web_tool/ModelSessionPyTorchExample.py [222:278]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
                predictions = F.softmax(predictions)

                predictions = predictions.cpu().numpy()
                features = features.cpu().numpy()

            predictions = np.rollaxis(predictions, 1, 4)
            features = np.rollaxis(features, 1, 4)

            model_output.append(predictions)
            model_feature_output.append(features)

        model_output = np.concatenate(model_output, axis=0)
        model_feature_output = np.concatenate(model_feature_output, axis=0)

        for i, (y, x) in enumerate(batch_indices):
            output[y:y+self.input_size, x:x+self.input_size] += model_output[i] * kernel[..., np.newaxis]
            output_features[y:y+self.input_size, x:x+self.input_size] += model_feature_output[i] * kernel[..., np.newaxis]
            counts[y:y+self.input_size, x:x+self.input_size] += kernel

        output = output / counts[..., np.newaxis]
        output_features = output_features / counts[..., np.newaxis]

        return output, output_features

    def save_state_to(self, directory):
        raise NotImplementedError()
        # np.save(os.path.join(directory, "augment_x_train.npy"), np.array(self.augment_x_train))
        # np.save(os.path.join(directory, "augment_y_train.npy"), np.array(self.augment_y_train))

        # joblib.dump(self.augment_model, os.path.join(directory, "augment_model.p"))

        # if self.augment_model_trained:
        #     with open(os.path.join(directory, "trained.txt"), "w") as f:
        #         f.write("")

        return {
            "message": "Saved model state", 
            "success": True
        }

    def load_state_from(self, directory):
        raise NotImplementedError()
        # self.augment_x_train = []
        # self.augment_y_train = []

        # for sample in np.load(os.path.join(directory, "augment_x_train.npy")):
        #     self.augment_x_train.append(sample)
        # for sample in np.load(os.path.join(directory, "augment_y_train.npy")):
        #     self.augment_y_train.append(sample)

        # self.augment_model = joblib.load(os.path.join(directory, "augment_model.p"))
        # self.augment_model_trained = os.path.exists(os.path.join(directory, "trained.txt"))

        return {
            "message": "Loaded model state", 
            "success": True
        }
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



web_tool/ModelSessionPyTorchSegmentationModel.py [233:289]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
                predictions = F.softmax(predictions)

                predictions = predictions.cpu().numpy()
                features = features.cpu().numpy()

            predictions = np.rollaxis(predictions, 1, 4)
            features = np.rollaxis(features, 1, 4)

            model_output.append(predictions)
            model_feature_output.append(features)

        model_output = np.concatenate(model_output, axis=0)
        model_feature_output = np.concatenate(model_feature_output, axis=0)
        
        for i, (y, x) in enumerate(batch_indices):
            output[y:y+self.input_size, x:x+self.input_size] += model_output[i] * kernel[..., np.newaxis]
            output_features[y:y+self.input_size, x:x+self.input_size] += model_feature_output[i] * kernel[..., np.newaxis]
            counts[y:y+self.input_size, x:x+self.input_size] += kernel

        output = output / counts[..., np.newaxis]
        output_features = output_features / counts[..., np.newaxis]

        return output, output_features

    def save_state_to(self, directory):
        raise NotImplementedError()
        # np.save(os.path.join(directory, "augment_x_train.npy"), np.array(self.augment_x_train))
        # np.save(os.path.join(directory, "augment_y_train.npy"), np.array(self.augment_y_train))

        # joblib.dump(self.augment_model, os.path.join(directory, "augment_model.p"))

        # if self.augment_model_trained:
        #     with open(os.path.join(directory, "trained.txt"), "w") as f:
        #         f.write("")

        return {
            "message": "Saved model state", 
            "success": True
        }

    def load_state_from(self, directory):
        raise NotImplementedError()
        # self.augment_x_train = []
        # self.augment_y_train = []

        # for sample in np.load(os.path.join(directory, "augment_x_train.npy")):
        #     self.augment_x_train.append(sample)
        # for sample in np.load(os.path.join(directory, "augment_y_train.npy")):
        #     self.augment_y_train.append(sample)

        # self.augment_model = joblib.load(os.path.join(directory, "augment_model.p"))
        # self.augment_model_trained = os.path.exists(os.path.join(directory, "trained.txt"))

        return {
            "message": "Loaded model state", 
            "success": True
        }
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -