engine.py

# Copyright (c) 2015-present, Facebook, Inc.
# All rights reserved.
"""
Train and eval functions used in main.py
"""
import math
import sys
from typing import Iterable, Optional, Callable, Dict

import torch

from timm.data import Mixup
from timm.utils import accuracy, ModelEma

from losses import DistillationLoss
import utils


def train_one_epoch(model: torch.nn.Module, criterion: DistillationLoss,
                    data_loader: Iterable, optimizer: torch.optim.Optimizer,
                    device: torch.device, epoch: int, loss_scaler, max_norm: float = 0,
                    model_ema: Optional[ModelEma] = None, mixup_fn: Optional[Mixup] = None,
                    set_training_mode=True, args=None, ext_logger: Optional[Callable[[Dict, int], None]] = None):
    model.train(set_training_mode)
    metric_logger = utils.MetricLogger(delimiter="  ")
    metric_logger.add_meter('lr', utils.SmoothedValue(window_size=1, fmt='{value:.6f}'))
    header = 'Epoch: [{}]'.format(epoch)
    print_freq = 10

    for samples, targets in metric_logger.log_every(data_loader, print_freq, header):
        coords = None
        if isinstance(samples, list):
            samples, coords = samples
            coords = coords.to(device, non_blocking=True)
        samples = samples.to(device, non_blocking=True)
        targets = targets.to(device, non_blocking=True)

        if args.bce_loss:
            targets = targets.gt(0.0).type(targets.dtype)

        with torch.cuda.amp.autocast():
            outputs = model(samples, coords)
            loss = criterion(samples, outputs, targets)

        loss_value = loss.item()

        if not math.isfinite(loss_value):
            print("Loss is {}, stopping training".format(loss_value))
            sys.exit(1)

        optimizer.zero_grad()

        # this attribute is added by timm on one optimizer (adahessian)
        is_second_order = hasattr(optimizer, 'is_second_order') and optimizer.is_second_order
        loss_scaler(loss, optimizer, clip_grad=max_norm,
                    parameters=model.parameters(), create_graph=is_second_order)

        torch.cuda.synchronize()
        if model_ema is not None:
            model_ema.update(model)

        metric_logger.update(loss=loss_value)
        metric_logger.update(lr=optimizer.param_groups[0]["lr"])
    # gather the stats from all processes
    metric_logger.synchronize_between_processes()
    print("Averaged stats:", metric_logger)
    if ext_logger is not None:
        ext_logger({
            'Train/Loss': metric_logger.loss.global_avg,
            'Train/LR': metric_logger.lr.global_avg
        }, epoch)
    return {k: meter.global_avg for k, meter in metric_logger.meters.items()}


@torch.no_grad()
def evaluate(data_loader, model, device, epoch=None, ext_logger: Optional[Callable[[Dict, int], None]] = None):
    criterion = torch.nn.CrossEntropyLoss()

    metric_logger = utils.MetricLogger(delimiter="  ")
    header = 'Test:'

    # switch to evaluation mode
    model.eval()

    for images, target in metric_logger.log_every(data_loader, 10, header):
        coords = None
        keeps = None
        if isinstance(images, list):
            images, coords, keeps = images
            coords = coords.to(device, non_blocking=True)
            keeps = keeps.to(device, non_blocking=True) if keeps is not None else None
        images = images.to(device, non_blocking=True)
        target = target.to(device, non_blocking=True)

        # compute output
        with torch.cuda.amp.autocast():
            output = model(images, coords, keeps)
            loss = criterion(output, target)

        acc1, acc5 = accuracy(output, target, topk=(1, 5))

        batch_size = images.shape[0]
        metric_logger.update(loss=loss.item())
        metric_logger.meters['acc1'].update(acc1.item(), n=batch_size)
        metric_logger.meters['acc5'].update(acc5.item(), n=batch_size)
    # gather the stats from all processes
    metric_logger.synchronize_between_processes()
    print('* Acc@1 {top1.global_avg:.3f} Acc@5 {top5.global_avg:.3f} loss {losses.global_avg:.3f}'
          .format(top1=metric_logger.acc1, top5=metric_logger.acc5, losses=metric_logger.loss))
    if ext_logger is not None:
        ext_logger({
            'Eval/Acc@1': metric_logger.acc1.global_avg,
            'Eval/Acc@5': metric_logger.acc5.global_avg,
            'Eval/Loss': metric_logger.loss.global_avg
        }, epoch)

    return {k: meter.global_avg for k, meter in metric_logger.meters.items()}