├── README.md
├── annotation.py
├── annotations
    └── emty
├── celebrityDB
    └── images here
├── data
    ├── label.pbtxt
    └── ssd_mobilenet_v1_coco.config
├── generate_tfrecords.py
├── it5.jpg
├── lbpcascade_frontalface_improved.xml
├── split_data.ipynb
├── ssd_mobilenet_v1_coco_2018_01_28
    └── ssd.md
└── xml_to_csv.py


/README.md:
--------------------------------------------------------------------------------
  1 | # Face Recognition with the TensorFlow Object Detection API
  2 | ![Screenshot](it5.jpg)
  3 | 
  4 | ### Note: Tensorflow object detection is an accurate machine learning API capable of localizing and identifying multiple objects in a single image. You can use the API for multiple use cases like object detection , person recognition, text detection, etc..
  5 | ### [This tutorial needs tensorflow api installation](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/installation.md)
  6 | Today, we will see together how tensorflow can recognize people. In this post I'll outline the steps I took to get from a collection of Celebrities images (crawled from the internet)
  7 | 
  8 | 1. Data Set Download
  9 | 2. Image Annotation
 10 | 3. Xml to CSV
 11 | 4. TF-Record Creation 
 12 | 5. Label Map preparation
 13 | 6. Pipeline Configuration
 14 | 7. Training
 15 | 8. Exporting Graph
 16 | 
 17 | 
 18 | 
 19 | ## 1. Data set Download
 20 | 
 21 | You can crawl celebrity pictures from google images if you don't have a ready Data set. Try to order the data set as bellow:
 22 | 
 23 | ```
 24 | 
 25 | CelebrityDB/
 26 |     TomHanks/
 27 |         img001.jpg
 28 |         tomhanks.jpg
 29 |         ...
 30 |     WillSmith/
 31 |         willsmith1.jpg
 32 |         will-smith-pic.jpg
 33 |         ...
 34 |     ... 
 35 | 
 36 | ```
 37 | ## 2. Image Annotation
 38 | you can annotate the images using an annotation tool like labelImg. But it will take a lot of time. That's why i created a script to generate xml files(exactly like PASCAL VOC). I used opencv to detect faces but, you can change it with any other tool( i recommend dlib or a neural network face detection model which are much more accurate than opencv).
 39 | #### Use [python annotation script to generate xml annotations](annotation.py)
 40 | The Xml files should look like :
 41 | ```xml
 42 | <annotation verified="yes">
 43 |     <folder>celebrityDB</folder><filename>1d9k49.jpeg</filename>
 44 |     <path>/media/emna/datapartition/tutos/celebrityDB/dewayneJohnson/1d9k49.jpeg</path>
 45 |     <source>
 46 |         <database>Emna Amor</database>
 47 |     </source>
 48 |     <size>
 49 |         <width>104</width>
 50 |         <height>142</height>
 51 |         <depth>3</depth>
 52 |     </size>
 53 |     <segmented>0</segmented>
 54 |     <object>
 55 |         <name>dewayneJohnson</name>
 56 |         <pose>Unspecified</pose>
 57 |         <truncated>0</truncated>
 58 |         <difficult>0</difficult>
 59 |         <bndbox>
 60 |             <xmin>2</xmin>
 61 |             <ymin>34</ymin>
 62 |             <xmax>133</xmax>
 63 |             <ymax>99</ymax>
 64 |         </bndbox>
 65 |     </object>
 66 | </annotation>
 67 | ```
 68 | ## 3. XML tO CSV
 69 | After annotating the pictures we have, We need to generate a csv file containing all pictures details / classes. 
 70 |  #### Use  [python xml to csv script to generate the csv file](xml_to_csv.py)
 71 | The csv File should look like this:
 72 | 
 73 | 
 74 | | filename | width | height | class | xmin | ymin | xmax | ymax | 
 75 | | --- | --- | --- | --- | --- | --- | --- | --- |
 76 | | 1jth1461.jpeg | 76 | 105 | kimKardashian | 10 | 30 | 89 | 59 |
 77 | | wenn33850496.jpg | 470 | 654 | AndySerkis | 52 | 142 | 494 | 352 |
 78 | 
 79 | Then we need to split the data into train and test using [this python notebook ](split_data.ipynb.ipynb).
 80 | 
 81 | ### 4. TF-Record Creation 
 82 | To train our model, we need to convert the data to Tensorflow file format called Tfrecords. Most of the batch operations aren’t done directly from images, rather they are converted into a single tfrecord file (images which are numpy arrays and labels which are a list of strings).
 83 | #### WHAT IS TFRECORD?
 84 | ```
 85 |  “… TFRECORD is an approach that convert whatever data you have into a supported format. This approach makes it easier to mix and match data sets and network architectures. The recommended format for TensorFlow is a TFRecords file containing tf.train.Example protocol buffers (which contain Features as a field).“
 86 | ```
 87 | Use this [python script](generate_tfrecords.py) to generate te tf records files (train.record  and test.record) 
 88 | ```
 89 | Usage:
 90 |   # Create train data:
 91 |   python generate_tfrecords.py --csv_input=train.csv  --output_path=data/train.record
 92 | 
 93 |   # Create test data:
 94 |   python generate_tfrecords.py --csv_input=test.csv  --output_path=data/test.record
 95 | ```
 96 | ###### Note: do not forget to edit the generate_tfrecords file with your own labels.
 97 | 
 98 | ### 5. Label Map preparation
 99 | Use the same order you appended the labels in the generate_tfrecords python script 
100 | ###### Note: label map id should be diffirent to 0 !
101 | ```
102 | item {
103 |   id: 1
104 |   name: 'AndySerkis'
105 | }
106 | 
107 | item {
108 |   id: 2
109 |   name: 'dewayneJohnson'
110 | }
111 | item {
112 |   id: 3
113 |   name: 'drake'
114 | }
115 | 
116 | item {
117 |   id: 4
118 |   name: 'jayZ'
119 | }
120 | 
121 | item {
122 |   id: 5
123 |   name: 'justinBieber'
124 | }
125 | 
126 | item {
127 |   id: 6
128 |   name: 'kimKardashian'
129 | }
130 | 
131 | 
132 | item {
133 |   id: 7
134 |   name: 'kimKardashian'
135 | }
136 | 
137 | item {
138 |   id: 8
139 |   name: 'tomHanks'
140 | }
141 | 
142 | item {
143 |   id: 9
144 |   name: 'willSmith'
145 | }
146 | 
147 | ```
148 | ### Pipeline Configuration
149 | We will use [ssd_mobilenet_v1_coco](http://download.tensorflow.org/models/object_detection/ssd_mobilenet_v1_coco_2018_01_28.tar.gz) to train our face recognition model.
150 | #### Do not forget to edit [the ssd_mobilenet_v1_coco.config file](data/ssd_mobilenet_v1_coco.config) with the number of classes( 9 in my case) , ssd_mobilenet_v1_coco model.ckpt under ssd_mobilenet_v1_coco_2018_01_28, the train record path , test record path and the label.pbtxt path.
151 | 
152 | ### 7. Training
153 | 
154 | ```
155 | # From the tensorflow/models/research/ directory
156 | 
157 | PIPELINE_CONFIG_PATH={path to pipeline config file}/data/ssd_mobilenet_v1_coco.config
158 | MODEL_DIR={path to model directory}/ssd_mobilenet_v1_coco_2018_01_28
159 | NUM_TRAIN_STEPS=50000
160 | SAMPLE_1_OF_N_EVAL_EXAMPLES=1
161 | python object_detection/model_main.py \
162 |     --pipeline_config_path=${PIPELINE_CONFIG_PATH} \
163 |     --model_dir=${MODEL_DIR} \
164 |     --num_train_steps=${NUM_TRAIN_STEPS} \
165 |     --sample_1_of_n_eval_examples=$SAMPLE_1_OF_N_EVAL_EXAMPLES \
166 |     --alsologtostderr
167 |     
168 | - to keep training infinetly, remove NUM_TRAIN_STEPS=50000
169 | 
170 | ```
171 | ### 8. Exporting Graph
172 | ```
173 | 
174 | # From tensorflow/models/research/
175 | INPUT_TYPE=image_tensor
176 | PIPELINE_CONFIG_PATH={path to pipeline config file}/data/ssd_mobilenet_v1_coco.config
177 | TRAINED_CKPT_PREFIX={path to model directory}/ssd_mobilenet_v1_coco_2018_01_28/model.ckpt-num
178 | EXPORT_DIR={path to folder that will be used for export}
179 | python object_detection/export_inference_graph.py \
180 |     --input_type=${INPUT_TYPE} \
181 |     --pipeline_config_path=${PIPELINE_CONFIG_PATH} \
182 |     --trained_checkpoint_prefix=${TRAINED_CKPT_PREFIX} \
183 |     --output_directory=${EXPORT_DIR}
184 | ```
185 | You can use the object detection notebook to make prediction (change the model download section with your model path)
186 | 


--------------------------------------------------------------------------------
/annotation.py:
--------------------------------------------------------------------------------
 1 | import xml.etree.cElementTree as ET
 2 | import os
 3 | import cv2
 4 | import numpy as np
 5 | #from detect_faces.src import detect_faces, show_bboxes
 6 | from PIL import Image
 7 | 
 8 | path, filename = os.path.split(os.getcwd()+'/celebrityDB')
 9 | face_cascade = cv2.CascadeClassifier('lbpcascade_frontalface_improved.xml')
10 | 
11 | for subdir, dirs, files in os.walk(os.getcwd()):
12 |     for file in files:
13 | 
14 |         
15 |         img_path=os.path.join(subdir, file)
16 |         img_name=os.path.basename(img_path)
17 | 
18 |         if img_path.lower().endswith(('.png', '.jpg', '.jpeg')):
19 | 
20 |             img = cv2.imread(img_path)
21 |             gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
22 |             h,w,bpp = np.shape(img)
23 |             imgp = Image.open(img_path)
24 | #            b, landmarks = detect_faces(imgp)
25 |             faces = face_cascade.detectMultiScale(gray, 1.3, 5)
26 | 
27 | 
28 | #            g=show_bboxes(imgp, b, landmarks)
29 |             #print(len(faces))
30 |             if len(faces) == 1 :
31 |                 try :
32 | 
33 |                     for bounding_boxes in faces:
34 |                         face = img[int(bounding_boxes[1]):int(bounding_boxes[3]),
35 |                         int(bounding_boxes[0]):int(bounding_boxes[2])]
36 | 
37 |                         
38 |                         subdir_path, subdir_name = os.path.split(subdir)
39 | 
40 |                         root = ET.Element("annotation",verified="yes")
41 |                         ET.SubElement(root, "folder").text=filename
42 | 
43 |                         ET.SubElement(root, "filename").text = img_name
44 |                         ET.SubElement(root, "path").text = img_path
45 | 
46 |                         source=ET.SubElement(root, "source")
47 |                         ET.SubElement(source, "database").text = "Emna Amor"
48 | 
49 |                         size=ET.SubElement(root, "size")
50 |                         ET.SubElement(size, "width").text = str(w)
51 |                         ET.SubElement(size, "height").text = str(h)
52 |                         ET.SubElement(size, "depth").text = str(bpp)
53 | 
54 |                         ET.SubElement(root, "segmented").text = "0"
55 | 
56 |                         obj=ET.SubElement(root, "object")
57 |                         ET.SubElement(obj, "name").text = subdir_name
58 |                         ET.SubElement(obj, "pose").text = "Unspecified"
59 |                         ET.SubElement(obj, "truncated").text = "0"
60 |                         ET.SubElement(obj, "difficult").text = "0"
61 | 
62 |                         box=ET.SubElement(obj, "bndbox")
63 |                         ET.SubElement(box, "xmin").text = str(int(bounding_boxes[0]))
64 |                         ET.SubElement(box, "ymin").text = str(int(bounding_boxes[1]))
65 |                         ET.SubElement(box, "xmax").text = str(int(bounding_boxes[2])+int(bounding_boxes[1]))
66 |                         ET.SubElement(box, "ymax").text = str(int(bounding_boxes[3]))
67 | 
68 |                         tree = ET.ElementTree(root)
69 |                     tree.write(os.path.join(os.getcwd()+'/annotations', os.path.splitext(img_name)[0] + '.xml'))
70 |                 except RuntimeError :
71 |                     with open("delete.txt", "a") as myfile:
72 |                         myfile.write(img_path+"\n")
73 |                 	#print(img_path)
74 | 
75 |             else :
76 |                 with open("delete.txt", "a") as myfile:
77 |                         myfile.write(img_path+"\n")
78 |                 #print('no face')
79 | 


--------------------------------------------------------------------------------
/annotations/emty:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/EmnamoR/Face-recognition-Tensorflow-object-detection-api/bf0a8563252a30106ec763310bc1fd41bcae4167/annotations/emty


--------------------------------------------------------------------------------
/celebrityDB/images here:
--------------------------------------------------------------------------------
1 | images in this folder
2 | 
3 | 


--------------------------------------------------------------------------------
/data/label.pbtxt:
--------------------------------------------------------------------------------
 1 | item {
 2 |   id: 1
 3 |   name: 'AndySerkis'
 4 | }
 5 | 
 6 | item {
 7 |   id: 2
 8 |   name: 'dewayneJohnson'
 9 | }
10 | item {
11 |   id: 3
12 |   name: 'drake'
13 | }
14 | 
15 | item {
16 |   id: 4
17 |   name: 'jayZ'
18 | }
19 | 
20 | item {
21 |   id: 5
22 |   name: 'justinBieber'
23 | }
24 | 
25 | item {
26 |   id: 6
27 |   name: 'kimKardashian'
28 | }
29 | 
30 | 
31 | item {
32 |   id: 7
33 |   name: 'kimKardashian'
34 | }
35 | 
36 | item {
37 |   id: 8
38 |   name: 'tomHanks'
39 | }
40 | 
41 | item {
42 |   id: 9
43 |   name: 'willSmith'
44 | }
45 | 


--------------------------------------------------------------------------------
/data/ssd_mobilenet_v1_coco.config:
--------------------------------------------------------------------------------
  1 | # SSD with Mobilenet v1 configuration for MSCOCO Dataset.
  2 | # Users should configure the fine_tune_checkpoint field in the train config as
  3 | # well as the label_map_path and input_path fields in the train_input_reader and
  4 | # eval_input_reader. Search for "PATH_TO_BE_CONFIGURED" to find the fields that
  5 | # should be configured.
