template.runner.apply_model package

Submodules

template.runner.apply_model.apply_model module

This file is the template for the boilerplate of train/test of a DNN

There are a lot of parameter which can be specified to modify the behaviour and they should be used instead of hard-coding stuff.

@authors: Vinaychandran Pondenkandath , Michele Alberti

class template.runner.apply_model.apply_model.ApplyModel

Bases: object

static single_run(writer, current_log_folder, model_name, lr, output_channels, classify, **kwargs)

This is the main routine where train(), validate() and test() are called.

Parameters

• writer (Tensorboard SummaryWriter) – Responsible for writing logs in Tensorboard compatible format.

• current_log_folder (string) – Path to where logs/checkpoints are saved

• model_name (string) – Name of the model

• lr (float) – Value for learning rate

• kwargs (dict) – Any additional arguments.

• output_channels (int) – Specify shape of final layer of network.

• classify (boolean) – Specifies whether to generate a classification report for the data or not.

Returns

None – None

Return type

None

template.runner.apply_model.evaluate module

template.runner.apply_model.evaluate.feature_extract(data_loader, model, writer, epoch, no_cuda, log_interval, classify, **kwargs)

The evaluation routine

Parameters

• data_loader (torch.utils.data.DataLoader) – The dataloader of the evaluation set

• model (torch.nn.module) – The network model being used

• writer (tensorboardX.writer.SummaryWriter) – The tensorboard writer object. Used to log values on file for the tensorboard visualization.

• epoch (int) – Number of the epoch (for logging purposes)

• no_cuda (boolean) – Specifies whether the GPU should be used or not. A value of ‘True’ means the CPU will be used.

• log_interval (int) – Interval limiting the logging of mini-batches. Default value of 10.

• classify (boolean) – Specifies whether to generate a classification report for the data or not.

Returns

Return type

None

template.runner.apply_model.setup module

template.runner.apply_model.setup.set_up_dataloader(model_expected_input_size, dataset_folder, batch_size, workers, inmem, multi_crop, classify, **kwargs)

Set up the dataloaders for the specified datasets.

Parameters

• param model_expected_input_size: tuple – Specify the height and width that the model expects.

• param dataset_folder: string – Path string that points to the three folder train/val/test. Example: ~/../../data/svhn

• param batch_size: int – Number of datapoints to process at once

• param workers: int – Number of workers to use for the dataloaders

• param inmem: boolean – Flag: if False, the dataset is loaded in an online fashion i.e. only file names are stored and images are loaded on demand. This is slower than storing everything in memory.

• param multi_crop: int – if None, the MultiCrop transform is not applied to the data. Otherwise, multi_crop contains an integer which specifies how many crops to make from each image.

• param classify : boolean – Specifies whether to generate a classification report for the data or not.

• param kwargs: dict – Any additional arguments.

• return: dataloader, dataloader, dataloader, int – Three dataloaders for train, val and test. Number of classes for the model.

Module contents

class template.runner.apply_model.ApplyModel

Bases: object

static single_run(writer, current_log_folder, model_name, lr, output_channels, classify, **kwargs)

This is the main routine where train(), validate() and test() are called.

Parameters

• writer (Tensorboard SummaryWriter) – Responsible for writing logs in Tensorboard compatible format.

• current_log_folder (string) – Path to where logs/checkpoints are saved

• model_name (string) – Name of the model

• lr (float) – Value for learning rate

• kwargs (dict) – Any additional arguments.

• output_channels (int) – Specify shape of final layer of network.

• classify (boolean) – Specifies whether to generate a classification report for the data or not.

Returns

None – None

Return type

None