 177 | test/08_025_6.npy 178 | test/08_025_7.npy 179 | test/08_025_8.npy 180 | test/08_025_9.npy 181 | test/08_028_0.npy 182 | test/08_028_1.npy 183 | test/08_028_2.npy 184 | test/08_028_3.npy 185 | test/08_028_4.npy 186 | test/08_028_5.npy 187 | test/08_028_6.npy 188 | test/08_028_7.npy 189 | test/08_028_8.npy 190 | test/08_028_9.npy 191 | test/08_029_0.npy 192 | test/08_029_1.npy 193 | test/08_029_2.npy 194 | test/08_029_3.npy 195 | test/08_029_4.npy 196 | test/08_029_5.npy 197 | test/08_029_6.npy 198 | test/08_029_7.npy 199 | test/08_029_8.npy 200 | test/08_029_9.npy 201 | test/08_030_0.npy 202 | test/08_030_1.npy 203 | test/08_030_2.npy 204 | test/08_030_3.npy 205 | test/08_030_4.npy 206 | test/08_030_5.npy 207 | test/08_030_6.npy 208 | test/08_030_7.npy 209 | test/08_030_8.npy 210 | test/08_030_9.npy 211 | test/08_032_0.npy 212 | test/08_032_1.npy 213 | test/08_032_2.npy 214 | test/08_032_3.npy 215 | test/08_032_4.npy 216 | test/08_032_5.npy 217 | test/08_032_6.npy 218 | test/08_032_7.npy 219 | test/08_032_8.npy 220 | test/08_032_9.npy 221 | test/08_034_0.npy 222 | test/08_034_1.npy 223 | test/08_034_2.npy 224 | test/08_034_3.npy 225 | test/08_034_4.npy 226 | test/08_034_5.npy 227 | test/08_034_6.npy 228 | test/08_034_7.npy 229 | test/08_034_8.npy 230 | test/08_034_9.npy 231 | test/08_037_0.npy 232 | test/08_037_1.npy 233 | test/08_037_2.npy 234 | test/08_037_3.npy 235 | test/08_037_4.npy 236 | test/08_037_5.npy 237 | test/08_037_6.npy 238 | test/08_037_7.npy 239 | test/08_037_8.npy 240 | test/08_037_9.npy 241 | test/08_043_0.npy 242 | test/08_043_1.npy 243 | test/08_043_2.npy 244 | test/08_043_3.npy 245 | test/08_043_4.npy 246 | test/08_043_5.npy 247 | test/08_043_6.npy 248 | test/08_043_7.npy 249 | test/08_043_8.npy 250 | test/08_043_9.npy 251 | test/08_045_0.npy 252 | test/08_045_1.npy 253 | test/08_045_2.npy 254 | test/08_045_3.npy 255 | test/08_045_4.npy 256 | test/08_045_5.npy 257 | test/08_045_6.npy 258 | test/08_045_7.npy 259 | test/08_045_8.npy 260 | test/08_045_9.npy 261 | test/08_046_0.npy 262 | test/08_046_1.npy 263 | test/08_046_2.npy 264 | test/08_046_3.npy 265 | test/08_046_4.npy 266 | test/08_046_5.npy 267 | test/08_046_6.npy 268 | test/08_046_7.npy 269 | test/08_046_8.npy 270 | test/08_046_9.npy 271 | test/08_047_0.npy 272 | test/08_047_1.npy 273 | test/08_047_2.npy 274 | test/08_047_3.npy 275 | test/08_047_4.npy 276 | test/08_047_5.npy 277 | test/08_047_6.npy 278 | test/08_047_7.npy 279 | test/08_047_8.npy 280 | test/08_047_9.npy 281 | test/08_048_0.npy 282 | test/08_048_1.npy 283 | test/08_048_2.npy 284 | test/08_048_3.npy 285 | test/08_048_4.npy 286 | test/08_048_5.npy 287 | test/08_048_6.npy 288 | test/08_048_7.npy 289 | test/08_048_8.npy 290 | test/08_048_9.npy 291 | test/11_003_0.npy 292 | test/11_003_1.npy 293 | test/11_003_2.npy 294 | test/11_003_3.npy 295 | test/11_003_4.npy 296 | test/11_003_5.npy 297 | test/11_003_6.npy 298 | test/11_003_7.npy 299 | test/11_003_8.npy 300 | test/11_003_9.npy 301 | test/11_004_0.npy 302 | test/11_004_1.npy 303 | test/11_004_2.npy 304 | test/11_004_3.npy 305 | test/11_004_4.npy 306 | test/11_004_5.npy 307 | test/11_004_6.npy 308 | test/11_004_7.npy 309 | test/11_004_8.npy 310 | test/11_004_9.npy 311 | test/11_005_0.npy 312 | test/11_005_1.npy 313 | test/11_005_2.npy 314 | test/11_005_3.npy 315 | test/11_005_4.npy 316 | test/11_005_5.npy 317 | test/11_005_6.npy 318 | test/11_005_7.npy 319 | test/11_005_8.npy 320 | test/11_005_9.npy 321 | test/11_006_0.npy 322 | test/11_006_1.npy 323 | test/11_006_2.npy 324 | test/11_006_3.npy 325 | test/11_006_4.npy 326 | test/11_006_5.npy 327 | test/11_006_6.npy 328 | test/11_006_7.npy 329 | test/11_006_8.npy 330 | test/11_006_9.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_clients/client4.list: -------------------------------------------------------------------------------- 1 | test/05_0017_0.npy 2 | test/05_0017_1.npy 3 | test/05_0017_2.npy 4 | test/05_0017_3.npy 5 | test/05_0017_4.npy 6 | test/05_0017_5.npy 7 | test/05_0017_6.npy 8 | test/05_0017_7.npy 9 | test/05_0017_8.npy 10 | test/05_0017_9.npy 11 | test/05_0019_0.npy 12 | test/05_0019_1.npy 13 | test/05_0019_2.npy 14 | test/05_0019_3.npy 15 | test/05_0019_4.npy 16 | test/05_0019_5.npy 17 | test/05_0019_6.npy 18 | test/05_0019_7.npy 19 | test/05_0019_8.npy 20 | test/05_0019_9.npy 21 | test/05_0023_0.npy 22 | test/05_0023_1.npy 23 | test/05_0023_2.npy 24 | test/05_0023_3.npy 25 | test/05_0023_4.npy 26 | test/05_0023_5.npy 27 | test/05_0023_6.npy 28 | test/05_0023_7.npy 29 | test/05_0023_8.npy 30 | test/05_0023_9.npy 31 | test/05_004_0.npy 32 | test/05_004_1.npy 33 | test/05_004_2.npy 34 | test/05_004_3.npy 35 | test/05_004_4.npy 36 | test/05_004_5.npy 37 | test/05_004_6.npy 38 | test/05_004_7.npy 39 | test/05_004_8.npy 40 | test/05_004_9.npy 41 | test/05_005_0.npy 42 | test/05_005_1.npy 43 | test/05_005_2.npy 44 | test/05_005_3.npy 45 | test/05_005_4.npy 46 | test/05_005_5.npy 47 | test/05_005_6.npy 48 | test/05_005_7.npy 49 | test/05_005_8.npy 50 | test/05_005_9.npy 51 | test/05_009_0.npy 52 | test/05_009_1.npy 53 | test/05_009_2.npy 54 | test/05_009_3.npy 55 | test/05_009_4.npy 56 | test/05_009_5.npy 57 | test/05_009_6.npy 58 | test/05_009_7.npy 59 | test/05_009_8.npy 60 | test/05_009_9.npy 61 | test/05_010_0.npy 62 | test/05_010_1.npy 63 | test/05_010_2.npy 64 | test/05_010_3.npy 65 | test/05_010_4.npy 66 | test/05_010_5.npy 67 | test/05_010_6.npy 68 | test/05_010_7.npy 69 | test/05_010_8.npy 70 | test/05_010_9.npy 71 | test/05_013_0.npy 72 | test/05_013_1.npy 73 | test/05_013_2.npy 74 | test/05_013_3.npy 75 | test/05_013_4.npy 76 | test/05_013_5.npy 77 | test/05_013_6.npy 78 | test/05_013_7.npy 79 | test/05_013_8.npy 80 | test/05_013_9.npy 81 | test/05_016_0.npy 82 | test/05_016_1.npy 83 | test/05_016_2.npy 84 | test/05_016_3.npy 85 | test/05_016_4.npy 86 | test/05_016_5.npy 87 | test/05_016_6.npy 88 | test/05_016_7.npy 89 | test/05_016_8.npy 90 | test/05_016_9.npy 91 | test/05_018_0.npy 92 | test/05_018_1.npy 93 | test/05_018_2.npy 94 | test/05_018_3.npy 95 | test/05_018_4.npy 96 | test/05_018_5.npy 97 | test/05_018_6.npy 98 | test/05_018_7.npy 99 | test/05_018_8.npy 100 | test/05_018_9.npy 101 | test/05_021_0.npy 102 | test/05_021_1.npy 103 | test/05_021_2.npy 104 | test/05_021_3.npy 105 | test/05_021_4.npy 106 | test/05_021_5.npy 107 | test/05_021_6.npy 108 | test/05_021_7.npy 109 | test/05_021_8.npy 110 | test/05_021_9.npy 111 | test/05_023_0.npy 112 | test/05_023_1.npy 113 | test/05_023_2.npy 114 | test/05_023_3.npy 115 | test/05_023_4.npy 116 | test/05_023_5.npy 117 | test/05_023_6.npy 118 | test/05_023_7.npy 119 | test/05_023_8.npy 120 | test/05_023_9.npy 121 | test/05_025_0.npy 122 | test/05_025_1.npy 123 | test/05_025_2.npy 124 | test/05_025_3.npy 125 | test/05_025_4.npy 126 | test/05_025_5.npy 127 | test/05_025_6.npy 128 | test/05_025_7.npy 129 | test/05_025_8.npy 130 | test/05_025_9.npy 131 | test/05_026_0.npy 132 | test/05_026_1.npy 133 | test/05_026_2.npy 134 | test/05_026_3.npy 135 | test/05_026_4.npy 136 | test/05_026_5.npy 137 | test/05_026_6.npy 138 | test/05_026_7.npy 139 | test/05_026_8.npy 140 | test/05_026_9.npy 141 | test/05_028_0.npy 142 | test/05_028_1.npy 143 | test/05_028_2.npy 144 | test/05_028_3.npy 145 | test/05_028_4.npy 146 | test/05_028_5.npy 147 | test/05_028_6.npy 148 | test/05_028_7.npy 149 | test/05_028_8.npy 150 | test/05_028_9.npy 151 | test/05_030_0.npy 152 | test/05_030_1.npy 153 | test/05_030_2.npy 154 | test/05_030_3.npy 155 | test/05_030_4.npy 156 | test/05_030_5.npy 157 | test/05_030_6.npy 158 | test/05_030_7.npy 159 | test/05_030_8.npy 160 | test/05_030_9.npy 161 | test/05_033_0.npy 162 | test/05_033_1.npy 163 | test/05_033_2.npy 164 | test/05_033_3.npy 165 | test/05_033_4.npy 166 | test/05_033_5.npy 167 | test/05_033_6.npy 168 | test/05_033_7.npy 169 | test/05_033_8.npy 170 | test/05_033_9.npy 171 | test/05_034_0.npy 172 | test/05_034_1.npy 173 | test/05_034_2.npy 174 | test/05_034_3.npy 175 | test/05_034_4.npy 176 | test/05_034_5.npy 177 | test/05_034_6.npy 178 | test/05_034_7.npy 179 | test/05_034_8.npy 180 | test/05_034_9.npy 181 | test/05_038_0.npy 182 | test/05_038_1.npy 183 | test/05_038_2.npy 184 | test/05_038_3.npy 185 | test/05_038_4.npy 186 | test/05_038_5.npy 187 | test/05_038_6.npy 188 | test/05_038_7.npy 189 | test/05_038_8.npy 190 | test/05_038_9.npy 191 | test/05_039_0.npy 192 | test/05_039_1.npy 193 | test/05_039_2.npy 194 | test/05_039_3.npy 195 | test/05_039_4.npy 196 | test/05_039_5.npy 197 | test/05_039_6.npy 198 | test/05_039_7.npy 199 | test/05_039_8.npy 200 | test/05_039_9.npy 201 | test/05_041_0.npy 202 | test/05_041_1.npy 203 | test/05_041_2.npy 204 | test/05_041_3.npy 205 | test/05_041_4.npy 206 | test/05_041_5.npy 207 | test/05_041_6.npy 208 | test/05_041_7.npy 209 | test/05_041_8.npy 210 | test/05_041_9.npy 211 | test/05_043_0.npy 212 | test/05_043_1.npy 213 | test/05_043_2.npy 214 | test/05_043_3.npy 215 | test/05_043_4.npy 216 | test/05_043_5.npy 217 | test/05_043_6.npy 218 | test/05_043_7.npy 219 | test/05_043_8.npy 220 | test/05_043_9.npy 221 | test/05_044_0.npy 222 | test/05_044_1.npy 223 | test/05_044_2.npy 224 | test/05_044_3.npy 225 | test/05_044_4.npy 226 | test/05_044_5.npy 227 | test/05_044_6.npy 228 | test/05_044_7.npy 229 | test/05_044_8.npy 230 | test/05_044_9.npy 231 | test/05_045_0.npy 232 | test/05_045_1.npy 233 | test/05_045_2.npy 234 | test/05_045_3.npy 235 | test/05_045_4.npy 236 | test/05_045_5.npy 237 | test/05_045_6.npy 238 | test/05_045_7.npy 239 | test/05_045_8.npy 240 | test/05_045_9.npy 241 | test/05_048_0.npy 242 | test/05_048_1.npy 243 | test/05_048_2.npy 244 | test/05_048_3.npy 245 | test/05_048_4.npy 246 | test/05_048_5.npy 247 | test/05_048_6.npy 248 | test/05_048_7.npy 249 | test/05_048_8.npy 250 | test/05_048_9.npy 251 | test/05_049_0.npy 252 | test/05_049_1.npy 253 | test/05_049_2.npy 254 | test/05_049_3.npy 255 | test/05_049_4.npy 256 | test/05_049_5.npy 257 | test/05_049_6.npy 258 | test/05_049_7.npy 259 | test/05_049_8.npy 260 | test/05_049_9.npy 261 | test/05_054_0.npy 262 | test/05_054_1.npy 263 | test/05_054_2.npy 264 | test/05_054_3.npy 265 | test/05_054_4.npy 266 | test/05_054_5.npy 267 | test/05_054_6.npy 268 | test/05_054_7.npy 269 | test/05_054_8.npy 270 | test/05_054_9.npy 271 | test/05_056_0.npy 272 | test/05_056_1.npy 273 | test/05_056_2.npy 274 | test/05_056_3.npy 275 | test/05_056_4.npy 276 | test/05_056_5.npy 277 | test/05_056_6.npy 278 | test/05_056_7.npy 279 | test/05_056_8.npy 280 | test/05_056_9.npy 281 | test/05_057_0.npy 282 | test/05_057_1.npy 283 | test/05_057_2.npy 284 | test/05_057_3.npy 285 | test/05_057_4.npy 286 | test/05_057_5.npy 287 | test/05_057_6.npy 288 | test/05_057_7.npy 289 | test/05_057_8.npy 290 | test/05_057_9.npy 291 | test/05_058_0.npy 292 | test/05_058_1.npy 293 | test/05_058_2.npy 294 | test/05_058_3.npy 295 | test/05_058_4.npy 296 | test/05_058_5.npy 297 | test/05_058_6.npy 298 | test/05_058_7.npy 299 | test/05_058_8.npy 300 | test/05_058_9.npy 301 | test/05_060_0.npy 302 | test/05_060_1.npy 303 | test/05_060_2.npy 304 | test/05_060_3.npy 305 | test/05_060_4.npy 306 | test/05_060_5.npy 307 | test/05_060_6.npy 308 | test/05_060_7.npy 309 | test/05_060_8.npy 310 | test/05_060_9.npy 311 | test/05_061_0.npy 312 | test/05_061_1.npy 313 | test/05_061_2.npy 314 | test/05_061_3.npy 315 | test/05_061_4.npy 316 | test/05_061_5.npy 317 | test/05_061_6.npy 318 | test/05_061_7.npy 319 | test/05_061_8.npy 320 | test/05_061_9.npy 321 | test/05_062_0.npy 322 | test/05_062_1.npy 323 | test/05_062_2.npy 324 | test/05_062_3.npy 325 | test/05_062_4.npy 326 | test/05_062_5.npy 327 | test/05_062_6.npy 328 | test/05_062_7.npy 329 | test/05_062_8.npy 330 | test/05_062_9.npy 331 | test/05_064_0.npy 332 | test/05_064_1.npy 333 | test/05_064_2.npy 334 | test/05_064_3.npy 335 | test/05_064_4.npy 336 | test/05_064_5.npy 337 | test/05_064_6.npy 338 | test/05_064_7.npy 339 | test/05_064_8.npy 340 | test/05_064_9.npy 341 | test/05_065_0.npy 342 | test/05_065_1.npy 343 | test/05_065_2.npy 344 | test/05_065_3.npy 345 | test/05_065_4.npy 346 | test/05_065_5.npy 347 | test/05_065_6.npy 348 | test/05_065_7.npy 349 | test/05_065_8.npy 350 | test/05_065_9.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/new_split/client4.list: -------------------------------------------------------------------------------- 1 | train/05_001_0.npy 0.0 2 | train/05_001_1.npy 0.0 3 | train/05_001_2.npy 0.0 4 | train/05_001_3.npy 0.0 5 | train/05_001_4.npy 0.0 6 | train/05_001_5.npy 0.0 7 | train/05_001_6.npy 0.0 8 | train/05_001_7.npy 0.0 9 | train/05_001_8.npy 0.0 10 | train/05_001_9.npy 0.0 11 | train/05_002_0.npy 0.0 12 | train/05_002_1.npy 0.0 13 | train/05_002_2.npy 0.0 14 | train/05_002_3.npy 0.0 15 | train/05_002_4.npy 0.0 16 | train/05_002_5.npy 0.0 17 | train/05_002_6.npy 0.0 18 | train/05_002_7.npy 0.0 19 | train/05_002_8.npy 0.0 20 | train/05_002_9.npy 0.0 21 | train/05_0020_0.npy 1.0 22 | train/05_0020_1.npy 1.0 23 | train/05_0020_2.npy 1.0 24 | train/05_0020_3.npy 1.0 25 | train/05_0020_4.npy 1.0 26 | train/05_0020_5.npy 1.0 27 | train/05_0020_6.npy 1.0 28 | train/05_0020_7.npy 1.0 29 | train/05_0020_8.npy 1.0 30 | train/05_0020_9.npy 1.0 31 | train/05_0021_0.npy 1.0 32 | train/05_0021_1.npy 1.0 33 | train/05_0021_2.npy 1.0 34 | train/05_0021_3.npy 1.0 35 | train/05_0021_4.npy 1.0 36 | train/05_0021_5.npy 1.0 37 | train/05_0021_6.npy 1.0 38 | train/05_0021_7.npy 1.0 39 | train/05_0021_8.npy 1.0 40 | train/05_0021_9.npy 1.0 41 | train/05_0022_0.npy 1.0 42 | train/05_0022_1.npy 1.0 43 | train/05_0022_2.npy 1.0 44 | train/05_0022_3.npy 1.0 45 | train/05_0022_4.npy 1.0 46 | train/05_0022_5.npy 1.0 47 | train/05_0022_6.npy 1.0 48 | train/05_0022_7.npy 1.0 49 | train/05_0022_8.npy 1.0 50 | train/05_0022_9.npy 1.0 51 | train/05_0024_0.npy 1.0 52 | train/05_0024_1.npy 1.0 53 | train/05_0024_2.npy 1.0 54 | train/05_0024_3.npy 1.0 55 | train/05_0024_4.npy 1.0 56 | train/05_0024_5.npy 1.0 57 | train/05_0024_6.npy 1.0 58 | train/05_0024_7.npy 1.0 59 | train/05_0024_8.npy 1.0 60 | train/05_0024_9.npy 1.0 61 | train/05_003_0.npy 0.0 62 | train/05_003_1.npy 0.0 63 | train/05_003_2.npy 0.0 64 | train/05_003_3.npy 0.0 65 | train/05_003_4.npy 0.0 66 | train/05_003_5.npy 0.0 67 | train/05_003_6.npy 0.0 68 | train/05_003_7.npy 0.0 69 | train/05_003_8.npy 0.0 70 | train/05_003_9.npy 0.0 71 | train/05_006_0.npy 0.0 72 | train/05_006_1.npy 0.0 73 | train/05_006_2.npy 0.0 74 | train/05_006_3.npy 0.0 75 | train/05_006_4.npy 0.0 76 | train/05_006_5.npy 0.0 77 | train/05_006_6.npy 0.0 78 | train/05_006_7.npy 0.0 79 | train/05_006_8.npy 0.0 80 | train/05_006_9.npy 0.0 81 | train/05_007_0.npy 0.0 82 | train/05_007_1.npy 0.0 83 | train/05_007_2.npy 0.0 84 | train/05_007_3.npy 0.0 85 | train/05_007_4.npy 0.0 86 | train/05_007_5.npy 0.0 87 | train/05_007_6.npy 0.0 88 | train/05_007_7.npy 0.0 89 | train/05_007_8.npy 0.0 90 | train/05_007_9.npy 0.0 91 | train/05_008_0.npy 0.0 92 | train/05_008_1.npy 0.0 93 | train/05_008_2.npy 0.0 94 | train/05_008_3.npy 0.0 95 | train/05_008_4.npy 0.0 96 | train/05_008_5.npy 0.0 97 | train/05_008_6.npy 0.0 98 | train/05_008_7.npy 0.0 99 | train/05_008_8.npy 0.0 100 | train/05_008_9.npy 0.0 101 | train/05_011_0.npy 0.0 102 | train/05_011_1.npy 0.0 103 | train/05_011_2.npy 0.0 104 | train/05_011_3.npy 0.0 105 | train/05_011_4.npy 0.0 106 | train/05_011_5.npy 0.0 107 | train/05_011_6.npy 0.0 108 | train/05_011_7.npy 0.0 109 | train/05_011_8.npy 0.0 110 | train/05_011_9.npy 0.0 111 | train/05_014_0.npy 0.0 112 | train/05_014_1.npy 0.0 113 | train/05_014_2.npy 0.0 114 | train/05_014_3.npy 0.0 115 | train/05_014_4.npy 0.0 116 | train/05_014_5.npy 0.0 117 | train/05_014_6.npy 0.0 118 | train/05_014_7.npy 0.0 119 | train/05_014_8.npy 0.0 120 | train/05_014_9.npy 0.0 121 | train/05_017_0.npy 0.0 122 | train/05_017_1.npy 0.0 123 | train/05_017_2.npy 0.0 124 | train/05_017_3.npy 0.0 125 | train/05_017_4.npy 0.0 126 | train/05_017_5.npy 0.0 127 | train/05_017_6.npy 0.0 128 | train/05_017_7.npy 0.0 129 | train/05_017_8.npy 0.0 130 | train/05_017_9.npy 0.0 131 | train/05_019_0.npy 0.0 132 | train/05_019_1.npy 0.0 133 | train/05_019_2.npy 0.0 134 | train/05_019_3.npy 0.0 135 | train/05_019_4.npy 0.0 136 | train/05_019_5.npy 0.0 137 | train/05_019_6.npy 0.0 138 | train/05_019_7.npy 0.0 139 | train/05_019_8.npy 0.0 140 | train/05_019_9.npy 0.0 141 | train/05_020_0.npy 0.0 142 | train/05_020_1.npy 0.0 143 | train/05_020_2.npy 0.0 144 | train/05_020_3.npy 0.0 145 | train/05_020_4.npy 0.0 146 | train/05_020_5.npy 0.0 147 | train/05_020_6.npy 0.0 148 | train/05_020_7.npy 0.0 149 | train/05_020_8.npy 0.0 150 | train/05_020_9.npy 0.0 151 | train/05_022_0.npy 0.0 152 | train/05_022_1.npy 0.0 153 | train/05_022_2.npy 0.0 154 | train/05_022_3.npy 0.0 155 | train/05_022_4.npy 0.0 156 | train/05_022_5.npy 0.0 157 | train/05_022_6.npy 0.0 158 | train/05_022_7.npy 0.0 159 | train/05_022_8.npy 0.0 160 | train/05_022_9.npy 0.0 161 | train/05_024_0.npy 0.0 162 | train/05_024_1.npy 0.0 163 | train/05_024_2.npy 0.0 164 | train/05_024_3.npy 0.0 165 | train/05_024_4.npy 0.0 166 | train/05_024_5.npy 0.0 167 | train/05_024_6.npy 0.0 168 | train/05_024_7.npy 0.0 169 | train/05_024_8.npy 0.0 170 | train/05_024_9.npy 0.0 171 | train/05_027_0.npy 0.0 172 | train/05_027_1.npy 0.0 173 | train/05_027_2.npy 0.0 174 | train/05_027_3.npy 0.0 175 | train/05_027_4.npy 0.0 176 | train/05_027_5.npy 0.0 177 | train/05_027_6.npy 0.0 178 | train/05_027_7.npy 0.0 179 | train/05_027_8.npy 0.0 180 | train/05_027_9.npy 0.0 181 | train/05_029_0.npy 0.0 182 | train/05_029_1.npy 0.0 183 | train/05_029_2.npy 0.0 184 | train/05_029_3.npy 0.0 185 | train/05_029_4.npy 0.0 186 | train/05_029_5.npy 0.0 187 | train/05_029_6.npy 0.0 188 | train/05_029_7.npy 0.0 189 | train/05_029_8.npy 0.0 190 | train/05_029_9.npy 0.0 191 | train/05_031_0.npy 0.0 192 | train/05_031_1.npy 0.0 193 | train/05_031_2.npy 0.0 194 | train/05_031_3.npy 0.0 195 | train/05_031_4.npy 0.0 196 | train/05_031_5.npy 0.0 197 | train/05_031_6.npy 0.0 198 | train/05_031_7.npy 0.0 199 | train/05_031_8.npy 0.0 200 | train/05_031_9.npy 0.0 201 | train/05_032_0.npy 0.0 202 | train/05_032_1.npy 0.0 203 | train/05_032_2.npy 0.0 204 | train/05_032_3.npy 0.0 205 | train/05_032_4.npy 0.0 206 | train/05_032_5.npy 0.0 207 | train/05_032_6.npy 0.0 208 | train/05_032_7.npy 0.0 209 | train/05_032_8.npy 0.0 210 | train/05_032_9.npy 0.0 211 | train/05_035_0.npy 0.0 212 | train/05_035_1.npy 0.0 213 | train/05_035_2.npy 0.0 214 | train/05_035_3.npy 0.0 215 | train/05_035_4.npy 0.0 216 | train/05_035_5.npy 0.0 217 | train/05_035_6.npy 0.0 218 | train/05_035_7.npy 0.0 219 | train/05_035_8.npy 0.0 220 | train/05_035_9.npy 0.0 221 | train/05_036_0.npy 0.0 222 | train/05_036_1.npy 0.0 223 | train/05_036_2.npy 0.0 224 | train/05_036_3.npy 0.0 225 | train/05_036_4.npy 0.0 226 | train/05_036_5.npy 0.0 227 | train/05_036_6.npy 0.0 228 | train/05_036_7.npy 0.0 229 | train/05_036_8.npy 0.0 230 | train/05_036_9.npy 0.0 231 | train/05_040_0.npy 0.0 232 | train/05_040_1.npy 0.0 233 | train/05_040_2.npy 0.0 234 | train/05_040_3.npy 0.0 235 | train/05_040_4.npy 0.0 236 | train/05_040_5.npy 0.0 237 | train/05_040_6.npy 0.0 238 | train/05_040_7.npy 0.0 239 | train/05_040_8.npy 0.0 240 | train/05_040_9.npy 0.0 241 | train/05_042_0.npy 0.0 242 | train/05_042_1.npy 0.0 243 | train/05_042_2.npy 0.0 244 | train/05_042_3.npy 0.0 245 | train/05_042_4.npy 0.0 246 | train/05_042_5.npy 0.0 247 | train/05_042_6.npy 0.0 248 | train/05_042_7.npy 0.0 249 | train/05_042_8.npy 0.0 250 | train/05_042_9.npy 0.0 251 | train/05_046_0.npy 0.0 252 | train/05_046_1.npy 0.0 253 | train/05_046_2.npy 0.0 254 | train/05_046_3.npy 0.0 255 | train/05_046_4.npy 0.0 256 | train/05_046_5.npy 0.0 257 | train/05_046_6.npy 0.0 258 | train/05_046_7.npy 0.0 259 | train/05_046_8.npy 0.0 260 | train/05_046_9.npy 0.0 261 | train/05_047_0.npy 0.0 262 | train/05_047_1.npy 0.0 263 | train/05_047_2.npy 0.0 264 | train/05_047_3.npy 0.0 265 | train/05_047_4.npy 0.0 266 | train/05_047_5.npy 0.0 267 | train/05_047_6.npy 0.0 268 | train/05_047_7.npy 0.0 269 | train/05_047_8.npy 0.0 270 | train/05_047_9.npy 0.0 271 | train/05_050_0.npy 0.0 272 | train/05_050_1.npy 0.0 273 | train/05_050_2.npy 0.0 274 | train/05_050_3.npy 0.0 275 | train/05_050_4.npy 0.0 276 | train/05_050_5.npy 0.0 277 | train/05_050_6.npy 0.0 278 | train/05_050_7.npy 0.0 279 | train/05_050_8.npy 0.0 280 | train/05_050_9.npy 0.0 281 | train/05_051_0.npy 0.0 282 | train/05_051_1.npy 0.0 283 | train/05_051_2.npy 0.0 284 | train/05_051_3.npy 0.0 285 | train/05_051_4.npy 0.0 286 | train/05_051_5.npy 0.0 287 | train/05_051_6.npy 0.0 288 | train/05_051_7.npy 0.0 289 | train/05_051_8.npy 0.0 290 | train/05_051_9.npy 0.0 291 | train/05_052_0.npy 0.0 292 | train/05_052_1.npy 0.0 293 | train/05_052_2.npy 0.0 294 | train/05_052_3.npy 0.0 295 | train/05_052_4.npy 0.0 296 | train/05_052_5.npy 0.0 297 | train/05_052_6.npy 0.0 298 | train/05_052_7.npy 0.0 299 | train/05_052_8.npy 0.0 300 | train/05_052_9.npy 0.0 301 | train/05_059_0.npy 0.0 302 | train/05_059_1.npy 0.0 303 | train/05_059_2.npy 0.0 304 | train/05_059_3.npy 0.0 305 | train/05_059_4.npy 0.0 306 | train/05_059_5.npy 0.0 307 | train/05_059_6.npy 0.0 308 | train/05_059_7.npy 0.0 309 | train/05_059_8.npy 0.0 310 | train/05_059_9.npy 0.0 311 | train/05_063_0.npy 0.0 312 | train/05_063_1.npy 0.0 313 | train/05_063_2.npy 0.0 314 | train/05_063_3.npy 0.0 315 | train/05_063_4.npy 0.0 316 | train/05_063_5.npy 0.0 317 | train/05_063_6.npy 0.0 318 | train/05_063_7.npy 0.0 319 | train/05_063_8.npy 0.0 320 | train/05_063_9.npy 0.0 -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/new_split/client6.list: -------------------------------------------------------------------------------- 1 | train/08_003_0.npy 0.0 2 | train/08_003_1.npy 0.0 3 | train/08_003_2.npy 0.0 4 | train/08_003_3.npy 0.0 5 | train/08_003_4.npy 0.0 6 | train/08_003_5.npy 0.0 7 | train/08_003_6.npy 0.0 8 | train/08_003_7.npy 0.0 9 | train/08_003_8.npy 0.0 10 | train/08_003_9.npy 0.0 11 | train/08_0044_0.npy 1.0 12 | train/08_0044_1.npy 1.0 13 | train/08_0044_2.npy 1.0 14 | train/08_0044_3.npy 1.0 15 | train/08_0044_4.npy 1.0 16 | train/08_0044_5.npy 1.0 17 | train/08_0044_6.npy 1.0 18 | train/08_0044_7.npy 1.0 19 | train/08_0044_8.npy 1.0 20 | train/08_0044_9.npy 1.0 21 | train/08_005_0.npy 0.0 22 | train/08_005_1.npy 0.0 23 | train/08_005_2.npy 0.0 24 | train/08_005_3.npy 0.0 25 | train/08_005_4.npy 0.0 26 | train/08_005_5.npy 0.0 27 | train/08_005_6.npy 0.0 28 | train/08_005_7.npy 0.0 29 | train/08_005_8.npy 0.0 30 | train/08_005_9.npy 0.0 31 | train/08_0078_0.npy 1.0 32 | train/08_0078_1.npy 1.0 33 | train/08_0078_2.npy 1.0 34 | train/08_0078_3.npy 1.0 35 | train/08_0078_4.npy 1.0 36 | train/08_0078_5.npy 1.0 37 | train/08_0078_6.npy 1.0 38 | train/08_0078_7.npy 1.0 39 | train/08_0078_8.npy 1.0 40 | train/08_0078_9.npy 1.0 41 | train/08_0079_0.npy 1.0 42 | train/08_0079_1.npy 1.0 43 | train/08_0079_2.npy 1.0 44 | train/08_0079_3.npy 1.0 45 | train/08_0079_4.npy 1.0 46 | train/08_0079_5.npy 1.0 47 | train/08_0079_6.npy 1.0 48 | train/08_0079_7.npy 1.0 49 | train/08_0079_8.npy 1.0 50 | train/08_0079_9.npy 1.0 51 | train/08_009_0.npy 0.0 52 | train/08_009_1.npy 0.0 53 | train/08_009_2.npy 0.0 54 | train/08_009_3.npy 0.0 55 | train/08_009_4.npy 0.0 56 | train/08_009_5.npy 0.0 57 | train/08_009_6.npy 0.0 58 | train/08_009_7.npy 0.0 59 | train/08_009_8.npy 0.0 60 | train/08_009_9.npy 0.0 61 | train/08_012_0.npy 0.0 62 | train/08_012_1.npy 0.0 63 | train/08_012_2.npy 0.0 64 | train/08_012_3.npy 0.0 65 | train/08_012_4.npy 0.0 66 | train/08_012_5.npy 0.0 67 | train/08_012_6.npy 0.0 68 | train/08_012_7.npy 0.0 69 | train/08_012_8.npy 0.0 70 | train/08_012_9.npy 0.0 71 | train/08_013_0.npy 0.0 72 | train/08_013_1.npy 0.0 73 | train/08_013_2.npy 0.0 74 | train/08_013_3.npy 0.0 75 | train/08_013_4.npy 0.0 76 | train/08_013_5.npy 0.0 77 | train/08_013_6.npy 0.0 78 | train/08_013_7.npy 0.0 79 | train/08_013_8.npy 0.0 80 | train/08_013_9.npy 0.0 81 | train/08_014_0.npy 0.0 82 | train/08_014_1.npy 0.0 83 | train/08_014_2.npy 0.0 84 | train/08_014_3.npy 0.0 85 | train/08_014_4.npy 0.0 86 | train/08_014_5.npy 0.0 87 | train/08_014_6.npy 0.0 88 | train/08_014_7.npy 0.0 89 | train/08_014_8.npy 0.0 90 | train/08_014_9.npy 0.0 91 | train/08_0159_0.npy 1.0 92 | train/08_0159_1.npy 1.0 93 | train/08_0159_2.npy 1.0 94 | train/08_0159_3.npy 1.0 95 | train/08_0159_4.npy 1.0 96 | train/08_0159_5.npy 1.0 97 | train/08_0159_6.npy 1.0 98 | train/08_0159_7.npy 1.0 99 | train/08_0159_8.npy 1.0 100 | train/08_0159_9.npy 1.0 101 | train/08_017_0.npy 0.0 102 | train/08_017_1.npy 0.0 103 | train/08_017_2.npy 0.0 104 | train/08_017_3.npy 0.0 105 | train/08_017_4.npy 0.0 106 | train/08_017_5.npy 0.0 107 | train/08_017_6.npy 0.0 108 | train/08_017_7.npy 0.0 109 | train/08_017_8.npy 0.0 110 | train/08_017_9.npy 0.0 111 | train/08_0179_0.npy 1.0 112 | train/08_0179_1.npy 1.0 113 | train/08_0179_2.npy 1.0 114 | train/08_0179_3.npy 1.0 115 | train/08_0179_4.npy 1.0 116 | train/08_0179_5.npy 1.0 117 | train/08_0179_6.npy 1.0 118 | train/08_0179_7.npy 1.0 119 | train/08_0179_8.npy 1.0 120 | train/08_0179_9.npy 1.0 121 | train/08_018_0.npy 0.0 122 | train/08_018_1.npy 0.0 123 | train/08_018_2.npy 0.0 124 | train/08_018_3.npy 0.0 125 | train/08_018_4.npy 0.0 126 | train/08_018_5.npy 0.0 127 | train/08_018_6.npy 0.0 128 | train/08_018_7.npy 0.0 129 | train/08_018_8.npy 0.0 130 | train/08_018_9.npy 0.0 131 | train/08_019_0.npy 0.0 132 | train/08_019_1.npy 0.0 133 | train/08_019_2.npy 0.0 134 | train/08_019_3.npy 0.0 135 | train/08_019_4.npy 0.0 136 | train/08_019_5.npy 0.0 137 | train/08_019_6.npy 0.0 138 | train/08_019_7.npy 0.0 139 | train/08_019_8.npy 0.0 140 | train/08_019_9.npy 0.0 141 | train/08_022_0.npy 0.0 142 | train/08_022_1.npy 0.0 143 | train/08_022_2.npy 0.0 144 | train/08_022_3.npy 0.0 145 | train/08_022_4.npy 0.0 146 | train/08_022_5.npy 0.0 147 | train/08_022_6.npy 0.0 148 | train/08_022_7.npy 0.0 149 | train/08_022_8.npy 0.0 150 | train/08_022_9.npy 0.0 151 | train/08_026_0.npy 0.0 152 | train/08_026_1.npy 0.0 153 | train/08_026_2.npy 0.0 154 | train/08_026_3.npy 0.0 155 | train/08_026_4.npy 0.0 156 | train/08_026_5.npy 0.0 157 | train/08_026_6.npy 0.0 158 | train/08_026_7.npy 0.0 159 | train/08_026_8.npy 0.0 160 | train/08_026_9.npy 0.0 161 | train/08_027_0.npy 0.0 162 | train/08_027_1.npy 0.0 163 | train/08_027_2.npy 0.0 164 | train/08_027_3.npy 0.0 165 | train/08_027_4.npy 0.0 166 | train/08_027_5.npy 0.0 167 | train/08_027_6.npy 0.0 168 | train/08_027_7.npy 0.0 169 | train/08_027_8.npy 0.0 170 | train/08_027_9.npy 0.0 171 | train/08_033_0.npy 0.0 172 | train/08_033_1.npy 0.0 173 | train/08_033_2.npy 0.0 174 | train/08_033_3.npy 0.0 175 | train/08_033_4.npy 0.0 176 | train/08_033_5.npy 0.0 177 | train/08_033_6.npy 0.0 178 | train/08_033_7.npy 0.0 179 | train/08_033_8.npy 0.0 180 | train/08_033_9.npy 0.0 181 | train/08_035_0.npy 0.0 182 | train/08_035_1.npy 0.0 183 | train/08_035_2.npy 0.0 184 | train/08_035_3.npy 0.0 185 | train/08_035_4.npy 0.0 186 | train/08_035_5.npy 0.0 187 | train/08_035_6.npy 0.0 188 | train/08_035_7.npy 0.0 189 | train/08_035_8.npy 0.0 190 | train/08_035_9.npy 0.0 191 | train/08_036_0.npy 0.0 192 | train/08_036_1.npy 0.0 193 | train/08_036_2.npy 0.0 194 | train/08_036_3.npy 0.0 195 | train/08_036_4.npy 0.0 196 | train/08_036_5.npy 0.0 197 | train/08_036_6.npy 0.0 198 | train/08_036_7.npy 0.0 199 | train/08_036_8.npy 0.0 200 | train/08_036_9.npy 0.0 201 | train/08_038_0.npy 0.0 202 | train/08_038_1.npy 0.0 203 | train/08_038_2.npy 0.0 204 | train/08_038_3.npy 0.0 205 | train/08_038_4.npy 0.0 206 | train/08_038_5.npy 0.0 207 | train/08_038_6.npy 0.0 208 | train/08_038_7.npy 0.0 209 | train/08_038_8.npy 0.0 210 | train/08_038_9.npy 0.0 211 | train/08_039_0.npy 0.0 212 | train/08_039_1.npy 0.0 213 | train/08_039_2.npy 0.0 214 | train/08_039_3.npy 0.0 215 | train/08_039_4.npy 0.0 216 | train/08_039_5.npy 0.0 217 | train/08_039_6.npy 0.0 218 | train/08_039_7.npy 0.0 219 | train/08_039_8.npy 0.0 220 | train/08_039_9.npy 0.0 221 | train/08_041_0.npy 0.0 222 | train/08_041_1.npy 0.0 223 | train/08_041_2.npy 0.0 224 | train/08_041_3.npy 0.0 225 | train/08_041_4.npy 0.0 226 | train/08_041_5.npy 0.0 227 | train/08_041_6.npy 0.0 228 | train/08_041_7.npy 0.0 229 | train/08_041_8.npy 0.0 230 | train/08_041_9.npy 0.0 231 | train/08_042_0.npy 0.0 232 | train/08_042_1.npy 0.0 233 | train/08_042_2.npy 0.0 234 | train/08_042_3.npy 0.0 235 | train/08_042_4.npy 0.0 236 | train/08_042_5.npy 0.0 237 | train/08_042_6.npy 0.0 238 | train/08_042_7.npy 0.0 239 | train/08_042_8.npy 0.0 240 | train/08_042_9.npy 0.0 241 | train/08_044_0.npy 0.0 242 | train/08_044_1.npy 0.0 243 | train/08_044_2.npy 0.0 244 | train/08_044_3.npy 0.0 245 | train/08_044_4.npy 0.0 246 | train/08_044_5.npy 0.0 247 | train/08_044_6.npy 0.0 248 | train/08_044_7.npy 0.0 249 | train/08_044_8.npy 0.0 250 | train/08_044_9.npy 0.0 251 | train/08_049_0.npy 0.0 252 | train/08_049_1.npy 0.0 253 | train/08_049_2.npy 0.0 254 | train/08_049_3.npy 0.0 255 | train/08_049_4.npy 0.0 256 | train/08_049_5.npy 0.0 257 | train/08_049_6.npy 0.0 258 | train/08_049_7.npy 0.0 259 | train/08_049_8.npy 0.0 260 | train/08_049_9.npy 0.0 261 | train/11_001_0.npy 0.0 262 | train/11_001_1.npy 0.0 263 | train/11_001_2.npy 0.0 264 | train/11_001_3.npy 0.0 265 | train/11_001_4.npy 0.0 266 | train/11_001_5.npy 0.0 267 | train/11_001_6.npy 0.0 268 | train/11_001_7.npy 0.0 269 | train/11_001_8.npy 0.0 270 | train/11_001_9.npy 0.0 271 | train/11_002_0.npy 0.0 272 | train/11_002_1.npy 0.0 273 | train/11_002_2.npy 0.0 274 | train/11_002_3.npy 0.0 275 | train/11_002_4.npy 0.0 276 | train/11_002_5.npy 0.0 277 | train/11_002_6.npy 0.0 278 | train/11_002_7.npy 0.0 279 | train/11_002_8.npy 0.0 280 | train/11_002_9.npy 0.0 281 | train/11_007_0.npy 0.0 282 | train/11_007_1.npy 0.0 283 | train/11_007_2.npy 0.0 284 | train/11_007_3.npy 0.0 285 | train/11_007_4.npy 0.0 286 | train/11_007_5.npy 0.0 287 | train/11_007_6.npy 0.0 288 | train/11_007_7.npy 0.0 289 | train/11_007_8.npy 0.0 290 | train/11_007_9.npy 0.0 291 | train/11_008_0.npy 0.0 292 | train/11_008_1.npy 0.0 293 | train/11_008_2.npy 0.0 294 | train/11_008_3.npy 0.0 295 | train/11_008_4.npy 0.0 296 | train/11_008_5.npy 0.0 297 | train/11_008_6.npy 0.0 298 | train/11_008_7.npy 0.0 299 | train/11_008_8.npy 0.0 300 | train/11_008_9.npy 0.0 301 | train/11_010_0.npy 0.0 302 | train/11_010_1.npy 0.0 303 | train/11_010_2.npy 0.0 304 | train/11_010_3.npy 0.0 305 | train/11_010_4.npy 0.0 306 | train/11_010_5.npy 0.0 307 | train/11_010_6.npy 0.0 308 | train/11_010_7.npy 0.0 309 | train/11_010_8.npy 0.0 310 | train/11_010_9.npy 0.0 311 | train/11_0176_0.npy 1.0 312 | train/11_0176_1.npy 1.0 313 | train/11_0176_2.npy 1.0 314 | train/11_0176_3.npy 1.0 315 | train/11_0176_4.npy 1.0 316 | train/11_0176_5.npy 1.0 317 | train/11_0176_6.npy 1.0 318 | train/11_0176_7.npy 1.0 319 | train/11_0176_8.npy 1.0 320 | train/11_0176_9.npy 1.0 -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/new_split/client0.list: -------------------------------------------------------------------------------- 1 | train/01_036_0.npy 0.0 2 | train/01_036_1.npy 0.0 3 | train/01_036_2.npy 0.0 4 | train/01_036_3.npy 0.0 5 | train/01_036_4.npy 0.0 6 | train/01_036_5.npy 0.0 7 | train/01_036_6.npy 0.0 8 | train/01_036_7.npy 0.0 9 | train/01_036_8.npy 0.0 10 | train/01_036_9.npy 0.0 11 | train/01_005_0.npy 0.0 12 | train/01_005_1.npy 0.0 13 | train/01_005_2.npy 0.0 14 | train/01_005_3.npy 0.0 15 | train/01_005_4.npy 0.0 16 | train/01_005_5.npy 0.0 17 | train/01_005_6.npy 0.0 18 | train/01_005_7.npy 0.0 19 | train/01_005_8.npy 0.0 20 | train/01_005_9.npy 0.0 21 | train/01_0016_0.npy 1.0 22 | train/01_0016_1.npy 1.0 23 | train/01_0016_2.npy 1.0 24 | train/01_0016_3.npy 1.0 25 | train/01_0016_4.npy 1.0 26 | train/01_0016_5.npy 1.0 27 | train/01_0016_6.npy 1.0 28 | train/01_0016_7.npy 1.0 29 | train/01_0016_8.npy 1.0 30 | train/01_0016_9.npy 1.0 31 | train/01_049_0.npy 0.0 32 | train/01_049_1.npy 0.0 33 | train/01_049_2.npy 0.0 34 | train/01_049_3.npy 0.0 35 | train/01_049_4.npy 0.0 36 | train/01_049_5.npy 0.0 37 | train/01_049_6.npy 0.0 38 | train/01_049_7.npy 0.0 39 | train/01_049_8.npy 0.0 40 | train/01_049_9.npy 0.0 41 | train/01_013_0.npy 0.0 42 | train/01_013_1.npy 0.0 43 | train/01_013_2.npy 0.0 44 | train/01_013_3.npy 0.0 45 | train/01_013_4.npy 0.0 46 | train/01_013_5.npy 0.0 47 | train/01_013_6.npy 0.0 48 | train/01_013_7.npy 0.0 49 | train/01_013_8.npy 0.0 50 | train/01_013_9.npy 0.0 51 | train/01_011_0.npy 0.0 52 | train/01_011_1.npy 0.0 53 | train/01_011_2.npy 0.0 54 | train/01_011_3.npy 0.0 55 | train/01_011_4.npy 0.0 56 | train/01_011_5.npy 0.0 57 | train/01_011_6.npy 0.0 58 | train/01_011_7.npy 0.0 59 | train/01_011_8.npy 0.0 60 | train/01_011_9.npy 0.0 61 | train/01_010_0.npy 0.0 62 | train/01_010_1.npy 0.0 63 | train/01_010_2.npy 0.0 64 | train/01_010_3.npy 0.0 65 | train/01_010_4.npy 0.0 66 | train/01_010_5.npy 0.0 67 | train/01_010_6.npy 0.0 68 | train/01_010_7.npy 0.0 69 | train/01_010_8.npy 0.0 70 | train/01_010_9.npy 0.0 71 | train/01_0054_0.npy 1.0 72 | train/01_0054_1.npy 1.0 73 | train/01_0054_2.npy 1.0 74 | train/01_0054_3.npy 1.0 75 | train/01_0054_4.npy 1.0 76 | train/01_0054_5.npy 1.0 77 | train/01_0054_6.npy 1.0 78 | train/01_0054_7.npy 1.0 79 | train/01_0054_8.npy 1.0 80 | train/01_0054_9.npy 1.0 81 | train/01_069_0.npy 0.0 82 | train/01_069_1.npy 0.0 83 | train/01_069_2.npy 0.0 84 | train/01_069_3.npy 0.0 85 | train/01_069_4.npy 0.0 86 | train/01_069_5.npy 0.0 87 | train/01_069_6.npy 0.0 88 | train/01_069_7.npy 0.0 89 | train/01_069_8.npy 0.0 90 | train/01_069_9.npy 0.0 91 | train/01_0030_0.npy 1.0 92 | train/01_0030_1.npy 1.0 93 | train/01_0030_2.npy 1.0 94 | train/01_0030_3.npy 1.0 95 | train/01_0030_4.npy 1.0 96 | train/01_0030_5.npy 1.0 97 | train/01_0030_6.npy 1.0 98 | train/01_0030_7.npy 1.0 99 | train/01_0030_8.npy 1.0 100 | train/01_0030_9.npy 1.0 101 | train/01_042_0.npy 0.0 102 | train/01_042_1.npy 0.0 103 | train/01_042_2.npy 0.0 104 | train/01_042_3.npy 0.0 105 | train/01_042_4.npy 0.0 106 | train/01_042_5.npy 0.0 107 | train/01_042_6.npy 0.0 108 | train/01_042_7.npy 0.0 109 | train/01_042_8.npy 0.0 110 | train/01_042_9.npy 0.0 111 | train/01_062_0.npy 0.0 112 | train/01_062_1.npy 0.0 113 | train/01_062_2.npy 0.0 114 | train/01_062_3.npy 0.0 115 | train/01_062_4.npy 0.0 116 | train/01_062_5.npy 0.0 117 | train/01_062_6.npy 0.0 118 | train/01_062_7.npy 0.0 119 | train/01_062_8.npy 0.0 120 | train/01_062_9.npy 0.0 121 | train/03_006_0.npy 0.0 122 | train/03_006_1.npy 0.0 123 | train/03_006_2.npy 0.0 124 | train/03_006_3.npy 0.0 125 | train/03_006_4.npy 0.0 126 | train/03_006_5.npy 0.0 127 | train/03_006_6.npy 0.0 128 | train/03_006_7.npy 0.0 129 | train/03_006_8.npy 0.0 130 | train/03_006_9.npy 0.0 131 | train/02_005_0.npy 0.0 132 | train/02_005_1.npy 0.0 133 | train/02_005_2.npy 0.0 134 | train/02_005_3.npy 0.0 135 | train/02_005_4.npy 0.0 136 | train/02_005_5.npy 0.0 137 | train/02_005_6.npy 0.0 138 | train/02_005_7.npy 0.0 139 | train/02_005_8.npy 0.0 140 | train/02_005_9.npy 0.0 141 | train/03_0036_0.npy 1.0 142 | train/03_0036_1.npy 1.0 143 | train/03_0036_2.npy 1.0 144 | train/03_0036_3.npy 1.0 145 | train/03_0036_4.npy 1.0 146 | train/03_0036_5.npy 1.0 147 | train/03_0036_6.npy 1.0 148 | train/03_0036_7.npy 1.0 149 | train/03_0036_8.npy 1.0 150 | train/03_0036_9.npy 1.0 151 | train/02_004_0.npy 0.0 152 | train/02_004_1.npy 0.0 153 | train/02_004_2.npy 0.0 154 | train/02_004_3.npy 0.0 155 | train/02_004_4.npy 0.0 156 | train/02_004_5.npy 0.0 157 | train/02_004_6.npy 0.0 158 | train/02_004_7.npy 0.0 159 | train/02_004_8.npy 0.0 160 | train/02_004_9.npy 0.0 161 | train/04_0001_0.npy 1.0 162 | train/04_0001_1.npy 1.0 163 | train/04_0001_2.npy 1.0 164 | train/04_0001_3.npy 1.0 165 | train/04_0001_4.npy 1.0 166 | train/04_0001_5.npy 1.0 167 | train/04_0001_6.npy 1.0 168 | train/04_0001_7.npy 1.0 169 | train/04_0001_8.npy 1.0 170 | train/04_0001_9.npy 1.0 171 | train/04_001_0.npy 0.0 172 | train/04_001_1.npy 0.0 173 | train/04_001_2.npy 0.0 174 | train/04_001_3.npy 0.0 175 | train/04_001_4.npy 0.0 176 | train/04_001_5.npy 0.0 177 | train/04_001_6.npy 0.0 178 | train/04_001_7.npy 0.0 179 | train/04_001_8.npy 0.0 180 | train/04_001_9.npy 0.0 181 | train/04_004_0.npy 0.0 182 | train/04_004_1.npy 0.0 183 | train/04_004_2.npy 0.0 184 | train/04_004_3.npy 0.0 185 | train/04_004_4.npy 0.0 186 | train/04_004_5.npy 0.0 187 | train/04_004_6.npy 0.0 188 | train/04_004_7.npy 0.0 189 | train/04_004_8.npy 0.0 190 | train/04_004_9.npy 0.0 191 | train/04_0050_0.npy 1.0 192 | train/04_0050_1.npy 1.0 193 | train/04_0050_2.npy 1.0 194 | train/04_0050_3.npy 1.0 195 | train/04_0050_4.npy 1.0 196 | train/04_0050_5.npy 1.0 197 | train/04_0050_6.npy 1.0 198 | train/04_0050_7.npy 1.0 199 | train/04_0050_8.npy 1.0 200 | train/04_0050_9.npy 1.0 201 | train/07_0005_0.npy 1.0 202 | train/07_0005_1.npy 1.0 203 | train/07_0005_2.npy 1.0 204 | train/07_0005_3.npy 1.0 205 | train/07_0005_4.npy 1.0 206 | train/07_0005_5.npy 1.0 207 | train/07_0005_6.npy 1.0 208 | train/07_0005_7.npy 1.0 209 | train/07_0005_8.npy 1.0 210 | train/07_0005_9.npy 1.0 211 | train/05_066_0.npy 0.0 212 | train/05_066_1.npy 0.0 213 | train/05_066_2.npy 0.0 214 | train/05_066_3.npy 0.0 215 | train/05_066_4.npy 0.0 216 | train/05_066_5.npy 0.0 217 | train/05_066_6.npy 0.0 218 | train/05_066_7.npy 0.0 219 | train/05_066_8.npy 0.0 220 | train/05_066_9.npy 0.0 221 | train/05_0018_0.npy 1.0 222 | train/05_0018_1.npy 1.0 223 | train/05_0018_2.npy 1.0 224 | train/05_0018_3.npy 1.0 225 | train/05_0018_4.npy 1.0 226 | train/05_0018_5.npy 1.0 227 | train/05_0018_6.npy 1.0 228 | train/05_0018_7.npy 1.0 229 | train/05_0018_8.npy 1.0 230 | train/05_0018_9.npy 1.0 231 | train/05_055_0.npy 0.0 232 | train/05_055_1.npy 0.0 233 | train/05_055_2.npy 0.0 234 | train/05_055_3.npy 0.0 235 | train/05_055_4.npy 0.0 236 | train/05_055_5.npy 0.0 237 | train/05_055_6.npy 0.0 238 | train/05_055_7.npy 0.0 239 | train/05_055_8.npy 0.0 240 | train/05_055_9.npy 0.0 241 | train/05_012_0.npy 0.0 242 | train/05_012_1.npy 0.0 243 | train/05_012_2.npy 0.0 244 | train/05_012_3.npy 0.0 245 | train/05_012_4.npy 0.0 246 | train/05_012_5.npy 0.0 247 | train/05_012_6.npy 0.0 248 | train/05_012_7.npy 0.0 249 | train/05_012_8.npy 0.0 250 | train/05_012_9.npy 0.0 251 | train/05_053_0.npy 0.0 252 | train/05_053_1.npy 0.0 253 | train/05_053_2.npy 0.0 254 | train/05_053_3.npy 0.0 255 | train/05_053_4.npy 0.0 256 | train/05_053_5.npy 0.0 257 | train/05_053_6.npy 0.0 258 | train/05_053_7.npy 0.0 259 | train/05_053_8.npy 0.0 260 | train/05_053_9.npy 0.0 261 | train/05_037_0.npy 0.0 262 | train/05_037_1.npy 0.0 263 | train/05_037_2.npy 0.0 264 | train/05_037_3.npy 0.0 265 | train/05_037_4.npy 0.0 266 | train/05_037_5.npy 0.0 267 | train/05_037_6.npy 0.0 268 | train/05_037_7.npy 0.0 269 | train/05_037_8.npy 0.0 270 | train/05_037_9.npy 0.0 271 | train/05_015_0.npy 0.0 272 | train/05_015_1.npy 0.0 273 | train/05_015_2.npy 0.0 274 | train/05_015_3.npy 0.0 275 | train/05_015_4.npy 0.0 276 | train/05_015_5.npy 0.0 277 | train/05_015_6.npy 0.0 278 | train/05_015_7.npy 0.0 279 | train/05_015_8.npy 0.0 280 | train/05_015_9.npy 0.0 281 | train/06_016_0.npy 0.0 282 | train/06_016_1.npy 0.0 283 | train/06_016_2.npy 0.0 284 | train/06_016_3.npy 0.0 285 | train/06_016_4.npy 0.0 286 | train/06_016_5.npy 0.0 287 | train/06_016_6.npy 0.0 288 | train/06_016_7.npy 0.0 289 | train/06_016_8.npy 0.0 290 | train/06_016_9.npy 0.0 291 | train/06_025_0.npy 0.0 292 | train/06_025_1.npy 0.0 293 | train/06_025_2.npy 0.0 294 | train/06_025_3.npy 0.0 295 | train/06_025_4.npy 0.0 296 | train/06_025_5.npy 0.0 297 | train/06_025_6.npy 0.0 298 | train/06_025_7.npy 0.0 299 | train/06_025_8.npy 0.0 300 | train/06_025_9.npy 0.0 301 | train/06_013_0.npy 0.0 302 | train/06_013_1.npy 0.0 303 | train/06_013_2.npy 0.0 304 | train/06_013_3.npy 0.0 305 | train/06_013_4.npy 0.0 306 | train/06_013_5.npy 0.0 307 | train/06_013_6.npy 0.0 308 | train/06_013_7.npy 0.0 309 | train/06_013_8.npy 0.0 310 | train/06_013_9.npy 0.0 311 | train/09_009_0.npy 0.0 312 | train/09_009_1.npy 0.0 313 | train/09_009_2.npy 0.0 314 | train/09_009_3.npy 0.0 315 | train/09_009_4.npy 0.0 316 | train/09_009_5.npy 0.0 317 | train/09_009_6.npy 0.0 318 | train/09_009_7.npy 0.0 319 | train/09_009_8.npy 0.0 320 | train/09_009_9.npy 0.0 321 | train/06_001_0.npy 0.0 322 | train/06_001_1.npy 0.0 323 | train/06_001_2.npy 0.0 324 | train/06_001_3.npy 0.0 325 | train/06_001_4.npy 0.0 326 | train/06_001_5.npy 0.0 327 | train/06_001_6.npy 0.0 328 | train/06_001_7.npy 0.0 329 | train/06_001_8.npy 0.0 330 | train/06_001_9.npy 0.0 331 | train/08_0156_0.npy 1.0 332 | train/08_0156_1.npy 1.0 333 | train/08_0156_2.npy 1.0 334 | train/08_0156_3.npy 1.0 335 | train/08_0156_4.npy 1.0 336 | train/08_0156_5.npy 1.0 337 | train/08_0156_6.npy 1.0 338 | train/08_0156_7.npy 1.0 339 | train/08_0156_8.npy 1.0 340 | train/08_0156_9.npy 1.0 341 | train/08_040_0.npy 0.0 342 | train/08_040_1.npy 0.0 343 | train/08_040_2.npy 0.0 344 | train/08_040_3.npy 0.0 345 | train/08_040_4.npy 0.0 346 | train/08_040_5.npy 0.0 347 | train/08_040_6.npy 0.0 348 | train/08_040_7.npy 0.0 349 | train/08_040_8.npy 0.0 350 | train/08_040_9.npy 0.0 351 | train/08_031_0.npy 0.0 352 | train/08_031_1.npy 0.0 353 | train/08_031_2.npy 0.0 354 | train/08_031_3.npy 0.0 355 | train/08_031_4.npy 0.0 356 | train/08_031_5.npy 0.0 357 | train/08_031_6.npy 0.0 358 | train/08_031_7.npy 0.0 359 | train/08_031_8.npy 0.0 360 | train/08_031_9.npy 0.0 361 | train/08_020_0.npy 0.0 362 | train/08_020_1.npy 0.0 363 | train/08_020_2.npy 0.0 364 | train/08_020_3.npy 0.0 365 | train/08_020_4.npy 0.0 366 | train/08_020_5.npy 0.0 367 | train/08_020_6.npy 0.0 368 | train/08_020_7.npy 0.0 369 | train/08_020_8.npy 0.0 370 | train/08_020_9.npy 0.0 371 | train/08_015_0.npy 0.0 372 | train/08_015_1.npy 0.0 373 | train/08_015_2.npy 0.0 374 | train/08_015_3.npy 0.0 375 | train/08_015_4.npy 0.0 376 | train/08_015_5.npy 0.0 377 | train/08_015_6.npy 0.0 378 | train/08_015_7.npy 0.0 379 | train/08_015_8.npy 0.0 380 | train/08_015_9.npy 0.0 381 | train/08_0178_0.npy 1.0 382 | train/08_0178_1.npy 1.0 383 | train/08_0178_2.npy 1.0 384 | train/08_0178_3.npy 1.0 385 | train/08_0178_4.npy 1.0 386 | train/08_0178_5.npy 1.0 387 | train/08_0178_6.npy 1.0 388 | train/08_0178_7.npy 1.0 389 | train/08_0178_8.npy 1.0 390 | train/08_0178_9.npy 1.0 391 | train/11_009_0.npy 0.0 392 | train/11_009_1.npy 0.0 393 | train/11_009_2.npy 0.0 394 | train/11_009_3.npy 0.0 395 | train/11_009_4.npy 0.0 396 | train/11_009_5.npy 0.0 397 | train/11_009_6.npy 0.0 398 | train/11_009_7.npy 0.0 399 | train/11_009_8.npy 0.0 400 | train/11_009_9.npy 0.0 401 | train/10_0037_0.npy 1.0 402 | train/10_0037_1.npy 1.0 403 | train/10_0037_2.npy 1.0 404 | train/10_0037_3.npy 1.0 405 | train/10_0037_4.npy 1.0 406 | train/10_0037_5.npy 1.0 407 | train/10_0037_6.npy 1.0 408 | train/10_0037_7.npy 1.0 409 | train/10_0037_8.npy 1.0 410 | train/10_0037_9.npy 1.0 411 | train/10_003_0.npy 0.0 412 | train/10_003_1.npy 0.0 413 | train/10_003_2.npy 0.0 414 | train/10_003_3.npy 0.0 415 | train/10_003_4.npy 0.0 416 | train/10_003_5.npy 0.0 417 | train/10_003_6.npy 0.0 418 | train/10_003_7.npy 0.0 419 | train/10_003_8.npy 0.0 420 | train/10_003_9.npy 0.0 421 | train/12_006_0.npy 0.0 422 | train/12_006_1.npy 0.0 423 | train/12_006_2.npy 0.0 424 | train/12_006_3.npy 0.0 425 | train/12_006_4.npy 0.0 426 | train/12_006_5.npy 0.0 427 | train/12_006_6.npy 0.0 428 | train/12_006_7.npy 0.0 429 | train/12_006_8.npy 0.0 430 | train/12_006_9.npy 0.0 431 | train/12_005_0.npy 0.0 432 | train/12_005_1.npy 0.0 433 | train/12_005_2.npy 0.0 434 | train/12_005_3.npy 0.0 435 | train/12_005_4.npy 0.0 436 | train/12_005_5.npy 0.0 437 | train/12_005_6.npy 0.0 438 | train/12_005_7.npy 0.0 439 | train/12_005_8.npy 0.0 440 | train/12_005_9.npy 0.0 441 | train/12_0149_0.npy 1.0 442 | train/12_0149_1.npy 1.0 443 | train/12_0149_2.npy 1.0 444 | train/12_0149_3.npy 1.0 445 | train/12_0149_4.npy 1.0 446 | train/12_0149_5.npy 1.0 447 | train/12_0149_6.npy 1.0 448 | train/12_0149_7.npy 1.0 449 | train/12_0149_8.npy 1.0 450 | train/12_0149_9.npy 1.0 -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/test_clients/client1.list: -------------------------------------------------------------------------------- 1 | test/01_0015_0.npy 2 | test/01_0015_1.npy 3 | test/01_0015_2.npy 4 | test/01_0015_3.npy 5 | test/01_0015_4.npy 6 | test/01_0015_5.npy 7 | test/01_0015_6.npy 8 | test/01_0015_7.npy 9 | test/01_0015_8.npy 10 | test/01_0015_9.npy 11 | test/01_0025_0.npy 12 | test/01_0025_1.npy 13 | test/01_0025_2.npy 14 | test/01_0025_3.npy 15 | test/01_0025_4.npy 16 | test/01_0025_5.npy 17 | test/01_0025_6.npy 18 | test/01_0025_7.npy 19 | test/01_0025_8.npy 20 | test/01_0025_9.npy 21 | test/01_0027_0.npy 22 | test/01_0027_1.npy 23 | test/01_0027_2.npy 24 | test/01_0027_3.npy 25 | test/01_0027_4.npy 26 | test/01_0027_5.npy 27 | test/01_0027_6.npy 28 | test/01_0027_7.npy 29 | test/01_0027_8.npy 30 | test/01_0027_9.npy 31 | test/01_0028_0.npy 32 | test/01_0028_1.npy 33 | test/01_0028_2.npy 34 | test/01_0028_3.npy 35 | test/01_0028_4.npy 36 | test/01_0028_5.npy 37 | test/01_0028_6.npy 38 | test/01_0028_7.npy 39 | test/01_0028_8.npy 40 | test/01_0028_9.npy 41 | test/01_0051_0.npy 42 | test/01_0051_1.npy 43 | test/01_0051_2.npy 44 | test/01_0051_3.npy 45 | test/01_0051_4.npy 46 | test/01_0051_5.npy 47 | test/01_0051_6.npy 48 | test/01_0051_7.npy 49 | test/01_0051_8.npy 50 | test/01_0051_9.npy 51 | test/01_0052_0.npy 52 | test/01_0052_1.npy 53 | test/01_0052_2.npy 54 | test/01_0052_3.npy 55 | test/01_0052_4.npy 56 | test/01_0052_5.npy 57 | test/01_0052_6.npy 58 | test/01_0052_7.npy 59 | test/01_0052_8.npy 60 | test/01_0052_9.npy 61 | test/01_0053_0.npy 62 | test/01_0053_1.npy 63 | test/01_0053_2.npy 64 | test/01_0053_3.npy 65 | test/01_0053_4.npy 66 | test/01_0053_5.npy 67 | test/01_0053_6.npy 68 | test/01_0053_7.npy 69 | test/01_0053_8.npy 70 | test/01_0053_9.npy 71 | test/01_0055_0.npy 72 | test/01_0055_1.npy 73 | test/01_0055_2.npy 74 | test/01_0055_3.npy 75 | test/01_0055_4.npy 76 | test/01_0055_5.npy 77 | test/01_0055_6.npy 78 | test/01_0055_7.npy 79 | test/01_0055_8.npy 80 | test/01_0055_9.npy 81 | test/01_0056_0.npy 82 | test/01_0056_1.npy 83 | test/01_0056_2.npy 84 | test/01_0056_3.npy 85 | test/01_0056_4.npy 86 | test/01_0056_5.npy 87 | test/01_0056_6.npy 88 | test/01_0056_7.npy 89 | test/01_0056_8.npy 90 | test/01_0056_9.npy 91 | test/01_0064_0.npy 92 | test/01_0064_1.npy 93 | test/01_0064_2.npy 94 | test/01_0064_3.npy 95 | test/01_0064_4.npy 96 | test/01_0064_5.npy 97 | test/01_0064_6.npy 98 | test/01_0064_7.npy 99 | test/01_0064_8.npy 100 | test/01_0064_9.npy 101 | test/01_0135_0.npy 102 | test/01_0135_1.npy 103 | test/01_0135_2.npy 104 | test/01_0135_3.npy 105 | test/01_0135_4.npy 106 | test/01_0135_5.npy 107 | test/01_0135_6.npy 108 | test/01_0135_7.npy 109 | test/01_0135_8.npy 110 | test/01_0135_9.npy 111 | test/01_0136_0.npy 112 | test/01_0136_1.npy 113 | test/01_0136_2.npy 114 | test/01_0136_3.npy 115 | test/01_0136_4.npy 116 | test/01_0136_5.npy 117 | test/01_0136_6.npy 118 | test/01_0136_7.npy 119 | test/01_0136_8.npy 120 | test/01_0136_9.npy 121 | test/01_0139_0.npy 122 | test/01_0139_1.npy 123 | test/01_0139_2.npy 124 | test/01_0139_3.npy 125 | test/01_0139_4.npy 126 | test/01_0139_5.npy 127 | test/01_0139_6.npy 128 | test/01_0139_7.npy 129 | test/01_0139_8.npy 130 | test/01_0139_9.npy 131 | test/01_0141_0.npy 132 | test/01_0141_1.npy 133 | test/01_0141_2.npy 134 | test/01_0141_3.npy 135 | test/01_0141_4.npy 136 | test/01_0141_5.npy 137 | test/01_0141_6.npy 138 | test/01_0141_7.npy 139 | test/01_0141_8.npy 140 | test/01_0141_9.npy 141 | test/01_0162_0.npy 142 | test/01_0162_1.npy 143 | test/01_0162_2.npy 144 | test/01_0162_3.npy 145 | test/01_0162_4.npy 146 | test/01_0162_5.npy 147 | test/01_0162_6.npy 148 | test/01_0162_7.npy 149 | test/01_0162_8.npy 150 | test/01_0162_9.npy 151 | test/01_0177_0.npy 152 | test/01_0177_1.npy 153 | test/01_0177_2.npy 154 | test/01_0177_3.npy 155 | test/01_0177_4.npy 156 | test/01_0177_5.npy 157 | test/01_0177_6.npy 158 | test/01_0177_7.npy 159 | test/01_0177_8.npy 160 | test/01_0177_9.npy 161 | test/01_001_0.npy 162 | test/01_001_1.npy 163 | test/01_001_2.npy 164 | test/01_001_3.npy 165 | test/01_001_4.npy 166 | test/01_001_5.npy 167 | test/01_001_6.npy 168 | test/01_001_7.npy 169 | test/01_001_8.npy 170 | test/01_001_9.npy 171 | test/01_004_0.npy 172 | test/01_004_1.npy 173 | test/01_004_2.npy 174 | test/01_004_3.npy 175 | test/01_004_4.npy 176 | test/01_004_5.npy 177 | test/01_004_6.npy 178 | test/01_004_7.npy 179 | test/01_004_8.npy 180 | test/01_004_9.npy 181 | test/01_006_0.npy 182 | test/01_006_1.npy 183 | test/01_006_2.npy 184 | test/01_006_3.npy 185 | test/01_006_4.npy 186 | test/01_006_5.npy 187 | test/01_006_6.npy 188 | test/01_006_7.npy 189 | test/01_006_8.npy 190 | test/01_006_9.npy 191 | test/01_008_0.npy 192 | test/01_008_1.npy 193 | test/01_008_2.npy 194 | test/01_008_3.npy 195 | test/01_008_4.npy 196 | test/01_008_5.npy 197 | test/01_008_6.npy 198 | test/01_008_7.npy 199 | test/01_008_8.npy 200 | test/01_008_9.npy 201 | test/01_012_0.npy 202 | test/01_012_1.npy 203 | test/01_012_2.npy 204 | test/01_012_3.npy 205 | test/01_012_4.npy 206 | test/01_012_5.npy 207 | test/01_012_6.npy 208 | test/01_012_7.npy 209 | test/01_012_8.npy 210 | test/01_012_9.npy 211 | test/01_018_0.npy 212 | test/01_018_1.npy 213 | test/01_018_2.npy 214 | test/01_018_3.npy 215 | test/01_018_4.npy 216 | test/01_018_5.npy 217 | test/01_018_6.npy 218 | test/01_018_7.npy 219 | test/01_018_8.npy 220 | test/01_018_9.npy 221 | test/01_021_0.npy 222 | test/01_021_1.npy 223 | test/01_021_2.npy 224 | test/01_021_3.npy 225 | test/01_021_4.npy 226 | test/01_021_5.npy 227 | test/01_021_6.npy 228 | test/01_021_7.npy 229 | test/01_021_8.npy 230 | test/01_021_9.npy 231 | test/01_026_0.npy 232 | test/01_026_1.npy 233 | test/01_026_2.npy 234 | test/01_026_3.npy 235 | test/01_026_4.npy 236 | test/01_026_5.npy 237 | test/01_026_6.npy 238 | test/01_026_7.npy 239 | test/01_026_8.npy 240 | test/01_026_9.npy 241 | test/01_027_0.npy 242 | test/01_027_1.npy 243 | test/01_027_2.npy 244 | test/01_027_3.npy 245 | test/01_027_4.npy 246 | test/01_027_5.npy 247 | test/01_027_6.npy 248 | test/01_027_7.npy 249 | test/01_027_8.npy 250 | test/01_027_9.npy 251 | test/01_028_0.npy 252 | test/01_028_1.npy 253 | test/01_028_2.npy 254 | test/01_028_3.npy 255 | test/01_028_4.npy 256 | test/01_028_5.npy 257 | test/01_028_6.npy 258 | test/01_028_7.npy 259 | test/01_028_8.npy 260 | test/01_028_9.npy 261 | test/01_030_0.npy 262 | test/01_030_1.npy 263 | test/01_030_2.npy 264 | test/01_030_3.npy 265 | test/01_030_4.npy 266 | test/01_030_5.npy 267 | test/01_030_6.npy 268 | test/01_030_7.npy 269 | test/01_030_8.npy 270 | test/01_030_9.npy 271 | test/01_032_0.npy 272 | test/01_032_1.npy 273 | test/01_032_2.npy 274 | test/01_032_3.npy 275 | test/01_032_4.npy 276 | test/01_032_5.npy 277 | test/01_032_6.npy 278 | test/01_032_7.npy 279 | test/01_032_8.npy 280 | test/01_032_9.npy 281 | test/01_034_0.npy 282 | test/01_034_1.npy 283 | test/01_034_2.npy 284 | test/01_034_3.npy 285 | test/01_034_4.npy 286 | test/01_034_5.npy 287 | test/01_034_6.npy 288 | test/01_034_7.npy 289 | test/01_034_8.npy 290 | test/01_034_9.npy 291 | test/01_038_0.npy 292 | test/01_038_1.npy 293 | test/01_038_2.npy 294 | test/01_038_3.npy 295 | test/01_038_4.npy 296 | test/01_038_5.npy 297 | test/01_038_6.npy 298 | test/01_038_7.npy 299 | test/01_038_8.npy 300 | test/01_038_9.npy 301 | test/01_039_0.npy 302 | test/01_039_1.npy 303 | test/01_039_2.npy 304 | test/01_039_3.npy 305 | test/01_039_4.npy 306 | test/01_039_5.npy 307 | test/01_039_6.npy 308 | test/01_039_7.npy 309 | test/01_039_8.npy 310 | test/01_039_9.npy 311 | test/01_040_0.npy 312 | test/01_040_1.npy 313 | test/01_040_2.npy 314 | test/01_040_3.npy 315 | test/01_040_4.npy 316 | test/01_040_5.npy 317 | test/01_040_6.npy 318 | test/01_040_7.npy 319 | test/01_040_8.npy 320 | test/01_040_9.npy 321 | test/01_043_0.npy 322 | test/01_043_1.npy 323 | test/01_043_2.npy 324 | test/01_043_3.npy 325 | test/01_043_4.npy 326 | test/01_043_5.npy 327 | test/01_043_6.npy 328 | test/01_043_7.npy 329 | test/01_043_8.npy 330 | test/01_043_9.npy 331 | test/01_045_0.npy 332 | test/01_045_1.npy 333 | test/01_045_2.npy 334 | test/01_045_3.npy 335 | test/01_045_4.npy 336 | test/01_045_5.npy 337 | test/01_045_6.npy 338 | test/01_045_7.npy 339 | test/01_045_8.npy 340 | test/01_045_9.npy 341 | test/01_047_0.npy 342 | test/01_047_1.npy 343 | test/01_047_2.npy 344 | test/01_047_3.npy 345 | test/01_047_4.npy 346 | test/01_047_5.npy 347 | test/01_047_6.npy 348 | test/01_047_7.npy 349 | test/01_047_8.npy 350 | test/01_047_9.npy 351 | test/01_050_0.npy 352 | test/01_050_1.npy 353 | test/01_050_2.npy 354 | test/01_050_3.npy 355 | test/01_050_4.npy 356 | test/01_050_5.npy 357 | test/01_050_6.npy 358 | test/01_050_7.npy 359 | test/01_050_8.npy 360 | test/01_050_9.npy 361 | test/01_052_0.npy 362 | test/01_052_1.npy 363 | test/01_052_2.npy 364 | test/01_052_3.npy 365 | test/01_052_4.npy 366 | test/01_052_5.npy 367 | test/01_052_6.npy 368 | test/01_052_7.npy 369 | test/01_052_8.npy 370 | test/01_052_9.npy 371 | test/01_053_0.npy 372 | test/01_053_1.npy 373 | test/01_053_2.npy 374 | test/01_053_3.npy 375 | test/01_053_4.npy 376 | test/01_053_5.npy 377 | test/01_053_6.npy 378 | test/01_053_7.npy 379 | test/01_053_8.npy 380 | test/01_053_9.npy 381 | test/01_055_0.npy 382 | test/01_055_1.npy 383 | test/01_055_2.npy 384 | test/01_055_3.npy 385 | test/01_055_4.npy 386 | test/01_055_5.npy 387 | test/01_055_6.npy 388 | test/01_055_7.npy 389 | test/01_055_8.npy 390 | test/01_055_9.npy 391 | test/01_056_0.npy 392 | test/01_056_1.npy 393 | test/01_056_2.npy 394 | test/01_056_3.npy 395 | test/01_056_4.npy 396 | test/01_056_5.npy 397 | test/01_056_6.npy 398 | test/01_056_7.npy 399 | test/01_056_8.npy 400 | test/01_056_9.npy 401 | test/01_057_0.npy 402 | test/01_057_1.npy 403 | test/01_057_2.npy 404 | test/01_057_3.npy 405 | test/01_057_4.npy 406 | test/01_057_5.npy 407 | test/01_057_6.npy 408 | test/01_057_7.npy 409 | test/01_057_8.npy 410 | test/01_057_9.npy 411 | test/01_058_0.npy 412 | test/01_058_1.npy 413 | test/01_058_2.npy 414 | test/01_058_3.npy 415 | test/01_058_4.npy 416 | test/01_058_5.npy 417 | test/01_058_6.npy 418 | test/01_058_7.npy 419 | test/01_058_8.npy 420 | test/01_058_9.npy 421 | test/01_059_0.npy 422 | test/01_059_1.npy 423 | test/01_059_2.npy 424 | test/01_059_3.npy 425 | test/01_059_4.npy 426 | test/01_059_5.npy 427 | test/01_059_6.npy 428 | test/01_059_7.npy 429 | test/01_059_8.npy 430 | test/01_059_9.npy 431 | test/01_061_0.npy 432 | test/01_061_1.npy 433 | test/01_061_2.npy 434 | test/01_061_3.npy 435 | test/01_061_4.npy 436 | test/01_061_5.npy 437 | test/01_061_6.npy 438 | test/01_061_7.npy 439 | test/01_061_8.npy 440 | test/01_061_9.npy 441 | test/01_064_0.npy 442 | test/01_064_1.npy 443 | test/01_064_2.npy 444 | test/01_064_3.npy 445 | test/01_064_4.npy 446 | test/01_064_5.npy 447 | test/01_064_6.npy 448 | test/01_064_7.npy 449 | test/01_064_8.npy 450 | test/01_064_9.npy 451 | test/01_065_0.npy 452 | test/01_065_1.npy 453 | test/01_065_2.npy 454 | test/01_065_3.npy 455 | test/01_065_4.npy 456 | test/01_065_5.npy 457 | test/01_065_6.npy 458 | test/01_065_7.npy 459 | test/01_065_8.npy 460 | test/01_065_9.npy 461 | test/01_071_0.npy 462 | test/01_071_1.npy 463 | test/01_071_2.npy 464 | test/01_071_3.npy 465 | test/01_071_4.npy 466 | test/01_071_5.npy 467 | test/01_071_6.npy 468 | test/01_071_7.npy 469 | test/01_071_8.npy 470 | test/01_071_9.npy 471 | test/01_072_0.npy 472 | test/01_072_1.npy 473 | test/01_072_2.npy 474 | test/01_072_3.npy 475 | test/01_072_4.npy 476 | test/01_072_5.npy 477 | test/01_072_6.npy 478 | test/01_072_7.npy 479 | test/01_072_8.npy 480 | test/01_072_9.npy 481 | test/01_074_0.npy 482 | test/01_074_1.npy 483 | test/01_074_2.npy 484 | test/01_074_3.npy 485 | test/01_074_4.npy 486 | test/01_074_5.npy 487 | test/01_074_6.npy 488 | test/01_074_7.npy 489 | test/01_074_8.npy 490 | test/01_074_9.npy 491 | test/01_075_0.npy 492 | test/01_075_1.npy 493 | test/01_075_2.npy 494 | test/01_075_3.npy 495 | test/01_075_4.npy 496 | test/01_075_5.npy 497 | test/01_075_6.npy 498 | test/01_075_7.npy 499 | test/01_075_8.npy 500 | test/01_075_9.npy 501 | test/01_076_0.npy 502 | test/01_076_1.npy 503 | test/01_076_2.npy 504 | test/01_076_3.npy 505 | test/01_076_4.npy 506 | test/01_076_5.npy 507 | test/01_076_6.npy 508 | test/01_076_7.npy 509 | test/01_076_8.npy 510 | test/01_076_9.npy 511 | test/01_077_0.npy 512 | test/01_077_1.npy 513 | test/01_077_2.npy 514 | test/01_077_3.npy 515 | test/01_077_4.npy 516 | test/01_077_5.npy 517 | test/01_077_6.npy 518 | test/01_077_7.npy 519 | test/01_077_8.npy 520 | test/01_077_9.npy 521 | test/01_078_0.npy 522 | test/01_078_1.npy 523 | test/01_078_2.npy 524 | test/01_078_3.npy 525 | test/01_078_4.npy 526 | test/01_078_5.npy 527 | test/01_078_6.npy 528 | test/01_078_7.npy 529 | test/01_078_8.npy 530 | test/01_078_9.npy 531 | test/01_079_0.npy 532 | test/01_079_1.npy 533 | test/01_079_2.npy 534 | test/01_079_3.npy 535 | test/01_079_4.npy 536 | test/01_079_5.npy 537 | test/01_079_6.npy 538 | test/01_079_7.npy 539 | test/01_079_8.npy 540 | test/01_079_9.npy 541 | test/01_080_0.npy 542 | test/01_080_1.npy 543 | test/01_080_2.npy 544 | test/01_080_3.npy 545 | test/01_080_4.npy 546 | test/01_080_5.npy 547 | test/01_080_6.npy 548 | test/01_080_7.npy 549 | test/01_080_8.npy 550 | test/01_080_9.npy 551 | test/01_081_0.npy 552 | test/01_081_1.npy 553 | test/01_081_2.npy 554 | test/01_081_3.npy 555 | test/01_081_4.npy 556 | test/01_081_5.npy 557 | test/01_081_6.npy 558 | test/01_081_7.npy 559 | test/01_081_8.npy 560 | test/01_081_9.npy 561 | test/01_082_0.npy 562 | test/01_082_1.npy 563 | test/01_082_2.npy 564 | test/01_082_3.npy 565 | test/01_082_4.npy 566 | test/01_082_5.npy 567 | test/01_082_6.npy 568 | test/01_082_7.npy 569 | test/01_082_8.npy 570 | test/01_082_9.npy -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/ST/new_split/client1.list: -------------------------------------------------------------------------------- 1 | train/01_0014_0.npy 1.0 2 | train/01_0014_1.npy 1.0 3 | train/01_0014_2.npy 1.0 4 | train/01_0014_3.npy 1.0 5 | train/01_0014_4.npy 1.0 6 | train/01_0014_5.npy 1.0 7 | train/01_0014_6.npy 1.0 8 | train/01_0014_7.npy 1.0 9 | train/01_0014_8.npy 1.0 10 | train/01_0014_9.npy 1.0 11 | train/01_002_0.npy 0.0 12 | train/01_002_1.npy 0.0 13 | train/01_002_2.npy 0.0 14 | train/01_002_3.npy 0.0 15 | train/01_002_4.npy 0.0 16 | train/01_002_5.npy 0.0 17 | train/01_002_6.npy 0.0 18 | train/01_002_7.npy 0.0 19 | train/01_002_8.npy 0.0 20 | train/01_002_9.npy 0.0 21 | train/01_0026_0.npy 1.0 22 | train/01_0026_1.npy 1.0 23 | train/01_0026_2.npy 1.0 24 | train/01_0026_3.npy 1.0 25 | train/01_0026_4.npy 1.0 26 | train/01_0026_5.npy 1.0 27 | train/01_0026_6.npy 1.0 28 | train/01_0026_7.npy 1.0 29 | train/01_0026_8.npy 1.0 30 | train/01_0026_9.npy 1.0 31 | train/01_0029_0.npy 1.0 32 | train/01_0029_1.npy 1.0 33 | train/01_0029_2.npy 1.0 34 | train/01_0029_3.npy 1.0 35 | train/01_0029_4.npy 1.0 36 | train/01_0029_5.npy 1.0 37 | train/01_0029_6.npy 1.0 38 | train/01_0029_7.npy 1.0 39 | train/01_0029_8.npy 1.0 40 | train/01_0029_9.npy 1.0 41 | train/01_003_0.npy 0.0 42 | train/01_003_1.npy 0.0 43 | train/01_003_2.npy 0.0 44 | train/01_003_3.npy 0.0 45 | train/01_003_4.npy 0.0 46 | train/01_003_5.npy 0.0 47 | train/01_003_6.npy 0.0 48 | train/01_003_7.npy 0.0 49 | train/01_003_8.npy 0.0 50 | train/01_003_9.npy 0.0 51 | train/01_0063_0.npy 1.0 52 | train/01_0063_1.npy 1.0 53 | train/01_0063_2.npy 1.0 54 | train/01_0063_3.npy 1.0 55 | train/01_0063_4.npy 1.0 56 | train/01_0063_5.npy 1.0 57 | train/01_0063_6.npy 1.0 58 | train/01_0063_7.npy 1.0 59 | train/01_0063_8.npy 1.0 60 | train/01_0063_9.npy 1.0 61 | train/01_007_0.npy 0.0 62 | train/01_007_1.npy 0.0 63 | train/01_007_2.npy 0.0 64 | train/01_007_3.npy 0.0 65 | train/01_007_4.npy 0.0 66 | train/01_007_5.npy 0.0 67 | train/01_007_6.npy 0.0 68 | train/01_007_7.npy 0.0 69 | train/01_007_8.npy 0.0 70 | train/01_007_9.npy 0.0 71 | train/01_0073_0.npy 1.0 72 | train/01_0073_1.npy 1.0 73 | train/01_0073_2.npy 1.0 74 | train/01_0073_3.npy 1.0 75 | train/01_0073_4.npy 1.0 76 | train/01_0073_5.npy 1.0 77 | train/01_0073_6.npy 1.0 78 | train/01_0073_7.npy 1.0 79 | train/01_0073_8.npy 1.0 80 | train/01_0073_9.npy 1.0 81 | train/01_0076_0.npy 1.0 82 | train/01_0076_1.npy 1.0 83 | train/01_0076_2.npy 1.0 84 | train/01_0076_3.npy 1.0 85 | train/01_0076_4.npy 1.0 86 | train/01_0076_5.npy 1.0 87 | train/01_0076_6.npy 1.0 88 | train/01_0076_7.npy 1.0 89 | train/01_0076_8.npy 1.0 90 | train/01_0076_9.npy 1.0 91 | train/01_009_0.npy 0.0 92 | train/01_009_1.npy 0.0 93 | train/01_009_2.npy 0.0 94 | train/01_009_3.npy 0.0 95 | train/01_009_4.npy 0.0 96 | train/01_009_5.npy 0.0 97 | train/01_009_6.npy 0.0 98 | train/01_009_7.npy 0.0 99 | train/01_009_8.npy 0.0 100 | train/01_009_9.npy 0.0 101 | train/01_0129_0.npy 1.0 102 | train/01_0129_1.npy 1.0 103 | train/01_0129_2.npy 1.0 104 | train/01_0129_3.npy 1.0 105 | train/01_0129_4.npy 1.0 106 | train/01_0129_5.npy 1.0 107 | train/01_0129_6.npy 1.0 108 | train/01_0129_7.npy 1.0 109 | train/01_0129_8.npy 1.0 110 | train/01_0129_9.npy 1.0 111 | train/01_0130_0.npy 1.0 112 | train/01_0130_1.npy 1.0 113 | train/01_0130_2.npy 1.0 114 | train/01_0130_3.npy 1.0 115 | train/01_0130_4.npy 1.0 116 | train/01_0130_5.npy 1.0 117 | train/01_0130_6.npy 1.0 118 | train/01_0130_7.npy 1.0 119 | train/01_0130_8.npy 1.0 120 | train/01_0130_9.npy 1.0 121 | train/01_0131_0.npy 1.0 122 | train/01_0131_1.npy 1.0 123 | train/01_0131_2.npy 1.0 124 | train/01_0131_3.npy 1.0 125 | train/01_0131_4.npy 1.0 126 | train/01_0131_5.npy 1.0 127 | train/01_0131_6.npy 1.0 128 | train/01_0131_7.npy 1.0 129 | train/01_0131_8.npy 1.0 130 | train/01_0131_9.npy 1.0 131 | train/01_0132_0.npy 1.0 132 | train/01_0132_1.npy 1.0 133 | train/01_0132_2.npy 1.0 134 | train/01_0132_3.npy 1.0 135 | train/01_0132_4.npy 1.0 136 | train/01_0132_5.npy 1.0 137 | train/01_0132_6.npy 1.0 138 | train/01_0132_7.npy 1.0 139 | train/01_0132_8.npy 1.0 140 | train/01_0132_9.npy 1.0 141 | train/01_0133_0.npy 1.0 142 | train/01_0133_1.npy 1.0 143 | train/01_0133_2.npy 1.0 144 | train/01_0133_3.npy 1.0 145 | train/01_0133_4.npy 1.0 146 | train/01_0133_5.npy 1.0 147 | train/01_0133_6.npy 1.0 148 | train/01_0133_7.npy 1.0 149 | train/01_0133_8.npy 1.0 150 | train/01_0133_9.npy 1.0 151 | train/01_0134_0.npy 1.0 152 | train/01_0134_1.npy 1.0 153 | train/01_0134_2.npy 1.0 154 | train/01_0134_3.npy 1.0 155 | train/01_0134_4.npy 1.0 156 | train/01_0134_5.npy 1.0 157 | train/01_0134_6.npy 1.0 158 | train/01_0134_7.npy 1.0 159 | train/01_0134_8.npy 1.0 160 | train/01_0134_9.npy 1.0 161 | train/01_0138_0.npy 1.0 162 | train/01_0138_1.npy 1.0 163 | train/01_0138_2.npy 1.0 164 | train/01_0138_3.npy 1.0 165 | train/01_0138_4.npy 1.0 166 | train/01_0138_5.npy 1.0 167 | train/01_0138_6.npy 1.0 168 | train/01_0138_7.npy 1.0 169 | train/01_0138_8.npy 1.0 170 | train/01_0138_9.npy 1.0 171 | train/01_014_0.npy 0.0 172 | train/01_014_1.npy 0.0 173 | train/01_014_2.npy 0.0 174 | train/01_014_3.npy 0.0 175 | train/01_014_4.npy 0.0 176 | train/01_014_5.npy 0.0 177 | train/01_014_6.npy 0.0 178 | train/01_014_7.npy 0.0 179 | train/01_014_8.npy 0.0 180 | train/01_014_9.npy 0.0 181 | train/01_0140_0.npy 1.0 182 | train/01_0140_1.npy 1.0 183 | train/01_0140_2.npy 1.0 184 | train/01_0140_3.npy 1.0 185 | train/01_0140_4.npy 1.0 186 | train/01_0140_5.npy 1.0 187 | train/01_0140_6.npy 1.0 188 | train/01_0140_7.npy 1.0 189 | train/01_0140_8.npy 1.0 190 | train/01_0140_9.npy 1.0 191 | train/01_015_0.npy 0.0 192 | train/01_015_1.npy 0.0 193 | train/01_015_2.npy 0.0 194 | train/01_015_3.npy 0.0 195 | train/01_015_4.npy 0.0 196 | train/01_015_5.npy 0.0 197 | train/01_015_6.npy 0.0 198 | train/01_015_7.npy 0.0 199 | train/01_015_8.npy 0.0 200 | train/01_015_9.npy 0.0 201 | train/01_016_0.npy 0.0 202 | train/01_016_1.npy 0.0 203 | train/01_016_2.npy 0.0 204 | train/01_016_3.npy 0.0 205 | train/01_016_4.npy 0.0 206 | train/01_016_5.npy 0.0 207 | train/01_016_6.npy 0.0 208 | train/01_016_7.npy 0.0 209 | train/01_016_8.npy 0.0 210 | train/01_016_9.npy 0.0 211 | train/01_0163_0.npy 1.0 212 | train/01_0163_1.npy 1.0 213 | train/01_0163_2.npy 1.0 214 | train/01_0163_3.npy 1.0 215 | train/01_0163_4.npy 1.0 216 | train/01_0163_5.npy 1.0 217 | train/01_0163_6.npy 1.0 218 | train/01_0163_7.npy 1.0 219 | train/01_0163_8.npy 1.0 220 | train/01_0163_9.npy 1.0 221 | train/01_017_0.npy 0.0 222 | train/01_017_1.npy 0.0 223 | train/01_017_2.npy 0.0 224 | train/01_017_3.npy 0.0 225 | train/01_017_4.npy 0.0 226 | train/01_017_5.npy 0.0 227 | train/01_017_6.npy 0.0 228 | train/01_017_7.npy 0.0 229 | train/01_017_8.npy 0.0 230 | train/01_017_9.npy 0.0 231 | train/01_019_0.npy 0.0 232 | train/01_019_1.npy 0.0 233 | train/01_019_2.npy 0.0 234 | train/01_019_3.npy 0.0 235 | train/01_019_4.npy 0.0 236 | train/01_019_5.npy 0.0 237 | train/01_019_6.npy 0.0 238 | train/01_019_7.npy 0.0 239 | train/01_019_8.npy 0.0 240 | train/01_019_9.npy 0.0 241 | train/01_020_0.npy 0.0 242 | train/01_020_1.npy 0.0 243 | train/01_020_2.npy 0.0 244 | train/01_020_3.npy 0.0 245 | train/01_020_4.npy 0.0 246 | train/01_020_5.npy 0.0 247 | train/01_020_6.npy 0.0 248 | train/01_020_7.npy 0.0 249 | train/01_020_8.npy 0.0 250 | train/01_020_9.npy 0.0 251 | train/01_022_0.npy 0.0 252 | train/01_022_1.npy 0.0 253 | train/01_022_2.npy 0.0 254 | train/01_022_3.npy 0.0 255 | train/01_022_4.npy 0.0 256 | train/01_022_5.npy 0.0 257 | train/01_022_6.npy 0.0 258 | train/01_022_7.npy 0.0 259 | train/01_022_8.npy 0.0 260 | train/01_022_9.npy 0.0 261 | train/01_023_0.npy 0.0 262 | train/01_023_1.npy 0.0 263 | train/01_023_2.npy 0.0 264 | train/01_023_3.npy 0.0 265 | train/01_023_4.npy 0.0 266 | train/01_023_5.npy 0.0 267 | train/01_023_6.npy 0.0 268 | train/01_023_7.npy 0.0 269 | train/01_023_8.npy 0.0 270 | train/01_023_9.npy 0.0 271 | train/01_024_0.npy 0.0 272 | train/01_024_1.npy 0.0 273 | train/01_024_2.npy 0.0 274 | train/01_024_3.npy 0.0 275 | train/01_024_4.npy 0.0 276 | train/01_024_5.npy 0.0 277 | train/01_024_6.npy 0.0 278 | train/01_024_7.npy 0.0 279 | train/01_024_8.npy 0.0 280 | train/01_024_9.npy 0.0 281 | train/01_025_0.npy 0.0 282 | train/01_025_1.npy 0.0 283 | train/01_025_2.npy 0.0 284 | train/01_025_3.npy 0.0 285 | train/01_025_4.npy 0.0 286 | train/01_025_5.npy 0.0 287 | train/01_025_6.npy 0.0 288 | train/01_025_7.npy 0.0 289 | train/01_025_8.npy 0.0 290 | train/01_025_9.npy 0.0 291 | train/01_029_0.npy 0.0 292 | train/01_029_1.npy 0.0 293 | train/01_029_2.npy 0.0 294 | train/01_029_3.npy 0.0 295 | train/01_029_4.npy 0.0 296 | train/01_029_5.npy 0.0 297 | train/01_029_6.npy 0.0 298 | train/01_029_7.npy 0.0 299 | train/01_029_8.npy 0.0 300 | train/01_029_9.npy 0.0 301 | train/01_031_0.npy 0.0 302 | train/01_031_1.npy 0.0 303 | train/01_031_2.npy 0.0 304 | train/01_031_3.npy 0.0 305 | train/01_031_4.npy 0.0 306 | train/01_031_5.npy 0.0 307 | train/01_031_6.npy 0.0 308 | train/01_031_7.npy 0.0 309 | train/01_031_8.npy 0.0 310 | train/01_031_9.npy 0.0 311 | train/01_033_0.npy 0.0 312 | train/01_033_1.npy 0.0 313 | train/01_033_2.npy 0.0 314 | train/01_033_3.npy 0.0 315 | train/01_033_4.npy 0.0 316 | train/01_033_5.npy 0.0 317 | train/01_033_6.npy 0.0 318 | train/01_033_7.npy 0.0 319 | train/01_033_8.npy 0.0 320 | train/01_033_9.npy 0.0 321 | train/01_035_0.npy 0.0 322 | train/01_035_1.npy 0.0 323 | train/01_035_2.npy 0.0 324 | train/01_035_3.npy 0.0 325 | train/01_035_4.npy 0.0 326 | train/01_035_5.npy 0.0 327 | train/01_035_6.npy 0.0 328 | train/01_035_7.npy 0.0 329 | train/01_035_8.npy 0.0 330 | train/01_035_9.npy 0.0 331 | train/01_037_0.npy 0.0 332 | train/01_037_1.npy 0.0 333 | train/01_037_2.npy 0.0 334 | train/01_037_3.npy 0.0 335 | train/01_037_4.npy 0.0 336 | train/01_037_5.npy 0.0 337 | train/01_037_6.npy 0.0 338 | train/01_037_7.npy 0.0 339 | train/01_037_8.npy 0.0 340 | train/01_037_9.npy 0.0 341 | train/01_041_0.npy 0.0 342 | train/01_041_1.npy 0.0 343 | train/01_041_2.npy 0.0 344 | train/01_041_3.npy 0.0 345 | train/01_041_4.npy 0.0 346 | train/01_041_5.npy 0.0 347 | train/01_041_6.npy 0.0 348 | train/01_041_7.npy 0.0 349 | train/01_041_8.npy 0.0 350 | train/01_041_9.npy 0.0 351 | train/01_044_0.npy 0.0 352 | train/01_044_1.npy 0.0 353 | train/01_044_2.npy 0.0 354 | train/01_044_3.npy 0.0 355 | train/01_044_4.npy 0.0 356 | train/01_044_5.npy 0.0 357 | train/01_044_6.npy 0.0 358 | train/01_044_7.npy 0.0 359 | train/01_044_8.npy 0.0 360 | train/01_044_9.npy 0.0 361 | train/01_046_0.npy 0.0 362 | train/01_046_1.npy 0.0 363 | train/01_046_2.npy 0.0 364 | train/01_046_3.npy 0.0 365 | train/01_046_4.npy 0.0 366 | train/01_046_5.npy 0.0 367 | train/01_046_6.npy 0.0 368 | train/01_046_7.npy 0.0 369 | train/01_046_8.npy 0.0 370 | train/01_046_9.npy 0.0 371 | train/01_048_0.npy 0.0 372 | train/01_048_1.npy 0.0 373 | train/01_048_2.npy 0.0 374 | train/01_048_3.npy 0.0 375 | train/01_048_4.npy 0.0 376 | train/01_048_5.npy 0.0 377 | train/01_048_6.npy 0.0 378 | train/01_048_7.npy 0.0 379 | train/01_048_8.npy 0.0 380 | train/01_048_9.npy 0.0 381 | train/01_051_0.npy 0.0 382 | train/01_051_1.npy 0.0 383 | train/01_051_2.npy 0.0 384 | train/01_051_3.npy 0.0 385 | train/01_051_4.npy 0.0 386 | train/01_051_5.npy 0.0 387 | train/01_051_6.npy 0.0 388 | train/01_051_7.npy 0.0 389 | train/01_051_8.npy 0.0 390 | train/01_051_9.npy 0.0 391 | train/01_054_0.npy 0.0 392 | train/01_054_1.npy 0.0 393 | train/01_054_2.npy 0.0 394 | train/01_054_3.npy 0.0 395 | train/01_054_4.npy 0.0 396 | train/01_054_5.npy 0.0 397 | train/01_054_6.npy 0.0 398 | train/01_054_7.npy 0.0 399 | train/01_054_8.npy 0.0 400 | train/01_054_9.npy 0.0 401 | train/01_060_0.npy 0.0 402 | train/01_060_1.npy 0.0 403 | train/01_060_2.npy 0.0 404 | train/01_060_3.npy 0.0 405 | train/01_060_4.npy 0.0 406 | train/01_060_5.npy 0.0 407 | train/01_060_6.npy 0.0 408 | train/01_060_7.npy 0.0 409 | train/01_060_8.npy 0.0 410 | train/01_060_9.npy 0.0 411 | train/01_063_0.npy 0.0 412 | train/01_063_1.npy 0.0 413 | train/01_063_2.npy 0.0 414 | train/01_063_3.npy 0.0 415 | train/01_063_4.npy 0.0 416 | train/01_063_5.npy 0.0 417 | train/01_063_6.npy 0.0 418 | train/01_063_7.npy 0.0 419 | train/01_063_8.npy 0.0 420 | train/01_063_9.npy 0.0 421 | train/01_066_0.npy 0.0 422 | train/01_066_1.npy 0.0 423 | train/01_066_2.npy 0.0 424 | train/01_066_3.npy 0.0 425 | train/01_066_4.npy 0.0 426 | train/01_066_5.npy 0.0 427 | train/01_066_6.npy 0.0 428 | train/01_066_7.npy 0.0 429 | train/01_066_8.npy 0.0 430 | train/01_066_9.npy 0.0 431 | train/01_067_0.npy 0.0 432 | train/01_067_1.npy 0.0 433 | train/01_067_2.npy 0.0 434 | train/01_067_3.npy 0.0 435 | train/01_067_4.npy 0.0 436 | train/01_067_5.npy 0.0 437 | train/01_067_6.npy 0.0 438 | train/01_067_7.npy 0.0 439 | train/01_067_8.npy 0.0 440 | train/01_067_9.npy 0.0 441 | train/01_068_0.npy 0.0 442 | train/01_068_1.npy 0.0 443 | train/01_068_2.npy 0.0 444 | train/01_068_3.npy 0.0 445 | train/01_068_4.npy 0.0 446 | train/01_068_5.npy 0.0 447 | train/01_068_6.npy 0.0 448 | train/01_068_7.npy 0.0 449 | train/01_068_8.npy 0.0 450 | train/01_068_9.npy 0.0 451 | train/01_070_0.npy 0.0 452 | train/01_070_1.npy 0.0 453 | train/01_070_2.npy 0.0 454 | train/01_070_3.npy 0.0 455 | train/01_070_4.npy 0.0 456 | train/01_070_5.npy 0.0 457 | train/01_070_6.npy 0.0 458 | train/01_070_7.npy 0.0 459 | train/01_070_8.npy 0.0 460 | train/01_070_9.npy 0.0 461 | train/01_073_0.npy 0.0 462 | train/01_073_1.npy 0.0 463 | train/01_073_2.npy 0.0 464 | train/01_073_3.npy 0.0 465 | train/01_073_4.npy 0.0 466 | train/01_073_5.npy 0.0 467 | train/01_073_6.npy 0.0 468 | train/01_073_7.npy 0.0 469 | train/01_073_8.npy 0.0 470 | train/01_073_9.npy 0.0 471 | train/01_083_0.npy 0.0 472 | train/01_083_1.npy 0.0 473 | train/01_083_2.npy 0.0 474 | train/01_083_3.npy 0.0 475 | train/01_083_4.npy 0.0 476 | train/01_083_5.npy 0.0 477 | train/01_083_6.npy 0.0 478 | train/01_083_7.npy 0.0 479 | train/01_083_8.npy 0.0 480 | train/01_083_9.npy 0.0 -------------------------------------------------------------------------------- /Code/main.py: -------------------------------------------------------------------------------- 1 | from torch.utils.data import DataLoader 2 | import torch.optim as optim 3 | import torch 4 | import time 5 | import numpy as np 6 | import random 7 | from configs import build_config 8 | from utils import setup_seed,Visualizer 9 | from log import get_logger 10 | from model import XModel 11 | from dataset import * 12 | from train import train_func,public_train_func 13 | from test import test_func 14 | import argparse 15 | import copy 16 | import os 17 | import numpy as np 18 | import torch 19 | import torch.nn as nn 20 | from tensorboardX import SummaryWriter 21 | 22 | from cluster_utils import * 23 | from fusion import * 24 | from distill import * 25 | 26 | os.environ['CUDA_VISIBLE_DEVICES'] = '5' 27 | os.environ["TOKENIZERS_PARALLELISM"] = "false" 28 | log_dir = '' 29 | writer = SummaryWriter(log_dir) 30 | rounds = 102 31 | num_clients = 7 32 | device = torch.device("cuda") 33 | seed = 42 34 | torch.manual_seed(seed) 35 | 36 | def add_noise_to_state_dict(state_dict, noise_type, sigma, device): 37 | new_state_dict = {} 38 | for name, param in state_dict.items(): 39 | if noise_type == 1: 40 | # 拉普拉斯噪声 41 | noise = torch.tensor(np.random.laplace(0, sigma, param.size()), device=device) 42 | elif noise_type == 2: 43 | # 高斯噪声 44 | noise = torch.randn(param.size(), device=device) * sigma 45 | else: 46 | # 不添加噪声 47 | noise = torch.zeros(param.size(), device=device) 48 | 49 | new_state_dict[name] = param + noise 50 | return new_state_dict 51 | 52 | def average_weights(weight_list): 53 | device = torch.device("cuda" if torch.cuda.is_available() else "cpu") 54 | avg_weights = {} 55 | for key in weight_list[0].keys(): 56 | tensor_list = [client_weights[key].float().to(device) for client_weights in weight_list] 57 | avg_weights[key] = torch.stack(tensor_list).mean(dim=0) 58 | return avg_weights 59 | def load_checkpoint(model, ckpt_path, logger): 60 | if os.path.isfile(ckpt_path): 61 | logger.info('loading pretrained checkpoint from {}.'