  6 | 
  7 | model {
  8 |   ssd {
  9 |     num_classes: 9
 10 |     box_coder {
 11 |       faster_rcnn_box_coder {
 12 |         y_scale: 10.0
 13 |         x_scale: 10.0
 14 |         height_scale: 5.0
 15 |         width_scale: 5.0
 16 |       }
 17 |     }
 18 |     matcher {
 19 |       argmax_matcher {
 20 |         matched_threshold: 0.5
 21 |         unmatched_threshold: 0.5
 22 |         ignore_thresholds: false
 23 |         negatives_lower_than_unmatched: true
 24 |         force_match_for_each_row: true
 25 |       }
 26 |     }
 27 |     similarity_calculator {
 28 |       iou_similarity {
 29 |       }
 30 |     }
 31 |     anchor_generator {
 32 |       ssd_anchor_generator {
 33 |         num_layers: 6
 34 |         min_scale: 0.2
 35 |         max_scale: 0.95
 36 |         aspect_ratios: 1.0
 37 |         aspect_ratios: 2.0
 38 |         aspect_ratios: 0.5
 39 |         aspect_ratios: 3.0
 40 |         aspect_ratios: 0.3333
 41 |       }
 42 |     }
 43 |     image_resizer {
 44 |       fixed_shape_resizer {
 45 |         height: 300
 46 |         width: 300
 47 |       }
 48 |     }
 49 |     box_predictor {
 50 |       convolutional_box_predictor {
 51 |         min_depth: 0
 52 |         max_depth: 0
 53 |         num_layers_before_predictor: 0
 54 |         use_dropout: false
 55 |         dropout_keep_probability: 0.8
 56 |         kernel_size: 1
 57 |         box_code_size: 4
 58 |         apply_sigmoid_to_scores: false
 59 |         conv_hyperparams {
 60 |           activation: RELU_6,
 61 |           regularizer {
 62 |             l2_regularizer {
 63 |               weight: 0.00004
 64 |             }
 65 |           }
 66 |           initializer {
 67 |             truncated_normal_initializer {
 68 |               stddev: 0.03
 69 |               mean: 0.0
 70 |             }
 71 |           }
 72 |           batch_norm {
 73 |             train: true,
 74 |             scale: true,
 75 |             center: true,
 76 |             decay: 0.9997,
 77 |             epsilon: 0.001,
 78 |           }
 79 |         }
 80 |       }
 81 |     }
 82 |     feature_extractor {
 83 |       type: 'ssd_mobilenet_v1'
 84 |       min_depth: 16
 85 |       depth_multiplier: 1.0
 86 |       conv_hyperparams {
 87 |         activation: RELU_6,
 88 |         regularizer {
 89 |           l2_regularizer {
 90 |             weight: 0.00004
 91 |           }
 92 |         }
 93 |         initializer {
 94 |           truncated_normal_initializer {
 95 |             stddev: 0.03
 96 |             mean: 0.0
 97 |           }
 98 |         }
 99 |         batch_norm {
100 |           train: true,
101 |           scale: true,
102 |           center: true,
103 |           decay: 0.9997,
104 |           epsilon: 0.001,
105 |         }
106 |       }
107 |     }
108 |     loss {
109 |       classification_loss {
110 |         weighted_sigmoid {
111 |         }
112 |       }
113 |       localization_loss {
114 |         weighted_smooth_l1 {
115 |         }
116 |       }
117 |       hard_example_miner {
118 |         num_hard_examples: 3000
119 |         iou_threshold: 0.99
120 |         loss_type: CLASSIFICATION
121 |         max_negatives_per_positive: 3
122 |         min_negatives_per_image: 0
123 |       }
124 |       classification_weight: 1.0
125 |       localization_weight: 1.0
126 |     }
127 |     normalize_loss_by_num_matches: true
128 |     post_processing {
129 |       batch_non_max_suppression {
130 |         score_threshold: 1e-8
131 |         iou_threshold: 0.6
132 |         max_detections_per_class: 100
133 |         max_total_detections: 100
134 |       }
135 |       score_converter: SIGMOID
136 |     }
137 |   }
138 | }
139 | 
140 | train_config: {
141 |   batch_size: 24
142 |   optimizer {
143 |     rms_prop_optimizer: {
144 |       learning_rate: {
145 |         exponential_decay_learning_rate {
146 |           initial_learning_rate: 0.004
147 |           decay_steps: 800720
148 |           decay_factor: 0.95
149 |         }
150 |       }
151 |       momentum_optimizer_value: 0.9
152 |       decay: 0.9
153 |       epsilon: 1.0
154 |     }
155 |   }
156 |   fine_tune_checkpoint: "PATH_TO_BE_CONFIGURED(the ssd folder we just downloaded)/model.ckpt"
157 |   from_detection_checkpoint: true
158 |   # Note: The below line limits the training process to 200K steps, which we
159 |   # empirically found to be sufficient enough to train the pets dataset. This
160 |   # effectively bypasses the learning rate schedule (the learning rate will
161 |   # never decay). Remove the below line to train indefinitely.
162 |   num_steps: 200000
163 |   data_augmentation_options {
164 |     random_horizontal_flip {
165 |     }
166 |   }
167 |   data_augmentation_options {
168 |     ssd_random_crop {
169 |     }
170 |   }
171 | }
172 | 
173 | train_input_reader: {
174 |   tf_record_input_reader {
175 |     input_path: "PATH_TO_BE_CONFIGURED/train.record"
176 |   }
177 |   label_map_path: "PATH_TO_BE_CONFIGURED/label.pbtxt"
178 | }
179 | 
180 | eval_config: {
181 |   num_examples: 8000
182 |   # Note: The below line limits the evaluation process to 10 evaluations.
183 |   # Remove the below line to evaluate indefinitely.
184 |   max_evals: 10
185 | }
186 | 
187 | eval_input_reader: {
188 |   tf_record_input_reader {
189 |     input_path: "PATH_TO_BE_CONFIGURED/test.record"
190 |   }
191 |   label_map_path: "PATH_TO_BE_CONFIGURED/label.pbtxt"
192 |   shuffle: false
193 |   num_readers: 1
194 | }
195 | 


--------------------------------------------------------------------------------
/generate_tfrecords.py:
--------------------------------------------------------------------------------
  1 | """
  2 | Usage:
  3 |   # From tensorflow/models/
  4 |   # Create train data:
  5 |   python generate_tfrecord.py --csv_input=train.csv  --output_path=data/train.record
  6 | 
  7 |   # Create test data:
  8 |   python generate_tfrecord.py --csv_input=test.csv  --output_path=data/test.record
  9 | """
 10 | from __future__ import division
 11 | from __future__ import print_function
 12 | from __future__ import absolute_import
 13 | 
 14 | import os
 15 | import io
 16 | import pandas as pd
 17 | import tensorflow as tf
 18 | 
 19 | from PIL import Image
 20 | from models.research.object_detection.utils import dataset_util
 21 | from collections import namedtuple, OrderedDict
 22 | 
 23 | flags = tf.app.flags
 24 | flags.DEFINE_string('csv_input', '', 'Path to the CSV input')
 25 | flags.DEFINE_string('output_path', '', 'Path to output TFRecord')
 26 | FLAGS = flags.FLAGS
 27 | 
 28 | 
 29 | # TO-DO replace this with label map
 30 | def class_text_to_int(row_label):
 31 |     if row_label == 'AndySerkis':
 32 |         return 1
 33 |     elif row_label == 'dewayneJohnson':
 34 |         return 2
 35 |     elif row_label == 'drake':
 36 |         return 3
 37 |     elif row_label == 'jayZ':
 38 |         return 4
 39 |     elif row_label == 'justinBieber':
 40 |         return 5
 41 |     elif row_label == 'kimKardashian':
 42 |         return 6
 43 |     elif row_label == 'tomHanks':
 44 |         return 7
 45 |     elif row_label == 'willSmith':
 46 |         return 8
 47 |     else:
 48 |         None
 49 | 
 50 | 
 51 | def split(df, group):
 52 |     data = namedtuple('data', ['filename', 'object'])
 53 |     gb = df.groupby(group)
 54 |     return [data(filename, gb.get_group(x)) for filename, x in zip(gb.groups.keys(), gb.groups)]
 55 | 
 56 | 
 57 | def create_tf_example(group, path):
 58 |     #print(os.path.join(path, '{}'.format(group.filename)))
 59 |     with tf.gfile.GFile(os.path.join(path, '{}'.format(group.filename)), 'rb') as fid:
 60 |         encoded_jpg = fid.read()
 61 |     encoded_jpg_io = io.BytesIO(encoded_jpg)
 62 |     image = Image.open(encoded_jpg_io)
 63 |     width, height = image.size
 64 | 
 65 |     filename = group.filename.encode('utf8')
 66 |     image_format = b'jpg'
 67 |     xmins = []
 68 |     xmaxs = []
 69 |     ymins = []
 70 |     ymaxs = []
 71 |     classes_text = []
 72 |     classes = []
 73 | 
 74 |     for index, row in group.object.iterrows():
 75 |         xmins.append(row['xmin'] / width)
 76 |         xmaxs.append(row['xmax'] / width)
 77 |         ymins.append(row['ymin'] / height)
 78 |         ymaxs.append(row['ymax'] / height)
 79 |         classes_text.append(row['class'].encode('utf8'))
 80 |         classes.append(class_text_to_int(row['class']))
 81 | 
 82 |     tf_example = tf.train.Example(features=tf.train.Features(feature={
 83 |         'image/height': dataset_util.int64_feature(height),
 84 |         'image/width': dataset_util.int64_feature(width),
 85 |         'image/filename': dataset_util.bytes_feature(filename),
 86 |         'image/source_id': dataset_util.bytes_feature(filename),
 87 |         'image/encoded': dataset_util.bytes_feature(encoded_jpg),
 88 |         'image/format': dataset_util.bytes_feature(image_format),
 89 |         'image/object/bbox/xmin': dataset_util.float_list_feature(xmins),
 90 |         'image/object/bbox/xmax': dataset_util.float_list_feature(xmaxs),
 91 |         'image/object/bbox/ymin': dataset_util.float_list_feature(ymins),
 92 |         'image/object/bbox/ymax': dataset_util.float_list_feature(ymaxs),
 93 |         'image/object/class/text': dataset_util.bytes_list_feature(classes_text),
 94 |         'image/object/class/label': dataset_util.int64_list_feature(classes),
 95 |     }))
 96 |     return tf_example
 97 | 
 98 | 
 99 | def main(_):
100 |     writer = tf.python_io.TFRecordWriter(FLAGS.output_path)
101 |     path = os.path.join(os.getcwd(), 'images')
102 |     examples = pd.read_csv(FLAGS.csv_input)
103 |     grouped = split(examples, 'filename')
104 |     grouped2 = split(examples, 'class')
105 |     for dirpath,dirs,filenames in os.walk(path):
106 |         for i, val in enumerate(dirs):
107 |             #print(dirs[i])
108 |             #print((i,val))
109 | 
110 |             for d in dirs :
111 |                 path2=os.path.join(dirpath,d)
112 |                 for group in grouped:
113 | 
114 |                 
115 |                     name=group[0].split('-')
116 |                     if name[0] == d :
117 |                         tf_example = create_tf_example(group, path2)
118 |                         t=path2+'/'+group[0]
119 |                         print(os.path.join(group,t))
120 |                         writer.write(tf_example.SerializeToString())
121 |             
122 |     writer.close()
123 |     output_path = os.path.join(os.getcwd(), FLAGS.output_path)
124 |     print('Successfully created the TFRecords: {}'.format(output_path))
125 | 
126 | if __name__ == '__main__':
127 |     tf.app.run()
128 | 


--------------------------------------------------------------------------------
/it5.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/EmnamoR/Face-recognition-Tensorflow-object-detection-api/bf0a8563252a30106ec763310bc1fd41bcae4167/it5.jpg


--------------------------------------------------------------------------------
/lbpcascade_frontalface_improved.xml:
--------------------------------------------------------------------------------
   1 | <?xml version="1.0"?>
   2 | <!--
   3 | //////////////////////////////////////////////////////////////////////////
   4 | | Contributors License Agreement
   5 | | IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
   6 | |   By downloading, copying, installing or using the software you agree
   7 | |   to this license. If you do not agree to this license, do not download,
   8 | |   install, copy or use the software.
   9 | |
  10 | | Copyright (c) 2017, Puttemans Steven, Can Ergun and Toon Goedeme
  11 | | (KU Leuven, EAVISE Research Group, Jan Pieter De Nayerlaan 5,
  12 | | Sint-Katelijne-Waver, Belgium).
  13 | | All rights reserved.
  14 | |
  15 | | Redistribution and use in source and binary forms, with or without
  16 | | modification, are permitted provided that the following conditions are
  17 | | met:
  18 | |
  19 | |    * Redistributions of source code must retain the above copyright
  20 | |       notice, this list of conditions and the following disclaimer.
  21 | |    * Redistributions in binary form must reproduce the above
  22 | |      copyright notice, this list of conditions and the following
  23 | |      disclaimer in the documentation and/or other materials provided
  24 | |      with the distribution.
  25 | |    * The name of Contributor may not used to endorse or promote products
  26 | |      derived from this software without specific prior written permission.
  27 | |
  28 | | THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
  29 | | "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  30 | | LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
  31 | | A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
  32 | | CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
  33 | | EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
  34 | | PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
  35 | | PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
  36 | | LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
  37 | | NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
  38 | | SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  39 | //////////////////////////////////////////////////////////////////////////
  40 | 
  41 | =====================================================================
  42 | Improving Open Source Face Detection by Combining an Adapted Cascade
  43 | Classification Pipeline and Active Learning
  44 | =====================================================================
  45 | by Puttemans Steven, Can Ergun and Toon Goedeme (KU Leuven, EAVISE, Belgium)
  46 | 
  47 | This model is the best performing IterativeHardPositives+ frontal face detection model
  48 | used in the research paper presented at VISAPP2017, Porto, Portugal.
