├── models ├── discriminator.py ├── mlp.py └── graphcnn.py ├── README.md ├── .gitignore ├── evaluate ├── compute_silhouette.py ├── compute_robustness.py ├── visualize_saliency.py ├── plot_latent.py └── plot_saliency_nii.py ├── dataset.py ├── util.py ├── main.py └── LICENSE /models/discriminator.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import torch.nn as nn 3 | 4 | 5 | class Discriminator(nn.Module): 6 | def __init__(self, n_h): 7 | super(Discriminator, self).__init__() 8 | self.f_k = nn.Bilinear(n_h, n_h, 1) 9 | 10 | for m in self.modules(): 11 | self.weights_init(m) 12 | 13 | def weights_init(self, m): 14 | if isinstance(m, nn.Bilinear): 15 | torch.nn.init.xavier_uniform_(m.weight.data) 16 | if m.bias is not None: 17 | m.bias.data.fill_(0.0) 18 | 19 | def forward(self, c, h_pl, h_mi, s_bias1=None, s_bias2=None): 20 | # c_x = torch.unsqueeze(c, 1) 21 | c_x = c 22 | c_x_list=[] 23 | for c in c_x: 24 | c_x_list.append(c.expand([h_pl.shape[0]//c_x.shape[0],h_pl.shape[1]])) 25 | 26 | c_x = torch.cat(c_x_list, 0) 27 | 28 | sc_1= self.f_k(h_pl, c_x) 29 | sc_2 = self.f_k(h_mi, c_x) 30 | 31 | if s_bias1 is not None: 32 | sc_1 += s_bias1 33 | if s_bias2 is not None: 34 | sc_2 += s_bias2 35 | 36 | logits = torch.cat((sc_1, sc_2), 0) 37 | 38 | return logits 39 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Graph Neural Mapping 2 | 3 | ## Notice 4 | PyTorch implementation of the paper [Understanding Graph Isomorphism Network for rs-fMRI Functional Connectivity Analysis](https://www.frontiersin.org/articles/10.3389/fnins.2020.00630) by Byung-Hoon Kim and Jong Chul Ye. 5 | 6 | The model code is based on the official implementation of the 7 | - Graph Isomorphism Network: [paper](https://arxiv.org/abs/1810.00826), [github](https://github.com/weihua916/powerful-gnns) 8 | - Deep Graph Infomax: [paper](https://arxiv.org/abs/1809.10341), [github](https://github.com/PetarV-/DGI) 9 | 10 | ## Resources 11 | Dataset: 12 | - The Human Connectome Project: [paper](https://www.sciencedirect.com/science/article/pii/S1053811913005351), [web](https://www.humanconnectome.org/) 13 | 14 | Processing: 15 | - FSL: [paper](https://www.sciencedirect.com/science/article/pii/S1053811911010603), [web](https://fsl.fmrib.ox.ac.uk/fsl/fslwiki) 16 | - GRETNA: [paper](https://www.frontiersin.org/articles/10.3389/fnhum.2015.00386/full), [web](https://www.nitrc.org/projects/gretna/) 17 | 18 | Atlas: 19 | - Schaefer et al.: [paper](https://academic.oup.com/cercor/article/28/9/3095/3978804), [github](https://github.com/ThomasYeoLab/CBIG/tree/master/stable_projects/brain_parcellation/Schaefer2018_LocalGlobal) 20 | 21 | Visualization: 22 | - MRIcroGL: [github](https://github.com/rordenlab/MRIcroGL12), [web](https://www.nitrc.org/plugins/mwiki/index.php/mricrogl:MainPage) 23 | 24 | ## Requirements 25 | Python3 with following packages 26 | - `pytorch >= 1.4.0` 27 | - `scikit-learn >= 0.21.3` 28 | - `nilearn >= 0.5.2` 29 | - `nibabel >= 2.5.0` 30 | - `tqdm` 31 | -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | # Byte-compiled / optimized / DLL files 2 | __pycache__/ 3 | *.py[cod] 4 | *$py.class 5 | 6 | # C extensions 7 | *.so 8 | 9 | # Distribution / packaging 10 | .Python 11 | build/ 12 | develop-eggs/ 13 | dist/ 14 | downloads/ 15 | eggs/ 16 | .eggs/ 17 | lib/ 18 | lib64/ 19 | parts/ 20 | sdist/ 21 | var/ 22 | wheels/ 23 | *.egg-info/ 24 | .installed.cfg 25 | *.egg 26 | MANIFEST 27 | 28 | # PyInstaller 29 | # Usually these files are written by a python script from a template 30 | # before PyInstaller builds the exe, so as to inject date/other infos into it. 31 | *.manifest 32 | *.spec 33 | 34 | # Installer logs 35 | pip-log.txt 36 | pip-delete-this-directory.txt 37 | 38 | # Unit test / coverage reports 39 | htmlcov/ 40 | .tox/ 41 | .coverage 42 | .coverage.* 43 | .cache 44 | nosetests.xml 45 | coverage.xml 46 | *.cover 47 | .hypothesis/ 48 | .pytest_cache/ 49 | 50 | # Translations 51 | *.mo 52 | *.pot 53 | 54 | # Django stuff: 55 | *.log 56 | local_settings.py 57 | db.sqlite3 58 | 59 | # Flask stuff: 60 | instance/ 61 | .webassets-cache 62 | 63 | # Scrapy stuff: 64 | .scrapy 65 | 66 | # Sphinx documentation 67 | docs/_build/ 68 | 69 | # PyBuilder 70 | target/ 71 | 72 | # Jupyter Notebook 73 | .ipynb_checkpoints 74 | 75 | # pyenv 76 | .python-version 77 | 78 | # celery beat schedule file 79 | celerybeat-schedule 80 | 81 | # SageMath parsed files 82 | *.sage.py 83 | 84 | # Environments 85 | .env 86 | .venv 87 | env/ 88 | venv/ 89 | ENV/ 90 | env.bak/ 91 | venv.bak/ 92 | 93 | # Spyder project settings 94 | .spyderproject 95 | .spyproject 96 | 97 | # Rope project settings 98 | .ropeproject 99 | 100 | # mkdocs documentation 101 | /site 102 | 103 | # mypy 104 | .mypy_cache/ 105 | 106 | data/ 107 | results/ 108 | -------------------------------------------------------------------------------- /models/mlp.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import torch.nn as nn 3 | import torch.nn.functional as F 4 | 5 | ###MLP with lienar output 6 | class MLP(nn.Module): 7 | def __init__(self, num_layers, input_dim, hidden_dim, output_dim): 8 | ''' 9 | num_layers: number of layers in the neural networks (EXCLUDING the input layer). If num_layers=1, this reduces to linear model. 10 | input_dim: dimensionality of input features 11 | hidden_dim: dimensionality of hidden units at ALL layers 12 | output_dim: number of classes for prediction 13 | device: which device to use 14 | ''' 15 | 16 | super(MLP, self).__init__() 17 | 18 | self.linear_or_not = True #default is linear model 19 | self.num_layers = num_layers 20 | 21 | if num_layers < 1: 22 | raise ValueError("number of layers should be positive!") 23 | elif num_layers == 1: 24 | #Linear model 25 | self.linear = nn.Linear(input_dim, output_dim) 26 | else: 27 | #Multi-layer model 28 | self.linear_or_not = False 29 | self.linears = torch.nn.ModuleList() 30 | self.batch_norms = torch.nn.ModuleList() 31 | 32 | self.linears.append(nn.Linear(input_dim, hidden_dim)) 33 | for layer in range(num_layers - 2): 34 | self.linears.append(nn.Linear(hidden_dim, hidden_dim)) 35 | self.linears.append(nn.Linear(hidden_dim, output_dim)) 36 | 37 | for layer in range(num_layers - 1): 38 | self.batch_norms.append(nn.BatchNorm1d((hidden_dim))) 39 | 40 | def forward(self, x): 41 | if self.linear_or_not: 42 | #If linear model 43 | return self.linear(x) 44 | else: 45 | #If MLP 46 | h = x 47 | for layer in range(self.num_layers - 1): 48 | h = F.relu(self.batch_norms[layer](self.linears[layer](h))) 49 | return self.linears[self.num_layers - 1](h) 50 | -------------------------------------------------------------------------------- /evaluate/compute_silhouette.py: -------------------------------------------------------------------------------- 1 | import os 2 | import argparse 3 | import numpy as np 4 | from sklearn.metrics import silhouette_score 5 | 6 | 7 | def main(): 8 | parser = argparse.ArgumentParser(description='Compute the silhouette score of the latent space') 9 | parser.add_argument('--expdir', type=str, default='results/graph_neural_mapping', help='path to the experiment results') 10 | parser.add_argument('--latentdir', type=str, default='latent', help='path containing the latent_space_*.npy') 11 | parser.add_argument('--savedir', type=str, default='silhouette', help='path to save the silhouette value within the expdir') 12 | parser.add_argument('--fold_idx', nargs='+', default=['0','1','2','3','4','5','6','7','8','9'], help='fold indices') 13 | 14 | opt = parser.parse_args() 15 | os.makedirs(os.path.join(opt.expdir, opt.savedir), exist_ok=True) 16 | 17 | latent_space_initial = [] 18 | latent_space = [] 19 | labels = [] 20 | 21 | for current_fold in opt.fold_idx: 22 | latent_space_initial.append(np.load(os.path.join(opt.expdir, opt.latentdir, str(current_fold), 'latent_space_initial.npy'))) 23 | latent_space.append(np.load(os.path.join(opt.expdir, opt.latentdir, str(current_fold), 'latent_space.npy'))) 24 | labels.append(np.load(os.path.join(opt.expdir, opt.latentdir, str(current_fold), 'labels.npy')).squeeze()) 25 | 26 | initial_silhouette = [] 27 | silhouette = [] 28 | 29 | for idx, _ in enumerate(labels): 30 | initial_silhouette.append(silhouette_score(latent_space_initial[idx], labels[idx], sample_size=latent_space_initial[idx].shape[0])) 31 | silhouette.append(silhouette_score(latent_space[idx], labels[idx], sample_size=latent_space[idx].shape[0])) 32 | 33 | with open(os.path.join(opt.expdir, opt.savedir,'silhouette_score.csv'), 'w') as f: 34 | f.write("fold_idx,initial_silhouette_score,silhouette_score\n") 35 | for idx, (init_sil, sil) in enumerate(zip(initial_silhouette, silhouette)): 36 | f.write("{},{},{}\n".format(idx, init_sil, sil)) 37 | f.write("mean,{},{}\n".format(np.mean(initial_silhouette), np.mean(silhouette))) 38 | 39 | 40 | if __name__=='__main__': 41 | main() 42 | -------------------------------------------------------------------------------- /evaluate/compute_robustness.py: -------------------------------------------------------------------------------- 1 | import os 2 | import argparse 3 | import numpy as np 4 | import pandas as pd 5 | 6 | 7 | def main(): 8 | parser = argparse.ArgumentParser(description='compute robustness of the saliency mapping') 9 | parser.add_argument('--expdir', type=str, default='results/graph_neural_mapping', help='path to the experiment results') 10 | parser.add_argument('--saliency', type=str, default='saliency_female', help='saliency type') 11 | parser.add_argument('--topk', type=int, default=20, help='top k items to compare robustness') 12 | opt = parser.parse_args() 13 | 14 | 15 | full_folds = pd.read_csv(os.path.join(opt.expdir, 'saliency_nii', f'{opt.saliency}.csv')) 16 | one_folds = [pd.read_csv(os.path.join(opt.expdir, 'saliency_nii_fold{}'.format(i), f'{opt.saliency}.csv')) for i in range(10)] 17 | five_folds = [pd.read_csv(os.path.join(opt.expdir, 'saliency_nii_fold{}'.format(i), f'{opt.saliency}.csv')) for i in ['01234', '56789']] 18 | 19 | full_one_match = count_matches(full_folds, one_folds, opt.topk) 20 | full_one_match_mean = np.mean(full_one_match) 21 | full_one_match_std = np.std(full_one_match) 22 | one_folds_robustness = 100*full_one_match_mean / opt.topk 23 | one_folds_robustness_std = 100*full_one_match_std / opt.topk 24 | 25 | 26 | full_five_match = count_matches(full_folds, five_folds, opt.topk) 27 | full_five_match_mean = np.mean(full_five_match) 28 | full_five_match_std = np.std(full_five_match) 29 | five_folds_robustness = 100*full_five_match_mean / opt.topk 30 | five_folds_robustness_std = 100*full_five_match_std / opt.topk 31 | 32 | print('===='*12) 33 | print('===='*12) 34 | print(f'one fold matches {full_one_match_mean} out of {opt.topk}. robustness: {one_folds_robustness}+{one_folds_robustness_std:.2f}%') 35 | print(f'five fold matches {full_five_match_mean} out of {opt.topk}. robustness: {five_folds_robustness}+{five_folds_robustness_std:.2f}%') 36 | print('===='*12) 37 | print('===='*12) 38 | 39 | 40 | def count_matches(full_folds, partial_folds, topk): 41 | full_fold_rois = full_folds['roi'][:topk].to_list() 42 | partial_fold_rois = [fold['roi'][:topk].to_list() for fold in partial_folds] 