Loading deepchem/metrics/__init__.py +8 −4 Original line number Diff line number Diff line Loading @@ -73,6 +73,7 @@ def pearson_r2_score(y, y_pred): """Computes Pearson R^2 (square of Pearson correlation).""" return pearsonr(y, y_pred)[0]**2 def jaccard_index(y, y_pred): """Computes Jaccard Index which is the Intersection Over Union metric which is commonly used in image segmentation tasks Loading @@ -84,6 +85,7 @@ def jaccard_index(y, y_pred): """ return jaccard_similarity_score(y, y_pred) def pixel_error(y, y_pred): """defined as 1 - the maximal F-score of pixel similarity, or squared Euclidean distance between the original and the result labels. Loading @@ -95,6 +97,7 @@ def pixel_error(y, y_pred): """ return 1 - f1_score(y, y_pred) def prc_auc_score(y, y_pred): """Compute area under precision-recall curve""" assert y_pred.shape == y.shape Loading Loading @@ -135,13 +138,14 @@ def kappa_score(y_true, y_pred): yt = np.asarray(y_true, dtype=int) yp = np.asarray(y_pred, dtype=int) assert np.array_equal( np.unique(yt), [0, 1]), ('Class labels must be binary: %s' % np.unique(yt)) np.unique(yt), [0, 1]), ('Class labels must be binary: %s' % np.unique(yt)) observed_agreement = np.true_divide( np.count_nonzero(np.equal(yt, yp)), len(yt)) expected_agreement = np.true_divide( np.count_nonzero(yt == 1) * np.count_nonzero(yp == 1) + np.count_nonzero(yt == 0) * np.count_nonzero(yp == 0), len(yt)**2) np.count_nonzero(yt == 0) * np.count_nonzero(yp == 0), len(yt)**2) kappa = np.true_divide(observed_agreement - expected_agreement, 1.0 - expected_agreement) return kappa Loading deepchem/models/tensorgraph/models/unet.py +136 −39 Original line number Diff line number Diff line Loading @@ -12,6 +12,7 @@ from deepchem.models import Sequential from deepchem.models.tensorgraph.layers import Conv2D, MaxPool2D, Conv2DTranspose, Concat, Feature from deepchem.models import TensorGraph class UNet(TensorGraph): """ U-Net architecture implementation. Loading @@ -22,7 +23,12 @@ class UNet(TensorGraph): img_cols: int number of columns of the image """ def __init__(self,img_rows=512, img_cols=512, model=dc.models.TensorGraph(), **kwargs): def __init__(self, img_rows=512, img_cols=512, model=dc.models.TensorGraph(), **kwargs): super(UNet, self).__init__(use_queue=False, **kwargs) self.img_cols = img_cols self.img_rows = img_rows Loading @@ -30,45 +36,136 @@ class UNet(TensorGraph): input = Feature(shape=(None, self.img_rows, self.img_cols)) conv1 = Conv2D(num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[input]) conv1 = Conv2D(num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[conv1]) conv1 = Conv2D( num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[input]) conv1 = Conv2D( num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[conv1]) pool1 = MaxPool2D(ksize=2, in_layers=[conv1]) conv2 = Conv2D(num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[pool1]) conv2 = Conv2D(num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[conv2]) conv2 = Conv2D( num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[pool1]) conv2 = Conv2D( num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[conv2]) pool2 = MaxPool2D(ksize=2, in_layers=[conv2]) conv3 = Conv2D(num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[pool2]) conv3 = Conv2D(num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[conv3]) conv3 = Conv2D( num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[pool2]) conv3 = Conv2D( num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[conv3]) pool3 = MaxPool2D(ksize=2, in_layers=[conv3]) conv4 = Conv2D(num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[pool3]) conv4 = Conv2D(num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[conv4]) conv4 = Conv2D( num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[pool3]) conv4 = Conv2D( num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[conv4]) pool4 = MaxPool2D(ksize=2, in_layers=[conv4]) conv5 = Conv2D(num_outputs=1024, kernel_size=3, activation='relu', padding='same', in_layers=[pool4]) conv5 = Conv2D(num_outputs=1024, kernel_size=3, activation='relu', padding='same', in_layers=[conv5]) conv5 = Conv2D( num_outputs=1024, kernel_size=3, activation='relu', padding='same', in_layers=[pool4]) conv5 = Conv2D( num_outputs=1024, kernel_size=3, activation='relu', padding='same', in_layers=[conv5]) up6 = Conv2DTranspose(num_outputs=512, kernel_size=2, in_layers=[conv5]) concat6 = Concat(in_layers=[conv4, up6], axis=1) conv6 = Conv2D(num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[concat6]) conv6 = Conv2D(num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[conv6]) conv6 = Conv2D( num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[concat6]) conv6 = Conv2D( num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[conv6]) up7 = Conv2DTranspose(num_outputs=256, kernel_size=2, in_layers=[conv6]) concat7 = Concat(in_layers=[conv3, up7], axis=1) conv7 = Conv2D(num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[concat7]) conv7 = Conv2D(num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[conv7]) conv7 = Conv2D( num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[concat7]) conv7 = Conv2D( num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[conv7]) up8 = Conv2DTranspose(num_outputs=128, kernel_size=2, in_layers=[conv7]) concat8 = Concat(in_layers=[conv2, up8], axis=1) conv8 = Conv2D(num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[concat8]) conv8 = Conv2D(num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[conv8]) conv8 = Conv2D( num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[concat8]) conv8 = Conv2D( num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[conv8]) up9 = Conv2DTranspose(num_outputs=64, kernel_size=2, in_layers=[conv8]) concat9 = Concat(in_layers=[conv1, up9], axis=1) conv9 = Conv2D(num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[concat9]) conv9 = Conv2D(num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[conv9]) conv9 = Conv2D( num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[concat9]) conv9 = Conv2D( num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[conv9]) conv10 = Conv2D(num_outputs=1, kernel_size=1, activation='sigmoid', in_layers=[conv9]) conv10 = Conv2D( num_outputs=1, kernel_size=1, activation='sigmoid', in_layers=[conv9]) model.add_output(conv10) Loading
deepchem/metrics/__init__.py +8 −4 Original line number Diff line number Diff line Loading @@ -73,6 +73,7 @@ def pearson_r2_score(y, y_pred): """Computes Pearson R^2 (square of Pearson correlation).""" return pearsonr(y, y_pred)[0]**2 def jaccard_index(y, y_pred): """Computes Jaccard Index which is the Intersection Over Union metric which is commonly used in image segmentation tasks Loading @@ -84,6 +85,7 @@ def jaccard_index(y, y_pred): """ return jaccard_similarity_score(y, y_pred) def pixel_error(y, y_pred): """defined as 1 - the maximal F-score of pixel similarity, or squared Euclidean distance between the original and the result labels. Loading @@ -95,6 +97,7 @@ def pixel_error(y, y_pred): """ return 1 - f1_score(y, y_pred) def prc_auc_score(y, y_pred): """Compute area under precision-recall curve""" assert y_pred.shape == y.shape Loading Loading @@ -135,13 +138,14 @@ def kappa_score(y_true, y_pred): yt = np.asarray(y_true, dtype=int) yp = np.asarray(y_pred, dtype=int) assert np.array_equal( np.unique(yt), [0, 1]), ('Class labels must be binary: %s' % np.unique(yt)) np.unique(yt), [0, 1]), ('Class labels must be binary: %s' % np.unique(yt)) observed_agreement = np.true_divide( np.count_nonzero(np.equal(yt, yp)), len(yt)) expected_agreement = np.true_divide( np.count_nonzero(yt == 1) * np.count_nonzero(yp == 1) + np.count_nonzero(yt == 0) * np.count_nonzero(yp == 0), len(yt)**2) np.count_nonzero(yt == 0) * np.count_nonzero(yp == 0), len(yt)**2) kappa = np.true_divide(observed_agreement - expected_agreement, 1.0 - expected_agreement) return kappa Loading
deepchem/models/tensorgraph/models/unet.py +136 −39 Original line number Diff line number Diff line Loading @@ -12,6 +12,7 @@ from deepchem.models import Sequential from deepchem.models.tensorgraph.layers import Conv2D, MaxPool2D, Conv2DTranspose, Concat, Feature from deepchem.models import TensorGraph class UNet(TensorGraph): """ U-Net architecture implementation. Loading @@ -22,7 +23,12 @@ class UNet(TensorGraph): img_cols: int number of columns of the image """ def __init__(self,img_rows=512, img_cols=512, model=dc.models.TensorGraph(), **kwargs): def __init__(self, img_rows=512, img_cols=512, model=dc.models.TensorGraph(), **kwargs): super(UNet, self).__init__(use_queue=False, **kwargs) self.img_cols = img_cols self.img_rows = img_rows Loading @@ -30,45 +36,136 @@ class UNet(TensorGraph): input = Feature(shape=(None, self.img_rows, self.img_cols)) conv1 = Conv2D(num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[input]) conv1 = Conv2D(num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[conv1]) conv1 = Conv2D( num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[input]) conv1 = Conv2D( num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[conv1]) pool1 = MaxPool2D(ksize=2, in_layers=[conv1]) conv2 = Conv2D(num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[pool1]) conv2 = Conv2D(num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[conv2]) conv2 = Conv2D( num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[pool1]) conv2 = Conv2D( num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[conv2]) pool2 = MaxPool2D(ksize=2, in_layers=[conv2]) conv3 = Conv2D(num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[pool2]) conv3 = Conv2D(num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[conv3]) conv3 = Conv2D( num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[pool2]) conv3 = Conv2D( num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[conv3]) pool3 = MaxPool2D(ksize=2, in_layers=[conv3]) conv4 = Conv2D(num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[pool3]) conv4 = Conv2D(num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[conv4]) conv4 = Conv2D( num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[pool3]) conv4 = Conv2D( num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[conv4]) pool4 = MaxPool2D(ksize=2, in_layers=[conv4]) conv5 = Conv2D(num_outputs=1024, kernel_size=3, activation='relu', padding='same', in_layers=[pool4]) conv5 = Conv2D(num_outputs=1024, kernel_size=3, activation='relu', padding='same', in_layers=[conv5]) conv5 = Conv2D( num_outputs=1024, kernel_size=3, activation='relu', padding='same', in_layers=[pool4]) conv5 = Conv2D( num_outputs=1024, kernel_size=3, activation='relu', padding='same', in_layers=[conv5]) up6 = Conv2DTranspose(num_outputs=512, kernel_size=2, in_layers=[conv5]) concat6 = Concat(in_layers=[conv4, up6], axis=1) conv6 = Conv2D(num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[concat6]) conv6 = Conv2D(num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[conv6]) conv6 = Conv2D( num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[concat6]) conv6 = Conv2D( num_outputs=512, kernel_size=3, activation='relu', padding='same', in_layers=[conv6]) up7 = Conv2DTranspose(num_outputs=256, kernel_size=2, in_layers=[conv6]) concat7 = Concat(in_layers=[conv3, up7], axis=1) conv7 = Conv2D(num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[concat7]) conv7 = Conv2D(num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[conv7]) conv7 = Conv2D( num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[concat7]) conv7 = Conv2D( num_outputs=256, kernel_size=3, activation='relu', padding='same', in_layers=[conv7]) up8 = Conv2DTranspose(num_outputs=128, kernel_size=2, in_layers=[conv7]) concat8 = Concat(in_layers=[conv2, up8], axis=1) conv8 = Conv2D(num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[concat8]) conv8 = Conv2D(num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[conv8]) conv8 = Conv2D( num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[concat8]) conv8 = Conv2D( num_outputs=128, kernel_size=3, activation='relu', padding='same', in_layers=[conv8]) up9 = Conv2DTranspose(num_outputs=64, kernel_size=2, in_layers=[conv8]) concat9 = Concat(in_layers=[conv1, up9], axis=1) conv9 = Conv2D(num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[concat9]) conv9 = Conv2D(num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[conv9]) conv9 = Conv2D( num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[concat9]) conv9 = Conv2D( num_outputs=64, kernel_size=3, activation='relu', padding='same', in_layers=[conv9]) conv10 = Conv2D(num_outputs=1, kernel_size=1, activation='sigmoid', in_layers=[conv9]) conv10 = Conv2D( num_outputs=1, kernel_size=1, activation='sigmoid', in_layers=[conv9]) model.add_output(conv10)