  49 | 
  50 | RESEARCHERS:
  51 | If you are using the improved face detection model or involved ideas please cite
  52 | this paper (available at http://eavise.be/publications_lirias.php):
  53 | 
  54 | @InProceedings{PuttemansVISAPP2017,
  55 |   author =       "Puttemans Steven, Can Ergun and Toon Goedeme",
  56 |   title =        "Improving Open Source Face Detection by Combining an Adapted Cascade Classification Pipeline and Active Learning"
  57 |   booktitle =    "12th International Conference on Computer Vision Theory and Applications"
  58 |   year =         "2017",
  59 |   month =        "February"
  60 | }
  61 | 
  62 | COMMERCIAL:
  63 | If you have any commercial interest in this work please contact
  64 | steven.puttemans@kuleuven.be or toon.goedeme@kuleuven.be
  65 | -->
  66 | 
  67 | <opencv_storage>
  68 | <cascade>
  69 |   <stageType>BOOST</stageType>
  70 |   <featureType>LBP</featureType>
  71 |   <height>45</height>
  72 |   <width>45</width>
  73 |   <stageParams>
  74 |     <boostType>GAB</boostType>
  75 |     <minHitRate>9.9500000476837158e-001</minHitRate>
  76 |     <maxFalseAlarm>5.0000000000000000e-001</maxFalseAlarm>
  77 |     <weightTrimRate>9.4999999999999996e-001</weightTrimRate>
  78 |     <maxDepth>1</maxDepth>
  79 |     <maxWeakCount>100</maxWeakCount></stageParams>
  80 |   <featureParams>
  81 |     <maxCatCount>256</maxCatCount>
  82 |     <featSize>1</featSize></featureParams>
  83 |   <stageNum>19</stageNum>
  84 |   <stages>
  85 |     <!-- stage 0 -->
  86 |     <_>
  87 |       <maxWeakCount>6</maxWeakCount>
  88 |       <stageThreshold>-4.1617846488952637e+000</stageThreshold>
  89 |       <weakClassifiers>
  90 |         <_>
  91 |           <internalNodes>
  92 |             0 -1 26 -1 -1 -17409 -1 -1 -1 -1 -1</internalNodes>
  93 |           <leafValues>
  94 |             -9.9726462364196777e-001 -3.8938775658607483e-001</leafValues></_>
  95 |         <_>
  96 |           <internalNodes>
  97 |             0 -1 18 -1 -1 -21569 -20545 -1 -1 -20545 -1</internalNodes>
  98 |           <leafValues>
  99 |             -9.8648911714553833e-001 -2.5386649370193481e-001</leafValues></_>
 100 |         <_>
 101 |           <internalNodes>
 102 |             0 -1 30 -21569 -16449 1006578219 -20801 -16449 -1 -21585 -1</internalNodes>
 103 |           <leafValues>
 104 |             -9.6436238288879395e-001 -1.4039695262908936e-001</leafValues></_>
 105 |         <_>
 106 |           <internalNodes>
 107 |             0 -1 54 -1 -1 -16402 -4370 -1 -1 -1053010 -4456466</internalNodes>
 108 |           <leafValues>
 109 |             -8.4081345796585083e-001 3.8321062922477722e-001</leafValues></_>
 110 |         <_>
 111 |           <internalNodes>
 112 |             0 -1 29 -184747280 -705314819 1326353 1364574079 -131073 -5
 113 |             2147481147 -1</internalNodes>
 114 |           <leafValues>
 115 |             -8.1084597110748291e-001 4.3495711684226990e-001</leafValues></_>
 116 |         <_>
 117 |           <internalNodes>
 118 |             0 -1 89 -142618625 -4097 -37269 -20933 872350430 -268476417
 119 |             1207894255 2139032115</internalNodes>
 120 |           <leafValues>
 121 |             -7.3140043020248413e-001 4.3799084424972534e-001</leafValues></_></weakClassifiers></_>
 122 |     <!-- stage 1 -->
 123 |     <_>
 124 |       <maxWeakCount>6</maxWeakCount>
 125 |       <stageThreshold>-4.0652265548706055e+000</stageThreshold>
 126 |       <weakClassifiers>
 127 |         <_>
 128 |           <internalNodes>
 129 |             0 -1 19 -1 -1 -17409 -1 -1 -1 -1 -1</internalNodes>
 130 |           <leafValues>
 131 |             -9.9727255105972290e-001 -7.2050148248672485e-001</leafValues></_>
 132 |         <_>
 133 |           <internalNodes>
 134 |             0 -1 38 -1 1073741823 -1 -1 -1 -1 -1 -1</internalNodes>
 135 |           <leafValues>
 136 |             -9.8717331886291504e-001 -5.3031939268112183e-001</leafValues></_>
 137 |         <_>
 138 |           <internalNodes>
 139 |             0 -1 28 -16385 -1 -21569 -20545 -1 -1 -21569 -1</internalNodes>
 140 |           <leafValues>
 141 |             -9.3442338705062866e-001 6.5213099122047424e-002</leafValues></_>
 142 |         <_>
 143 |           <internalNodes>
 144 |             0 -1 112 -2097153 -1 -1 -1 -1 -8193 -1 -35467</internalNodes>
 145 |           <leafValues>
 146 |             -7.9567342996597290e-001 4.2883640527725220e-001</leafValues></_>
 147 |         <_>
 148 |           <internalNodes>
 149 |             0 -1 48 -134239573 -16465 58663467 -1079022929 -1073758273
 150 |             -81937 -8412501 -404766817</internalNodes>
 151 |           <leafValues>
 152 |             -7.1264797449111938e-001 4.1050794720649719e-001</leafValues></_>
 153 |         <_>
 154 |           <internalNodes>
 155 |             0 -1 66 -17047555 -1099008003 2147479551 -1090584581 -69633
 156 |             -1342177281 -1090650121 -1472692240</internalNodes>
 157 |           <leafValues>
 158 |             -7.6119172573089600e-001 4.2042696475982666e-001</leafValues></_></weakClassifiers></_>
 159 |     <!-- stage 2 -->
 160 |     <_>
 161 |       <maxWeakCount>7</maxWeakCount>
 162 |       <stageThreshold>-4.6904473304748535e+000</stageThreshold>
 163 |       <weakClassifiers>
 164 |         <_>
 165 |           <internalNodes>
 166 |             0 -1 12 -1 -1 -17409 -1 -1 -1 -1 -1</internalNodes>
 167 |           <leafValues>
 168 |             -9.9725550413131714e-001 -8.3142280578613281e-001</leafValues></_>
 169 |         <_>
 170 |           <internalNodes>
 171 |             0 -1 31 -1 -168429569 -1 -1 -1 -1 -1 -1</internalNodes>
 172 |           <leafValues>
 173 |             -9.8183268308639526e-001 -3.6373397707939148e-001</leafValues></_>
 174 |         <_>
 175 |           <internalNodes>
 176 |             0 -1 38 -1 1073741759 -1 -1 -1 -1 -1 -1</internalNodes>
 177 |           <leafValues>
 178 |             -9.1890293359756470e-001 7.8322596848011017e-002</leafValues></_>
 179 |         <_>
 180 |           <internalNodes>
 181 |             0 -1 27 -17409 -2097153 -134372726 -21873 -65 -536870913
 182 |             -161109 -4215889</internalNodes>
 183 |           <leafValues>
 184 |             -8.0752444267272949e-001 1.9565649330615997e-001</leafValues></_>
 185 |         <_>
 186 |           <internalNodes>
 187 |             0 -1 46 -469779457 -286371842 -33619971 -212993 -1 -41943049
 188 |             -134217731 -1346863620</internalNodes>
 189 |           <leafValues>
 190 |             -6.9232726097106934e-001 3.8141927123069763e-001</leafValues></_>
 191 |         <_>
 192 |           <internalNodes>
 193 |             0 -1 125 -1896950780 -1964839052 -9 707723004 -34078727
 194 |             -1074266122 -536872969 -262145</internalNodes>
 195 |           <leafValues>
 196 |             -8.1760478019714355e-001 3.4172961115837097e-001</leafValues></_>
 197 |         <_>
 198 |           <internalNodes>
 199 |             0 -1 80 -402657501 654311423 -419533278 -452984853
 200 |             1979676215 -1208090625 -167772569 -524289</internalNodes>
 201 |           <leafValues>
 202 |             -6.3433408737182617e-001 4.3154156208038330e-001</leafValues></_></weakClassifiers></_>
 203 |     <!-- stage 3 -->
 204 |     <_>
 205 |       <maxWeakCount>8</maxWeakCount>
 206 |       <stageThreshold>-4.2590322494506836e+000</stageThreshold>
 207 |       <weakClassifiers>
 208 |         <_>
 209 |           <internalNodes>
 210 |             0 -1 42 -1 -655361 -1 -1 -1 -1 -1 -1</internalNodes>
 211 |           <leafValues>
 212 |             -9.9715477228164673e-001 -8.6178696155548096e-001</leafValues></_>
 213 |         <_>
 214 |           <internalNodes>
 215 |             0 -1 40 -1 -705300491 -1 -1 -1 -1 -1 -1</internalNodes>
 216 |           <leafValues>
 217 |             -9.8356908559799194e-001 -5.7423096895217896e-001</leafValues></_>
 218 |         <_>
 219 |           <internalNodes>
 220 |             0 -1 43 -65 872413111 -2049 -1 -1 -1 -1 -1</internalNodes>
 221 |           <leafValues>
 222 |             -9.2525935173034668e-001 -1.3835857808589935e-001</leafValues></_>
 223 |         <_>
 224 |           <internalNodes>
 225 |             0 -1 111 -1 -5242881 -1 -524289 -4194305 -1 -1 -43148</internalNodes>
 226 |           <leafValues>
 227 |             -7.8076487779617310e-001 1.8362471461296082e-001</leafValues></_>
 228 |         <_>
 229 |           <internalNodes>
 230 |             0 -1 25 -145227841 868203194 -1627394049 935050171
 231 |             2147483647 1006600191 -268439637 1002437615</internalNodes>
 232 |           <leafValues>
 233 |             -7.2554033994674683e-001 3.3393219113349915e-001</leafValues></_>
 234 |         <_>
 235 |           <internalNodes>
 236 |             0 -1 116 -214961408 50592514 -2128 1072162674 -1077940293
 237 |             -1084489966 -134219854 -1074790401</internalNodes>
 238 |           <leafValues>
 239 |             -6.1547595262527466e-001 3.9214438199996948e-001</leafValues></_>
 240 |         <_>
 241 |           <internalNodes>
 242 |             0 -1 3 -294987948 -1124421633 -73729 -268435841 -33654928
 243 |             2122317823 -268599297 -33554945</internalNodes>
 244 |           <leafValues>
 245 |             -6.4863425493240356e-001 3.8784855604171753e-001</leafValues></_>
 246 |         <_>
 247 |           <internalNodes>
 248 |             0 -1 22 -525585 -26738821 -17895690 1123482236 1996455758
 249 |             -8519849 -252182980 -461898753</internalNodes>
 250 |           <leafValues>
 251 |             -5.5464369058609009e-001 4.4275921583175659e-001</leafValues></_></weakClassifiers></_>
 252 |     <!-- stage 4 -->
 253 |     <_>
 254 |       <maxWeakCount>8</maxWeakCount>
 255 |       <stageThreshold>-4.0009465217590332e+000</stageThreshold>
 256 |       <weakClassifiers>
 257 |         <_>
 258 |           <internalNodes>
 259 |             0 -1 82 -1 -1 -1 -1 -33685505 -1 -1 -1</internalNodes>
 260 |           <leafValues>
 261 |             -9.9707120656967163e-001 -8.9196771383285522e-001</leafValues></_>
 262 |         <_>
 263 |           <internalNodes>
 264 |             0 -1 84 -1 -1 -1 -1 2147446783 -1 -1 -1</internalNodes>
 265 |           <leafValues>
 266 |             -9.8670446872711182e-001 -7.5064390897750854e-001</leafValues></_>
 267 |         <_>
 268 |           <internalNodes>
 269 |             0 -1 79 -1 -1 -262145 -1 -252379137 -1 -1 -1</internalNodes>
 270 |           <leafValues>
 271 |             -8.9446705579757690e-001 7.0268943905830383e-002</leafValues></_>
 272 |         <_>
 273 |           <internalNodes>
 274 |             0 -1 61 -1 -8201 -1 -2097153 -16777217 -513 -16777217
 275 |             -1162149889</internalNodes>
 276 |           <leafValues>
 277 |             -7.2166109085083008e-001 2.9786801338195801e-001</leafValues></_>
 278 |         <_>
 279 |           <internalNodes>
 280 |             0 -1 30 -21569 -1069121 1006578211 -134238545 -16450
 281 |             -268599297 -21617 -14680097</internalNodes>
 282 |           <leafValues>
 283 |             -6.2449234724044800e-001 3.8551881909370422e-001</leafValues></_>
 284 |         <_>
 285 |           <internalNodes>
 286 |             0 -1 75 -268701913 -1999962377 1995165474 -453316822
 287 |             1744684853 -2063597697 -134226057 -50336769</internalNodes>
 288 |           <leafValues>
 289 |             -5.5207914113998413e-001 4.2211884260177612e-001</leafValues></_>
 290 |         <_>
 291 |           <internalNodes>
 292 |             0 -1 21 -352321825 -526489 -420020626 -486605074 1155483470
 293 |             -110104705 -587840772 -25428801</internalNodes>
 294 |           <leafValues>
 295 |             -5.3324747085571289e-001 4.4535955786705017e-001</leafValues></_>
 296 |         <_>
 297 |           <internalNodes>
 298 |             0 -1 103 70270772 2012790229 -16810020 -245764 -1208090635
 299 |             -753667 -1073741828 -1363662420</internalNodes>
 300 |           <leafValues>
 301 |             -6.4402890205383301e-001 3.8995954394340515e-001</leafValues></_></weakClassifiers></_>
 302 |     <!-- stage 5 -->
 303 |     <_>
 304 |       <maxWeakCount>8</maxWeakCount>
 305 |       <stageThreshold>-4.6897511482238770e+000</stageThreshold>
 306 |       <weakClassifiers>
 307 |         <_>
 308 |           <internalNodes>
 309 |             0 -1 97 -1 -1 -1 -1 -524289 -524289 -1 -1</internalNodes>
 310 |           <leafValues>
 311 |             -9.9684870243072510e-001 -8.8232177495956421e-001</leafValues></_>
 312 |         <_>
 313 |           <internalNodes>
 314 |             0 -1 84 -1 -1 -1 -1 2147438591 -1 -1 -1</internalNodes>
 315 |           <leafValues>