.format(ckpt_path)) 62 | weight_dict = torch.load(ckpt_path) 63 | model_dict = model.state_dict() 64 | for name, param in weight_dict.items(): 65 | if 'module' in name: 66 | name = '.'.join(name.split('.')[1:]) 67 | if name in model_dict: 68 | if param.size() == model_dict[name].size(): 69 | model_dict[name].copy_(param) 70 | else: 71 | logger.info('{} size mismatch: load {} given {}'.format( 72 | name, param.size(), model_dict[name].size())) 73 | else: 74 | logger.info('{} not found in model dict.'.format(name)) 75 | else: 76 | logger.info('Not found pretrained checkpoint file.') 77 | 78 | def train(model, train_loader, test_loader, client_idx,gt, logger): 79 | if not os.path.exists(cfg.save_dir): 80 | os.makedirs(cfg.save_dir) 81 | 82 | criterion = torch.nn.BCELoss() 83 | criterion2 = torch.nn.KLDivLoss(reduction='batchmean') 84 | dl_loss_function = nn.MSELoss(reduction='mean') 85 | optimizer = optim.Adam(model.parameters(), lr=cfg.lr) 86 | scheduler = optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=60, eta_min=0) 87 | 88 | best_model_wts = copy.deepcopy(model.state_dict()) 89 | best_auc = 0.0 90 | auc_far = 0.0 91 | 92 | st = time.time() 93 | 94 | best_epoch = 0 95 | 96 | for epoch in range(cfg.max_epoch): 97 | print("epoch = {}".format(epoch)) 98 | loss1, loss2 = train_func(train_loader, model, optimizer, criterion, criterion2,dl_loss_function,client_idx, cfg.lamda) 99 | 100 | scheduler.step() 101 | 102 | auc, far = test_func(test_loader, model, gt, cfg.dataset) 103 | if auc >= best_auc: 104 | best_auc = auc 105 | auc_far = far 106 | best_epoch = epoch 107 | best_model_wts = copy.deepcopy(model.state_dict()) 108 | 109 | logger.info('[Epoch:{}/{}]: loss1:{:.4f} loss2:{:.4f} | AUC:{:.4f} FAR:{:.5f}'.format( 110 | epoch + 1, cfg.max_epoch, loss1, loss2, auc, far)) 111 | 112 | print("best_epoch = {}".format(best_epoch)) 113 | print("best_auc = {}".format(best_auc)) 114 | 115 | time_elapsed = time.time() - st 116 | model.load_state_dict(best_model_wts) 117 | logger.info('Training completes in {:.0f}m {:.0f}s | best {}:{:.4f} FAR:{:.5f}\n'. 118 | format(time_elapsed // 60, time_elapsed % 60, cfg.metrics, best_auc, auc_far)) 119 | return best_model_wts,best_auc 120 | 121 | def double_cluster(local_weights): 122 | num_clusters = 3 123 | feature_size = 1207347 # model parameter size 124 | bert_feature_size = 768 # BERT feature dim 125 | bert_features = [] 126 | for i in range(1,8): 127 | path = 'each client object target feature path.' 128 | fea = np.load(path) 129 | bert_features.append(fea) 130 | params_matrix = [] 131 | for w in local_weights: 132 | params_vector = torch.cat([v.flatten() for v in w.values()]) 133 | params_matrix.append(params_vector.detach().cpu().numpy()) 134 | params_matrix = np.array(params_matrix) 135 | # Randomly initialize cluster centers 136 | cluster_centers_w = torch.randn(num_clusters, feature_size) 137 | cluster_centers_o = torch.randn(num_clusters, bert_feature_size) 138 | 139 | alpha = 0.4 140 | beta = 0.6 141 | 142 | # iterative process 143 | num_iterations = 10 144 | for iteration in range(num_iterations): 145 | # Step 1: Assign clusters 146 | r_ik = cluster_assignment_update(params_matrix, bert_features, cluster_centers_w, cluster_centers_o, alpha, beta) 147 | 148 | # Step 2: Update cluster centers 149 | cluster_centers_w, cluster_centers_o = update_cluster_centers(r_ik, params_matrix, bert_features) 150 | 151 | # Step 3: Retrieve the cluster indices for each data point 152 | cluster_indices = torch.argmax(r_ik, dim=1) 153 | 154 | print("iteration process is {}".format(iteration)) 155 | print("Final cluster centers for weights:\n", cluster_centers_w) 156 | print("Final cluster centers for distributions:\n", cluster_centers_o) 157 | print("Cluster indices for each data point:\n", cluster_indices) 158 | print("cluster end !!!") 159 | c = [] 160 | c0 = [] 161 | c1 = [] 162 | c2 = [] 163 | c3 = [] 164 | c4 = [] 165 | classes = cluster_indices 166 | for i in range(0,len(classes)): 167 | if classes[i] == 0: 168 | c0.append(i) 169 | if classes[i] == 1: 170 | c1.append(i) 171 | if classes[i] == 2: 172 | c2.append(i) 173 | if classes[i] == 3: 174 | c3.append(i) 175 | if classes[i] == 4: 176 | c4.append(i) 177 | if len(c0) != 0: 178 | c.append(c0) 179 | if len(c1) != 0: 180 | c.append(c1) 181 | if len(c2) != 0: 182 | c.append(c2) 183 | if len(c3) != 0: 184 | c.append(c3) 185 | if len(c4) != 0: 186 | c.append(c4) 187 | return c 188 | 189 | def main(cfg,global_model): 190 | global_model = global_model 191 | print(global_model) 192 | logger = get_logger(cfg.logs_dir) 193 | setup_seed(cfg.seed) 194 | for round in range(rounds): 195 | local_weights = [] 196 | auc_list = [] 197 | print(f"Round {round + 1} of {rounds}") 198 | flag = False 199 | for client in range(1,num_clients+1): 200 | print("client = {}".format(client)) 201 | if cfg.dataset == 'shanghaiTech': 202 | cfg.train_list = 'each client train list path.' 203 | cfg.test_list = 'each client test list path.' 204 | cfg.gt = 'each client gt list path.' 205 | elif cfg.dataset == 'ucf-crime': 206 | cfg.train_list = 'each client train list path.' 207 | cfg.test_list = 'each client test list path.' 208 | cfg.gt = 'each client gt list path.' 209 | else: 210 | cfg.train_list = 'each client train list path.' 211 | cfg.test_list = 'each client test list path.' 212 | cfg.gt = 'each client gt list path.' 213 | print("client = {},dataset = {},train_list = {}".format(client,cfg.dataset,cfg.train_list)) 214 | logger.info('Config:{}'.format(cfg.__dict__)) 215 | if cfg.dataset == 'ucf-crime': 216 | train_data = UCFDataset(cfg, test_mode=False) 217 | test_data = UCFDataset(cfg, test_mode=True) 218 | flag = True 219 | elif cfg.dataset == 'xd-violence': 220 | train_data = XDataset(cfg, test_mode=False) 221 | test_data = XDataset(cfg, test_mode=True) 222 | elif cfg.dataset == 'shanghaiTech': 223 | train_data = SHDataset(cfg, test_mode=False) 224 | test_data = SHDataset(cfg, test_mode=True) 225 | else: 226 | raise RuntimeError("Do not support this dataset!") 227 | print("train_data_length = {}".format(len(train_data))) 228 | print("test_data_length = {}".format(len(test_data))) 229 | print(cfg.train_list) 230 | train_loader = DataLoader(train_data, batch_size=cfg.train_bs, shuffle=True, 231 | num_workers=cfg.workers, pin_memory=True) 232 | print("train_loader!!!") 233 | test_loader = DataLoader(test_data, batch_size=cfg.test_bs, shuffle=False, 234 | num_workers=cfg.workers, pin_memory=True) 235 | 236 | local_model = XModel(cfg) 237 | """ ================================ Add distillation operation ====================================""" 238 | if round != 0: 239 | local_model.load_state_dict(global_group_weights_dicts[client-1]) 240 | else: 241 | local_model.load_state_dict(global_model.state_dict()) 242 | """ ================================ Not add distillation operation ====================================""" 243 | gt = np.load(cfg.gt) 244 | device = torch.device("cuda") 245 | local_model = local_model.to(device) 246 | 247 | param = sum(p.numel() for p in local_model.parameters()) 248 | logger.info('total params:{:.4f}M'.format(param / (1000 ** 2))) 249 | if args.mode == 'train': 250 | logger.info('Training Mode') 251 | client_idx = client 252 | best_model_wts,best_auc = train(local_model, train_loader, test_loader, client_idx,gt, logger) 253 | noise_type = 2 254 | sigma = 0.01 255 | # print(best_model_wts == local_model.parameters()) 256 | update_bese_model_parameters = add_noise_to_state_dict(best_model_wts,noise_type,sigma,device) 257 | local_weights.append(update_bese_model_parameters) 258 | # auc_list.append(best_auc) 259 | # local_weights.append(best_model_wts) 260 | auc_list.append(best_auc) 261 | total_auc = sum(auc_list) / len(auc_list) 262 | 263 | writer.add_scalar("AUC", total_auc,round) 264 | 265 | global_group_weights_dicts = {} 266 | if round % 5 == 0: 267 | global_weights = average_weights(local_weights) 268 | for i in range(0,13): 269 | global_group_weights_dicts[i] = global_weights 270 | else: 271 | tags = double_cluster(local_weights) 272 | global_group_weights = [] 273 | print(len(tags)) 274 | for tag in tags: 275 | group_k = [] 276 | print(tag) 277 | public_train_list_path = get_public_aug_train_list(tag) 278 | for k in range(0,len(tag)): 279 | group_k.append(local_weights[tag[k]]) 280 | global_weights = average_weights(group_k) 281 | global_model.load_state_dict(global_weights) 282 | global_model_weights_k = distill(global_model,global_weights,cfg,public_train_list_path) 283 | global_group_weights.append(global_model_weights_k) 284 | for k in tag: 285 | global_group_weights_dicts[k] = global_model_weights_k 286 | 287 | def distill(global_model,global_weights,args,public_train_list_path): 288 | 289 | global_model.load_state_dict(global_weights) 290 | visual_model_checkpoints = 'PEL4VAD pretrain visual model checkpoints path.' 291 | args.dataset = 'shanghaiTech' 292 | cfg = build_config(args.dataset) 293 | visual_model = XModel(cfg) 294 | visual_model.to("cuda:0") 295 | checkpoint = torch.load(visual_model_checkpoints) 296 | visual_model.load_state_dict(checkpoint, strict=False) 297 | 298 | cfg.train_list = public_train_list_path 299 | train_data = SHDataset(cfg, test_mode=False) 300 | 301 | train_loader = DataLoader(train_data, batch_size=cfg.train_bs, shuffle=True, 302 | num_workers=cfg.workers, pin_memory=True) 303 | global_model_weights = public_train(global_model,visual_model,train_loader) 304 | 305 | return global_model_weights 306 | 307 | def public_train(global_model,visual_model,train_loader): 308 | global_model.train() 309 | visual_model.eval() 310 | 311 | criterion = torch.nn.BCELoss() 312 | criterion2 = torch.nn.KLDivLoss(reduction='batchmean') 313 | dl_loss_function = nn.MSELoss(reduction='mean') 314 | optimizer = optim.Adam(global_model.parameters(), lr=cfg.lr) 315 | 316 | for epoch in range(1): 317 | print("epoch = {}".format(epoch)) 318 | global_model_weights = public_train_func(train_loader, global_model,visual_model, optimizer, criterion, criterion2,dl_loss_function, cfg.lamda) 319 | return global_model_weights 320 | 321 | if __name__ == '__main__': 322 | parser = argparse.ArgumentParser(description='WeaklySupAnoDet') 323 | parser.add_argument('--dataset', default='ucf', help='anomaly video dataset') 324 | parser.add_argument('--mode', default='train', help='model status: (train or infer)') 325 | args = parser.parse_args() 326 | cfg = build_config(args.dataset) 327 | global_model = XModel(cfg) 328 | main(cfg,global_model) -------------------------------------------------------------------------------- /other/blip2_utils.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "markdown", 5 | "metadata": {}, 6 | "source": [ 7 | "#### Large RAM is required to load the larger models. Running on GPU can optimize inference speed." 8 | ] 9 | }, 10 | { 11 | "cell_type": "code", 12 | "execution_count": null, 13 | "metadata": {}, 14 | "outputs": [], 15 | "source": [ 16 | "import sys\n", 17 | "!pip3 install salesforce-lavis" 18 | ] 19 | }, 20 | { 21 | "cell_type": "code", 22 | "execution_count": null, 23 | "metadata": {}, 24 | "outputs": [], 25 | "source": [ 26 | "import torch\n", 27 | "from PIL import Image\n", 28 | "import requests\n", 29 | "from lavis.models import load_model_and_preprocess" 30 | ] 31 | }, 32 | { 33 | "cell_type": "markdown", 34 | "metadata": {}, 35 | "source": [ 36 | "#### Load an example image" 37 | ] 38 | }, 39 | { 40 | "cell_type": "code", 41 | "execution_count": null, 42 | "metadata": {}, 43 | "outputs": [], 44 | "source": [ 45 | "img_url = 'https://storage.googleapis.com/sfr-vision-language-research/LAVIS/assets/merlion.png' \n", 46 | "raw_image = Image.open(requests.get(img_url, stream=True).raw).convert('RGB') \n", 47 | "display(raw_image.resize((596, 437)))" 48 | ] 49 | }, 50 | { 51 | "cell_type": "code", 52 | "execution_count": null, 53 | "metadata": {}, 54 | "outputs": [], 55 | "source": [ 56 | "# setup device to use\n", 57 | "device = torch.device(\"cuda:7\") if torch.cuda.is_available() else \"cpu\"" 58 | ] 59 | }, 60 | { 61 | "cell_type": "markdown", 62 | "metadata": {}, 63 | "source": [ 64 | "#### Load pretrained/finetuned BLIP2 captioning model" 65 | ] 66 | }, 67 | { 68 | "cell_type": "code", 69 | "execution_count": null, 70 | "metadata": {}, 71 | "outputs": [], 72 | "source": [ 73 | "# we associate a model with its preprocessors to make it easier for inference.