43 | 44 | partial_fold_counts = [] 45 | for rois in partial_fold_rois: 46 | count = 0 47 | for roi in rois: 48 | if roi in full_fold_rois: 49 | count += 1 50 | partial_fold_counts.append(count) 51 | return partial_fold_counts 52 | 53 | 54 | if __name__=='__main__': 55 | main() 56 | -------------------------------------------------------------------------------- /evaluate/visualize_saliency.py: -------------------------------------------------------------------------------- 1 | # mricrogl>=1.2 python script 2 | # run example: 3 | # >> MRIcroGL evaluate/visualize_saliency.py 4 | 5 | import os 6 | import gl 7 | import argparse 8 | 9 | min = 0.75 10 | max = 1.0 11 | colorname = '4hot' 12 | 13 | def main(): 14 | EXP_DIR='results/graph_neural_mapping' 15 | NII_DIR='saliency_nii' 16 | SAVE_DIR='saliency_img' 17 | 18 | os.makedirs(os.path.join(EXP_DIR, SAVE_DIR), exist_ok=True) 19 | 20 | gl.resetdefaults() 21 | gl.loadimage('mni152') 22 | gl.overlayloadsmooth(True) 23 | gl.opacity(0, 50) 24 | gl.colorbarposition(0) 25 | 26 | visualize_axial(os.path.join(EXP_DIR, NII_DIR), os.path.join(EXP_DIR, SAVE_DIR)) 27 | visualize_sagittal(os.path.join(EXP_DIR, NII_DIR), os.path.join(EXP_DIR, SAVE_DIR)) 28 | visualize_colorbar(os.path.join(EXP_DIR, SAVE_DIR)) 29 | 30 | 31 | def visualize_axial(niidir, savedir, min=0.75, max=1.0): 32 | gl.viewaxial(1) 33 | for method in ['grad', 'cam']: 34 | for network in ['Vis', 'SomMot', 'DorsAttn', 'SalVentAttn', 'Limbic', 'Cont', 'Default']: 35 | for gender in ['female', 'male']: 36 | gl.overlayload(os.path.join(niidir, 'network', 'saliency_{}_{}_top20_{}_lh'.format(method, gender, network))) 37 | gl.overlayload(os.path.join(niidir, 'network', 'saliency_{}_{}_top20_{}_rh'.format(method, gender, network))) 38 | gl.minmax(1, min, max) 39 | gl.minmax(2, min, max) 40 | gl.colorname(1, colorname) 41 | gl.colorname(2, colorname) 42 | 43 | gl.savebmp(os.path.join(savedir, '{}_{}_{}_axial.png'.format(method, network, gender))) 44 | gl.overlaycloseall() 45 | 46 | def visualize_sagittal(niidir, savedir, min=0.75, max=1.0): 47 | for hemisphere in ['lh', 'rh']: 48 | if hemisphere =='lh': gl.clipazimuthelevation(0.49, 90, 0) 49 | elif hemisphere =='rh': gl.clipazimuthelevation(0.49, 270, 0) 50 | for network in ['Vis', 'SomMot', 'DorsAttn', 'SalVentAttn', 'Limbic', 'Cont', 'Default']: 51 | for gender in ['female', 'male']: 52 | gl.overlayload(os.path.join(niidir, 'network', 'saliency_{}_top20_{}_{}'.format(method, gender, network, hemisphere))) 53 | gl.minmax(1, min, max) 54 | gl.colorname(1, colorname) 55 | 56 | gl.viewsagittal(1) 57 | gl.savebmp(os.path.join(savedir, '{}_{}_sagittal_{}_lt.png'.format(network, gender, hemisphere))) 58 | gl.viewsagittal(0) 59 | gl.savebmp(os.path.join(savedir, '{}_{}_sagittal_{}_rt.png'.format(network, gender, hemisphere))) 60 | gl.overlaycloseall() 61 | 62 | def visualize_colorbar(savedir, min=0.75, max=1.0): 63 | gl.resetdefaults() 64 | gl.minmax(0, min, max) 65 | gl.colorname(0, colorname) 66 | gl.opacity(0, 0) 67 | gl.colorbarposition(1) 68 | gl.savebmp(os.path.join(savedir, 'colorbar.png')) 69 | 70 | 71 | if __name__=='__main__': 72 | main() 73 | -------------------------------------------------------------------------------- /evaluate/plot_latent.py: -------------------------------------------------------------------------------- 1 | import os 2 | import argparse 3 | import numpy as np 4 | import matplotlib.pyplot as plt 5 | 6 | from matplotlib.ticker import NullFormatter 7 | from sklearn import manifold 8 | from sklearn.metrics import silhouette_score 9 | 10 | 11 | def main(): 12 | parser = argparse.ArgumentParser(description='Plot the t-SNE embedding the latent space') 13 | parser.add_argument('--expdir', type=str, default='results/graph_neural_mapping', help='path to the experiment results') 14 | parser.add_argument('--latentdir', type=str, default='latent', help='path containing the latent_space_*.npy') 15 | parser.add_argument('--savedir', type=str, default='tsne', help='path to save the plotted tsne files within the expdir') 16 | parser.add_argument('--fold_idx', nargs='+', default=['0','1','2','3','4','5','6','7','8','9'], help='fold indices') 17 | parser.add_argument('--perplexities', type=int, nargs='+', default=[50], help='tsne perplexities') 18 | parser.add_argument('--random_state', type=int, default=0, help='tsne random state') 19 | 20 | opt = parser.parse_args() 21 | 22 | os.makedirs(os.path.join(opt.expdir, opt.savedir), exist_ok=True) 23 | 24 | latent_space_initial = [] 25 | latent_space = [] 26 | labels = [] 27 | 28 | for current_fold in opt.fold_idx: 29 | latent_space_initial.append(np.load(os.path.join(opt.expdir, opt.latentdir, str(current_fold), 'latent_space_initial.npy'))) 30 | latent_space.append(np.load(os.path.join(opt.expdir, opt.latentdir, str(current_fold), 'latent_space.npy'))) 31 | labels.append(np.load(os.path.join(opt.expdir, opt.latentdir, str(current_fold), 'labels.npy'))) 32 | 33 | for perplexity in opt.perplexities: 34 | print('PLOTTING PER-FOLD LATENT SPACE PERPLEXITY: {}'.format(perplexity)) 35 | for idx, _ in enumerate(labels): 36 | plot_tsne(latent_space_initial[idx], labels[idx], perplexity, os.path.join(opt.expdir, opt.savedir), 'initial_latent_perp{}_fold{}'.format(perplexity, idx), opt.random_state) 37 | plot_tsne(latent_space[idx], labels[idx], perplexity, os.path.join(opt.expdir, opt.savedir), 'latent_perp{}_fold{}'.format(perplexity, idx), opt.random_state) 38 | 39 | 40 | def plot_tsne(latent, label, perplexity, savedir, savename, random_state=0): 41 | n_samples = label.shape[0] 42 | n_components = label.max()+1 43 | 44 | X, y = latent, label.squeeze() 45 | 46 | female = y == 0 47 | male = y == 1 48 | 49 | fig, ax = plt.subplots() 50 | fig.set_size_inches((8,8)) 51 | ax.axis('off') 52 | tsne = manifold.TSNE(n_components=n_components, init='random', 53 | random_state=random_state, perplexity=perplexity) 54 | Y = tsne.fit_transform(X) 55 | plt.scatter(Y[female, 0], Y[female, 1], c="r", s=1000, linewidth=3, edgecolors='black') 56 | plt.scatter(Y[male, 0], Y[male, 1], c="b", s=1000, linewidth=3, edgecolors='black') 57 | plt.axis('tight') 58 | plt.savefig(os.path.join(savedir, savename)) 59 | plt.clf() 60 | plt.close() 61 | 62 | if __name__=='__main__': 63 | main() 64 | -------------------------------------------------------------------------------- /dataset.py: -------------------------------------------------------------------------------- 1 | import os 2 | import numpy as np 3 | import pandas as pd 4 | 5 | 6 | # Class of behavioral measurments 7 | class DataBehavioral(object): 8 | def __init__(self, sourcedir): 9 | super(DataBehavioral, self).__init__() 10 | self.df = pd.read_csv(os.path.join(sourcedir, 'behavioral', 'hcp.csv')).set_index('Subject') 11 | 12 | def get_feature(self, feature): 13 | behavioral_features = self.df[feature].to_dict() 14 | behavioral_label = {} 15 | 16 | for f in feature: 17 | behavioral_label[f] = {} 18 | seen = [] 19 | for v in behavioral_features[f].values(): 20 | if v not in seen: 21 | seen.append(v) 22 | seen.sort() 23 | for k, v in behavioral_features[f].items(): 24 | label = seen.index(v) 25 | behavioral_label[f][k] = label 26 | 27 | return behavioral_features, behavioral_label # dict {subject: feature_string} / dict {subject: label} 28 | 29 | 30 | # Class of nodes, i.e. ROI features 31 | class DataNodes(object): 32 | def __init__(self, sourcedir): 33 | super(DataNodes, self).__init__() 34 | self.df = pd.read_csv(os.path.join(sourcedir, 'roi', '7_400.txt'), index_col=0, header=None, delimiter='\t') 35 | self.features = self.df[1].str.split("_", expand=True) 36 | self.features.columns = ['YeoNetwork', 'Hemisphere', 'Network', 'Region', 'Index'] 37 | self.df_coord = pd.read_csv(os.path.join(sourcedir, 'roi', '7_400_coord.csv'), index_col=0)[1:] 38 | 39 | for i in self.features.index: 40 | row = self.features.loc[i] 41 | if row.isnull().any(): 42 | row[4] = row[3] 43 | row[3] = row[2] 44 | 45 | def __call__(self, subject): 46 | self.df_timeseries = pd.read_csv(os.path.join(sourcedir, 'timeseries', f'{subject}.txt'), index_col=False, header=None, delimiter='\t').dropna(axis='columns').to_numpy() 47 | 48 | def get_feature(self, type): # List of 'YeoNetwork', 'Hemisphere', 'Network', 'Region', 'Index' 49 | feature=['Hemisphere', 'Region', 'Network', 'Index'] 50 | if type=='one_hot': 51 | filtered_features = self.features[feature] 52 | node_features = filtered_features.apply(lambda x: '_'.join(x), axis='columns').to_dict() 53 | node_label = {} 54 | 55 | seen = [] 56 | for v in node_features.values(): 57 | if v not in seen: 58 | seen.append(v) 59 | for k, v in node_features.items(): 60 | label = seen.index(v) 61 | node_label[k] = label 62 | return node_features, node_label # dict {roi: feature_string} / dict {roi: label_value} 63 | 64 | elif type=='coordinate': 65 | filtered_features = self.df_coord[['R','A','S']] 66 | node_label_dict = filtered_features.to_dict() 67 | node_label = {} 68 | for k in node_label_dict['R'].keys(): 69 | node_label[k] = (node_label_dict['R'][k], node_label_dict['A'][k], node_label_dict['S'][k]) 70 | return node_label_dict, node_label # dict {R,A,S:{roi: coordinate}} / dict {roi: tuple (R,A,S) coordinate} 71 | 72 | elif type=='mean_bold': 73 | node_label_numpy = np.mean(self.df_timeseries, axis=0) 74 | node_label_numpy = (node_label_numpy - node_label_numpy.mean()) / (node_label_numpy.std() + 1e-8) 75 | node_label = {} 76 | for i, timeseries in enumerate(node_label_numpy): 77 | node_label[i] = tuple([timeseries]) 78 | return node_label_numpy, node_label 79 | 80 | else: 81 | raise Exception('unknown node feature type') 82 | 83 | 84 | # Class of edges, i.e. FC features 85 | class DataEdges(object): 86 | def __init__(self, sourcedir): 87 | super(DataEdges, self).__init__() 88 | self.sourcedir = sourcedir 89 | 90 | def __call__(self, subject): 91 | self.df = pd.read_csv(os.path.join(self.sourcedir, 'connectivity', f'r{subject}.txt'), index_col=False, header=None, delimiter='\t').dropna(axis='columns').to_numpy() 92 | 93 | def get_adjacency(self, threshold): 94 | mask = (self.df > np.percentile(self.df, threshold)).astype(np.uint8) 95 | nodes, neighbors = np.nonzero(mask) 96 | sparse_mask = {} 97 | for i, node in enumerate(nodes): 98 | if neighbors[i] > node: 99 | if not node in sparse_mask: sparse_mask[node] = [neighbors[i]] 100 | else: sparse_mask[node].append(neighbors[i]) 101 | return mask, sparse_mask # matrix adjacency / dict {roi: neighbor_roi} 102 | -------------------------------------------------------------------------------- /util.py: -------------------------------------------------------------------------------- 1 | import os 2 | import random 3 | import numpy as np 4 | import networkx as nx 5 | import torch 6 | from dataset import * 7 | from sklearn.model_selection import StratifiedKFold 8 | 9 | class S2VGraph(object): 10 | def __init__(self, g, label, node_tags=None, node_features=None): 11 | self.label = label 12 | self.g = g 13 | self.node_tags = node_tags 14 | self.neighbors = [] 15 | self.node_features = 0 16 | self.edge_mat = 0 17 | self.max_neighbor = 0 18 | 19 | 20 | def load_data(sourcedir, threshold, type): 21 | subject_list = [subject.split('.')