 316 |             -9.8677414655685425e-001 -7.8965580463409424e-001</leafValues></_>
 317 |         <_>
 318 |           <internalNodes>
 319 |             0 -1 113 -1 -1 -1 -1 -1048577 -262149 -1048577 -35339</internalNodes>
 320 |           <leafValues>
 321 |             -9.2621946334838867e-001 -2.9984828829765320e-001</leafValues></_>
 322 |         <_>
 323 |           <internalNodes>
 324 |             0 -1 33 -2249 867434291 -32769 -33562753 -1 -1073758209
 325 |             -4165 -1</internalNodes>
 326 |           <leafValues>
 327 |             -7.2429555654525757e-001 2.2348840534687042e-001</leafValues></_>
 328 |         <_>
 329 |           <internalNodes>
 330 |             0 -1 98 1659068671 -142606337 587132538 -67108993 577718271
 331 |             -294921 -134479873 -129</internalNodes>
 332 |           <leafValues>
 333 |             -5.5495566129684448e-001 3.5419258475303650e-001</leafValues></_>
 334 |         <_>
 335 |           <internalNodes>
 336 |             0 -1 100 -268441813 788267007 -286265494 -486576145 -8920251
 337 |             2138505075 -151652570 -2050</internalNodes>
 338 |           <leafValues>
 339 |             -5.3362584114074707e-001 3.9479774236679077e-001</leafValues></_>
 340 |         <_>
 341 |           <internalNodes>
 342 |             0 -1 51 -1368387212 -537102978 -98305 -163843 1065109500
 343 |             -16777217 -67321939 -1141359619</internalNodes>
 344 |           <leafValues>
 345 |             -5.6162708997726440e-001 3.8008108735084534e-001</leafValues></_>
 346 |         <_>
 347 |           <internalNodes>
 348 |             0 -1 127 -268435550 1781120906 -251658720 -143130698
 349 |             -1048605 -1887436825 1979700688 -1008730125</internalNodes>
 350 |           <leafValues>
 351 |             -5.1167154312133789e-001 4.0678605437278748e-001</leafValues></_></weakClassifiers></_>
 352 |     <!-- stage 6 -->
 353 |     <_>
 354 |       <maxWeakCount>10</maxWeakCount>
 355 |       <stageThreshold>-4.2179841995239258e+000</stageThreshold>
 356 |       <weakClassifiers>
 357 |         <_>
 358 |           <internalNodes>
 359 |             0 -1 97 -1 -1 -1 -1 -524289 -524289 -1 -1</internalNodes>
 360 |           <leafValues>
 361 |             -9.9685418605804443e-001 -8.8037383556365967e-001</leafValues></_>
 362 |         <_>
 363 |           <internalNodes>
 364 |             0 -1 90 -1 -1 -1 -1 -8912897 -524297 -8912897 -1</internalNodes>
 365 |           <leafValues>
 366 |             -9.7972750663757324e-001 -5.7626229524612427e-001</leafValues></_>
 367 |         <_>
 368 |           <internalNodes>
 369 |             0 -1 96 -1 -1 -1 -1 -1 -65 -1 -2249</internalNodes>
 370 |           <leafValues>
 371 |             -9.0239793062210083e-001 -1.7454113066196442e-001</leafValues></_>
 372 |         <_>
 373 |           <internalNodes>
 374 |             0 -1 71 -1 -4097 -1 -513 -16777217 -268468483 -16797697
 375 |             -1430589697</internalNodes>
 376 |           <leafValues>
 377 |             -7.4346423149108887e-001 9.4165161252021790e-002</leafValues></_>
 378 |         <_>
 379 |           <internalNodes>
 380 |             0 -1 37 1364588304 -581845274 -536936460 -3 -308936705
 381 |             -1074331649 -4196865 -134225953</internalNodes>
 382 |           <leafValues>
 383 |             -6.8877440690994263e-001 2.7647304534912109e-001</leafValues></_>
 384 |         <_>
 385 |           <internalNodes>
 386 |             0 -1 117 -37765187 -540675 -3 -327753 -1082458115 -65537
 387 |             1071611901 536827253</internalNodes>
 388 |           <leafValues>
 389 |             -5.7555085420608521e-001 3.4339720010757446e-001</leafValues></_>
 390 |         <_>
 391 |           <internalNodes>
 392 |             0 -1 85 -269490650 -1561395522 -1343312090 -857083986
 393 |             -1073750223 -369098755 -50856110 -2065</internalNodes>
 394 |           <leafValues>
 395 |             -5.4036927223205566e-001 4.0065473318099976e-001</leafValues></_>
 396 |         <_>
 397 |           <internalNodes>
 398 |             0 -1 4 -425668880 -34427164 1879048177 -269570140 790740912
 399 |             -196740 2138535839 -536918145</internalNodes>
 400 |           <leafValues>
 401 |             -4.8439365625381470e-001 4.4630467891693115e-001</leafValues></_>
 402 |         <_>
 403 |           <internalNodes>
 404 |             0 -1 92 74726960 -1246482434 -1 -246017 -1078607916
 405 |             -1073947163 -1644231687 -1359211496</internalNodes>
 406 |           <leafValues>
 407 |             -5.6686979532241821e-001 3.6671569943428040e-001</leafValues></_>
 408 |         <_>
 409 |           <internalNodes>
 410 |             0 -1 11 -135274809 -1158173459 -353176850 540195262
 411 |             2139086600 2071977814 -546898600 -96272673</internalNodes>
 412 |           <leafValues>
 413 |             -5.1499199867248535e-001 4.0788397192955017e-001</leafValues></_></weakClassifiers></_>
 414 |     <!-- stage 7 -->
 415 |     <_>
 416 |       <maxWeakCount>9</maxWeakCount>
 417 |       <stageThreshold>-4.0345416069030762e+000</stageThreshold>
 418 |       <weakClassifiers>
 419 |         <_>
 420 |           <internalNodes>
 421 |             0 -1 78 -1 -1 -1 -1 -8912897 -1 -8912897 -1</internalNodes>
 422 |           <leafValues>
 423 |             -9.9573624134063721e-001 -8.5452395677566528e-001</leafValues></_>
 424 |         <_>
 425 |           <internalNodes>
 426 |             0 -1 93 -1 -1 -1 -1 -148635649 -524297 -8912897 -1</internalNodes>
 427 |           <leafValues>
 428 |             -9.7307401895523071e-001 -5.2884924411773682e-001</leafValues></_>
 429 |         <_>
 430 |           <internalNodes>
 431 |             0 -1 77 -1 -8209 -1 -257 -772734977 -1 -201850881 -1</internalNodes>
 432 |           <leafValues>
 433 |             -8.6225658655166626e-001 4.3712578713893890e-002</leafValues></_>
 434 |         <_>
 435 |           <internalNodes>
 436 |             0 -1 68 -570427393 -16649 -69633 -131073 -536944677 -1 -8737
 437 |             -1435828225</internalNodes>
 438 |           <leafValues>
 439 |             -6.8078064918518066e-001 2.5120577216148376e-001</leafValues></_>
 440 |         <_>
 441 |           <internalNodes>
 442 |             0 -1 50 -1179697 -34082849 -3278356 -37429266 -1048578
 443 |             -555753474 -1015551096 -37489685</internalNodes>
 444 |           <leafValues>
 445 |             -6.1699724197387695e-001 3.0963841080665588e-001</leafValues></_>
 446 |         <_>
 447 |           <internalNodes>
 448 |             0 -1 129 -1931606992 -17548804 -16842753 -1075021827
 449 |             1073667572 -81921 -1611073620 -1415047752</internalNodes>
 450 |           <leafValues>
 451 |             -6.0499197244644165e-001 3.0735063552856445e-001</leafValues></_>
 452 |         <_>
 453 |           <internalNodes>
 454 |             0 -1 136 -269754813 1761591286 -1073811523 2130378623 -17580
 455 |             -1082294665 -159514800 -1026883840</internalNodes>
 456 |           <leafValues>
 457 |             -5.6772041320800781e-001 3.5023149847984314e-001</leafValues></_>
 458 |         <_>
 459 |           <internalNodes>
 460 |             0 -1 65 2016561683 1528827871 -10258447 960184191 125476830
 461 |             -8511618 -1078239365 187648611</internalNodes>
 462 |           <leafValues>
 463 |             -5.5894804000854492e-001 3.4856522083282471e-001</leafValues></_>
 464 |         <_>
 465 |           <internalNodes>
 466 |             0 -1 13 -207423502 -333902 2013200231 -202348848 1042454451
 467 |             -16393 1073117139 2004162321</internalNodes>
 468 |           <leafValues>
 469 |             -5.7197356224060059e-001 3.2818377017974854e-001</leafValues></_></weakClassifiers></_>
 470 |     <!-- stage 8 -->
 471 |     <_>
 472 |       <maxWeakCount>9</maxWeakCount>
 473 |       <stageThreshold>-3.4892759323120117e+000</stageThreshold>
 474 |       <weakClassifiers>
 475 |         <_>
 476 |           <internalNodes>
 477 |             0 -1 78 -1 -1 -1 -1 -8912897 -1 -8912897 -1</internalNodes>
 478 |           <leafValues>
 479 |             -9.8917990922927856e-001 -7.3812037706375122e-001</leafValues></_>
 480 |         <_>
 481 |           <internalNodes>
 482 |             0 -1 93 -1 -1 -1 -1 -148635649 -524297 -8912897 -1</internalNodes>
 483 |           <leafValues>
 484 |             -9.3414896726608276e-001 -2.6945295929908752e-001</leafValues></_>
 485 |         <_>
 486 |           <internalNodes>
 487 |             0 -1 83 -1 -524289 -1 -1048577 1879011071 -32769 -524289
 488 |             -3178753</internalNodes>
 489 |           <leafValues>
 490 |             -7.6891708374023438e-001 5.2568886429071426e-002</leafValues></_>
 491 |         <_>
 492 |           <internalNodes>
 493 |             0 -1 9 -352329729 -17891329 -16810117 -486871042 -688128841
 494 |             -1358954675 -16777218 -219217968</internalNodes>
 495 |           <leafValues>
 496 |             -6.2337344884872437e-001 2.5143685936927795e-001</leafValues></_>
 497 |         <_>
 498 |           <internalNodes>
 499 |             0 -1 130 -2157 -1548812374 -1343233440 -418381854 -953155613
 500 |             -836960513 -713571200 -709888014</internalNodes>
 501 |           <leafValues>
 502 |             -4.7277018427848816e-001 3.9616456627845764e-001</leafValues></_>
 503 |         <_>
 504 |           <internalNodes>
 505 |             0 -1 121 -1094717701 -67240065 -65857 -32899 -5783756
 506 |             -136446081 -134285352 -2003298884</internalNodes>
 507 |           <leafValues>
 508 |             -5.1766264438629150e-001 3.5814732313156128e-001</leafValues></_>
 509 |         <_>
 510 |           <internalNodes>
 511 |             0 -1 23 -218830160 -119671186 5505075 1241491391 -1594469
 512 |             -2097185 2004828075 -67649541</internalNodes>
 513 |           <leafValues>
 514 |             -6.5394639968872070e-001 3.0377501249313354e-001</leafValues></_>
 515 |         <_>
 516 |           <internalNodes>
 517 |             0 -1 115 -551814749 2099511088 -1090732551 -2045546512
 518 |             -1086341441 1059848178 800042912 252705994</internalNodes>
 519 |           <leafValues>
 520 |             -5.2584588527679443e-001 3.3847147226333618e-001</leafValues></_>
 521 |         <_>
 522 |           <internalNodes>
 523 |             0 -1 99 -272651477 578776766 -285233490 -889225217
 524 |             2147448656 377454463 2012701952 -68157761</internalNodes>
 525 |           <leafValues>
 526 |             -6.1836904287338257e-001 2.8922611474990845e-001</leafValues></_></weakClassifiers></_>
 527 |     <!-- stage 9 -->
 528 |     <_>
 529 |       <maxWeakCount>9</maxWeakCount>
 530 |       <stageThreshold>-3.0220029354095459e+000</stageThreshold>
 531 |       <weakClassifiers>
 532 |         <_>
 533 |           <internalNodes>
 534 |             0 -1 36 -1 -570425345 -1 -570425345 -1 -50331649 -6291457 -1</internalNodes>
 535 |           <leafValues>
 536 |             -9.7703826427459717e-001 -6.2527233362197876e-001</leafValues></_>
 537 |         <_>
 538 |           <internalNodes>
 539 |             0 -1 124 -1430602241 -33619969 -1 -3 -1074003969 -1073758209
 540 |             -1073741825 -1073768705</internalNodes>
 541 |           <leafValues>
 542 |             -8.9538317918777466e-001 -3.1887885928153992e-001</leafValues></_>
 543 |         <_>
 544 |           <internalNodes>
 545 |             0 -1 88 -1 -268439625 -65601 -268439569 -393809 -270532609
 546 |             -42076889 -288361721</internalNodes>
 547 |           <leafValues>
 548 |             -6.8733429908752441e-001 1.2978810071945190e-001</leafValues></_>
 549 |         <_>
 550 |           <internalNodes>
 551 |             0 -1 132 -755049252 2042563807 1795096575 465121071
 552 |             -1090585188 -20609 -1459691784 539672495</internalNodes>
 553 |           <leafValues>
 554 |             -5.7038843631744385e-001 3.0220884084701538e-001</leafValues></_>
 555 |         <_>
 556 |           <internalNodes>
 557 |             0 -1 20 -94377762 -25702678 1694167798 -231224662 1079955016
 558 |             -346144140 2029995743 -536918961</internalNodes>
 559 |           <leafValues>
 560 |             -5.3204691410064697e-001 3.4054222702980042e-001</leafValues></_>
 561 |         <_>
 562 |           <internalNodes>
 563 |             0 -1 47 2143026943 -285278225 -3 -612438281 -16403 -131074
 564 |             -1 -1430749256</internalNodes>
 565 |           <leafValues>
 566 |             -4.6176829934120178e-001 4.1114711761474609e-001</leafValues></_>
 567 |         <_>
 568 |           <internalNodes>
 569 |             0 -1 74 203424336 -25378820 -35667973 1073360894 -1912815660
