train_step2.py

import argparse
import logging
import math
import os
import random
import time
from copy import deepcopy
from pathlib import Path
from threading import Thread
import yaml
from tqdm import tqdm

import numpy as np
import torch
import torch.distributed as dist
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
import torch.optim.lr_scheduler as lr_scheduler
from torch.cuda import amp
from torch.nn.parallel import DistributedDataParallel as DDP
from torch.utils.tensorboard import SummaryWriter

import partinet.DynamicDet
from partinet.DynamicDet import test  # import test.py to get mAP after each epoch
from partinet.DynamicDet.models.yolo import Model
from partinet.DynamicDet.utils.autoanchor import check_anchors
from partinet.DynamicDet.utils.datasets import create_dataloader
from partinet.DynamicDet.utils.general import labels_to_class_weights, increment_path, labels_to_image_weights, init_seeds, \
    fitness, strip_optimizer, get_latest_run, check_dataset, check_file, check_git_status, check_img_size, \
    set_logging, colorstr
from partinet.DynamicDet.utils.loss import ComputeLoss, ComputeLossOTADy
from partinet.DynamicDet.utils.plots import plot_images, plot_results, plot_lr_scheduler
from partinet.DynamicDet.utils.torch_utils import ModelEMA, select_device, intersect_dicts, torch_distributed_zero_first, is_parallel
from partinet.DynamicDet.utils.wandb_logging.wandb_utils import WandbLogger, check_wandb_resume
from partinet.DynamicDet.utils.checkpoint import get_state_dict


logger = logging.getLogger(__name__)

def train(hyp, opt, device, cfg, tb_writer=None):
    logger.info(colorstr('hyperparameters: ') + ', '.join(f'{k}={v}' for k, v in hyp.items()))
    save_dir, epochs, batch_size, total_batch_size, weight, rank, freeze = \
        Path(opt.save_dir), opt.epochs, opt.batch_size, opt.total_batch_size, opt.weight, opt.global_rank, opt.freeze

    # Directories
    wdir = save_dir / 'weights'
    wdir.mkdir(parents=True, exist_ok=True)  # make dir
    last = wdir / 'last.pt'
    best = wdir / 'best.pt'
    results_file = save_dir / 'results.txt'

    # Save run settings
    with open(save_dir / 'hyp.yaml', 'w') as f:
        yaml.dump(hyp, f, sort_keys=False)
    with open(save_dir / 'opt.yaml', 'w') as f:
        yaml.dump(vars(opt), f, sort_keys=False)
    if not opt.resume:
        with open(save_dir / 'cfg.yaml', 'w') as f_d:
            with open(cfg) as f_s:
                cfg_yaml = yaml.load(f_s, Loader=yaml.SafeLoader)
            yaml.dump(cfg_yaml, f_d, sort_keys=False)

    # Configure
    plots = True
    cuda = device.type != 'cpu'
    init_seeds(2 + rank)
    with open(opt.data) as f:
        data_dict = yaml.load(f, Loader=yaml.SafeLoader)  # data dict
    is_coco = opt.data.endswith('coco.yaml')

    # Logging- Doing this before checking the dataset. Might update data_dict
    loggers = {'wandb': None}  # loggers dict
    if rank in [-1, 0]:
        run_id = torch.load(weight, map_location='cpu')['wandb_id'] if weight.endswith('.pt') and os.path.isfile(weight) else None
        wandb_logger = WandbLogger(opt, Path(opt.save_dir).stem, run_id, data_dict)
        loggers['wandb'] = wandb_logger.wandb
        data_dict = wandb_logger.data_dict
        if wandb_logger.wandb:
            weight, epochs, hyp = opt.weight, opt.epochs, hyp  # WandbLogger might update weights, epochs if resuming

    nc = 1 if opt.single_cls else int(data_dict['nc'])  # number of classes
    names = ['item'] if opt.single_cls and len(data_dict['names']) != 1 else data_dict['names']  # class names
    assert len(names) == nc, '%g names found for nc=%g dataset in %s' % (len(names), nc, opt.data)  # check

    # Model
    pretrained = weight.endswith('.pt')
    if pretrained:
        ckpt = torch.load(weight, map_location='cpu')  # load checkpoint
        state_dict = ckpt['model']
        model = Model(cfg, ch=3, nc=nc)  # create
        exclude = ['anchor'] if (cfg or hyp.get('anchors')) and not opt.resume else []  # exclude keys
        state_dict = intersect_dicts(state_dict, model.state_dict(), exclude=exclude)  # intersect
        model.load_state_dict(state_dict, strict=False)  # load
        model.to(device)
        logger.info('Transferred %g/%g items from %s' % (len(state_dict), len(model.state_dict()), weight))  # report
    else:
        model = Model(cfg, ch=3, nc=nc).to(device)  # create
    with torch_distributed_zero_first(rank):
        check_dataset(data_dict)  # check
    train_path = data_dict['train']
    test_path = data_dict['val']

