attunet_trainer Module

../_images/attunet.jpg

AttunetTrainer class uses Attention U-Net model, which is originally proposed in this arXiv 1. Implemented trainer module uses this Git 2 code as reference work.

References

1

https://arxiv.org/abs/1804.03999

2

https://github.com/LeeJunHyun/Image_Segmentation

AttunetTrainer class

class farabio.models.segmentation.attunet.attunet_trainer.AttunetTrainer(config)[source]

Attention U-Net trainer class. Override with custom methods here. Attention U-Net cannot fit into one GPU. Supports only parallel mode.

Parameters
ConvnetTrainerBaseTrainer

Inherits ConvnetTrainer class

define_data_attr(*args)[source]

Define data related attributes here

define_model_attr(*args)[source]

Define model related attributes here

define_train_attr(*args)[source]

Define training related attributes here

define_test_attr(*args)[source]

Define training related attributes here

define_log_attr(*args)[source]

Define log related attributes here

define_compute_attr(*args)[source]

Define compute related attributes here

define_misc_attr(*args)[source]

Define miscellaneous attributes here

get_trainloader()[source]

Hook: Retreives training set of torch.utils.data.DataLoader class

get_testloader()[source]

Hook: Retreives test set of torch.utils.data.DataLoader class

build_model()[source]

Abstract method that builds model

build_parallel_model()[source]

Abstract method that builds multi-GPU model in parallel

show_model_summary(*args)[source]
load_model()[source]

Hook: load model

load_parallel_model()[source]

Hook: load parallel model

start_logger()[source]

Hook: Starts logger

on_train_epoch_start()[source]

Hook: On epoch start

on_start_training_batch(args)[source]

Hook: On training batch start

optimizer_zero_grad()[source]

Hook: Zero gradients of optimizer

loss_backward()[source]

Hook: Loss back-propagation

optimizer_step()[source]

Hook: Optimizer step

training_step()[source]

Hook: During training batch

on_end_training_batch()[source]

Hook: On end of training batch

on_train_epoch_end()[source]

Hook: On end of training epoch

on_evaluate_epoch_start()[source]

Hook: on evaluation start

evaluate_batch(args)[source]

Hook: batch of evaluation loop

on_evaluate_batch_end()[source]

Hook: On evaluate batch end

on_evaluate_epoch_end()[source]
on_epoch_end()[source]

Hook: on epoch end

save_model()[source]

Hook: saves model

save_parallel_model()[source]

Hook: saves parallel model

on_test_start()[source]

Hook: on test start

test_step(args)[source]

Test action (Put test here)

generate_result_img(*args)[source]

Generate image from batch: one by one

on_end_test_batch()[source]

Hook: on end of batch test