 570 |             -573444 -356583491 -1365235056</internalNodes>
 571 |           <leafValues>
 572 |             -4.9911966919898987e-001 3.5335537791252136e-001</leafValues></_>
 573 |         <_>
 574 |           <internalNodes>
 575 |             0 -1 6 -1056773 -1508430 -558153 -102747408 2133997491
 576 |             -269043865 2004842231 -8947721</internalNodes>
 577 |           <leafValues>
 578 |             -4.0219521522521973e-001 4.3947893381118774e-001</leafValues></_>
 579 |         <_>
 580 |           <internalNodes>
 581 |             0 -1 70 -880809694 -1070282769 -1363162108 -838881281
 582 |             -680395161 -2064124929 -34244753 1173880701</internalNodes>
 583 |           <leafValues>
 584 |             -5.3891533613204956e-001 3.2062566280364990e-001</leafValues></_></weakClassifiers></_>
 585 |     <!-- stage 10 -->
 586 |     <_>
 587 |       <maxWeakCount>8</maxWeakCount>
 588 |       <stageThreshold>-2.5489892959594727e+000</stageThreshold>
 589 |       <weakClassifiers>
 590 |         <_>
 591 |           <internalNodes>
 592 |             0 -1 39 -1 -572522497 -8519681 -570425345 -4195329 -50333249
 593 |             -1 -1</internalNodes>
 594 |           <leafValues>
 595 |             -9.4647216796875000e-001 -3.3662387728691101e-001</leafValues></_>
 596 |         <_>
 597 |           <internalNodes>
 598 |             0 -1 124 -1430735362 -33619971 -8201 -3 -1677983745
 599 |             -1073762817 -1074003969 -1142979329</internalNodes>
 600 |           <leafValues>
 601 |             -8.0300611257553101e-001 -3.8466516882181168e-002</leafValues></_>
 602 |         <_>
 603 |           <internalNodes>
 604 |             0 -1 91 -67113217 -524289 -671482265 -786461 1677132031
 605 |             -268473345 -68005889 -70291765</internalNodes>
 606 |           <leafValues>
 607 |             -5.8367580175399780e-001 2.6507318019866943e-001</leafValues></_>
 608 |         <_>
 609 |           <internalNodes>
 610 |             0 -1 17 -277872641 -553910292 -268435458 -16843010
 611 |             1542420439 -1342178311 -143132940 -2834</internalNodes>
 612 |           <leafValues>
 613 |             -4.6897178888320923e-001 3.7864661216735840e-001</leafValues></_>
 614 |         <_>
 615 |           <internalNodes>
 616 |             0 -1 137 -1312789 -290527285 -286326862 -5505280 -1712335966
 617 |             -2045979188 1165423617 -709363723</internalNodes>
 618 |           <leafValues>
 619 |             -4.6382644772529602e-001 3.6114525794982910e-001</leafValues></_>
 620 |         <_>
 621 |           <internalNodes>
 622 |             0 -1 106 1355856590 -109445156 -96665606 2066939898
 623 |             1356084692 1549031917 -30146561 -16581701</internalNodes>
 624 |           <leafValues>
 625 |             -6.3095021247863770e-001 2.9294869303703308e-001</leafValues></_>
 626 |         <_>
 627 |           <internalNodes>
 628 |             0 -1 104 -335555328 118529 1860167712 -810680357 -33558656
 629 |             -1368391795 -402663552 -1343225921</internalNodes>
 630 |           <leafValues>
 631 |             -5.9658926725387573e-001 2.7228885889053345e-001</leafValues></_>
 632 |         <_>
 633 |           <internalNodes>
 634 |             0 -1 76 217581168 -538349634 1062631419 1039868926
 635 |             -1090707460 -2228359 -1078042693 -1147128518</internalNodes>
 636 |           <leafValues>
 637 |             -4.5812287926673889e-001 3.7063929438591003e-001</leafValues></_></weakClassifiers></_>
 638 |     <!-- stage 11 -->
 639 |     <_>
 640 |       <maxWeakCount>9</maxWeakCount>
 641 |       <stageThreshold>-2.5802578926086426e+000</stageThreshold>
 642 |       <weakClassifiers>
 643 |         <_>
 644 |           <internalNodes>
 645 |             0 -1 35 -513 -706873891 -270541825 1564475391 -120602625
 646 |             -118490145 -3162113 -1025</internalNodes>
 647 |           <leafValues>
 648 |             -8.9068460464477539e-001 -1.6470588743686676e-001</leafValues></_>
 649 |         <_>
 650 |           <internalNodes>
 651 |             0 -1 41 -1025 872144563 -2105361 -1078076417 -1048577
 652 |             -1145061461 -87557413 -1375993973</internalNodes>
 653 |           <leafValues>
 654 |             -7.1808964014053345e-001 2.2022204473614693e-002</leafValues></_>
 655 |         <_>
 656 |           <internalNodes>
 657 |             0 -1 95 -42467849 967946223 -811601986 1030598351
 658 |             -1212430676 270856533 -1392539508 147705039</internalNodes>
 659 |           <leafValues>
 660 |             -4.9424821138381958e-001 3.0048963427543640e-001</leafValues></_>
 661 |         <_>
 662 |           <internalNodes>
 663 |             0 -1 10 -218116370 -637284625 -87373174 -521998782
 664 |             -805355450 -615023745 -814267322 -12069282</internalNodes>
 665 |           <leafValues>
 666 |             -5.5306458473205566e-001 2.9137542843818665e-001</leafValues></_>
 667 |         <_>
 668 |           <internalNodes>
 669 |             0 -1 105 -275849241 -527897 -11052049 -69756067 -15794193
 670 |             -1141376839 -564771 -287095455</internalNodes>
 671 |           <leafValues>
 672 |             -4.6759819984436035e-001 3.6638516187667847e-001</leafValues></_>
 673 |         <_>
 674 |           <internalNodes>
 675 |             0 -1 24 -1900898096 -18985228 -44056577 -24675 -1074880639
 676 |             -283998 796335613 -1079041957</internalNodes>
 677 |           <leafValues>
 678 |             -4.2737138271331787e-001 3.9243003726005554e-001</leafValues></_>
 679 |         <_>
 680 |           <internalNodes>
 681 |             0 -1 139 -555790844 410735094 -32106513 406822863 -897632192
 682 |             -912830145 -117771560 -1204027649</internalNodes>
 683 |           <leafValues>
 684 |             -4.1896930336952209e-001 3.6744937300682068e-001</leafValues></_>
 685 |         <_>
 686 |           <internalNodes>
 687 |             0 -1 0 -1884822366 -1406613148 1135342180 -1979127580
 688 |             -68174862 246469804 1001386992 -708885872</internalNodes>
 689 |           <leafValues>
 690 |             -5.7093089818954468e-001 2.9880744218826294e-001</leafValues></_>
 691 |         <_>
 692 |           <internalNodes>
 693 |             0 -1 45 -469053950 1439068142 2117758841 2004671078
 694 |             207931006 1265321675 970353931 1541343047</internalNodes>
 695 |           <leafValues>
 696 |             -6.0491901636123657e-001 2.4652053415775299e-001</leafValues></_></weakClassifiers></_>
 697 |     <!-- stage 12 -->
 698 |     <_>
 699 |       <maxWeakCount>9</maxWeakCount>
 700 |       <stageThreshold>-2.2425732612609863e+000</stageThreshold>
 701 |       <weakClassifiers>
 702 |         <_>
 703 |           <internalNodes>
 704 |             0 -1 58 1481987157 282547485 -14952129 421131223 -391065352
 705 |             -24212488 -100094241 -1157907473</internalNodes>
 706 |           <leafValues>
 707 |             -8.2822084426879883e-001 -2.1619293093681335e-001</leafValues></_>
 708 |         <_>
 709 |           <internalNodes>
 710 |             0 -1 126 -134217889 -543174305 -75497474 -16851650 -6685738
 711 |             -75834693 -2097200 -262146</internalNodes>
 712 |           <leafValues>
 713 |             -5.4628932476043701e-001 2.7662658691406250e-001</leafValues></_>
 714 |         <_>
 715 |           <internalNodes>
 716 |             0 -1 133 -220728227 -604288517 -661662214 413104863
 717 |             -627323700 -251915415 -626200872 -1157958657</internalNodes>
 718 |           <leafValues>
 719 |             -4.1643124818801880e-001 4.1700571775436401e-001</leafValues></_>
 720 |         <_>
 721 |           <internalNodes>
 722 |             0 -1 2 -186664033 -44236961 -1630262774 -65163606 -103237330
 723 |             -3083265 -1003729 2053105955</internalNodes>
 724 |           <leafValues>
 725 |             -5.4847818613052368e-001 2.9710745811462402e-001</leafValues></_>
 726 |         <_>
 727 |           <internalNodes>
 728 |             0 -1 62 -256115886 -237611873 -620250696 387061799
 729 |             1437882671 274878849 -8684449 1494294023</internalNodes>
 730 |           <leafValues>
 731 |             -4.6202757954597473e-001 3.3915829658508301e-001</leafValues></_>
 732 |         <_>
 733 |           <internalNodes>
 734 |             0 -1 1 -309400577 -275864640 -1056864869 1737132756
 735 |             -272385089 1609671419 1740601343 1261376789</internalNodes>
 736 |           <leafValues>
 737 |             -4.6158722043037415e-001 3.3939516544342041e-001</leafValues></_>
 738 |         <_>
 739 |           <internalNodes>
 740 |             0 -1 102 818197248 -196324552 286970589 -573270699
 741 |             -1174099579 -662077381 -1165157895 -1626859296</internalNodes>
 742 |           <leafValues>
 743 |             -4.6193107962608337e-001 3.2456985116004944e-001</leafValues></_>
 744 |         <_>
 745 |           <internalNodes>
 746 |             0 -1 69 -1042550357 14675409 1367955200 -841482753
 747 |             1642443255 8774277 1941304147 1099949563</internalNodes>
 748 |           <leafValues>
 749 |             -4.9091196060180664e-001 3.3870378136634827e-001</leafValues></_>
 750 |         <_>
 751 |           <internalNodes>
 752 |             0 -1 72 -639654997 1375720439 -2129542805 1614801090
 753 |             -626787937 -5779294 1488699183 -525406458</internalNodes>
 754 |           <leafValues>
 755 |             -4.9073097109794617e-001 3.0637946724891663e-001</leafValues></_></weakClassifiers></_>
 756 |     <!-- stage 13 -->
 757 |     <_>
 758 |       <maxWeakCount>9</maxWeakCount>
 759 |       <stageThreshold>-1.2258235216140747e+000</stageThreshold>
 760 |       <weakClassifiers>
 761 |         <_>
 762 |           <internalNodes>
 763 |             0 -1 118 302046707 -16744240 1360106207 -543735387
 764 |             1025700851 -1079408512 1796961263 -6334981</internalNodes>
 765 |           <leafValues>
 766 |             -6.1358314752578735e-001 2.3539231717586517e-001</leafValues></_>
 767 |         <_>
 768 |           <internalNodes>
 769 |             0 -1 5 -144765953 -116448726 -653851877 1934829856 722021887
 770 |             856564834 1933919231 -540838029</internalNodes>
 771 |           <leafValues>
 772 |             -5.1209545135498047e-001 3.2506987452507019e-001</leafValues></_>
 773 |         <_>
 774 |           <internalNodes>
 775 |             0 -1 140 -170132825 -1438923874 1879300370 -1689337194
 776 |             -695606496 285911565 -1044188928 -154210028</internalNodes>
 777 |           <leafValues>
 778 |             -5.1769560575485229e-001 3.2290914654731750e-001</leafValues></_>
 779 |         <_>
 780 |           <internalNodes>
 781 |             0 -1 131 -140776261 -355516414 822178224 -1039743806
 782 |             -1012208926 134887424 1438876097 -908591660</internalNodes>
 783 |           <leafValues>
 784 |             -5.0321841239929199e-001 3.0263835191726685e-001</leafValues></_>
 785 |         <_>
 786 |           <internalNodes>
 787 |             0 -1 64 -2137211696 -1634281249 1464325973 498569935
 788 |             -1580152080 -2001687927 721783561 265096035</internalNodes>
 789 |           <leafValues>
 790 |             -4.6532225608825684e-001 3.4638473391532898e-001</leafValues></_>
 791 |         <_>
 792 |           <internalNodes>
 793 |             0 -1 101 -255073589 -211824417 -972195129 -1063415417
 794 |             1937994261 1363165220 -754733105 1967602541</internalNodes>
 795 |           <leafValues>
 796 |             -4.9611270427703857e-001 3.3260712027549744e-001</leafValues></_>
 797 |         <_>
 798 |           <internalNodes>
 799 |             0 -1 81 -548146862 -655567194 -2062466596 1164562721
 800 |             416408236 -1591631712 -83637777 975344427</internalNodes>
 801 |           <leafValues>
 802 |             -4.9862930178642273e-001 3.2003280520439148e-001</leafValues></_>
 803 |         <_>
 804 |           <internalNodes>
 805 |             0 -1 55 -731904652 2147179896 2147442687 2112830847 -65604
 806 |             -131073 -42139667 -1074907393</internalNodes>
 807 |           <leafValues>
 808 |             -3.6636069416999817e-001 4.5651626586914063e-001</leafValues></_>
 809 |         <_>
 810 |           <internalNodes>
 811 |             0 -1 67 1885036886 571985932 -1784930633 724431327
 812 |             1940422257 -1085746880 964888398 731867951</internalNodes>
 813 |           <leafValues>
 814 |             -5.2619713544845581e-001 3.2635414600372314e-001</leafValues></_></weakClassifiers></_>
 815 |     <!-- stage 14 -->
 816 |     <_>
 817 |       <maxWeakCount>9</maxWeakCount>