\n", 74 | "model, vis_processors, _ = load_model_and_preprocess(\n", 75 | " name=\"blip2_t5\", model_type=\"pretrain_flant5xl\", is_eval=True, device=device\n", 76 | ")\n", 77 | "vis_processors.keys()" 78 | ] 79 | }, 80 | { 81 | "cell_type": "markdown", 82 | "metadata": {}, 83 | "source": [ 84 | "#### prepare the image as model input using the associated processors" 85 | ] 86 | }, 87 | { 88 | "cell_type": "code", 89 | "execution_count": null, 90 | "metadata": {}, 91 | "outputs": [], 92 | "source": [ 93 | "image = vis_processors[\"eval\"](raw_image).unsqueeze(0).to(device)" 94 | ] 95 | }, 96 | { 97 | "cell_type": "markdown", 98 | "metadata": {}, 99 | "source": [ 100 | "#### generate caption using beam search" 101 | ] 102 | }, 103 | { 104 | "cell_type": "code", 105 | "execution_count": null, 106 | "metadata": {}, 107 | "outputs": [], 108 | "source": [ 109 | "model.generate({\"image\": image})" 110 | ] 111 | }, 112 | { 113 | "cell_type": "markdown", 114 | "metadata": {}, 115 | "source": [ 116 | "#### generate multiple captions using nucleus sampling" 117 | ] 118 | }, 119 | { 120 | "cell_type": "code", 121 | "execution_count": null, 122 | "metadata": {}, 123 | "outputs": [], 124 | "source": [ 125 | "# due to the non-determinstic nature of necleus sampling, you may get different captions.\n", 126 | "model.generate({\"image\": image}, use_nucleus_sampling=True, num_captions=3)" 127 | ] 128 | }, 129 | { 130 | "cell_type": "markdown", 131 | "metadata": {}, 132 | "source": [ 133 | "#### instructed zero-shot vision-to-language generation" 134 | ] 135 | }, 136 | { 137 | "cell_type": "code", 138 | "execution_count": null, 139 | "metadata": {}, 140 | "outputs": [], 141 | "source": [ 142 | "model.generate({\"image\": image, \"prompt\": \"Question: which city is this? Answer:\"})" 143 | ] 144 | }, 145 | { 146 | "cell_type": "code", 147 | "execution_count": null, 148 | "metadata": {}, 149 | "outputs": [], 150 | "source": [ 151 | "model.generate({\n", 152 | " \"image\": image,\n", 153 | " \"prompt\": \"Question: which city is this? Answer: singapore. Question: why?\"})" 154 | ] 155 | }, 156 | { 157 | "cell_type": "code", 158 | "execution_count": null, 159 | "metadata": {}, 160 | "outputs": [], 161 | "source": [ 162 | "context = [\n", 163 | " (\"which city is this?\", \"singapore\"),\n", 164 | " (\"why?\", \"it has a statue of a merlion\"),\n", 165 | "]\n", 166 | "question = \"where is the name merlion coming from?\"\n", 167 | "template = \"Question: {} Answer: {}.\"\n", 168 | "\n", 169 | "prompt = \" \".join([template.format(context[i][0], context[i][1]) for i in range(len(context))]) + \" Question: \" + question + \" Answer:\"\n", 170 | "\n", 171 | "print(prompt)" 172 | ] 173 | }, 174 | { 175 | "cell_type": "code", 176 | "execution_count": null, 177 | "metadata": {}, 178 | "outputs": [], 179 | "source": [ 180 | "model.generate(\n", 181 | " {\n", 182 | " \"image\": image,\n", 183 | " \"prompt\": prompt\n", 184 | " },\n", 185 | " use_nucleus_sampling=False,\n", 186 | ")" 187 | ] 188 | }, 189 | { 190 | "cell_type": "code", 191 | "execution_count": null, 192 | "metadata": {}, 193 | "outputs": [], 194 | "source": [ 195 | "def get_files_sorted_by_number(folder_path):\n", 196 | " files = [f for f in os.listdir(folder_path) if os.path.isfile(os.path.join(folder_path, f))]\n", 197 | " sorted_files = sorted(files, key=lambda x: int(x.split('_')[1].split('.')[0]))\n", 198 | " return sorted_files" 199 | ] 200 | }, 201 | { 202 | "cell_type": "code", 203 | "execution_count": null, 204 | "metadata": {}, 205 | "outputs": [], 206 | "source": [ 207 | "import os\n", 208 | "\n", 209 | "def get_subfolders(folder_path):\n", 210 | " subfolders = [f for f in os.listdir(folder_path) if os.path.isdir(os.path.join(folder_path, f))]\n", 211 | " return subfolders" 212 | ] 213 | }, 214 | { 215 | "cell_type": "code", 216 | "execution_count": null, 217 | "metadata": {}, 218 | "outputs": [], 219 | "source": [ 220 | "folder_path = ''\n", 221 | "subfolders_list = get_subfolders(folder_path)\n", 222 | "print(subfolders_list)\n", 223 | "print(len(subfolders_list))" 224 | ] 225 | }, 226 | { 227 | "cell_type": "code", 228 | "execution_count": null, 229 | "metadata": {}, 230 | "outputs": [], 231 | "source": [ 232 | "\n", 233 | "import csv\n", 234 | "base_path = ''\n", 235 | "csv_file_path = ''\n", 236 | "\n", 237 | "data_list = [\n", 238 | " ['video_name','descirbe_list_1']\n", 239 | "]\n", 240 | "\n", 241 | "for subfolder in subfolders_list:\n", 242 | " print(subfolder)\n", 243 | " files = get_files_sorted_by_number(base_path + '/' + subfolder)\n", 244 | " print(files)\n", 245 | "\n", 246 | " describe_list_1 = []\n", 247 | " describe_list_2 = []\n", 248 | " \n", 249 | " for f in files:\n", 250 | " image_path = base_path + '/' + subfolder + '/' + f\n", 251 | " # print(image_path)\n", 252 | " \n", 253 | " raw_image = Image.open(image_path).convert(\"RGB\")\n", 254 | " image1 = vis_processors[\"eval\"](raw_image).unsqueeze(0).to(device)\n", 255 | " # print(type(image1))\n", 256 | " data = model.generate({\n", 257 | " \"image\": image1,\n", 258 | " \"prompt\": \"Question: Please provide a detailed description of the content in the picture, taking into consideration the scene information and the actions and behavior of the people depicted in the image.\"}) \n", 259 | " describe_list_1.append(data[0])\n", 260 | " print(describe_list_1)\n", 261 | " print(set(describe_list_1))\n", 262 | " data_list.append([subfolder,describe_list_1,set(describe_list_1)])\n", 263 | " # break\n", 264 | "with open(csv_file_path, 'w', newline='', encoding='utf-8') as csv_file:\n", 265 | " csv_writer = csv.writer(csv_file)\n", 266 | " for row in data_list:\n", 267 | " csv_writer.writerow(row)\n", 268 | "\n", 269 | "print(\"数据已成功写入CSV文件。\")" 270 | ] 271 | }, 272 | { 273 | "cell_type": "code", 274 | "execution_count": null, 275 | "metadata": {}, 276 | "outputs": [], 277 | "source": [ 278 | "\"\"\" =========================== Filter ========================= \"\"\"" 279 | ] 280 | }, 281 | { 282 | "cell_type": "code", 283 | "execution_count": null, 284 | "metadata": {}, 285 | "outputs": [], 286 | "source": [ 287 | "from collections import OrderedDict\n", 288 | "\n", 289 | "def remove_duplicates_and_preserve_order(input_list):\n", 290 | " unique_sentences = OrderedDict()\n", 291 | "\n", 292 | " for sentence in input_list:\n", 293 | " unique_sentences[sentence] = None\n", 294 | "\n", 295 | " return list(unique_sentences.keys())" 296 | ] 297 | }, 298 | { 299 | "cell_type": "code", 300 | "execution_count": null, 301 | "metadata": {}, 302 | "outputs": [], 303 | "source": [ 304 | "csv_path = ''\n", 305 | "filter_csv_path = ''" 306 | ] 307 | }, 308 | { 309 | "cell_type": "code", 310 | "execution_count": null, 311 | "metadata": {}, 312 | "outputs": [], 313 | "source": [ 314 | "import csv\n", 315 | "import sys\n", 316 | "csv.field_size_limit(sys.maxsize)\n", 317 | "# 打开CSV文件\n", 318 | "with open(csv_path, 'r', newline='', encoding='utf-8') as csvfile:\n", 319 | " reader = csv.reader(csvfile)\n", 320 | " \n", 321 | " # 读取CSV文件的第一行，获取列标题\n", 322 | " header = next(reader)\n", 323 | " \n", 324 | " # 选择您想要提取的列的索引（假设您想要提取第二列和第四列）\n", 325 | " column_indices = [0,1,2] # 这里使用索引从0开始\n", 326 | " \n", 327 | " # 创建一个字典来存储按列提取的数据\n", 328 | " extracted_data = {header[i]: [] for i in column_indices}\n", 329 | " \n", 330 | " # 遍历CSV文件中的每一行，并按列提取数据\n", 331 | " for row in reader:\n", 332 | " for i in column_indices:\n", 333 | " extracted_data[header[i]].append(row[i])\n", 334 | "# 打印按列提取的数据\n", 335 | "print(len(extracted_data['video_name']))\n", 336 | "video_names = []\n", 337 | "idx1_s = []\n", 338 | "idx2_s = []\n", 339 | "for i in range(0,len(extracted_data['video_name'])):\n", 340 | " video_name = extracted_data['video_name'][i]\n", 341 | " idx1 = extracted_data['descirbe_list_1'][i]\n", 342 | "\n", 343 | " idx1_list = idx1.split('\\'')\n", 344 | " # print(idx1_list)\n", 345 | " idx11 = []\n", 346 | " for j in idx1_list:\n", 347 | " if len(j) <= 3:\n", 348 | " continue\n", 349 | " idx11.append(j)\n", 350 | " n_idx1 = remove_duplicates_and_preserve_order(idx11)\n", 351 | "\n", 352 | " video_names.append(video_name)\n", 353 | " idx1_s.append(n_idx1)\n", 354 | " # break\n", 355 | "print(video_names)\n", 356 | "print(idx1_s[2])\n", 357 | "print(len(idx1_s))" 358 | ] 359 | }, 360 | { 361 | "cell_type": "code", 362 | "execution_count": null, 363 | "metadata": {}, 364 | "outputs": [], 365 | "source": [ 366 | "\n", 367 | "with open('','w') as ta:\n", 368 | " for i in range(0,len(video_names)):\n", 369 | " name = video_names[i]\n", 370 | " des_list = idx1_s[i]\n", 371 | " total_des = ''\n", 372 | " for des in des_list:\n", 373 | " total_des = total_des + des + '.'\n", 374 | " print(total_des)\n", 375 | " ta.writelines(name +' '+ total_des + '\\n')" 376 | ] 377 | }, 378 | { 379 | "cell_type": "code", 380 | "execution_count": null, 381 | "metadata": {}, 382 | "outputs": [], 383 | "source": [ 384 | "names_list = []\n", 385 | "total_des_list = []\n", 386 | "for i in range(0,len(video_names)):\n", 387 | " name = video_names[i]\n", 388 | " des_list = idx1_s[i]\n", 389 | " total_des = ''\n", 390 | " for des in des_list:\n", 391 | " total_des = total_des + des + '.'\n", 392 | " names_list.append(name)\n", 393 | " total_des_list.append(total_des)\n", 394 | "\n", 395 | "print(names_list)\n", 396 | "print(total_des_list)" 397 | ] 398 | }, 399 | { 400 | "cell_type": "code", 401 | "execution_count": null, 402 | "metadata": {}, 403 | "outputs": [], 404 | "source": [ 405 | "new_names_list = []\n", 406 | "new_total_des_list = []\n", 407 | "for i in range(0,len(names_list)):\n", 408 | " name = names_list[i]\n", 409 | " des = total_des_list[i]\n", 410 | " for j in range(0,5):\n", 411 | " n = name + '_' + str(j)\n", 412 | " new_names_list.append(n)\n", 413 | " new_total_des_list.append(des)\n", 414 | "\n", 415 | "print(new_names_list)\n", 416 | "print(new_total_des_list)\n", 417 | "print(len(new_names_list))\n", 418 | "print(len(new_total_des_list))\n" 419 | ] 420 | }, 421 | { 422 | "cell_type": "code", 423 | "execution_count": null, 424 | "metadata": {}, 425 | "outputs": [], 426 | "source": [ 427 | "import csv\n", 428 | "\n", 429 | "# 假设这是你的两个列表\n", 430 | "# new_name_list = ['name1', 'name2', 'name3']\n", 431 | "# new_total_des_list = ['description1', 'description2', 'description3']\n", 432 | "\n", 433 | "# 确保两个列表长度相同\n", 434 | "if len(new_names_list) == len(new_total_des_list):\n", 435 | " # 创建一个字典，将两个列表对应的元素作为键值对\n", 436 | " combined_dict = dict(zip(new_names_list, new_total_des_list))\n", 437 | " \n", 438 | " # 指定CSV文件的名称\n", 439 | " csv_file = \"\"\n", 440 | "\n", 441 | " # 打开文件进行写入\n", 442 | " with open(csv_file, mode='w', newline='', encoding='utf-8') as file:\n", 443 | " writer = csv.writer(file)\n", 444 | " # 写入标题行\n", 445 | " writer.writerow(['Name', 'Description'])\n", 446 | " # 写入数据\n", 447 | " for key, value in combined_dict.items():\n", 448 | " writer.writerow([key, value])\n", 449 | " print(f\"file '{csv_file}' success。\")\n", 450 | "else:\n", 451 | " print(\"no equal\")\n" 452 | ] 453 | }, 454 | { 455 | "cell_type": "code", 456 | "execution_count": null, 457 | "metadata": {}, 458 | "outputs": [], 459 | "source": [] 460 | } 461 | ], 462 | "metadata": { 463 | "kernelspec": { 464 | "display_name": "Python 3 (ipykernel)", 465 | "language": "python", 466 | "name": "python3" 467 | }, 468 | "language_info": { 469 | "codemirror_mode": { 470 | "name": "ipython", 471 | "version": 3 472 | }, 473 | "file_extension": ".py", 474 | "mimetype": "text/x-python", 475 | "name": "python", 476 | "nbconvert_exporter": "python", 477 | "pygments_lexer": "ipython3", 478 | "version": "3.8.13" 479 | }, 480 | "vscode": { 481 | "interpreter": { 482 | "hash": "d4d1e4263499bec80672ea0156c357c1ee493ec2b1c70f0acce89fc37c4a6abe" 483 | } 484 | } 485 | }, 486 | "nbformat": 4, 487 | "nbformat_minor": 2 488 | } 489 | -------------------------------------------------------------------------------- /Client_Split/Weakly_Supervised/XD/test_clients/clients/client4.list: -------------------------------------------------------------------------------- 1 | test/City.of.God.2002__#00-37-20_00-38-02_label_B5-0-0__0.npy 2 | test/City.of.God.2002__#00-37-20_00-38-02_label_B5-0-0__1.npy 3 | test/City.of.God.2002__#00-37-20_00-38-02_label_B5-0-0__2.npy 4 | test/City.of.God.2002__#00-37-20_00-38-02_label_B5-0-0__3.npy 5 | test/City.of.God.2002__#00-37-20_00-38-02_label_B5-0-0__4.npy 6 | test/Death.Proof.2007__#00-45-05_00-47-36_label_B5-0-0__0.npy 7 | test/Death.Proof.2007__#00-45-05_00-47-36_label_B5-0-0__1.npy 8 | test/Death.Proof.2007__#00-45-05_00-47-36_label_B5-0-0__2.npy 9 | test/Death.Proof.2007__#00-45-05_00-47-36_label_B5-0-0__3.npy 10 | test/Death.Proof.2007__#00-45-05_00-47-36_label_B5-0-0__4.npy 11 | test/Death.Proof.2007__#01-40-41_01-42-17_label_B5-B6-0__0.npy 12 | test/Death.Proof.2007__#01-40-41_01-42-17_label_B5-B6-0__1.npy 13 | test/Death.Proof.2007__#01-40-41_01-42-17_label_B5-B6-0__2.npy 14 | test/Death.Proof.2007__#01-40-41_01-42-17_label_B5-B6-0__3.npy 15 | test/Death.Proof.2007__#01-40-41_01-42-17_label_B5-B6-0__4.npy 16 | test/Election.2005__#00-47-57_00-50-54_label_B5-0-0__0.npy 17 | test/Election.2005__#00-47-57_00-50-54_label_B5-0-0__1.npy 18 | test/Election.2005__#00-47-57_00-50-54_label_B5-0-0__2.npy 19 | test/Election.2005__#00-47-57_00-50-54_label_B5-0-0__3.npy 20 | test/Election.2005__#00-47-57_00-50-54_label_B5-0-0__4.npy 21 | test/Operation.Red.Sea.2018__#01-20-58_01-22-00_label_B5-0-0__0.npy 22 | test/Operation.Red.Sea.2018__#01-20-58_01-22-00_label_B5-0-0__1.npy 23 | test/Operation.Red.Sea.2018__#01-20-58_01-22-00_label_B5-0-0__2.npy 24 | test/Operation.Red.Sea.2018__#01-20-58_01-22-00_label_B5-0-0__3.npy 25 | test/Operation.Red.Sea.2018__#01-20-58_01-22-00_label_B5-0-0__4.npy 26 | test/Sin.City.2005__#0-22-04_0-22-18_label_B5-0-0__0.npy 27 | test/Sin.City.2005__#0-22-04_0-22-18_label_B5-0-0__1.npy 28 | test/Sin.City.2005__#0-22-04_0-22-18_label_B5-0-0__2.npy 29 | test/Sin.City.2005__#0-22-04_0-22-18_label_B5-0-0__3.npy 30 | test/Sin.City.2005__#0-22-04_0-22-18_label_B5-0-0__4.npy 31 | test/Taken.2.UNRATED.EXTENDED.2012__#00-13-42_00-14-16_label_B5-0-0__0.npy 32 | test/Taken.2.UNRATED.EXTENDED.2012__#00-13-42_00-14-16_label_B5-0-0__1.npy 33 | test/Taken.2.UNRATED.EXTENDED.2012__#00-13-42_00-14-16_label_B5-0-0__2.npy 34 | test/Taken.2.UNRATED.EXTENDED.2012__#00-13-42_00-14-16_label_B5-0-0__3.npy 35 | test/Taken.2.UNRATED.EXTENDED.2012__#00-13-42_00-14-16_label_B5-0-0__4.npy 36 | test/Yellow.Sea.2010__#01-58-30_01-59-27_label_B5-0-0__0.npy 37 | test/Yellow.Sea.2010__#01-58-30_01-59-27_label_B5-0-0__1.npy 38 | test/Yellow.Sea.2010__#01-58-30_01-59-27_label_B5-0-0__2.npy 39 | test/Yellow.Sea.2010__#01-58-30_01-59-27_label_B5-0-0__3.npy 40 | test/Yellow.Sea.2010__#01-58-30_01-59-27_label_B5-0-0__4.npy 41 | test/Mission.Impossible.II.2000__#00-11-37_00-15-46_label_A__0.npy 42 | test/Mission.Impossible.II.2000__#00-11-37_00-15-46_label_A__1.npy 43 | test/Mission.Impossible.II.2000__#00-11-37_00-15-46_label_A__2.npy 44 | test/Mission.Impossible.II.2000__#00-11-37_00-15-46_label_A__3.npy 45 | test/Mission.Impossible.II.2000__#00-11-37_00-15-46_label_A__4.npy 46 | test/v=SdYFFo_tsfk__#00-00-00_00-06-00_label_A__0.npy 47 | test/v=SdYFFo_tsfk__#00-00-00_00-06-00_label_A__1.npy 48 | test/v=SdYFFo_tsfk__#00-00-00_00-06-00_label_A__2.npy 49 | test/v=SdYFFo_tsfk__#00-00-00_00-06-00_label_A__3.npy 50 | test/v=SdYFFo_tsfk__#00-00-00_00-06-00_label_A__4.npy 51 | test/v=yZ2wlTAN49g__#00-03-00_00-06-00_label_A__0.npy 52 | test/v=yZ2wlTAN49g__#00-03-00_00-06-00_label_A__1.npy 53 | test/v=yZ2wlTAN49g__#00-03-00_00-06-00_label_A__2.npy 54 | test/v=yZ2wlTAN49g__#00-03-00_00-06-00_label_A__3.npy 55 | test/v=yZ2wlTAN49g__#00-03-00_00-06-00_label_A__4.npy 56 | test/A.Beautiful.Mind.2001__#00-40-52_00-42-01_label_A__0.npy 57 | test/A.Beautiful.Mind.2001__#00-40-52_00-42-01_label_A__1.npy 58 | test/A.Beautiful.Mind.2001__#00-40-52_00-42-01_label_A__2.npy 59 | test/A.Beautiful.Mind.2001__#00-40-52_00-42-01_label_A__3.npy 60 | test/A.Beautiful.Mind.2001__#00-40-52_00-42-01_label_A__4.npy 61 | test/v=KEh6EOQvQO0__#1_label_A__0.npy 62 | test/v=KEh6EOQvQO0__#1_label_A__1.npy 63 | test/v=KEh6EOQvQO0__#1_label_A__2.npy 64 | test/v=KEh6EOQvQO0__#1_label_A__3.npy 65 | test/v=KEh6EOQvQO0__#1_label_A__4.npy 66 | test/Death.Proof.2007__#00-02-04_00-05-00_label_A__0.npy 67 | test/Death.Proof.2007__#00-02-04_00-05-00_label_A__1.npy 68 | test/Death.Proof.2007__#00-02-04_00-05-00_label_A__2.npy 69 | test/Death.Proof.2007__#00-02-04_00-05-00_label_A__3.npy 70 | test/Death.Proof.2007__#00-02-04_00-05-00_label_A__4.npy 71 | test/Gladiator.2000__#02-43-00_02-43-59_label_A__0.npy 72 | test/Gladiator.2000__#02-43-00_02-43-59_label_A__1.npy 73 | test/Gladiator.2000__#02-43-00_02-43-59_label_A__2.npy 74 | test/Gladiator.2000__#02-43-00_02-43-59_label_A__3.npy 75 | test/Gladiator.2000__#02-43-00_02-43-59_label_A__4.npy 76 | test/Salt.2010__#01-16-00_01-16-50_label_A__0.npy 77 | test/Salt.2010__#01-16-00_01-16-50_label_A__1.npy 78 | test/Salt.2010__#01-16-00_01-16-50_label_A__2.npy 79 | test/Salt.2010__#01-16-00_01-16-50_label_A__3.npy 80 | test/Salt.2010__#01-16-00_01-16-50_label_A__4.npy 81 | test/v=pCyCi-q4MqY__#1_label_A__0.npy 82 | test/v=pCyCi-q4MqY__#1_label_A__1.npy 83 | test/v=pCyCi-q4MqY__#1_label_A__2.npy 84 | test/v=pCyCi-q4MqY__#1_label_A__3.npy 85 | test/v=pCyCi-q4MqY__#1_label_A__4.npy 86 | test/Skyfall.2012__#02-14-50_02-16-25_label_A__0.npy 87 | test/Skyfall.2012__#02-14-50_02-16-25_label_A__1.npy 88 | test/Skyfall.2012__#02-14-50_02-16-25_label_A__2.npy 89 | test/Skyfall.2012__#02-14-50_02-16-25_label_A__3.npy 90 | test/Skyfall.2012__#02-14-50_02-16-25_label_A__4.npy 91 | test/v=gGlqujwrhbI__#00-12-00_00-15-00_label_A__0.npy 92 | test/v=gGlqujwrhbI__#00-12-00_00-15-00_label_A__1.npy 93 | test/v=gGlqujwrhbI__#00-12-00_00-15-00_label_A__2.npy 94 | test/v=gGlqujwrhbI__#00-12-00_00-15-00_label_A__3.npy 95 | test/v=gGlqujwrhbI__#00-12-00_00-15-00_label_A__4.npy 96 | test/v=L2Qs4KyW-Co__#1_label_A__0.npy 97 | test/v=L2Qs4KyW-Co__#1_label_A__1.npy 98 | test/v=L2Qs4KyW-Co__#1_label_A__2.npy 99 | test/v=L2Qs4KyW-Co__#1_label_A__3.npy 100 | test/v=L2Qs4KyW-Co__#1_label_A__4.npy 101 | test/Be.with.You.2018__#00-04-20_00-05-35_label_A__0.npy 102 | test/Be.with.You.2018__#00-04-20_00-05-35_label_A__1.npy 103 | test/Be.with.You.2018__#00-04-20_00-05-35_label_A__2.npy 104 | test/Be.with.You.2018__#00-04-20_00-05-35_label_A__3.npy 105 | test/Be.with.You.2018__#00-04-20_00-05-35_label_A__4.npy 106 | test/Young.And.Dangerous.III.1996__#00-15-52_00-17-00_label_A__0.npy 107 | test/Young.And.Dangerous.III.1996__#00-15-52_00-17-00_label_A__1.npy 108 | test/Young.And.Dangerous.III.1996__#00-15-52_00-17-00_label_A__2.npy 109 | test/Young.And.Dangerous.III.1996__#00-15-52_00-17-00_label_A__3.npy 110 | test/Young.And.Dangerous.III.1996__#00-15-52_00-17-00_label_A__4.npy 111 | test/v=vFPQ_NiDBIU__#02-42-00_02-48-00_label_A__0.npy 112 | test/v=vFPQ_NiDBIU__#02-42-00_02-48-00_label_A__1.npy 113 | test/v=vFPQ_NiDBIU__#02-42-00_02-48-00_label_A__2.npy 114 | test/v=vFPQ_NiDBIU__#02-42-00_02-48-00_label_A__3.npy 115 | test/v=vFPQ_NiDBIU__#02-42-00_02-48-00_label_A__4.npy 116 | test/Love.Me.If.You.Dare.2003__#01-04-40_01-06-59_label_A__0.npy 117 | test/Love.Me.If.You.Dare.2003__#01-04-40_01-06-59_label_A__1.npy 118 | test/Love.Me.If.You.Dare.2003__#01-04-40_01-06-59_label_A__2.npy 119 | test/Love.Me.If.You.Dare.2003__#01-04-40_01-06-59_label_A__3.npy 120 | test/Love.Me.If.You.Dare.2003__#01-04-40_01-06-59_label_A__4.npy 121 | test/v=1EvRIhdsX1s__#1_label_A__0.npy 122 | test/v=1EvRIhdsX1s__#1_label_A__1.npy 123 | test/v=1EvRIhdsX1s__#1_label_A__2.npy 124 | test/v=1EvRIhdsX1s__#1_label_A__3.npy 125 | test/v=1EvRIhdsX1s__#1_label_A__4.npy 126 | test/Tropa.de.Elite.2007__#00-10-30_00-12-08_label_A__0.npy 127 | test/Tropa.de.Elite.2007__#00-10-30_00-12-08_label_A__1.npy 128 | test/Tropa.de.Elite.2007__#00-10-30_00-12-08_label_A__2.npy 129 | test/Tropa.de.Elite.2007__#00-10-30_00-12-08_label_A__3.npy 130 | test/Tropa.de.Elite.2007__#00-10-30_00-12-08_label_A__4.npy 131 | test/The.Hurt.Locker.2008__#0-29-25_0-30-15_label_A__0.npy 132 | test/The.Hurt.Locker.2008__#0-29-25_0-30-15_label_A__1.npy 133 | test/The.Hurt.Locker.2008__#0-29-25_0-30-15_label_A__2.npy 134 | test/The.Hurt.Locker.2008__#0-29-25_0-30-15_label_A__3.npy 135 | test/The.Hurt.Locker.2008__#0-29-25_0-30-15_label_A__4.npy 136 | test/Sin.City.2005__#01-43-21_01-44-21_label_A__0.npy 137 | test/Sin.City.2005__#01-43-21_01-44-21_label_A__1.npy 138 | test/Sin.City.2005__#01-43-21_01-44-21_label_A__2.npy 139 | test/Sin.City.2005__#01-43-21_01-44-21_label_A__3.npy 140 | test/Sin.City.2005__#01-43-21_01-44-21_label_A__4.npy 141 | test/Kingsman.The.Golden.Circle.2017__#02-09-30_02-10-45_label_A__0.npy 142 | test/Kingsman.The.Golden.Circle.2017__#02-09-30_02-10-45_label_A__1.npy 143 | test/Kingsman.The.Golden.Circle.2017__#02-09-30_02-10-45_label_A__2.npy 144 | test/Kingsman.The.Golden.Circle.2017__#02-09-30_02-10-45_label_A__3.npy 145 | test/Kingsman.The.Golden.Circle.2017__#02-09-30_02-10-45_label_A__4.npy 146 | test/v=9eME1y6V-T4__#01-18-00_01-24-00_label_A__0.npy 147 | test/v=9eME1y6V-T4__#01-18-00_01-24-00_label_A__1.npy 148 | test/v=9eME1y6V-T4__#01-18-00_01-24-00_label_A__2.npy 149 | test/v=9eME1y6V-T4__#01-18-00_01-24-00_label_A__3.npy 150 | test/v=9eME1y6V-T4__#01-18-00_01-24-00_label_A__4.npy 151 | test/v=es97CGK_5XQ__#1_label_A__0.npy 152 | test/v=es97CGK_5XQ__#1_label_A__1.npy 153 | test/v=es97CGK_5XQ__#1_label_A__2.npy 154 | test/v=es97CGK_5XQ__#1_label_A__3.npy 155 | test/v=es97CGK_5XQ__#1_label_A__4.npy 156 | test/Love.Actually.2003__#00-51-55_00-55-16_label_A__0.npy 157 | test/Love.Actually.2003__#00-51-55_00-55-16_label_A__1.npy 158 | test/Love.Actually.2003__#00-51-55_00-55-16_label_A__2.npy 159 | test/Love.Actually.2003__#00-51-55_00-55-16_label_A__3.npy 160 | test/Love.Actually.2003__#00-51-55_00-55-16_label_A__4.npy 161 | test/A.Beautiful.Mind.2001__#00-25-20_00-29-20_label_A__0.npy 162 | test/A.Beautiful.Mind.2001__#00-25-20_00-29-20_label_A__1.npy 163 | test/A.Beautiful.Mind.2001__#00-25-20_00-29-20_label_A__2.npy 164 | test/A.Beautiful.Mind.2001__#00-25-20_00-29-20_label_A__3.npy 165 | test/A.Beautiful.Mind.2001__#00-25-20_00-29-20_label_A__4.npy 166 | test/Mission.Impossible.Ghost.Protocol.2011__#01-07-24_01-09-09_label_A__0.npy 167 | test/Mission.Impossible.Ghost.Protocol.2011__#01-07-24_01-09-09_label_A__1.npy 168 | test/Mission.Impossible.Ghost.Protocol.2011__#01-07-24_01-09-09_label_A__2.npy 169 | test/Mission.Impossible.Ghost.Protocol.2011__#01-07-24_01-09-09_label_A__3.npy 170 | test/Mission.Impossible.Ghost.Protocol.2011__#01-07-24_01-09-09_label_A__4.npy 171 | test/Ip.Man.2008__#00-02-56_00-05-24_label_A__0.npy 172 | test/Ip.Man.2008__#00-02-56_00-05-24_label_A__1.npy 173 | test/Ip.Man.2008__#00-02-56_00-05-24_label_A__2.npy 174 | test/Ip.Man.2008__#00-02-56_00-05-24_label_A__3.npy 175 | test/Ip.Man.2008__#00-02-56_00-05-24_label_A__4.npy 176 | test/v=HksDvkX6Bjk__#1_label_A__0.npy 177 | test/v=HksDvkX6Bjk__#1_label_A__1.npy 178 | test/v=HksDvkX6Bjk__#1_label_A__2.npy 179 | test/v=HksDvkX6Bjk__#1_label_A__3.npy 180 | test/v=HksDvkX6Bjk__#1_label_A__4.npy 181 | test/Bullet.in.the.Head.1990__#01-58-25_01-59-30_label_A__0.npy 182 | test/Bullet.in.the.Head.1990__#01-58-25_01-59-30_label_A__1.npy 183 | test/Bullet.in.the.Head.1990__#01-58-25_01-59-30_label_A__2.npy 184 | test/Bullet.in.the.Head.1990__#01-58-25_01-59-30_label_A__3.npy 185 | test/Bullet.in.the.Head.1990__#01-58-25_01-59-30_label_A__4.npy 186 | test/v=NbNFfnVSIn8__#1_label_A__0.npy 187 | test/v=NbNFfnVSIn8__#1_label_A__1.npy 188 | test/v=NbNFfnVSIn8__#1_label_A__2.npy 189 | test/v=NbNFfnVSIn8__#1_label_A__3.npy 190 | test/v=NbNFfnVSIn8__#1_label_A__4.npy 191 | test/v=Ehh8ZdIMMj4__#1_label_A__0.npy 192 | test/v=Ehh8ZdIMMj4__#1_label_A__1.npy 193 | test/v=Ehh8ZdIMMj4__#1_label_A__2.npy 194 | test/v=Ehh8ZdIMMj4__#1_label_A__3.npy 195 | test/v=Ehh8ZdIMMj4__#1_label_A__4.npy 196 | test/Desperado.1995__#01-22-10_01-23-17_label_A__0.npy 197 | test/Desperado.1995__#01-22-10_01-23-17_label_A__1.npy 198 | test/Desperado.1995__#01-22-10_01-23-17_label_A__2.npy 199 | test/Desperado.1995__#01-22-10_01-23-17_label_A__3.npy 200 | test/Desperado.1995__#01-22-10_01-23-17_label_A__4.npy 201 | test/v=Hd_2Y29_FLU__#1_label_A__0.npy 202 | test/v=Hd_2Y29_FLU__#1_label_A__1.npy 203 | test/v=Hd_2Y29_FLU__#1_label_A__2.npy 204 | test/v=Hd_2Y29_FLU__#1_label_A__3.npy 205 | test/v=Hd_2Y29_FLU__#1_label_A__4.npy 206 | test/v=t_TZYpT6xjY__#1_label_A__0.npy 207 | test/v=t_TZYpT6xjY__#1_label_A__1.npy 208 | test/v=t_TZYpT6xjY__#1_label_A__2.npy 209 | test/v=t_TZYpT6xjY__#1_label_A__3.npy 210 | test/v=t_TZYpT6xjY__#1_label_A__4.npy 211 | test/v=HTEyaXBPf9U__#1_label_A__0.npy 212 | test/v=HTEyaXBPf9U__#1_label_A__1.npy 213 | test/v=HTEyaXBPf9U__#1_label_A__2.npy 214 | test/v=HTEyaXBPf9U__#1_label_A__3.npy 215 | test/v=HTEyaXBPf9U__#1_label_A__4.npy 216 | test/v=ldnLxTablA8__#1_label_A__0.npy 217 | test/v=ldnLxTablA8__#1_label_A__1.npy 218 | test/v=ldnLxTablA8__#1_label_A__2.npy 219 | test/v=ldnLxTablA8__#1_label_A__3.npy 220 | test/v=ldnLxTablA8__#1_label_A__4.npy 221 | test/IP.Man.2.2010__#00-54-00_00-57-12_label_A__0.npy 222 | test/IP.Man.2.2010__#00-54-00_00-57-12_label_A__1.npy 223 | test/IP.Man.2.2010__#00-54-00_00-57-12_label_A__2.npy 224 | test/IP.Man.2.2010__#00-54-00_00-57-12_label_A__3.npy 225 | test/IP.Man.2.2010__#00-54-00_00-57-12_label_A__4.npy 226 | test/Before.Sunrise.1995__#00-42-52_00-45-31_label_A__0.npy 227 | test/Before.Sunrise.1995__#00-42-52_00-45-31_label_A__1.npy 228 | test/Before.Sunrise.1995__#00-42-52_00-45-31_label_A__2.npy 229 | test/Before.Sunrise.1995__#00-42-52_00-45-31_label_A__3.npy 230 | test/Before.Sunrise.1995__#00-42-52_00-45-31_label_A__4.npy 231 | test/GoldenEye.1995__#01-02-11_01-03-43_label_A__0.npy 232 | test/GoldenEye.1995__#01-02-11_01-03-43_label_A__1.npy 233 | test/GoldenEye.1995__#01-02-11_01-03-43_label_A__2.npy 234 | test/GoldenEye.1995__#01-02-11_01-03-43_label_A__3.npy 235 | test/GoldenEye.1995__#01-02-11_01-03-43_label_A__4.npy 236 | test/v=2JJ9d97sd64__#1_label_A__0.npy 237 | test/v=2JJ9d97sd64__#1_label_A__1.npy 238 | test/v=2JJ9d97sd64__#1_label_A__2.npy 239 | test/v=2JJ9d97sd64__#1_label_A__3.npy 240 | test/v=2JJ9d97sd64__#1_label_A__4.npy 241 | test/Hear.Me.2009__#00-49-48_00-51-12_label_A__0.npy 242 | test/Hear.Me.2009__#00-49-48_00-51-12_label_A__1.npy 243 | test/Hear.Me.2009__#00-49-48_00-51-12_label_A__2.npy 244 | test/Hear.Me.2009__#00-49-48_00-51-12_label_A__3.npy 245 | test/Hear.Me.2009__#00-49-48_00-51-12_label_A__4.npy 246 | test/v=5l-nbDl0qvg__#1_label_A__0.npy 247 | test/v=5l-nbDl0qvg__#1_label_A__1.npy 248 | test/v=5l-nbDl0qvg__#1_label_A__2.npy 249 | test/v=5l-nbDl0qvg__#1_label_A__3.npy 250 | test/v=5l-nbDl0qvg__#1_label_A__4.npy 251 | test/v=DdIA1Owbcdc__#1_label_A__0.npy 252 | test/v=DdIA1Owbcdc__#1_label_A__1.npy 253 | test/v=DdIA1Owbcdc__#1_label_A__2.npy 254 | test/v=DdIA1Owbcdc__#1_label_A__3.npy 255 | test/v=DdIA1Owbcdc__#1_label_A__4.npy 256 | test/Black.Hawk.Down.2001__#00-06-44_00-07-39_label_A__0.npy 257 | test/Black.Hawk.Down.2001__#00-06-44_00-07-39_label_A__1.npy 258 | test/Black.Hawk.Down.2001__#00-06-44_00-07-39_label_A__2.npy 259 | test/Black.Hawk.Down.2001__#00-06-44_00-07-39_label_A__3.npy 260 | test/Black.Hawk.Down.2001__#00-06-44_00-07-39_label_A__4.npy 261 | test/v=vFPQ_NiDBIU__#01-48-00_01-54-00_label_A__0.npy 262 | test/v=vFPQ_NiDBIU__#01-48-00_01-54-00_label_A__1.npy 263 | test/v=vFPQ_NiDBIU__#01-48-00_01-54-00_label_A__2.npy 264 | test/v=vFPQ_NiDBIU__#01-48-00_01-54-00_label_A__3.npy 265 | test/v=vFPQ_NiDBIU__#01-48-00_01-54-00_label_A__4.npy 266 | test/v=9FxmNOXF25Q__#1_label_A__0.npy 267 | test/v=9FxmNOXF25Q__#1_label_A__1.npy 268 | test/v=9FxmNOXF25Q__#1_label_A__2.npy 269 | test/v=9FxmNOXF25Q__#1_label_A__3.npy 270 | test/v=9FxmNOXF25Q__#1_label_A__4.npy 271 | test/Taken.3.2014__#00-36-03_00-36-48_label_A__0.npy 272 | test/Taken.3.2014__#00-36-03_00-36-48_label_A__1.npy 273 | test/Taken.3.2014__#00-36-03_00-36-48_label_A__2.npy 274 | test/Taken.3.2014__#00-36-03_00-36-48_label_A__3.npy 275 | test/Taken.3.2014__#00-36-03_00-36-48_label_A__4.npy 276 | test/The.Bourne.Supremacy.2004__#01-34-52_01-35-50_label_A__0.npy 277 | test/The.Bourne.Supremacy.2004__#01-34-52_01-35-50_label_A__1.npy 278 | test/The.Bourne.Supremacy.2004__#01-34-52_01-35-50_label_A__2.npy 279 | test/The.Bourne.Supremacy.2004__#01-34-52_01-35-50_label_A__3.npy 280 | test/The.Bourne.Supremacy.2004__#01-34-52_01-35-50_label_A__4.npy 281 | test/Flipped.2010__#01-07-30_01-09-51_label_A__0.npy 282 | test/Flipped.2010__#01-07-30_01-09-51_label_A__1.npy 283 | test/Flipped.2010__#01-07-30_01-09-51_label_A__2.npy 284 | test/Flipped.2010__#01-07-30_01-09-51_label_A__3.npy 285 | test/Flipped.2010__#01-07-30_01-09-51_label_A__4.npy 286 | test/The.Secret.Life.of.Walter.Mitty.2013__#01-01-20_01-04-00_label_A__0.npy 287 | test/The.Secret.Life.of.Walter.Mitty.2013__#01-01-20_01-04-00_label_A__1.npy 288 | test/The.Secret.Life.of.Walter.Mitty.2013__#01-01-20_01-04-00_label_A__2.npy 289 | test/The.Secret.Life.of.Walter.Mitty.2013__#01-01-20_01-04-00_label_A__3.npy 290 | test/The.Secret.Life.of.Walter.Mitty.2013__#01-01-20_01-04-00_label_A__4.npy 291 | --------------------------------------------------------------------------------