[0][1:] for subject in os.listdir(os.path.join(sourcedir, 'connectivity'))] 22 | subject_list.sort() 23 | 24 | behav = DataBehavioral(sourcedir) 25 | roi = DataNodes(sourcedir) 26 | connectivity = DataEdges(sourcedir) 27 | 28 | _, behav_labels = behav.get_feature(['Gender']) 29 | 30 | g_list = [] 31 | label_dict = {} 32 | feat_dict = {} 33 | for i, subject in enumerate(subject_list): 34 | if 'bold' in type: roi(subject) 35 | _, node_labels = roi.get_feature(type) 36 | connectivity(subject) 37 | _, connection = connectivity.get_adjacency(100-threshold) 38 | n = node_labels 39 | l = behav_labels['Gender'][int(subject)] 40 | if not l in label_dict: 41 | mapped = len(label_dict) 42 | label_dict[l] = mapped 43 | g = nx.Graph() 44 | node_tags = [] 45 | node_features = [] 46 | n_edges = 0 47 | for j, node_label in enumerate(n.keys()): 48 | g.add_node(j) 49 | row = [node_labels[node_label]] 50 | if j in connection: 51 | row += [len(connection[j])] 52 | row += connection[j] 53 | else: 54 | row += [0] 55 | tmp = int(row[1]) + 2 56 | if tmp == len(row): 57 | attr = None 58 | else: 59 | row, attr = [int(w) for w in row[:tmp]], np.array([float(w) for w in row[tmp:]]) 60 | if type=='one_hot': 61 | if not row[0] in feat_dict: 62 | mapped = len(feat_dict) 63 | feat_dict[row[0]] = mapped 64 | node_tags.append(feat_dict[row[0]]) 65 | else: 66 | node_tags.append(row[0]) 67 | 68 | if tmp > len(row): 69 | node_features.append(attr) 70 | 71 | n_edges += row[1] 72 | for k in range(2, len(row)): 73 | g.add_edge(j, row[k]) 74 | 75 | if node_features != []: 76 | node_features = np.stack(node_features) 77 | node_feature_flag = True 78 | else: 79 | node_features = None 80 | node_feature_flag = False 81 | assert len(g) == len(n) 82 | 83 | g_list.append(S2VGraph(g, l, node_tags)) 84 | 85 | #add labels and edge_mat 86 | for g in g_list: 87 | g.neighbors = [[] for i in range(len(g.g))] 88 | for i, j in g.g.edges(): 89 | g.neighbors[i].append(j) 90 | g.neighbors[j].append(i) 91 | degree_list = [] 92 | for i in range(len(g.g)): 93 | g.neighbors[i] = g.neighbors[i] 94 | degree_list.append(len(g.neighbors[i])) 95 | g.max_neighbor = max(degree_list) 96 | 97 | g.label = label_dict[g.label] 98 | 99 | edges = [list(pair) for pair in g.g.edges()] 100 | edges.extend([[i, j] for j, i in edges]) 101 | 102 | deg_list = list(dict(g.g.degree(range(len(g.g)))).values()) 103 | g.edge_mat = torch.LongTensor(edges).transpose(0,1) 104 | 105 | #Extracting unique tag labels 106 | tagset = set([]) 107 | for g in g_list: 108 | tagset = tagset.union(set(g.node_tags)) 109 | 110 | tagset = list(tagset) 111 | tag2index = {tagset[i]:i for i in range(len(tagset))} 112 | 113 | for g in g_list: 114 | if type=='one_hot': 115 | g.node_features = torch.zeros(len(g.node_tags), len(tagset)) 116 | g.node_features[range(len(g.node_tags)), [tag2index[tag] for tag in g.node_tags]] = 1 117 | else: 118 | g.node_features = torch.zeros(len(g.node_tags), len(g.node_tags[0])) 119 | for i in range(len(g.node_tags)): 120 | for j in range(len(g.node_tags[0])): 121 | g.node_features[i, j] = g.node_tags[i][j] 122 | return g_list, len(label_dict) 123 | 124 | 125 | def separate_data(graph_list, seed, fold_idx): 126 | assert 0 <= fold_idx and fold_idx < 10, "fold_idx must be from 0 to 9." 127 | skf = StratifiedKFold(n_splits=10, shuffle = True, random_state = seed) 128 | 129 | labels = [graph.label for graph in graph_list] 130 | idx_list = [] 131 | for idx in skf.split(np.zeros(len(labels)), labels): 132 | idx_list.append(idx) 133 | train_idx, test_idx = idx_list[fold_idx] 134 | 135 | train_graph_list = [graph_list[i] for i in train_idx] 136 | test_graph_list = [graph_list[i] for i in test_idx] 137 | 138 | return train_graph_list, test_graph_list 139 | -------------------------------------------------------------------------------- /evaluate/plot_saliency_nii.py: -------------------------------------------------------------------------------- 1 | import os 2 | import argparse 3 | import numpy as np 4 | import pandas as pd 5 | import nibabel as nib 6 | import nilearn as nil 7 | 8 | 9 | def main(): 10 | parser = argparse.ArgumentParser(description='Plot the saliency map in the nifti format') 11 | parser.add_argument('--expdir', type=str, default='results/graph_neural_mapping', help='path containing the saliency_female.npy and the saliency_male.npy') 12 | parser.add_argument('--roidir', type=str, default='data/roi/Schaefer2018_400Parcels_7Networks_order_FSLMNI152_2mm.nii.gz', help='path containing the nifti ROI file') 13 | parser.add_argument('--roimetadir', type=str, default='data/roi/7_400.txt', help='path containing the metadata of the ROI file') 14 | parser.add_argument('--topk', type=int, default=20, help='top k rois to visualize') 15 | parser.add_argument('--savedir', type=str, default='saliency_nii', help='path to save the saliency nii files within the expdir') 16 | parser.add_argument('--fold_idx', nargs='+', default=['0','1','2','3','4','5','6','7','8','9'], help='fold indices') 17 | 18 | opt = parser.parse_args() 19 | 20 | os.makedirs(os.path.join(opt.expdir, opt.savedir), exist_ok=True) 21 | os.makedirs(os.path.join(opt.expdir, opt.savedir, 'network'), exist_ok=True) 22 | os.makedirs(os.path.join(opt.expdir, opt.savedir, 'description'), exist_ok=True) 23 | 24 | roiimg = nil.image.load_img(opt.roidir) 25 | roiimgarray = roiimg.get_fdata() 26 | roiimgaffine = roiimg.affine 27 | roimeta = pd.read_csv(opt.roimetadir, index_col=0, header=None, delimiter='\t') 28 | 29 | # plot proposed based saliency 30 | saliency0 = [] 31 | saliency1 = [] 32 | 33 | for current_fold in opt.fold_idx: 34 | saliency0.append(np.load(os.path.join(opt.expdir, 'saliency', str(current_fold), 'saliency_female.npy'))) 35 | saliency1.append(np.load(os.path.join(opt.expdir, 'saliency', str(current_fold), 'saliency_male.npy'))) 36 | 37 | saliency0 = np.mean(np.concatenate(saliency0, 0), axis=1) 38 | saliency1 = np.mean(np.concatenate(saliency1, 0), axis=1) 39 | 40 | saliency0subjects = [] 41 | saliency1subjects = [] 42 | 43 | for subjidx, (sal0, sal1) in enumerate(zip(saliency0, saliency1)): 44 | saliency0array = roiimgarray.copy() 45 | saliency1array = roiimgarray.copy() 46 | print("EXTRACTING SALIENCY SUBJECT: {}".format(subjidx)) 47 | for i, (s0, s1) in enumerate(zip(sal0, sal1)): 48 | roi_voxel_idx = np.where(roiimgarray==i+1) 49 | for j in range(roi_voxel_idx[0].shape[0]): 50 | saliency0array[roi_voxel_idx[0][j], roi_voxel_idx[1][j], roi_voxel_idx[2][j]] = s0 51 | saliency1array[roi_voxel_idx[0][j], roi_voxel_idx[1][j], roi_voxel_idx[2][j]] = s1 52 | 53 | saliency0subjects.append(saliency0array) 54 | saliency1subjects.append(saliency1array) 55 | 56 | plot_nii(saliency0subjects, opt.topk, roiimgaffine, roiimgarray, roimeta, os.path.join(opt.expdir, opt.savedir), 'female') 57 | plot_nii(saliency1subjects, opt.topk, roiimgaffine, roiimgarray, roimeta, os.path.join(opt.expdir, opt.savedir), 'male') 58 | 59 | 60 | def plot_nii(subject_list, topk, roiimgaffine, roiimgarray, roimeta, savepath, desc): 61 | saliency_array = np.mean(np.stack(subject_list), axis=0) 62 | 63 | saliency_array_normalized = saliency_array.copy() 64 | saliency_array_normalized -= saliency_array_normalized.min() 65 | saliency_array_normalized /= saliency_array_normalized.max() 66 | 67 | if topk: 68 | values = np.unique(saliency_array_normalized) 69 | topk_idx = np.argsort(values)[-topk] 70 | topk_value = values[topk_idx] 71 | saliency_array_normalized_topk = saliency_array_normalized.copy() 72 | saliency_array_normalized_topk[saliency_array_normalized_topkthreshold).astype(np.uint8) 108 | else: 109 | normalized_idx = (roiimgarray!=0).astype(np.uint8) 110 | idx_tuple = np.nonzero(normalized_idx) 111 | rois = [] 112 | values = [] 113 | abs_values = [] 114 | labels = [] 115 | for i in range(len(idx_tuple[0])): 116 | roi = roiimgarray[idx_tuple[0][i],idx_tuple[1][i],idx_tuple[2][i]] 117 | value = normalized_array[idx_tuple[0][i],idx_tuple[1][i],idx_tuple[2][i]] 118 | if str(roi) not in rois: 119 | assert value not in values 120 | rois.append(str(roi)) 121 | values.append(str(value)) 122 | abs_values.append(str(abs(value))) 123 | labels.append(str(roimeta[1][roi])) 124 | zipped = list(zip(abs_values, rois, labels, values)) 125 | zipped.sort(reverse=True) 126 | 127 | with open(os.path.join(savepath, 'description', 'saliency_{}.csv'.format(desc)), 'w') as f: 128 | f.write('abs_value,roi,label,value\n') 129 | for item in zipped: 130 | f.write(','.join(item)) 131 | f.write('\n') 132 | 133 | 134 | if __name__ == '__main__': 135 | main() 136 | -------------------------------------------------------------------------------- /main.py: -------------------------------------------------------------------------------- 1 | import os 2 | import csv 3 | import argparse 4 | import numpy as np 5 | import torch 6 | import torch.nn as nn 7 | import torch.optim as optim 8 | 9 | from models.graphcnn import * 10 | from util import load_data, separate_data 11 | from tqdm import tqdm 12 | from sklearn import metrics 13 | from torch.utils.tensorboard import SummaryWriter 14 | 15 | 16 | c_criterion = nn.CrossEntropyLoss() 17 | d_criterion = nn.BCEWithLogitsLoss() 18 | 19 | def train(args, model, device, train_graphs, optimizer, beta, epoch): 20 | model.train() 21 | 22 | num_rois = train_graphs[0].node_features.shape[1] 23 | total_iters = args.iters_per_epoch 24 | loss_accum = 0 25 | for pos in range(total_iters): 26 | selected_idx = np.random.permutation(len(train_graphs))[:args.batch_size] 27 | 28 | batch_graph = [train_graphs[idx] for idx in selected_idx] 29 | c_logit, d_logit = model(batch_graph) 30 | 31 | c_labels = torch.LongTensor([graph.label for graph in batch_graph]).to(device) 32 | d_labels = torch.cat([torch.ones(args.batch_size*num_rois, 1), torch.zeros(args.batch_size*num_rois, 1)], 0).to(device) 33 | 34 | d_loss = d_criterion(d_logit, d_labels) 35 | c_loss = c_criterion(c_logit, c_labels) 36 | 37 | loss = c_loss + beta*d_loss 38 | 39 | optimizer.zero_grad() 40 | loss.backward() 41 | optimizer.step() 42 | 43 | loss = loss.detach().cpu().numpy() 44 | loss_accum += loss 45 | 46 | average_loss = loss_accum/total_iters 47 | return average_loss 48 | 49 | def pass_data_iteratively(model, graphs): 50 | model.eval() 51 | c_logit_list = [] 52 | d_logit_list = [] 53 | for g in graphs: 54 | c_logit, d_logit = model([g]) 55 | c_logit_list.append(c_logit.detach()) 56 | d_logit_list.append(d_logit) 57 | return torch.cat(c_logit_list, 0), torch.cat(d_logit_list, 0) 58 | 59 | 60 | def get_saliency_map(model, graphs, cls): 61 | model.eval() 62 | saliency_maps = [] 63 | for graph in graphs: 64 | saliency_map = model.compute_saliency([graph], cls) 65 | saliency_maps.append(saliency_map.detach().cpu().numpy()) 66 | 67 | saliency_maps = np.stack(saliency_maps, axis=0) 68 | return saliency_maps 69 | 70 | 71 | def get_latent_space(model, graphs): 72 | model.eval() 73 | output_list = [] 74 | label_list = [] 75 | for g in graphs: 76 | latent = model([g], latent=True) 77 | label = np.array([g.label]) 78 | output_list.append(latent) 79 | label_list.append(label) 80 | latent_space = np.concatenate(output_list, axis=0) 81 | labels = np.stack(label_list, axis=0) 82 | return latent_space, labels 83 | 84 | 85 | def test(args, model, device, graphs): 86 | model.eval() 87 | output, _ = pass_data_iteratively(model, graphs) 88 | labels = torch.LongTensor([graph.label for graph in graphs]).to(device) 89 | 90 | pred = output.max(1, keepdim=True)[1] 91 | pred = pred.detach().cpu().numpy() 92 | labels = labels.detach().cpu().numpy() 93 | accuracy = metrics.accuracy_score(labels, pred) 94 | precision = metrics.precision_score(labels, pred) 95 | recall = metrics.recall_score(labels, pred) 96 | return accuracy, precision, recall 97 | 98 | 99 | def main(): 100 | # Training settings 101 | # Note: Hyper-parameters need to be tuned in order to obtain results reported in the paper. 