 818 |       <stageThreshold>-1.3604533672332764e+000</stageThreshold>
 819 |       <weakClassifiers>
 820 |         <_>
 821 |           <internalNodes>
 822 |             0 -1 8 -287609985 -965585953 -2146397793 -492129894
 823 |             -729029645 -544619901 -645693256 -6565484</internalNodes>
 824 |           <leafValues>
 825 |             -4.5212322473526001e-001 3.8910505175590515e-001</leafValues></_>
 826 |         <_>
 827 |           <internalNodes>
 828 |             0 -1 122 -102903523 -145031013 536899675 688195859
 829 |             -645291520 -1165359094 -905565928 171608223</internalNodes>
 830 |           <leafValues>
 831 |             -4.9594074487686157e-001 3.4109055995941162e-001</leafValues></_>
 832 |         <_>
 833 |           <internalNodes>
 834 |             0 -1 134 -790640459 487931983 1778450522 1036604041
 835 |             -904752984 -954040118 -2134707506 304866043</internalNodes>
 836 |           <leafValues>
 837 |             -4.1148442029953003e-001 3.9666590094566345e-001</leafValues></_>
 838 |         <_>
 839 |           <internalNodes>
 840 |             0 -1 141 -303829117 1726939070 922189815 -827983123
 841 |             1567883042 1324809852 292710260 -942678754</internalNodes>
 842 |           <leafValues>
 843 |             -3.5154473781585693e-001 4.8011952638626099e-001</leafValues></_>
 844 |         <_>
 845 |           <internalNodes>
 846 |             0 -1 59 -161295376 -159215460 -1858041315 2140644499
 847 |             -2009065472 -133804007 -2003265301 1263206851</internalNodes>
 848 |           <leafValues>
 849 |             -4.2808216810226440e-001 3.9841541647911072e-001</leafValues></_>
 850 |         <_>
 851 |           <internalNodes>
 852 |             0 -1 34 -264248081 -667846464 1342624856 1381160835
 853 |             -2104716852 1342865409 -266612310 -165954877</internalNodes>
 854 |           <leafValues>
 855 |             -4.3293288350105286e-001 4.0339657664299011e-001</leafValues></_>
 856 |         <_>
 857 |           <internalNodes>
 858 |             0 -1 32 -1600388464 -40369901 285344639 1394344275
 859 |             -255680312 -100532214 -1031663944 -7471079</internalNodes>
 860 |           <leafValues>
 861 |             -4.1385015845298767e-001 4.5087572932243347e-001</leafValues></_>
 862 |         <_>
 863 |           <internalNodes>
 864 |             0 -1 15 1368521651 280207469 35779199 -105983261 1208124819
 865 |             -565870452 -1144024288 -591535344</internalNodes>
 866 |           <leafValues>
 867 |             -4.2956474423408508e-001 4.2176279425621033e-001</leafValues></_>
 868 |         <_>
 869 |           <internalNodes>
 870 |             0 -1 109 1623607527 -661513115 -1073217263 -2142994420
 871 |             -1339883309 -89816956 436308899 1426178059</internalNodes>
 872 |           <leafValues>
 873 |             -4.7764992713928223e-001 3.7551075220108032e-001</leafValues></_></weakClassifiers></_>
 874 |     <!-- stage 15 -->
 875 |     <_>
 876 |       <maxWeakCount>9</maxWeakCount>
 877 |       <stageThreshold>-4.2518746852874756e-001</stageThreshold>
 878 |       <weakClassifiers>
 879 |         <_>
 880 |           <internalNodes>
 881 |             0 -1 135 -116728032 -1154420809 -1350582273 746061691
 882 |             -1073758277 2138570623 2113797566 -138674182</internalNodes>
 883 |           <leafValues>
 884 |             -1.7125381529331207e-001 6.5421247482299805e-001</leafValues></_>
 885 |         <_>
 886 |           <internalNodes>
 887 |             0 -1 63 -453112432 -1795354691 -1342242964 494112553
 888 |             209458404 -2114697500 1316830362 259213855</internalNodes>
 889 |           <leafValues>
 890 |             -3.9870172739028931e-001 4.5807033777236938e-001</leafValues></_>
 891 |         <_>
 892 |           <internalNodes>
 893 |             0 -1 52 -268172036 294715533 268575185 486785157 -1065303920
 894 |             -360185856 -2147476808 134777113</internalNodes>
 895 |           <leafValues>
 896 |             -5.3581339120864868e-001 3.5815808176994324e-001</leafValues></_>
 897 |         <_>
 898 |           <internalNodes>
 899 |             0 -1 86 -301996882 -345718921 1877946252 -940720129
 900 |             -58737369 -721944585 -92954835 -530449</internalNodes>
 901 |           <leafValues>
 902 |             -3.9938014745712280e-001 4.9603295326232910e-001</leafValues></_>
 903 |         <_>
 904 |           <internalNodes>
 905 |             0 -1 14 -853281886 -756895766 2130706352 -9519120
 906 |             -1921059862 394133373 2138453959 -538200841</internalNodes>
 907 |           <leafValues>
 908 |             -4.0230083465576172e-001 4.9537116289138794e-001</leafValues></_>
 909 |         <_>
 910 |           <internalNodes>
 911 |             0 -1 128 -2133448688 -641138493 1078022185 294060066
 912 |             -327122776 -2130640896 -2147466247 -1910634326</internalNodes>
 913 |           <leafValues>
 914 |             -5.8290809392929077e-001 3.4102553129196167e-001</leafValues></_>
 915 |         <_>
 916 |           <internalNodes>
 917 |             0 -1 53 587265978 -2071658479 1108361221 -578448765
 918 |             -1811905899 -2008965119 33900729 762301595</internalNodes>
 919 |           <leafValues>
 920 |             -4.5518967509269714e-001 4.7242793440818787e-001</leafValues></_>
 921 |         <_>
 922 |           <internalNodes>
 923 |             0 -1 138 -1022189373 -2139094976 16658 -1069445120
 924 |             -1073555454 -1073577856 1096068 -978351488</internalNodes>
 925 |           <leafValues>
 926 |             -4.7530207037925720e-001 4.3885371088981628e-001</leafValues></_>
 927 |         <_>
 928 |           <internalNodes>
 929 |             0 -1 7 -395352441 -1073541103 -1056964605 1053186 269111298
 930 |             -2012184576 1611208714 -360415095</internalNodes>
 931 |           <leafValues>
 932 |             -5.0448113679885864e-001 4.1588482260704041e-001</leafValues></_></weakClassifiers></_>
 933 |     <!-- stage 16 -->
 934 |     <_>
 935 |       <maxWeakCount>7</maxWeakCount>
 936 |       <stageThreshold>2.7163455262780190e-002</stageThreshold>
 937 |       <weakClassifiers>
 938 |         <_>
 939 |           <internalNodes>
 940 |             0 -1 49 783189748 -137429026 -257 709557994 2130460236
 941 |             -196611 -9580 585428708</internalNodes>
 942 |           <leafValues>
 943 |             -2.0454545319080353e-001 7.9608374834060669e-001</leafValues></_>
 944 |         <_>
 945 |           <internalNodes>
 946 |             0 -1 108 1284360448 1057423155 1592696573 -852672655
 947 |             1547382714 -1642594369 125705358 797134398</internalNodes>
 948 |           <leafValues>
 949 |             -3.6474677920341492e-001 6.0925579071044922e-001</leafValues></_>
 950 |         <_>
 951 |           <internalNodes>
 952 |             0 -1 94 1347680270 -527720448 1091567712 1073745933
 953 |             -1073180671 0 285745154 -511192438</internalNodes>
 954 |           <leafValues>
 955 |             -4.6406838297843933e-001 5.5626088380813599e-001</leafValues></_>
 956 |         <_>
 957 |           <internalNodes>
 958 |             0 -1 73 1705780944 -145486260 -115909 -281793505 -418072663
 959 |             -1681064068 1877454127 -1912330993</internalNodes>
 960 |           <leafValues>
 961 |             -4.7043186426162720e-001 5.8430361747741699e-001</leafValues></_>
 962 |         <_>
 963 |           <internalNodes>
 964 |             0 -1 110 -2118142016 339509033 -285260567 1417764573
 965 |             68144392 -468879483 -2033291636 231451911</internalNodes>
 966 |           <leafValues>
 967 |             -4.8700931668281555e-001 5.4639810323715210e-001</leafValues></_>
 968 |         <_>
 969 |           <internalNodes>
 970 |             0 -1 119 -1888051818 489996135 -65539 849536890 2146716845
 971 |             -1107542088 -1275615746 -1119617586</internalNodes>
 972 |           <leafValues>
 973 |             -4.3356490135192871e-001 6.5175366401672363e-001</leafValues></_>
 974 |         <_>
 975 |           <internalNodes>
 976 |             0 -1 44 -1879021438 336830528 1073766659 1477541961 8560696
 977 |             -1207369568 8462472 1493893448</internalNodes>
 978 |           <leafValues>
 979 |             -5.4343086481094360e-001 5.2777874469757080e-001</leafValues></_></weakClassifiers></_>
 980 |     <!-- stage 17 -->
 981 |     <_>
 982 |       <maxWeakCount>7</maxWeakCount>
 983 |       <stageThreshold>4.9174150824546814e-001</stageThreshold>
 984 |       <weakClassifiers>
 985 |         <_>
 986 |           <internalNodes>
 987 |             0 -1 57 644098 15758324 1995964260 -463011882 893285175
 988 |             83156983 2004317989 16021237</internalNodes>
 989 |           <leafValues>
 990 |             -1.7073170840740204e-001 9.0782123804092407e-001</leafValues></_>
 991 |         <_>
 992 |           <internalNodes>
 993 |             0 -1 123 268632845 -2147450864 -2143240192 -2147401728
 994 |             8523937 -1878523840 16777416 616824984</internalNodes>
 995 |           <leafValues>
 996 |             -4.8744434118270874e-001 7.3311311006546021e-001</leafValues></_>
 997 |         <_>
 998 |           <internalNodes>
 999 |             0 -1 120 -2110735872 803880886 989739810 1673281312 91564930
1000 |             -277454958 997709514 -581366443</internalNodes>
1001 |           <leafValues>
1002 |             -4.0291741490364075e-001 8.2450771331787109e-001</leafValues></_>
1003 |         <_>
1004 |           <internalNodes>
1005 |             0 -1 87 941753434 -1067128905 788512753 -1074450460
1006 |             779101657 -1346552460 938805167 -2050424642</internalNodes>
1007 |           <leafValues>
1008 |             -3.6246949434280396e-001 8.7103593349456787e-001</leafValues></_>
1009 |         <_>
1010 |           <internalNodes>
1011 |             0 -1 60 208 1645217920 130 538263552 33595552 -1475870592
1012 |             16783361 1375993867</internalNodes>
1013 |           <leafValues>
1014 |             -6.1472141742706299e-001 5.9707164764404297e-001</leafValues></_>
1015 |         <_>
1016 |           <internalNodes>
1017 |             0 -1 114 1860423179 1034692624 -285213187 -986681712
1018 |             1576755092 -1408205463 -127714 -1246035687</internalNodes>
1019 |           <leafValues>
1020 |             -4.5621752738952637e-001 8.9482426643371582e-001</leafValues></_>
1021 |         <_>
1022 |           <internalNodes>
1023 |             0 -1 107 33555004 -1861746688 1073807361 -754909184
1024 |             645922856 8388608 134250648 419635458</internalNodes>
1025 |           <leafValues>
1026 |             -5.2466005086898804e-001 7.1834069490432739e-001</leafValues></_></weakClassifiers></_>
1027 |     <!-- stage 18 -->
1028 |     <_>
1029 |       <maxWeakCount>2</maxWeakCount>
1030 |       <stageThreshold>1.9084988832473755e+000</stageThreshold>
1031 |       <weakClassifiers>
1032 |         <_>
1033 |           <internalNodes>
1034 |             0 -1 16 536064 131072 -20971516 524288 576 1048577 0 40960</internalNodes>
1035 |           <leafValues>
1036 |             -8.0000001192092896e-001 9.8018401861190796e-001</leafValues></_>
1037 |         <_>
1038 |           <internalNodes>
1039 |             0 -1 56 67108864 0 4096 1074003968 8192 536870912 4 262144</internalNodes>
1040 |           <leafValues>
1041 |             -9.6610915660858154e-001 9.2831486463546753e-001</leafValues></_></weakClassifiers></_></stages>
1042 |   <features>
1043 |     <_>
1044 |       <rect>
1045 |         0 0 1 1</rect></_>
1046 |     <_>
1047 |       <rect>
1048 |         0 0 3 2</rect></_>
1049 |     <_>
1050 |       <rect>
1051 |         0 1 13 6</rect></_>
1052 |     <_>
1053 |       <rect>
1054 |         0 2 3 14</rect></_>
1055 |     <_>
1056 |       <rect>
1057 |         0 2 4 2</rect></_>
1058 |     <_>
1059 |       <rect>
1060 |         0 6 2 3</rect></_>
1061 |     <_>
1062 |       <rect>
1063 |         0 6 3 2</rect></_>
1064 |     <_>
1065 |       <rect>
1066 |         0 16 1 3</rect></_>
1067 |     <_>
1068 |       <rect>
1069 |         0 20 3 3</rect></_>
1070 |     <_>
1071 |       <rect>
1072 |         0 22 2 3</rect></_>
1073 |     <_>
1074 |       <rect>
1075 |         0 28 4 4</rect></_>
1076 |     <_>
1077 |       <rect>
1078 |         0 35 2 3</rect></_>
1079 |     <_>
1080 |       <rect>
1081 |         1 0 14 7</rect></_>
1082 |     <_>
1083 |       <rect>
1084 |         1 5 3 2</rect></_>
1085 |     <_>
1086 |       <rect>
1087 |         1 6 2 1</rect></_>
1088 |     <_>
1089 |       <rect>
1090 |         1 14 10 9</rect></_>
1091 |     <_>
1092 |       <rect>
1093 |         1 21 4 4</rect></_>
1094 |     <_>
1095 |       <rect>
1096 |         1 23 4 2</rect></_>