    # Freeze
    freeze = [f'model.{x}.' for x in (freeze if len(freeze) > 1 else range(freeze[0]))]  # parameter names to freeze (full or partial)
    for k, v in model.named_parameters():
        v.requires_grad = True  # train all layers
        if any(x in k for x in freeze):
            print('freezing %s' % k)
            v.requires_grad = False

    # freeze the parameters of the cascaded detector
    for k, v in model.named_parameters():
        if 'router' not in k:
            print('freezing %s' % k)
            v.requires_grad = False

    # Optimizer
    nbs = 64  # nominal batch size
    accumulate = max(round(nbs / total_batch_size), 1)  # accumulate loss before optimizing
    hyp['weight_decay'] *= total_batch_size * accumulate / nbs  # scale weight_decay
    logger.info(f"Scaled weight_decay = {hyp['weight_decay']}")

    pg0, pg1, pg2 = [], [], []  # optimizer parameter groups
    for k, v in model.named_modules():
        if 'router' in k:
            if hasattr(v, 'bias') and isinstance(v.bias, nn.Parameter):
                pg2.append(v.bias)  # biases
            if isinstance(v, nn.BatchNorm2d):
                pg0.append(v.weight)  # no decay
            elif hasattr(v, 'weight') and isinstance(v.weight, nn.Parameter):
                pg1.append(v.weight)  # apply decay

    if opt.adam:
        optimizer = optim.AdamW(pg1, lr=hyp['lr0'], weight_decay=hyp['weight_decay'], betas=(hyp['momentum'], 0.999))  # adjust beta1 to momentum
    else:
        optimizer = optim.SGD(pg1, lr=hyp['lr0'], weight_decay=hyp['weight_decay'], momentum=hyp['momentum'], nesterov=True)
    if len(pg0):
        optimizer.add_param_group({'params': pg0, 'weight_decay': 0})  # add pg0 without weight_decay
    if len(pg2):
        optimizer.add_param_group({'params': pg2, 'weight_decay': 0})  # add pg2 (biases)
    logger.info('Optimizer groups: %g .bias, %g conv.weight, %g other' % (len(pg2), len(pg1), len(pg0)))
    del pg0, pg1, pg2

    # Scheduler https://arxiv.org/pdf/1812.01187.pdf
    # https://pytorch.org/docs/stable/_modules/torch/optim/lr_scheduler.html#OneCycleLR
    lf = lambda x: (1 - x / (epochs - 1)) * (1.0 - hyp['lrf']) + hyp['lrf']  # linear
    scheduler = lr_scheduler.LambdaLR(optimizer, lr_lambda=lf)
    plot_lr_scheduler(optimizer, scheduler, epochs, save_dir)

    # EMA
    ema_decay = 0.
    ema = ModelEMA(model, decay=ema_decay) if rank in [-1, 0] else None

    # Resume
    start_epoch, best_fitness = 0, 0.0
    if pretrained:
        # Optimizer
        if ckpt['optimizer'] is not None:
            optimizer.load_state_dict(ckpt['optimizer'])
            best_fitness = ckpt['best_fitness']

        # EMA
        if ema and ckpt.get('ema'):
            ema.ema.load_state_dict(ckpt['ema'])
            ema.updates = ckpt['updates']

        # Results
        if ckpt.get('training_results') is not None:
            results_file.write_text(ckpt['training_results'])  # write results.txt

        # Epochs
        start_epoch = ckpt['epoch'] + 1
        if opt.resume:
            assert start_epoch > 0, '%s training to %g epochs is finished, nothing to resume.' % (weight, epochs)
        if epochs < start_epoch:
            logger.info('%s has been trained for %g epochs. Fine-tuning for %g additional epochs.' %
                        (weight, ckpt['epoch'], epochs))
            epochs += ckpt['epoch']  # finetune additional epochs

        del ckpt

    # Image sizes
    gs = max(int(model.stride.max()), 32)  # grid size (max stride)
    nl = model.model_h2[-1].nl  # number of detection layers (used for scaling hyp['obj'])
    imgsz, imgsz_test = [check_img_size(x, gs) for x in opt.img_size]  # verify imgsz are gs-multiples

    # DP mode
    if cuda and rank == -1 and torch.cuda.device_count() > 1:
        model = torch.nn.DataParallel(model)

    # SyncBatchNorm
    if opt.sync_bn and cuda and rank != -1:
        model = torch.nn.SyncBatchNorm.convert_sync_batchnorm(model).to(device)
        logger.info('Using SyncBatchNorm()')