102 | parser = argparse.ArgumentParser(description='PyTorch GIN fMRI') 103 | parser.add_argument('--device', type=int, default=0, help='which gpu to use if any') 104 | parser.add_argument('--sourcedir', type=str, default='data', help='path to the data directory') 105 | parser.add_argument('--sparsity', type=int, default=30, help='sparsity M of graph adjacency') 106 | parser.add_argument('--input_feature', type=str, default='one_hot', help='input feature type', choices=['one_hot', 'coordinate', 'mean_bold']) 107 | parser.add_argument('--batch_size', type=int, default=32, help='input minibatch size for training') 108 | parser.add_argument('--iters_per_epoch', type=int, default=50, help='number of iterations per each epoch') 109 | parser.add_argument('--epochs', type=int, default=150, help='number of epochs to train') 110 | parser.add_argument('--lr', type=float, default=0.005, help='initial learning rate') 111 | parser.add_argument('--lr_step', type=int, default=5, help='learning rate decay step') 112 | parser.add_argument('--lr_rate', type=float, default=0.8, help='learning rate decay rate') 113 | parser.add_argument('--fold_seed', type=int, default=0, help='random seed for splitting the dataset') 114 | parser.add_argument('--fold_idx', type=int, default=0, help='indices of fold in 10-fold validation.') 115 | parser.add_argument('--num_layers', type=int, default=5, help='number of the GNN layers') 116 | parser.add_argument('--num_mlp_layers', type=int, default=2, help='number of layers for the MLP. 1 means linear model.') 117 | parser.add_argument('--hidden_dim', type=int, default=64, help='number of hidden units') 118 | parser.add_argument('--beta', type=float, default=0.05, help='coefficient for infograph regularizer') 119 | parser.add_argument('--final_dropout', type=float, default=0.5, help='final layer dropout') 120 | parser.add_argument('--graph_pooling_type', type=str, default="sum", choices=["sum", "average"], help='Pooling for over nodes in a graph: sum or average') 121 | parser.add_argument('--neighbor_pooling_type', type=str, default="sum", choices=["sum", "average", "max"], help='Pooling for over neighboring nodes: sum, average or max') 122 | parser.add_argument('--learn_eps', action="store_true", help='whether to learn the epsilon weighting for the center nodes. Does not affect training accuracy though.') 123 | parser.add_argument('--exp', type = str, default = "graph_neural_mapping", help='experiment name') 124 | args = parser.parse_args() 125 | 126 | device = torch.device("cuda" if torch.cuda.is_available() else "cpu") 127 | graphs, num_classes = load_data(args.sourcedir, args.sparsity, args.input_feature) 128 | 129 | os.makedirs('results/{}/saliency/{}'.format(args.exp, args.fold_idx), exist_ok=True) 130 | os.makedirs('results/{}/latent/{}'.format(args.exp, args.fold_idx), exist_ok=True) 131 | os.makedirs('results/{}/model/{}'.format(args.exp, args.fold_idx), exist_ok=True) 132 | 133 | train_graphs, test_graphs = separate_data(graphs, args.fold_seed, args.fold_idx) 134 | 135 | model = GIN_InfoMaxReg(args.num_layers, args.num_mlp_layers, train_graphs[0].node_features.shape[1], args.hidden_dim, num_classes, args.final_dropout, args.learn_eps, args.graph_pooling_type, args.neighbor_pooling_type, device).to(device) 136 | optimizer = optim.Adam(model.parameters(), lr=args.lr) 137 | scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=args.lr_step, gamma=args.lr_rate) 138 | 139 | train_summary_writer = SummaryWriter('results/{}/summary/{}/train'.format(args.exp, args.fold_idx), flush_secs=1, max_queue=1) 140 | test_summary_writer = SummaryWriter('results/{}/summary/{}/test'.format(args.exp, args.fold_idx), flush_secs=1, max_queue=1) 141 | with open('results/{}/argv.csv'.format(args.exp), 'a', newline='') as f: 142 | writer = csv.writer(f) 143 | writer.writerows(vars(args).items()) 144 | 145 | latent_space_initial, labels = get_latent_space(model, test_graphs) 146 | np.save('results/{}/latent/{}/latent_space_initial.npy'.format(args.exp, args.fold_idx), latent_space_initial) 147 | np.save('results/{}/latent/{}/labels.npy'.format(args.exp, args.fold_idx), labels) 148 | del latent_space_initial 149 | del labels 150 | 151 | for epoch in tqdm(range(args.epochs), ncols=50, desc=f'{args.fold_idx}'): 152 | loss_train = train(args, model, device, train_graphs, optimizer, args.beta, epoch) 153 | scheduler.step() 154 | acc_train, precision_train, recall_train = test(args, model, device, train_graphs) 155 | 156 | train_summary_writer.add_scalar('loss/total', loss_train, epoch) 157 | train_summary_writer.add_scalar('metrics/accuracy', acc_train, epoch) 158 | train_summary_writer.add_scalar('metrics/precision', precision_train, epoch) 159 | train_summary_writer.add_scalar('metrics/recall', recall_train, epoch) 160 | 161 | acc_test, precision_test, recall_test = test(args, model, device, test_graphs) 162 | test_summary_writer.add_scalar('metrics/accuracy', acc_test, epoch) 163 | test_summary_writer.add_scalar('metrics/precision', precision_test, epoch) 164 | test_summary_writer.add_scalar('metrics/recall', recall_test, epoch) 165 | 166 | torch.save(model.state_dict(), 'results/{}/model/{}/model.pt'.format(args.exp, args.fold_idx)) 167 | latent_space, labels = get_latent_space(model, test_graphs) 168 | saliency_map_0 = get_saliency_map(model, test_graphs, 0) 169 | saliency_map_1 = get_saliency_map(model, test_graphs, 1) 170 | np.save('results/{}/latent/{}/latent_space.npy'.format(args.exp, args.fold_idx), latent_space) 171 | np.save('results/{}/saliency/{}/saliency_female.npy'.format(args.exp, args.fold_idx), saliency_map_0) 172 | np.save('results/{}/saliency/{}/saliency_male.npy'.format(args.exp, args.fold_idx), saliency_map_1) 173 | 174 | 175 | if __name__ == '__main__': 176 | main() 177 | -------------------------------------------------------------------------------- /models/graphcnn.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import torch.nn as nn 3 | import torch.nn.functional as F 4 | import numpy as np 5 | 6 | import sys 7 | sys.path.append("models/") 8 | from mlp import MLP 9 | from discriminator import Discriminator 10 | 11 | 12 | class GIN_InfoMaxReg(nn.Module): 13 | def __init__(self, num_layers, num_mlp_layers, input_dim, hidden_dim, output_dim, final_dropout, learn_eps, graph_pooling_type, neighbor_pooling_type, device): 14 | ''' 15 | num_layers: number of layers in the neural networks (INCLUDING the input layer) 16 | num_mlp_layers: number of layers in mlps (EXCLUDING the input layer) 17 | input_dim: dimensionality of input features 18 | hidden_dim: dimensionality of hidden units at ALL layers 19 | output_dim: number of classes for prediction 20 | final_dropout: dropout ratio on the final linear layer 21 | learn_eps: If True, learn epsilon to distinguish center nodes from neighboring nodes. If False, aggregate neighbors and center nodes altogether. 22 | neighbor_pooling_type: how to aggregate neighbors (mean, average, or max) 23 | graph_pooling_type: how to aggregate entire nodes in a graph (mean, average) 24 | device: which device to use 25 | ''' 26 | 27 | super(GIN_InfoMaxReg, self).__init__() 28 | 29 | self.disc = Discriminator(hidden_dim*num_layers) 30 | self.sigm = nn.Sigmoid() 31 | self.relu = nn.ReLU() 32 | 33 | self.final_dropout = final_dropout 34 | self.device = device 35 | self.num_layers = num_layers 36 | self.graph_pooling_type = graph_pooling_type 37 | self.neighbor_pooling_type = neighbor_pooling_type 38 | self.learn_eps = learn_eps 39 | self.eps = nn.Parameter(torch.zeros(num_layers)) 40 | 41 | self.mlps = torch.nn.ModuleList() 42 | self.batch_norms = torch.nn.ModuleList() 43 | self.linears_prediction = torch.nn.ModuleList() 44 | 45 | for layer in range(num_layers): 46 | if layer == 0: 47 | self.mlps.append(MLP(num_mlp_layers, input_dim, hidden_dim, hidden_dim)) 48 | else: 49 | self.mlps.append(MLP(num_mlp_layers, hidden_dim, hidden_dim, hidden_dim)) 50 | 51 | self.batch_norms.append(nn.BatchNorm1d(hidden_dim)) 52 | self.linears_prediction.append(nn.Linear(hidden_dim, output_dim)) 53 | 54 | 55 | def __preprocess_neighbors_maxpool(self, batch_graph): 56 | ###create padded_neighbor_list in concatenated graph 57 | 58 | #compute the maximum number of neighbors within the graphs in the current minibatch 59 | max_deg = max([graph.max_neighbor for graph in batch_graph]) 60 | 61 | padded_neighbor_list = [] 62 | start_idx = [0] 63 | 64 | 65 | for i, graph in enumerate(batch_graph): 66 | start_idx.append(start_idx[i] + len(graph.g)) 67 | padded_neighbors = [] 68 | for j in range(len(graph.neighbors)): 69 | #add off-set values to the neighbor indices 70 | pad = [n + start_idx[i] for n in graph.neighbors[j]] 71 | #padding, dummy data is assumed to be stored in -1 72 | pad.extend([-1]*(max_deg - len(pad))) 73 | 74 | #Add center nodes in the maxpooling if learn_eps is False, i.e., aggregate center nodes and neighbor nodes altogether. 75 | if not self.learn_eps: 76 | pad.append(j + start_idx[i]) 77 | 78 | padded_neighbors.append(pad) 79 | padded_neighbor_list.extend(padded_neighbors) 80 | 81 | return torch.LongTensor(padded_neighbor_list) 82 | 83 | 84 | def __preprocess_neighbors_sumavepool(self, batch_graph): 85 | ###create block diagonal sparse matrix 86 | 87 | edge_mat_list = [] 88 | start_idx = [0] 89 | for i, graph in enumerate(batch_graph): 90 | start_idx.append(start_idx[i] + len(graph.g)) 91 | edge_mat_list.append(graph.edge_mat + start_idx[i]) 92 | Adj_block_idx = torch.cat(edge_mat_list, 1) 93 | Adj_block_elem = torch.ones(Adj_block_idx.shape[1]) 94 | 95 | #Add self-loops in the adjacency matrix if learn_eps is False, i.e., aggregate center nodes and neighbor nodes altogether. 