1097 |     <_>
1098 |       <rect>
1099 |         2 0 13 7</rect></_>
1100 |     <_>
1101 |       <rect>
1102 |         2 0 14 7</rect></_>
1103 |     <_>
1104 |       <rect>
1105 |         2 33 5 4</rect></_>
1106 |     <_>
1107 |       <rect>
1108 |         2 36 4 3</rect></_>
1109 |     <_>
1110 |       <rect>
1111 |         2 39 3 2</rect></_>
1112 |     <_>
1113 |       <rect>
1114 |         3 1 13 11</rect></_>
1115 |     <_>
1116 |       <rect>
1117 |         3 2 3 2</rect></_>
1118 |     <_>
1119 |       <rect>
1120 |         4 0 7 8</rect></_>
1121 |     <_>
1122 |       <rect>
1123 |         4 0 13 7</rect></_>
1124 |     <_>
1125 |       <rect>
1126 |         5 0 12 6</rect></_>
1127 |     <_>
1128 |       <rect>
1129 |         5 0 13 7</rect></_>
1130 |     <_>
1131 |       <rect>
1132 |         5 1 10 13</rect></_>
1133 |     <_>
1134 |       <rect>
1135 |         5 1 12 7</rect></_>
1136 |     <_>
1137 |       <rect>
1138 |         5 2 7 13</rect></_>
1139 |     <_>
1140 |       <rect>
1141 |         5 4 2 1</rect></_>
1142 |     <_>
1143 |       <rect>
1144 |         5 8 7 4</rect></_>
1145 |     <_>
1146 |       <rect>
1147 |         5 39 3 2</rect></_>
1148 |     <_>
1149 |       <rect>
1150 |         6 3 5 2</rect></_>
1151 |     <_>
1152 |       <rect>
1153 |         6 3 6 2</rect></_>
1154 |     <_>
1155 |       <rect>
1156 |         6 5 4 12</rect></_>
1157 |     <_>
1158 |       <rect>
1159 |         6 9 6 3</rect></_>
1160 |     <_>
1161 |       <rect>
1162 |         7 3 5 2</rect></_>
1163 |     <_>
1164 |       <rect>
1165 |         7 3 6 13</rect></_>
1166 |     <_>
1167 |       <rect>
1168 |         7 5 6 4</rect></_>
1169 |     <_>
1170 |       <rect>
1171 |         7 7 6 10</rect></_>
1172 |     <_>
1173 |       <rect>
1174 |         7 8 6 4</rect></_>
1175 |     <_>
1176 |       <rect>
1177 |         7 32 5 4</rect></_>
1178 |     <_>
1179 |       <rect>
1180 |         7 33 5 4</rect></_>
1181 |     <_>
1182 |       <rect>
1183 |         8 0 1 1</rect></_>
1184 |     <_>
1185 |       <rect>
1186 |         8 0 2 1</rect></_>
1187 |     <_>
1188 |       <rect>
1189 |         8 2 10 7</rect></_>
1190 |     <_>
1191 |       <rect>
1192 |         9 0 6 2</rect></_>
1193 |     <_>
1194 |       <rect>
1195 |         9 2 9 3</rect></_>
1196 |     <_>
1197 |       <rect>
1198 |         9 4 1 1</rect></_>
1199 |     <_>
1200 |       <rect>
1201 |         9 6 2 1</rect></_>
1202 |     <_>
1203 |       <rect>
1204 |         9 28 6 4</rect></_>
1205 |     <_>
1206 |       <rect>
1207 |         10 0 9 3</rect></_>
1208 |     <_>
1209 |       <rect>
1210 |         10 3 1 1</rect></_>
1211 |     <_>
1212 |       <rect>
1213 |         10 10 11 11</rect></_>
1214 |     <_>
1215 |       <rect>
1216 |         10 15 4 3</rect></_>
1217 |     <_>
1218 |       <rect>
1219 |         11 4 2 1</rect></_>
1220 |     <_>
1221 |       <rect>
1222 |         11 27 4 3</rect></_>
1223 |     <_>
1224 |       <rect>
1225 |         11 36 8 2</rect></_>
1226 |     <_>
1227 |       <rect>
1228 |         12 0 2 2</rect></_>
1229 |     <_>
1230 |       <rect>
1231 |         12 23 4 3</rect></_>
1232 |     <_>
1233 |       <rect>
1234 |         12 25 4 3</rect></_>
1235 |     <_>
1236 |       <rect>
1237 |         12 29 5 3</rect></_>
1238 |     <_>
1239 |       <rect>
1240 |         12 33 3 4</rect></_>
1241 |     <_>
1242 |       <rect>
1243 |         13 0 2 2</rect></_>
1244 |     <_>
1245 |       <rect>
1246 |         13 36 8 3</rect></_>
1247 |     <_>
1248 |       <rect>
1249 |         14 0 2 2</rect></_>
1250 |     <_>
1251 |       <rect>
1252 |         15 15 2 2</rect></_>
1253 |     <_>
1254 |       <rect>
1255 |         16 13 3 4</rect></_>
1256 |     <_>
1257 |       <rect>
1258 |         17 0 1 3</rect></_>
1259 |     <_>
1260 |       <rect>
1261 |         17 1 3 3</rect></_>
1262 |     <_>
1263 |       <rect>
1264 |         17 31 5 3</rect></_>
1265 |     <_>
1266 |       <rect>
1267 |         17 35 3 1</rect></_>
1268 |     <_>
1269 |       <rect>
1270 |         18 13 2 3</rect></_>
1271 |     <_>
1272 |       <rect>
1273 |         18 39 2 1</rect></_>
1274 |     <_>
1275 |       <rect>
1276 |         19 0 7 15</rect></_>
1277 |     <_>
1278 |       <rect>
1279 |         19 2 7 2</rect></_>
1280 |     <_>
1281 |       <rect>
1282 |         19 3 7 13</rect></_>
1283 |     <_>
1284 |       <rect>
1285 |         19 14 2 2</rect></_>
1286 |     <_>
1287 |       <rect>
1288 |         19 24 7 4</rect></_>
1289 |     <_>
1290 |       <rect>
1291 |         20 1 6 13</rect></_>
1292 |     <_>
1293 |       <rect>
1294 |         20 8 7 3</rect></_>
1295 |     <_>
1296 |       <rect>
1297 |         20 9 7 3</rect></_>
1298 |     <_>
1299 |       <rect>
1300 |         20 13 1 1</rect></_>
1301 |     <_>
1302 |       <rect>
1303 |         20 14 2 3</rect></_>
1304 |     <_>
1305 |       <rect>
1306 |         20 30 3 2</rect></_>
1307 |     <_>
1308 |       <rect>
1309 |         21 0 3 4</rect></_>
1310 |     <_>
1311 |       <rect>
1312 |         21 0 6 8</rect></_>
1313 |     <_>
1314 |       <rect>
1315 |         21 3 6 2</rect></_>
1316 |     <_>
1317 |       <rect>
1318 |         21 6 6 4</rect></_>
1319 |     <_>
1320 |       <rect>
1321 |         21 37 2 1</rect></_>
1322 |     <_>
1323 |       <rect>
1324 |         22 3 6 2</rect></_>
1325 |     <_>
1326 |       <rect>
1327 |         22 13 1 2</rect></_>
1328 |     <_>
1329 |       <rect>
1330 |         22 22 4 3</rect></_>
1331 |     <_>
1332 |       <rect>
1333 |         23 0 2 3</rect></_>
1334 |     <_>
1335 |       <rect>
1336 |         23 3 6 2</rect></_>
1337 |     <_>
1338 |       <rect>
1339 |         23 9 5 4</rect></_>
1340 |     <_>
1341 |       <rect>
1342 |         23 11 1 1</rect></_>
1343 |     <_>
1344 |       <rect>
1345 |         23 15 1 1</rect></_>
1346 |     <_>
1347 |       <rect>
1348 |         23 16 3 2</rect></_>
1349 |     <_>
1350 |       <rect>
1351 |         23 35 2 1</rect></_>
1352 |     <_>
1353 |       <rect>
1354 |         23 36 1 1</rect></_>
1355 |     <_>
1356 |       <rect>
1357 |         23 39 6 2</rect></_>
1358 |     <_>
1359 |       <rect>
1360 |         24 0 2 3</rect></_>
1361 |     <_>
1362 |       <rect>
1363 |         24 8 6 11</rect></_>
1364 |     <_>
1365 |       <rect>
1366 |         24 28 2 2</rect></_>
1367 |     <_>
1368 |       <rect>
1369 |         24 33 4 4</rect></_>
1370 |     <_>
1371 |       <rect>
1372 |         25 16 4 3</rect></_>
1373 |     <_>
1374 |       <rect>
1375 |         25 31 5 3</rect></_>
1376 |     <_>
1377 |       <rect>
1378 |         26 0 1 2</rect></_>
1379 |     <_>
1380 |       <rect>
1381 |         26 0 2 2</rect></_>
1382 |     <_>
1383 |       <rect>
1384 |         26 0 3 2</rect></_>
1385 |     <_>
1386 |       <rect>
1387 |         26 24 4 4</rect></_>
1388 |     <_>
1389 |       <rect>
1390 |         27 30 4 5</rect></_>
1391 |     <_>
1392 |       <rect>
1393 |         27 36 5 3</rect></_>
1394 |     <_>
1395 |       <rect>
1396 |         28 0 2 2</rect></_>
1397 |     <_>
1398 |       <rect>
1399 |         28 4 2 1</rect></_>
1400 |     <_>
1401 |       <rect>
1402 |         28 21 2 5</rect></_>
1403 |     <_>
1404 |       <rect>
1405 |         29 8 2 1</rect></_>
1406 |     <_>
1407 |       <rect>
1408 |         33 0 2 1</rect></_>
1409 |     <_>
1410 |       <rect>
1411 |         33 0 4 2</rect></_>
1412 |     <_>
1413 |       <rect>
1414 |         33 0 4 6</rect></_>
1415 |     <_>
1416 |       <rect>
1417 |         33 3 1 1</rect></_>
1418 |     <_>
1419 |       <rect>
1420 |         33 6 4 12</rect></_>
1421 |     <_>
1422 |       <rect>
1423 |         33 21 4 2</rect></_>
1424 |     <_>
1425 |       <rect>
1426 |         33 36 4 3</rect></_>
1427 |     <_>
1428 |       <rect>
1429 |         35 1 2 2</rect></_>
1430 |     <_>
1431 |       <rect>
1432 |         36 5 1 1</rect></_>
1433 |     <_>
1434 |       <rect>
1435 |         36 29 3 4</rect></_>
1436 |     <_>
1437 |       <rect>
1438 |         36 39 2 2</rect></_>
1439 |     <_>
1440 |       <rect>
1441 |         37 5 2 2</rect></_>
1442 |     <_>
1443 |       <rect>
1444 |         38 6 2 1</rect></_>
1445 |     <_>
1446 |       <rect>
1447 |         38 6 2 2</rect></_>
1448 |     <_>
1449 |       <rect>
1450 |         39 1 2 12</rect></_>
1451 |     <_>
1452 |       <rect>
1453 |         39 24 1 2</rect></_>
1454 |     <_>
1455 |       <rect>
1456 |         39 36 2 2</rect></_>
1457 |     <_>
1458 |       <rect>
1459 |         40 39 1 2</rect></_>
1460 |     <_>
1461 |       <rect>
1462 |         42 4 1 1</rect></_>
1463 |     <_>
1464 |       <rect>
1465 |         42 20 1 2</rect></_>
1466 |     <_>
1467 |       <rect>
1468 |         42 29 1 2</rect></_></features></cascade>
1469 | </opencv_storage>
1470 | 


--------------------------------------------------------------------------------
/split_data.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": 1,
  6 |    "metadata": {},
  7 |    "outputs": [],
  8 |    "source": [
  9 |     "import sys\n",
 10 |     "sys.path.append('/home/emna/text/lib/python3.5/site-packages')\n",
 11 |     "import numpy as np\n",
 12 |     "import pandas as pd\n",
 13 |     "np.random.seed(1)"
 14 |    ]
 15 |   },
 16 |   {
 17 |    "cell_type": "code",
 18 |    "execution_count": 2,
 19 |    "metadata": {},
 20 |    "outputs": [],
 21 |    "source": [
 22 |     "full_labels = pd.read_csv('celebrities_labels.csv',index_col=0)"
 23 |    ]
 24 |   },
 25 |   {
 26 |    "cell_type": "code",
 27 |    "execution_count": 3,
 28 |    "metadata": {},
 29 |    "outputs": [
 30 |     {
 31 |      "data": {
 32 |       "text/html": [
 33 |        "<div>\n",
 34 |        "<style scoped>\n",
 35 |        "    .dataframe tbody tr th:only-of-type {\n",
 36 |        "        vertical-align: middle;\n",
 37 |        "    }\n",
 38 |        "\n",
 39 |        "    .dataframe tbody tr th {\n",
 40 |        "        vertical-align: top;\n",
 41 |        "    }\n",
 42 |        "\n",
 43 |        "    .dataframe thead th {\n",
 44 |        "        text-align: right;\n",
 45 |        "    }\n",
 46 |        "</style>\n",
 47 |        "<table border=\"1\" class=\"dataframe\">\n",
 48 |        "  <thead>\n",
 49 |        "    <tr style=\"text-align: right;\">\n",
 50 |        "      <th></th>\n",
 51 |        "      <th>width</th>\n",
 52 |        "      <th>height</th>\n",
 53 |        "      <th>class</th>\n",
 54 |        "      <th>xmin</th>\n",
 55 |        "      <th>ymin</th>\n",
 56 |        "      <th>xmax</th>\n",
 57 |        "      <th>ymax</th>\n",
 58 |        "    </tr>\n",
 59 |        "    <tr>\n",
 60 |        "      <th>filename</th>\n",
 61 |        "      <th></th>\n",
 62 |        "      <th></th>\n",
 63 |        "      <th></th>\n",
 64 |        "      <th></th>\n",
 65 |        "      <th></th>\n",
 66 |        "      <th></th>\n",
 67 |        "      <th></th>\n",
 68 |        "    </tr>\n",
 69 |        "  </thead>\n",
 70 |        "  <tbody>\n",
 71 |        "    <tr>\n",
 72 |        "      <th>wenn5937954.jpg</th>\n",
 73 |        "      <td>603</td>\n",
 74 |        "      <td>839</td>\n",
 75 |        "      <td>AndySerkis</td>\n",
 76 |        "      <td>79</td>\n",
 77 |        "      <td>186</td>\n",
 78 |        "      <td>642</td>\n",
 79 |        "      <td>456</td>\n",
 80 |        "    </tr>\n",
 81 |        "    <tr>\n",
 82 |        "      <th>2s6584.jpeg</th>\n",
 83 |        "      <td>125</td>\n",
 84 |        "      <td>165</td>\n",
 85 |        "      <td>willSmith</td>\n",
 86 |        "      <td>12</td>\n",
 87 |        "      <td>42</td>\n",
 88 |        "      <td>141</td>\n",
 89 |        "      <td>99</td>\n",
 90 |        "    </tr>\n",
 91 |        "    <tr>\n",
 92 |        "      <th>1s15927.jpeg</th>\n",
 93 |        "      <td>68</td>\n",
 94 |        "      <td>80</td>\n",
 95 |        "      <td>willSmith</td>\n",
 96 |        "      <td>6</td>\n",
 97 |        "      <td>16</td>\n",
 98 |        "      <td>72</td>\n",
 99 |        "      <td>56</td>\n",
100 |        "    </tr>\n",
101 |        "    <tr>\n",
102 |        "      <th>2s6372.jpeg</th>\n",
103 |        "      <td>85</td>\n",
104 |        "      <td>113</td>\n",
105 |        "      <td>willSmith</td>\n",
106 |        "      <td>14</td>\n",
107 |        "      <td>35</td>\n",
108 |        "      <td>100</td>\n",
109 |        "      <td>65</td>\n",
110 |        "    </tr>\n",
111 |        "    <tr>\n",
112 |        "      <th>1jth406.jpeg</th>\n",
113 |        "      <td>82</td>\n",
114 |        "      <td>115</td>\n",
115 |        "      <td>justinBieber</td>\n",
116 |        "      <td>11</td>\n",
117 |        "      <td>33</td>\n",
118 |        "      <td>97</td>\n",
119 |        "      <td>64</td>\n",
120 |        "    </tr>\n",
121 |        "  </tbody>\n",
122 |        "</table>\n",
123 |        "</div>"
124 |       ],
125 |       "text/plain": [
126 |        "                 width  height         class  xmin  ymin  xmax  ymax\n",
127 |        "filename                                                            \n",
128 |        "wenn5937954.jpg    603     839    AndySerkis    79   186   642   456\n",
129 |        "2s6584.jpeg        125     165     willSmith    12    42   141    99\n",
130 |        "1s15927.jpeg        68      80     willSmith     6    16    72    56\n",
131 |        "2s6372.jpeg         85     113     willSmith    14    35   100    65\n",
132 |        "1jth406.jpeg        82     115  justinBieber    11    33    97    64"
133 |       ]
134 |      },
135 |      "execution_count": 3,
136 |      "metadata": {},
137 |      "output_type": "execute_result"
138 |     }
139 |    ],
140 |    "source": [
141 |     "full_labels.head()\n"
142 |    ]
143 |   },
144 |   {
145 |    "cell_type": "code",
146 |    "execution_count": 4,
147 |    "metadata": {},
148 |    "outputs": [
149 |     {
150 |      "data": {
151 |       "text/plain": [
152 |        "<pandas.core.groupby.groupby.DataFrameGroupBy object at 0x7f63a321bfd0>"
153 |       ]
154 |      },
155 |      "execution_count": 4,
156 |      "metadata": {},
157 |      "output_type": "execute_result"
158 |     }
159 |    ],
160 |    "source": [
161 |     "grouped = full_labels.groupby('class')\n",
162 |     "grouped"
163 |    ]
164 |   },
165 |   {
166 |    "cell_type": "code",
167 |    "execution_count": 5,
168 |    "metadata": {},
169 |    "outputs": [
170 |     {
171 |      "data": {
172 |       "text/plain": [
173 |        "class\n",
174 |        "AndySerkis        200\n",
175 |        "dewayneJohnson     43\n",
176 |        "drake              61\n",
177 |        "jayZ               15\n",
178 |        "justinBieber      103\n",
179 |        "kimKardashian     128\n",
180 |        "tomHanks           79\n",
181 |        "willSmith          78\n",
182 |        "dtype: int64"
183 |       ]
184 |      },
185 |      "execution_count": 5,
186 |      "metadata": {},
187 |      "output_type": "execute_result"
188 |     }
189 |    ],
190 |    "source": [
191 |     "full_labels.groupby(['class']).size()"
192 |    ]
193 |   },
194 |   {
195 |    "cell_type": "code",
196 |    "execution_count": 6,
197 |    "metadata": {},
198 |    "outputs": [],
199 |    "source": [
200 |     "df = full_labels.groupby(['class']).size().to_frame('size')"
201 |    ]
202 |   },
203 |   {
204 |    "cell_type": "code",
205 |    "execution_count": 7,
206 |    "metadata": {},
207 |    "outputs": [
208 |     {
209 |      "data": {
210 |       "text/html": [
211 |        "<div>\n",
212 |        "<style scoped>\n",
213 |        "    .dataframe tbody tr th:only-of-type {\n",
214 |        "        vertical-align: middle;\n",
215 |        "    }\n",
216 |        "\n",
217 |        "    .dataframe tbody tr th {\n",
218 |        "        vertical-align: top;\n",
219 |        "    }\n",
220 |        "\n",
221 |        "    .dataframe thead th {\n",
222 |        "        text-align: right;\n",
223 |        "    }\n",
224 |        "</style>\n",
225 |        "<table border=\"1\" class=\"dataframe\">\n",
226 |        "  <thead>\n",
227 |        "    <tr style=\"text-align: right;\">\n",
228 |        "      <th></th>\n",
229 |        "      <th>size</th>\n",
230 |        "    </tr>\n",
231 |        "    <tr>\n",
232 |        "      <th>class</th>\n",
233 |        "      <th></th>\n",
234 |        "    </tr>\n",
235 |        "  </thead>\n",
236 |        "  <tbody>\n",
237 |        "    <tr>\n",
238 |        "      <th>AndySerkis</th>\n",
239 |        "      <td>200</td>\n",
240 |        "    </tr>\n",
241 |        "    <tr>\n",
242 |        "      <th>dewayneJohnson</th>\n",
243 |        "      <td>43</td>\n",
244 |        "    </tr>\n",
245 |        "    <tr>\n",
246 |        "      <th>drake</th>\n",
247 |        "      <td>61</td>\n",
248 |        "    </tr>\n",
249 |        "    <tr>\n",
250 |        "      <th>jayZ</th>\n",
251 |        "      <td>15</td>\n",
252 |        "    </tr>\n",
253 |        "    <tr>\n",
254 |        "      <th>justinBieber</th>\n",
255 |        "      <td>103</td>\n",
256 |        "    </tr>\n",
257 |        "    <tr>\n",
258 |        "      <th>kimKardashian</th>\n",
259 |        "      <td>128</td>\n",
260 |        "    </tr>\n",
261 |        "    <tr>\n",
262 |        "      <th>tomHanks</th>\n",
263 |        "      <td>79</td>\n",
264 |        "    </tr>\n",
265 |        "    <tr>\n",
266 |        "      <th>willSmith</th>\n",
267 |        "      <td>78</td>\n",
268 |        "    </tr>\n",
269 |        "  </tbody>\n",
270 |        "</table>\n",
271 |        "</div>"
272 |       ],
273 |       "text/plain": [
274 |        "                size\n",
275 |        "class               \n",
276 |        "AndySerkis       200\n",
277 |        "dewayneJohnson    43\n",
278 |        "drake             61\n",
279 |        "jayZ              15\n",
280 |        "justinBieber     103\n",
281 |        "kimKardashian    128\n",
282 |        "tomHanks          79\n",
283 |        "willSmith         78"
284 |       ]
285 |      },
286 |      "execution_count": 7,
287 |      "metadata": {},
288 |      "output_type": "execute_result"
289 |     }
290 |    ],
291 |    "source": [
292 |     "df"
293 |    ]
294 |   },
295 |   {
296 |    "cell_type": "markdown",
297 |    "metadata": {},
298 |    "source": [
299 |     "\n",
300 |     "### split each file into a group in a list\n"
301 |    ]
302 |   },
303 |   {
304 |    "cell_type": "code",
305 |    "execution_count": 8,
306 |    "metadata": {},
307 |    "outputs": [],
308 |    "source": [
309 |     "grouped = full_labels.groupby('class')\n"
310 |    ]
311 |   },
312 |   {
313 |    "cell_type": "code",
314 |    "execution_count": 9,
315 |    "metadata": {},
316 |    "outputs": [],
317 |    "source": [
318 |     "grouped_list = [grouped.get_group(x) for x in grouped.groups]"
319 |    ]
320 |   },
321 |   {
322 |    "cell_type": "code",
323 |    "execution_count": 10,
324 |    "metadata": {},
325 |    "outputs": [
326 |     {
327 |      "data": {
328 |       "text/plain": [
329 |        "8"
330 |       ]
331 |      },
332 |      "execution_count": 10,
333 |      "metadata": {},
334 |      "output_type": "execute_result"
335 |     }
336 |    ],
337 |    "source": [
338 |     "len(grouped_list)"
339 |    ]
340 |   },
341 |   {
342 |    "cell_type": "code",
343 |    "execution_count": 11,
344 |    "metadata": {},
345 |    "outputs": [],
346 |    "source": [
347 |     "from sklearn.model_selection import train_test_split"
348 |    ]
349 |   },
350 |   {
351 |    "cell_type": "code",
352 |    "execution_count": 12,
353 |    "metadata": {},
354 |    "outputs": [],
355 |    "source": [
356 |     "X_train, X_test, y_train, y_test = train_test_split(full_labels, full_labels['class'],test_size=0.2)"
357 |    ]
358 |   },
359 |   {
360 |    "cell_type": "code",
361 |    "execution_count": 13,
362 |    "metadata": {},
363 |    "outputs": [
364 |     {
365 |      "data": {
366 |       "text/plain": [
367 |        "(707, 7)"
368 |       ]
369 |      },
370 |      "execution_count": 13,
371 |      "metadata": {},
372 |      "output_type": "execute_result"
373 |     }
374 |    ],
375 |    "source": [
376 |     "full_labels.shape"
377 |    ]
378 |   },
379 |   {
380 |    "cell_type": "code",
381 |    "execution_count": 14,
382 |    "metadata": {},
383 |    "outputs": [
384 |     {
385 |      "data": {
386 |       "text/plain": [
387 |        "(565, 7)"
388 |       ]
389 |      },
390 |      "execution_count": 14,
391 |      "metadata": {},
392 |      "output_type": "execute_result"
393 |     }
394 |    ],
395 |    "source": [
396 |     "X_train.shape"
397 |    ]
398 |   },
399 |   {
400 |    "cell_type": "code",
401 |    "execution_count": 15,
402 |    "metadata": {},
403 |    "outputs": [
404 |     {
405 |      "data": {
406 |       "text/plain": [
407 |        "class\n",
408 |        "AndySerkis        159\n",
409 |        "dewayneJohnson     34\n",
410 |        "drake              49\n",
411 |        "jayZ               14\n",
412 |        "justinBieber       85\n",
413 |        "kimKardashian     108\n",
414 |        "tomHanks           59\n",
415 |        "willSmith          57\n",
416 |        "dtype: int64"
417 |       ]
418 |      },
419 |      "execution_count": 15,
420 |      "metadata": {},
421 |      "output_type": "execute_result"
422 |     }
423 |    ],
424 |    "source": [
425 |     "X_train.groupby(['class']).size()"
426 |    ]
427 |   },
428 |   {
429 |    "cell_type": "code",
430 |    "execution_count": 16,
431 |    "metadata": {},
432 |    "outputs": [
433 |     {
434 |      "data": {
435 |       "text/plain": [
436 |        "(142, 7)"
437 |       ]
438 |      },
439 |      "execution_count": 16,
440 |      "metadata": {},
441 |      "output_type": "execute_result"
442 |     }
443 |    ],
444 |    "source": [
445 |     "X_test.shape"
446 |    ]
447 |   },
448 |   {
449 |    "cell_type": "code",
450 |    "execution_count": 17,
451 |    "metadata": {},
452 |    "outputs": [
453 |     {
454 |      "data": {
455 |       "text/plain": [
456 |        "class\n",
457 |        "AndySerkis        41\n",
458 |        "dewayneJohnson     9\n",
459 |        "drake             12\n",
460 |        "jayZ               1\n",
461 |        "justinBieber      18\n",
462 |        "kimKardashian     20\n",
463 |        "tomHanks          20\n",
464 |        "willSmith         21\n",
465 |        "dtype: int64"
466 |       ]
467 |      },
468 |      "execution_count": 17,
469 |      "metadata": {},
470 |      "output_type": "execute_result"
471 |     }
472 |    ],
473 |    "source": [
474 |     "X_test.groupby(['class']).size()"
475 |    ]
476 |   },
477 |   {
478 |    "cell_type": "code",
479 |    "execution_count": 18,
480 |    "metadata": {},
481 |    "outputs": [],
482 |    "source": [
483 |     "X_train.to_csv('train.csv')\n",
484 |     "X_test.to_csv('test.csv')"
485 |    ]
486 |   },
487 |   {
488 |    "cell_type": "code",
489 |    "execution_count": null,
490 |    "metadata": {},
491 |    "outputs": [],
492 |    "source": []
493 |   }
494 |  ],
495 |  "metadata": {
496 |   "kernelspec": {
497 |    "display_name": "Python 3",
498 |    "language": "python",
499 |    "name": "python3"
500 |   },
501 |   "language_info": {
502 |    "codemirror_mode": {
503 |     "name": "ipython",
504 |     "version": 3
505 |    },
506 |    "file_extension": ".py",
507 |    "mimetype": "text/x-python",
508 |    "name": "python",
509 |    "nbconvert_exporter": "python",
510 |    "pygments_lexer": "ipython3",
511 |    "version": "3.5.2"
512 |   }
513 |  },
514 |  "nbformat": 4,
515 |  "nbformat_minor": 2
516 | }
517 | 


--------------------------------------------------------------------------------
/ssd_mobilenet_v1_coco_2018_01_28/ssd.md:
--------------------------------------------------------------------------------
1 | We will use [ssd_mobilenet_v1_coco](http://download.tensorflow.org/models/object_detection/ssd_mobilenet_v1_coco_2018_01_28.tar.gz) to train our face recognition model.
2 | 


--------------------------------------------------------------------------------
/xml_to_csv.py:
--------------------------------------------------------------------------------
 1 | import os
 2 | import glob
 3 | import pandas as pd
 4 | import xml.etree.ElementTree as ET
 5 | 
 6 | 
 7 | def xml_to_csv(path):
 8 |     xml_list = []
 9 |     for xml_file in glob.glob(path + '/*.xml'):
10 |         tree = ET.parse(xml_file)
11 |         root = tree.getroot()
12 |         for member in root.findall('object'):
13 |             value = (root.find('filename').text,
14 |                      int(root.find('size')[0].text),
15 |                      int(root.find('size')[1].text),
16 |                      member[0].text,
17 |                      int(member[4][0].text),
18 |                      int(member[4][1].text),
19 |                      int(member[4][2].text),
20 |                      int(member[4][3].text)
21 |                      )
22 |             xml_list.append(value)
23 |     column_name = ['filename', 'width', 'height', 'class', 'xmin', 'ymin', 'xmax', 'ymax']
24 |     xml_df = pd.DataFrame(xml_list, columns=column_name)
25 |     return xml_df
26 | 
27 | 
28 | def main():
29 |     image_path = os.path.join(os.getcwd(), 'annotations')
30 |     xml_df = xml_to_csv(image_path)
31 |     xml_df.to_csv('celebrities_labels.csv', index=None)
32 |     print('Successfully converted xml to csv.')
33 | 
34 | 
35 | main()
36 | 


--------------------------------------------------------------------------------