96 | 97 | if not self.learn_eps: 98 | num_node = start_idx[-1] 99 | self_loop_edge = torch.LongTensor([range(num_node), range(num_node)]) 100 | elem = torch.ones(num_node) 101 | Adj_block_idx = torch.cat([Adj_block_idx, self_loop_edge], 1) 102 | Adj_block_elem = torch.cat([Adj_block_elem, elem], 0) 103 | 104 | Adj_block = torch.sparse.FloatTensor(Adj_block_idx, Adj_block_elem, torch.Size([start_idx[-1],start_idx[-1]])) 105 | 106 | return Adj_block.to(self.device) 107 | 108 | 109 | def __preprocess_graphpool(self, batch_graph): 110 | ###create sum or average pooling sparse matrix over entire nodes in each graph (num graphs x num nodes) 111 | 112 | start_idx = [0] 113 | 114 | #compute the padded neighbor list 115 | for i, graph in enumerate(batch_graph): 116 | start_idx.append(start_idx[i] + len(graph.g)) 117 | 118 | idx = [] 119 | elem = [] 120 | for i, graph in enumerate(batch_graph): 121 | ###average pooling 122 | if self.graph_pooling_type == "average": 123 | elem.extend([1./len(graph.g)]*len(graph.g)) 124 | 125 | else: 126 | ###sum pooling 127 | elem.extend([1]*len(graph.g)) 128 | 129 | idx.extend([[i, j] for j in range(start_idx[i], start_idx[i+1], 1)]) 130 | elem = torch.FloatTensor(elem) 131 | idx = torch.LongTensor(idx).transpose(0,1) 132 | graph_pool = torch.sparse.FloatTensor(idx, elem, torch.Size([len(batch_graph), start_idx[-1]])) 133 | 134 | return graph_pool.to(self.device) 135 | 136 | 137 | def maxpool(self, h, padded_neighbor_list): 138 | ###Element-wise minimum will never affect max-pooling 139 | 140 | dummy = torch.min(h, dim = 0)[0] 141 | h_with_dummy = torch.cat([h, dummy.reshape((1, -1)).to(self.device)]) 142 | pooled_rep = torch.max(h_with_dummy[padded_neighbor_list], dim = 1)[0] 143 | return pooled_rep 144 | 145 | 146 | def next_layer_eps(self, h, layer, padded_neighbor_list = None, Adj_block = None): 147 | ###pooling neighboring nodes and center nodes separately by epsilon reweighting. 148 | 149 | if self.neighbor_pooling_type == "max": 150 | ##If max pooling 151 | pooled = self.maxpool(h, padded_neighbor_list) 152 | else: 153 | #If sum or average pooling 154 | pooled = torch.spmm(Adj_block, h) 155 | if self.neighbor_pooling_type == "average": 156 | #If average pooling 157 | degree = torch.spmm(Adj_block, torch.ones((Adj_block.shape[0], 1)).to(self.device)) 158 | pooled = pooled/degree 159 | 160 | #Reweights the center node representation when aggregating it with its neighbors 161 | pooled = pooled + (1 + self.eps[layer])*h 162 | pooled_rep = self.mlps[layer](pooled) 163 | h = self.batch_norms[layer](pooled_rep) 164 | 165 | #non-linearity 166 | h = self.relu(h) 167 | return h 168 | 169 | 170 | def next_layer(self, h, layer, padded_neighbor_list = None, Adj_block = None): 171 | ###pooling neighboring nodes and center nodes altogether 172 | 173 | if self.neighbor_pooling_type == "max": 174 | ##If max pooling 175 | pooled = self.maxpool(h, padded_neighbor_list) 176 | else: 177 | #If sum or average pooling 178 | pooled = torch.spmm(Adj_block, h) 179 | if self.neighbor_pooling_type == "average": 180 | #If average pooling 181 | degree = torch.spmm(Adj_block, torch.ones((Adj_block.shape[0], 1)).to(self.device)) 182 | pooled = pooled/degree 183 | 184 | #representation of neighboring and center nodes 185 | pooled_rep = self.mlps[layer](pooled) 186 | 187 | h = self.batch_norms[layer](pooled_rep) 188 | 189 | #non-linearity 190 | h = self.relu(h) 191 | return h 192 | 193 | 194 | def forward(self, batch_graph, latent=False): 195 | X_concat = torch.cat([graph.node_features for graph in batch_graph], 0).to(self.device) 196 | graph_pool = self.__preprocess_graphpool(batch_graph) 197 | 198 | idx = [] 199 | rand_seq = np.random.permutation(len(batch_graph)) 200 | for i in rand_seq: 201 | idx += [i]*len(batch_graph[0].node_features) 202 | 203 | if self.neighbor_pooling_type == "max": 204 | padded_neighbor_list = self.__preprocess_neighbors_maxpool(batch_graph) 205 | else: 206 | Adj_block = self.__preprocess_neighbors_sumavepool(batch_graph) 207 | 208 | #list of hidden representation at each layer (including input) 209 | hidden_rep = [] 210 | h = X_concat 211 | 212 | for layer in range(self.num_layers): 213 | if self.neighbor_pooling_type == "max" and self.learn_eps: 214 | h = self.next_layer_eps(h, layer, padded_neighbor_list = padded_neighbor_list) 215 | elif not self.neighbor_pooling_type == "max" and self.learn_eps: 216 | h = self.next_layer_eps(h, layer, Adj_block = Adj_block) 217 | elif self.neighbor_pooling_type == "max" and not self.learn_eps: 218 | h = self.next_layer(h, layer, padded_neighbor_list = padded_neighbor_list) 219 | elif not self.neighbor_pooling_type == "max" and not self.learn_eps: 220 | h = self.next_layer(h, layer, Adj_block = Adj_block) 221 | 222 | hidden_rep.append(h) 223 | 224 | c_logit = 0 225 | graph_latent = [] 226 | 227 | #perform pooling over all nodes in each graph in every layer 228 | for layer, h in enumerate(hidden_rep): 229 | pooled_h = torch.spmm(graph_pool, h) 230 | c_logit += F.dropout(self.linears_prediction[layer](pooled_h), self.final_dropout, training = self.training) # [32,2] 231 | graph_latent.append(pooled_h) 232 | 233 | n_f = torch.cat(hidden_rep, 1) 234 | g_f = torch.cat(graph_latent, 1) 235 | 236 | h_1 = n_f 237 | 238 | c = g_f 239 | c = self.sigm(c) 240 | 241 | idx = np.asarray(idx) 242 | shuf_n_f = n_f[idx, :] 243 | 244 | h_2 = shuf_n_f 245 | 246 | d_logit = self.disc(c, h_1, h_2, None, None) 247 | 248 | if latent: 249 | return g_f.detach().cpu().numpy() 250 | else: 251 | return c_logit, d_logit 252 | 253 | 254 | def compute_saliency(self, batch_graph, cls): 255 | self.eval() 256 | self.zero_grad() 257 | assert len(batch_graph)==1 258 | X_concat = torch.cat([graph.node_features for graph in batch_graph], 0).to(self.device) 259 | X_concat.requires_grad_() 260 | graph_pool = self.__preprocess_graphpool(batch_graph) 261 | 262 | # predicting 0 263 | predicting_class = torch.zeros([1,2]).to(self.device) 264 | predicting_class[0, cls] = 1 265 | 266 | if self.neighbor_pooling_type == "max": 267 | padded_neighbor_list = self.__preprocess_neighbors_maxpool(batch_graph) 268 | else: 269 | Adj_block = self.__preprocess_neighbors_sumavepool(batch_graph) 270 | 271 | #list of hidden representation at each layer (not including input) 272 | hidden_rep = [] 273 | h = X_concat 274 | 275 | for layer in range(self.num_layers): 276 | if self.neighbor_pooling_type == "max" and self.learn_eps: 277 | h = self.next_layer_eps(h, layer, padded_neighbor_list = padded_neighbor_list) 278 | elif not self.neighbor_pooling_type == "max" and self.learn_eps: 279 | h = self.next_layer_eps(h, layer, Adj_block = Adj_block) 280 | elif self.neighbor_pooling_type == "max" and not self.learn_eps: 281 | h = self.next_layer(h, layer, padded_neighbor_list = padded_neighbor_list) 282 | elif not self.neighbor_pooling_type == "max" and not self.learn_eps: 283 | h = self.next_layer(h, layer, Adj_block = Adj_block) 284 | h.retain_grad() 285 | hidden_rep.append(h) 286 | 287 | score_over_layer = 0 288 | class_activation = torch.zeros([X_concat.shape[0]]).to(self.device) 289 | grad_class_activation = torch.zeros([X_concat.shape[0]]).to(self.device) 290 | 291 | #perform pooling over all nodes in each graph in every layer 292 | for layer, h in enumerate(hidden_rep): 293 | pooled_h = torch.spmm(graph_pool, h) 294 | score_over_layer += F.dropout(self.linears_prediction[layer](pooled_h), self.final_dropout, training = self.training) 295 | 296 | score_over_layer.backward(predicting_class) 297 | saliency = X_concat.grad 298 | 299 | return saliency 300 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | GNU GENERAL PUBLIC LICENSE 2 | Version 3, 29 June 2007 3 | 4 | Copyright (C) 2007 Free Software Foundation, Inc. 5 | Everyone is permitted to copy and distribute verbatim copies 6 | of this license document, but changing it is not allowed. 7 | 8 | Preamble 9 | 10 | The GNU General Public License is a free, copyleft license for 11 | software and other kinds of works. 12 | 13 | The licenses for most software and other practical works are designed 14 | to take away your freedom to share and change the works. By contrast, 15 | the GNU General Public License is intended to guarantee your freedom to 16 | share and change all versions of a program--to make sure it remains free 17 | software for all its users. We, the Free Software Foundation, use the 18 | GNU General Public License for most of our software; it applies also to 19 | any other work released this way by its authors. You can apply it to 20 | your programs, too. 21 | 22 | When we speak of free software, we are referring to freedom, not 23 | price. Our General Public Licenses are designed to make sure that you 24 | have the freedom to distribute copies of free software (and charge for 25 | them if you wish), that you receive source code or can get it if you 26 | want it, that you can change the software or use pieces of it in new 27 | free programs, and that you know you can do these things. 28 | 29 | To protect your rights, we need to prevent others from denying you 30 | these rights or asking you to surrender the rights. Therefore, you have 31 | certain responsibilities if you distribute copies of the software, or if 32 | you modify it: responsibilities to respect the freedom of others. 33 | 34 | For example, if you distribute copies of such a program, whether 35 | gratis or for a fee, you must pass on to the recipients the same 36 | freedoms that you received. You must make sure that they, too, receive 37 | or can get the source code. And you must show them these terms so they 38 | know their rights. 39 | 40 | Developers that use the GNU GPL protect your rights with two steps: 41 | (1) assert copyright on the software, and (2) offer you this License 42 | giving you legal permission to copy, distribute and/or modify it. 43 | 44 | For the developers' and authors' protection, the GPL clearly explains 45 | that there is no warranty for this free software. For both users' and 46 | authors' sake, the GPL requires that modified versions be marked as 47 | changed, so that their problems will not be attributed erroneously to 48 | authors of previous versions. 49 | 50 | Some devices are designed to deny users access to install or run 51 | modified versions of the software inside them, although the manufacturer 52 | can do so. This is fundamentally incompatible with the aim of 53 | protecting users' freedom to change the software. The systematic 54 | pattern of such abuse occurs in the area of products for individuals to 55 | use, which is precisely where it is most unacceptable. Therefore, we 56 | have designed this version of the GPL to prohibit the practice for those 57 | products. If such problems arise substantially in other domains, we 58 | stand ready to extend this provision to those domains in future versions 59 | of the GPL, as needed to protect the freedom of users. 60 | 61 | Finally, every program is threatened constantly by software patents. 62 | States should not allow patents to restrict development and use of 63 | software on general-purpose computers, but in those that do, we wish to 64 | avoid the special danger that patents applied to a free program could 65 | make it effectively proprietary. To prevent this, the GPL assures that 66 | patents cannot be used to render the program non-free. 67 | 68 | The precise terms and conditions for copying, distribution and 69 | modification follow. 70 | 71 | TERMS AND CONDITIONS 72 | 73 | 0. Definitions. 74 | 75 | "This License" refers to version 3 of the GNU General Public License. 76 | 77 | "Copyright" also means copyright-like laws that apply to other kinds of 78 | works, such as semiconductor masks. 79 | 80 | "The Program" refers to any copyrightable work licensed under this 81 | License. Each licensee is addressed as "you". "Licensees" and 82 | "recipients" may be individuals or organizations. 83 | 84 | To "modify" a work means to copy from or adapt all or part of the work 85 | in a fashion requiring copyright permission, other than the making of an 86 | exact copy. The resulting work is called a "modified version" of the 87 | earlier work or a work "based on" the earlier work. 88 | 89 | A "covered work" means either the unmodified Program or a work based 90 | on the Program. 91 | 92 | To "propagate" a work means to do anything with it that, without 93 | permission, would make you directly or secondarily liable for 94 | infringement under applicable copyright law, except executing it on a 95 | computer or modifying a private copy. Propagation includes copying, 96 | distribution (with or without modification), making available to the 97 | public, and in some countries other activities as well. 98 | 99 | To "convey" a work means any kind of propagation that enables other 100 | parties to make or receive copies. Mere interaction with a user through 101 | a computer network, with no transfer of a copy, is not conveying. 102 | 103 | An interactive user interface displays "Appropriate Legal Notices" 104 | to the extent that it includes a convenient and prominently visible 105 | feature that (1) displays an appropriate copyright notice, and (2) 106 | tells the user that there is no warranty for the work (except to the 107 | extent that warranties are provided), that licensees may convey the 108 | work under this License, and how to view a copy of this License. If 109 | the interface presents a list of user commands or options, such as a 110 | menu, a prominent item in the list meets this criterion. 111 | 112 | 1. Source Code. 113 | 114 | The "source code" for a work means the preferred form of the work 115 | for making modifications to it. "Object code" means any non-source 116 | form of a work. 117 | 118 | A "Standard Interface" means an interface that either is an official 119 | standard defined by a recognized standards body, or, in the case of 120 | interfaces specified for a particular programming language, one that 121 | is widely used among developers working in that language. 122 | 123 | The "System Libraries" of an executable work include anything, other 124 | than the work as a whole, that (a) is included in the normal form of 125 | packaging a Major Component, but which is not part of that Major 126 | Component, and (b) serves only to enable use of the work with that 127 | Major Component, or to implement a Standard Interface for which an 128 | implementation is available to the public in source code form. A 129 | "Major Component", in this context, means a major essential component 130 | (kernel, window system, and so on) of the specific operating system 131 | (if any) on which the executable work runs, or a compiler used to 132 | produce the work, or an object code interpreter used to run it. 133 | 134 | The "Corresponding Source" for a work in object code form means all 135 | the source code needed to generate, install, and (for an executable 136 | work) run the object code and to modify the work, including scripts to 137 | control those activities. However, it does not include the work's 138 | System Libraries, or general-purpose tools or generally available free 139 | programs which are used unmodified in performing those activities but 140 | which are not part of the work. For example, Corresponding Source 141 | includes interface definition files associated with source files for 142 | the work, and the source code for shared libraries and dynamically 143 | linked subprograms that the work is specifically designed to require, 144 | such as by intimate data communication or control flow between those 145 | subprograms and other parts of the work. 146 | 147 | The Corresponding Source need not include anything that users 148 | can regenerate automatically from other parts of the Corresponding 149 | Source. 150 | 151 | The Corresponding Source for a work in source code form is that 152 | same work. 153 | 154 | 2. Basic Permissions. 155 | 156 | All rights granted under this License are granted for the term of 157 | copyright on the Program, and are irrevocable provided the stated 158 | conditions are met. This License explicitly affirms your unlimited 159 | permission to run the unmodified Program. The output from running a 160 | covered work is covered by this License only if the output, given its 161 | content, constitutes a covered work. This License acknowledges your 162 | rights of fair use or other equivalent, as provided by copyright law. 163 | 164 | You may make, run and propagate covered works that you do not 165 | convey, without conditions so long as your license otherwise remains 166 | in force. You may convey covered works to others for the sole purpose 167 | of having them make modifications exclusively for you, or provide you 168 | with facilities for running those works, provided that you comply with 169 | the terms of this License in conveying all material for which you do 170 | not control copyright. Those thus making or running the covered works 171 | for you must do so exclusively on your behalf, under your direction 172 | and control, on terms that prohibit them from making any copies of 173 | your copyrighted material outside their relationship with you. 174 | 175 | Conveying under any other circumstances is permitted solely under 176 | the conditions stated below. Sublicensing is not allowed; section 10 177 | makes it unnecessary. 178 | 179 | 3. Protecting Users' Legal Rights From Anti-Circumvention Law. 180 | 181 | No covered work shall be deemed part of an effective technological 182 | measure under any applicable law fulfilling obligations under article 183 | 11 of the WIPO copyright treaty adopted on 20 December 1996, or 184 | similar laws prohibiting or restricting circumvention of such 185 | measures. 186 | 187 | When you convey a covered work, you waive any legal power to forbid 188 | circumvention of technological measures to the extent such circumvention 189 | is effected by exercising rights under this License with respect to 190 | the covered work, and you disclaim any intention to limit operation or 191 | modification of the work as a means of enforcing, against the work's 192 | users, your or third parties' legal rights to forbid circumvention of 193 | technological measures. 194 | 195 | 4. Conveying Verbatim Copies. 196 | 197 | You may convey verbatim copies of the Program's source code as you 198 | receive it, in any medium, provided that you conspicuously and 199 | appropriately publish on each copy an appropriate copyright notice; 200 | keep intact all notices stating that this License and any 201 | non-permissive terms added in accord with section 7 apply to the code; 202 | keep intact all notices of the absence of any warranty; and give all 203 | recipients a copy of this License along with the Program. 204 | 205 | You may charge any price or no price for each copy that you convey, 206 | and you may offer support or warranty protection for a fee. 207 | 208 | 5. Conveying Modified Source Versions. 209 | 210 | You may convey a work based on the Program, or the modifications to 211 | produce it from the Program, in the form of source code under the 212 | terms of section 4, provided that you also meet all of these conditions: 213 | 214 | a) The work must carry prominent notices stating that you modified 215 | it, and giving a relevant date. 216 | 217 | b) The work must carry prominent notices stating that it is 218 | released under this License and any conditions added under section 219 | 7. This requirement modifies the requirement in section 4 to 220 | "keep intact all notices". 221 | 222 | c) You must license the entire work, as a whole, under this 223 | License to anyone who comes into possession of a copy. This 224 | License will therefore apply, along with any applicable section 7 225 | additional terms, to the whole of the work, and all its parts, 226 | regardless of how they are packaged. This License gives no 227 | permission to license the work in any other way, but it does not 228 | invalidate such permission if you have separately received it. 229 | 230 | d) If the work has interactive user interfaces, each must display 231 | Appropriate Legal Notices; however, if the Program has interactive 232 | interfaces that do not display Appropriate Legal Notices, your 233 | work need not make them do so. 234 | 235 | A compilation of a covered work with other separate and independent 236 | works, which are not by their nature extensions of the covered work, 237 | and which are not combined with it such as to form a larger program, 238 | in or on a volume of a storage or distribution medium, is called an 239 | "aggregate" if the compilation and its resulting copyright are not 240 | used to limit the access or legal rights of the compilation's users 241 | beyond what the individual works permit. Inclusion of a covered work 242 | in an aggregate does not cause this License to apply to the other 243 | parts of the aggregate. 244 | 245 | 6. Conveying Non-Source Forms. 246 | 247 | You may convey a covered work in object code form under the terms 248 | of sections 4 and 5, provided that you also convey the 249 | machine-readable Corresponding Source under the terms of this License, 250 | in one of these ways: 251 | 252 | a) Convey the object code in, or embodied in, a physical product 253 | (including a physical distribution medium), accompanied by the 254 | Corresponding Source fixed on a durable physical medium 255 | customarily used for software interchange. 256 | 257 | b) Convey the object code in, or embodied in, a physical product 258 | (including a physical distribution medium), accompanied by a 259 | written offer, valid for at least three years and valid for as 260 | long as you offer spare parts or customer support for that product 261 | model, to give anyone who possesses the object code either (1) a 262 | copy of the Corresponding Source for all the software in the 263 | product that is covered by this License, on a durable physical 264 | medium customarily used for software interchange, for a price no 265 | more than your reasonable cost of physically performing this 266 | conveying of source, or (2) access to copy the 267 | Corresponding Source from a network server at no charge. 268 | 269 | c) Convey individual copies of the object code with a copy of the 270 | written offer to provide the Corresponding Source. This 271 | alternative is allowed only occasionally and noncommercially, and 272 | only if you received the object code with such an offer, in accord 273 | with subsection 6b. 274 | 275 | d) Convey the object code by offering access from a designated 276 | place (gratis or for a charge), and offer equivalent access to the 277 | Corresponding Source in the same way through the same place at no 278 | further charge. You need not require recipients to copy the 279 | Corresponding Source along with the object code. If the place to 280 | copy the object code is a network server, the Corresponding Source 281 | may be on a different server (operated by you or a third party) 282 | that supports equivalent copying facilities, provided you maintain 283 | clear directions next to the object code saying where to find the 284 | Corresponding Source. Regardless of what server hosts the 285 | Corresponding Source, you remain obligated to ensure that it is 286 | available for as long as needed to satisfy these requirements. 287 | 288 | e) Convey the object code using peer-to-peer transmission, provided 289 | you inform other peers where the object code and Corresponding 290 | Source of the work are being offered to the general public at no 291 | charge under subsection 6d. 292 | 293 | A separable portion of the object code, whose source code is excluded 294 | from the Corresponding Source as a System Library, need not be 295 | included in conveying the object code work. 296 | 297 | A "User Product" is either (1) a "consumer product", which means any 298 | tangible personal property which is normally used for personal, family, 299 | or household purposes, or (2) anything designed or sold for incorporation 300 | into a dwelling. In determining whether a product is a consumer product, 301 | doubtful cases shall be resolved in favor of coverage. For a particular 302 | product received by a particular user, "normally used" refers to a 303 | typical or common use of that class of product, regardless of the status 304 | of the particular user or of the way in which the particular user 305 | actually uses, or expects or is expected to use, the product. A product 306 | is a consumer product regardless of whether the product has substantial 307 | commercial, industrial or non-consumer uses, unless such uses represent 308 | the only significant mode of use of the product. 309 | 310 | "Installation Information" for a User Product means any methods, 311 | procedures, authorization keys, or other information required to install 312 | and execute modified versions of a covered work in that User Product from 313 | a modified version of its Corresponding Source. The information must 314 | suffice to ensure that the continued functioning of the modified object 315 | code is in no case prevented or interfered with solely because 316 | modification has been made. 317 | 318 | If you convey an object code work under this section in, or with, or 319 | specifically for use in, a User Product, and the conveying occurs as 320 | part of a transaction in which the right of possession and use of the 321 | User Product is transferred to the recipient in perpetuity or for a 322 | fixed term (regardless of how the transaction is characterized), the 323 | Corresponding Source conveyed under this section must be accompanied 324 | by the Installation Information. But this requirement does not apply 325 | if neither you nor any third party retains the ability to install 326 | modified object code on the User Product (for example, the work has 327 | been installed in ROM). 328 | 329 | The requirement to provide Installation Information does not include a 330 | requirement to continue to provide support service, warranty, or updates 331 | for a work that has been modified or installed by the recipient, or for 332 | the User Product in which it has been modified or installed. Access to a 333 | network may be denied when the modification itself materially and 334 | adversely affects the operation of the network or violates the rules and 335 | protocols for communication across the network. 336 | 337 | Corresponding Source conveyed, and Installation Information provided, 338 | in accord with this section must be in a format that is publicly 339 | documented (and with an implementation available to the public in 340 | source code form), and must require no special password or key for 341 | unpacking, reading or copying. 342 | 343 | 7. Additional Terms. 344 | 345 | "Additional permissions" are terms that supplement the terms of this 346 | License by making exceptions from one or more of its conditions. 347 | Additional permissions that are applicable to the entire Program shall 348 | be treated as though they were included in this License, to the extent 349 | that they are valid under applicable law. If additional permissions 350 | apply only to part of the Program, that part may be used separately 351 | under those permissions, but the entire Program remains governed by 352 | this License without regard to the additional permissions. 353 | 354 | When you convey a copy of a covered work, you may at your option 355 | remove any additional permissions from that copy, or from any part of 356 | it. (Additional permissions may be written to require their own 357 | removal in certain cases when you modify the work.) You may place 358 | additional permissions on material, added by you to a covered work, 359 | for which you have or can give appropriate copyright permission. 360 | 361 | Notwithstanding any other provision of this License, for material you 362 | add to a covered work, you may (if authorized by the copyright holders of 363 | that material) supplement the terms of this License with terms: 364 | 365 | a) Disclaiming warranty or limiting liability differently from the 366 | terms of sections 15 and 16 of this License; or 367 | 368 | b) Requiring preservation of specified reasonable legal notices or 369 | author attributions in that material or in the Appropriate Legal 370 | Notices displayed by works containing it; or 371 | 372 | c) Prohibiting misrepresentation of the origin of that material, or 373 | requiring that modified versions of such material be marked in 374 | reasonable ways as different from the original version; or 375 | 376 | d) Limiting the use for publicity purposes of names of licensors or 377 | authors of the material; or 378 | 379 | e) Declining to grant rights under trademark law for use of some 380 | trade names, trademarks, or service marks; or 381 | 382 | f) Requiring indemnification of licensors and authors of that 383 | material by anyone who conveys the material (or modified versions of 384 | it) with contractual assumptions of liability to the recipient, for 385 | any liability that these contractual assumptions directly impose on 386 | those licensors and authors. 387 | 388 | All other non-permissive additional terms are considered "further 389 | restrictions" within the meaning of section 10. If the Program as you 390 | received it, or any part of it, contains a notice stating that it is 391 | governed by this License along with a term that is a further 392 | restriction, you may remove that term. If a license document contains 393 | a further restriction but permits relicensing or conveying under this 394 | License, you may add to a covered work material governed by the terms 395 | of that license document, provided that the further restriction does 396 | not survive such relicensing or conveying. 397 | 398 | If you add terms to a covered work in accord with this section, you 399 | must place, in the relevant source files, a statement of the 400 | additional terms that apply to those files, or a notice indicating 401 | where to find the applicable terms. 402 | 403 | Additional terms, permissive or non-permissive, may be stated in the 404 | form of a separately written license, or stated as exceptions; 405 | the above requirements apply either way. 406 | 407 | 8. Termination. 408 | 409 | You may not propagate or modify a covered work except as expressly 410 | provided under this License. Any attempt otherwise to propagate or 411 | modify it is void, and will automatically terminate your rights under 412 | this License (including any patent licenses granted under the third 413 | paragraph of section 11). 414 | 415 | However, if you cease all violation of this License, then your 416 | license from a particular copyright holder is reinstated (a) 417 | provisionally, unless and until the copyright holder explicitly and 418 | finally terminates your license, and (b) permanently, if the copyright 419 | holder fails to notify you of the violation by some reasonable means 420 | prior to 60 days after the cessation. 421 | 422 | Moreover, your license from a particular copyright holder is 423 | reinstated permanently if the copyright holder notifies you of the 424 | violation by some reasonable means, this is the first time you have 425 | received notice of violation of this License (for any work) from that 426 | copyright holder, and you cure the violation prior to 30 days after 427 | your receipt of the notice. 428 | 429 | Termination of your rights under this section does not terminate the 430 | licenses of parties who have received copies or rights from you under 431 | this License. If your rights have been terminated and not permanently 432 | reinstated, you do not qualify to receive new licenses for the same 433 | material under section 10. 434 | 435 | 9. Acceptance Not Required for Having Copies. 436 | 437 | You are not required to accept this License in order to receive or 438 | run a copy of the Program. Ancillary propagation of a covered work 439 | occurring solely as a consequence of using peer-to-peer transmission 440 | to receive a copy likewise does not require acceptance. However, 441 | nothing other than this License grants you permission to propagate or 442 | modify any covered work. These actions infringe copyright if you do 443 | not accept this License. Therefore, by modifying or propagating a 444 | covered work, you indicate your acceptance of this License to do so. 445 | 446 | 10. Automatic Licensing of Downstream Recipients. 447 | 448 | Each time you convey a covered work, the recipient automatically 449 | receives a license from the original licensors, to run, modify and 450 | propagate that work, subject to this License. You are not responsible 451 | for enforcing compliance by third parties with this License. 452 | 453 | An "entity transaction" is a transaction transferring control of an 454 | organization, or substantially all assets of one, or subdividing an 455 | organization, or merging organizations. If propagation of a covered 456 | work results from an entity transaction, each party to that 457 | transaction who receives a copy of the work also receives whatever 458 | licenses to the work the party's predecessor in interest had or could 459 | give under the previous paragraph, plus a right to possession of the 460 | Corresponding Source of the work from the predecessor in interest, if 461 | the predecessor has it or can get it with reasonable efforts. 462 | 463 | You may not impose any further restrictions on the exercise of the 464 | rights granted or affirmed under this License. For example, you may 465 | not impose a license fee, royalty, or other charge for exercise of 466 | rights granted under this License, and you may not initiate litigation 467 | (including a cross-claim or counterclaim in a lawsuit) alleging that 468 | any patent claim is infringed by making, using, selling, offering for 469 | sale, or importing the Program or any portion of it. 470 | 471 | 11. Patents. 472 | 473 | A "contributor" is a copyright holder who authorizes use under this 474 | License of the Program or a work on which the Program is based. The 475 | work thus licensed is called the contributor's "contributor version". 476 | 477 | A contributor's "essential patent claims" are all patent claims 478 | owned or controlled by the contributor, whether already acquired or 479 | hereafter acquired, that would be infringed by some manner, permitted 480 | by this License, of making, using, or selling its contributor version, 481 | but do not include claims that would be infringed only as a 482 | consequence of further modification of the contributor version. For 483 | purposes of this definition, "control" includes the right to grant 484 | patent sublicenses in a manner consistent with the requirements of 485 | this License. 486 | 487 | Each contributor grants you a non-exclusive, worldwide, royalty-free 488 | patent license under the contributor's essential patent claims, to 489 | make, use, sell, offer for sale, import and otherwise run, modify and 490 | propagate the contents of its contributor version. 491 | 492 | In the following three paragraphs, a "patent license" is any express 493 | agreement or commitment, however denominated, not to enforce a patent 494 | (such as an express permission to practice a patent or covenant not to 495 | sue for patent infringement). To "grant" such a patent license to a 496 | party means to make such an agreement or commitment not to enforce a 497 | patent against the party. 498 | 499 | If you convey a covered work, knowingly relying on a patent license, 500 | and the Corresponding Source of the work is not available for anyone 501 | to copy, free of charge and under the terms of this License, through a 502 | publicly available network server or other readily accessible means, 503 | then you must either (1) cause the Corresponding Source to be so 504 | available, or (2) arrange to deprive yourself of the benefit of the 505 | patent license for this particular work, or (3) arrange, in a manner 506 | consistent with the requirements of this License, to extend the patent 507 | license to downstream recipients. "Knowingly relying" means you have 508 | actual knowledge that, but for the patent license, your conveying the 509 | covered work in a country, or your recipient's use of the covered work 510 | in a country, would infringe one or more identifiable patents in that 511 | country that you have reason to believe are valid. 512 | 513 | If, pursuant to or in connection with a single transaction or 514 | arrangement, you convey, or propagate by procuring conveyance of, a 515 | covered work, and grant a patent license to some of the parties 516 | receiving the covered work authorizing them to use, propagate, modify 517 | or convey a specific copy of the covered work, then the patent license 518 | you grant is automatically extended to all recipients of the covered 519 | work and works based on it. 520 | 521 | A patent license is "discriminatory" if it does not include within 522 | the scope of its coverage, prohibits the exercise of, or is 523 | conditioned on the non-exercise of one or more of the rights that are 524 | specifically granted under this License. You may not convey a covered 525 | work if you are a party to an arrangement with a third party that is 526 | in the business of distributing software, under which you make payment 527 | to the third party based on the extent of your activity of conveying 528 | the work, and under which the third party grants, to any of the 529 | parties who would receive the covered work from you, a discriminatory 530 | patent license (a) in connection with copies of the covered work 531 | conveyed by you (or copies made from those copies), or (b) primarily 532 | for and in connection with specific products or compilations that 533 | contain the covered work, unless you entered into that arrangement, 534 | or that patent license was granted, prior to 28 March 2007. 535 | 536 | Nothing in this License shall be construed as excluding or limiting 537 | any implied license or other defenses to infringement that may 538 | otherwise be available to you under applicable patent law. 539 | 540 | 12. No Surrender of Others' Freedom. 541 | 542 | If conditions are imposed on you (whether by court order, agreement or 543 | otherwise) that contradict the conditions of this License, they do not 544 | excuse you from the conditions of this License. If you cannot convey a 545 | covered work so as to satisfy simultaneously your obligations under this 546 | License and any other pertinent obligations, then as a consequence you may 547 | not convey it at all. For example, if you agree to terms that obligate you 548 | to collect a royalty for further conveying from those to whom you convey 549 | the Program, the only way you could satisfy both those terms and this 550 | License would be to refrain entirely from conveying the Program. 551 | 552 | 13. Use with the GNU Affero General Public License. 553 | 554 | Notwithstanding any other provision of this License, you have 555 | permission to link or combine any covered work with a work licensed 556 | under version 3 of the GNU Affero General Public License into a single 557 | combined work, and to convey the resulting work. The terms of this 558 | License will continue to apply to the part which is the covered work, 559 | but the special requirements of the GNU Affero General Public License, 560 | section 13, concerning interaction through a network will apply to the 561 | combination as such. 562 | 563 | 14. Revised Versions of this License. 564 | 565 | The Free Software Foundation may publish revised and/or new versions of 566 | the GNU General Public License from time to time. Such new versions will 567 | be similar in spirit to the present version, but may differ in detail to 568 | address new problems or concerns. 569 | 570 | Each version is given a distinguishing version number. If the 571 | Program specifies that a certain numbered version of the GNU General 572 | Public License "or any later version" applies to it, you have the 573 | option of following the terms and conditions either of that numbered 574 | version or of any later version published by the Free Software 575 | Foundation. If the Program does not specify a version number of the 576 | GNU General Public License, you may choose any version ever published 577 | by the Free Software Foundation. 578 | 579 | If the Program specifies that a proxy can decide which future 580 | versions of the GNU General Public License can be used, that proxy's 581 | public statement of acceptance of a version permanently authorizes you 582 | to choose that version for the Program. 583 | 584 | Later license versions may give you additional or different 585 | permissions. However, no additional obligations are imposed on any 586 | author or copyright holder as a result of your choosing to follow a 587 | later version. 588 | 589 | 15. Disclaimer of Warranty. 590 | 591 | THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY 592 | APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT 593 | HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY 594 | OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, 595 | THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 596 | PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM 597 | IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF 598 | ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 599 | 600 | 16. Limitation of Liability. 601 | 602 | IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING 603 | WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS 604 | THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY 605 | GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE 606 | USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF 607 | DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD 608 | PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), 609 | EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF 610 | SUCH DAMAGES. 611 | 612 | 17. Interpretation of Sections 15 and 16. 613 | 614 | If the disclaimer of warranty and limitation of liability provided 615 | above cannot be given local legal effect according to their terms, 616 | reviewing courts shall apply local law that most closely approximates 617 | an absolute waiver of all civil liability in connection with the 618 | Program, unless a warranty or assumption of liability accompanies a 619 | copy of the Program in return for a fee. 620 | 621 | END OF TERMS AND CONDITIONS 622 | 623 | How to Apply These Terms to Your New Programs 624 | 625 | If you develop a new program, and you want it to be of the greatest 626 | possible use to the public, the best way to achieve this is to make it 627 | free software which everyone can redistribute and change under these terms. 628 | 629 | To do so, attach the following notices to the program. It is safest 630 | to attach them to the start of each source file to most effectively 631 | state the exclusion of warranty; and each file should have at least 632 | the "copyright" line and a pointer to where the full notice is found. 633 | 634 | 635 | Copyright (C) 636 | 637 | This program is free software: you can redistribute it and/or modify 638 | it under the terms of the GNU General Public License as published by 639 | the Free Software Foundation, either version 3 of the License, or 640 | (at your option) any later version. 641 | 642 | This program is distributed in the hope that it will be useful, 643 | but WITHOUT ANY WARRANTY; without even the implied warranty of 644 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 645 | GNU General Public License for more details. 646 | 647 | You should have received a copy of the GNU General Public License 648 | along with this program. If not, see . 649 | 650 | Also add information on how to contact you by electronic and paper mail. 651 | 652 | If the program does terminal interaction, make it output a short 653 | notice like this when it starts in an interactive mode: 654 | 655 | Copyright (C) 656 | This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'. 657 | This is free software, and you are welcome to redistribute it 658 | under certain conditions; type `show c' for details. 659 | 660 | The hypothetical commands `show w' and `show c' should show the appropriate 661 | parts of the General Public License. Of course, your program's commands 662 | might be different; for a GUI interface, you would use an "about box". 663 | 664 | You should also get your employer (if you work as a programmer) or school, 665 | if any, to sign a "copyright disclaimer" for the program, if necessary. 666 | For more information on this, and how to apply and follow the GNU GPL, see 667 | . 668 | 669 | The GNU General Public License does not permit incorporating your program 670 | into proprietary programs. If your program is a subroutine library, you 671 | may consider it more useful to permit linking proprietary applications with 672 | the library. If this is what you want to do, use the GNU Lesser General 673 | Public License instead of this License. But first, please read 674 | . 675 | --------------------------------------------------------------------------------