    # Trainloader
    dataloader, dataset = create_dataloader(train_path, imgsz, batch_size, gs, opt,
                                            hyp=hyp, augment=True, cache=opt.cache_images, rect=opt.rect, rank=rank,
                                            world_size=opt.world_size, workers=opt.workers,
                                            image_weights=opt.image_weights, quad=opt.quad, prefix=colorstr('train: '))
    mlc = np.concatenate(dataset.labels, 0)[:, 0].max()  # max label class
    nb = len(dataloader)  # number of batches
    assert mlc < nc, 'Label class %g exceeds nc=%g in %s. Possible class labels are 0-%g' % (mlc, nc, opt.data, nc - 1)

    # Process 0
    if rank in [-1, 0]:
        test_batch_size = 1
        testloader = create_dataloader(test_path, imgsz_test, test_batch_size, gs, opt,  # testloader
                                       hyp=hyp, cache=opt.cache_images and not opt.notest, rect=True, rank=-1,
                                       world_size=opt.world_size, workers=opt.workers,
                                       pad=0.5, prefix=colorstr('val: '))[0]

        if not opt.resume:
            labels = np.concatenate(dataset.labels, 0)
            c = torch.tensor(labels[:, 0])  # classes
            # cf = torch.bincount(c.long(), minlength=nc) + 1.  # frequency
            # model._initialize_biases(cf.to(device))
            if plots:
                if tb_writer:
                    tb_writer.add_histogram('classes', c, 0)

            # Anchors
            if not opt.noautoanchor:
                check_anchors(dataset, model=model, thr=hyp['anchor_t'], imgsz=imgsz)
            model.half().float()  # pre-reduce anchor precision

    # DDP mode
    if cuda and rank != -1:
        model = DDP(model, device_ids=[opt.local_rank], output_device=opt.local_rank,
                    # nn.MultiheadAttention incompatibility with DDP https://github.com/pytorch/pytorch/issues/26698
                    find_unused_parameters=any(isinstance(layer, nn.MultiheadAttention) for layer in model.modules()))

    # Model parameters
    hyp['box'] *= 3. / nl  # scale to layers
    hyp['cls'] *= nc / 80. * 3. / nl  # scale to classes and layers
    hyp['obj'] *= (imgsz / 640) ** 2 * 3. / nl  # scale to image size and layers
    hyp['label_smoothing'] = opt.label_smoothing
    model.nc = nc  # attach number of classes to model
    model.hyp = hyp  # attach hyperparameters to model
    model.gr = 1.0  # iou loss ratio (obj_loss = 1.0 or iou)
    model.class_weights = labels_to_class_weights(dataset.labels, nc).to(device) * nc  # attach class weights
    model.names = names

    # Start training
    t0 = time.time()
    nw = max(round(hyp['warmup_epochs'] * nb), 1000)  # number of warmup iterations, max(3 epochs, 1k iterations)
    # nw = min(nw, (epochs - start_epoch) / 2 * nb)  # limit warmup to < 1/2 of training
    maps = np.zeros(nc)  # mAP per class
    results = (0, 0, 0, 0, 0, 0, 0)  # P, R, mAP@.5, mAP@.5-.95, val_loss(box, obj, cls)
    scheduler.last_epoch = start_epoch - 1  # do not move
    scaler = amp.GradScaler(enabled=cuda)
    compute_loss_ota_dy = ComputeLossOTADy(model)
    compute_loss = ComputeLoss(model)  # init loss class
    logger.info(f'Image sizes {imgsz} train, {imgsz_test} test\n'
                f'Using {dataloader.num_workers} dataloader workers\n'
                f'Logging results to {save_dir}\n'
                f'Starting training for {epochs} epochs...')
    for epoch in range(start_epoch, epochs):  # epoch ------------------------------------------------------------------
        model.train()
        model.close_all_bn()

        # Update image weights (optional)
        if opt.image_weights:
            # Generate indices
            if rank in [-1, 0]:
                cw = model.class_weights.cpu().numpy() * (1 - maps) ** 2 / nc  # class weights
                iw = labels_to_image_weights(dataset.labels, nc=nc, class_weights=cw)  # image weights
                dataset.indices = random.choices(range(dataset.n), weights=iw, k=dataset.n)  # rand weighted idx
            # Broadcast if DDP
            if rank != -1:
                indices = (torch.tensor(dataset.indices) if rank == 0 else torch.zeros(dataset.n)).int()
                dist.broadcast(indices, 0)
                if rank != 0:
                    dataset.indices = indices.cpu().numpy()

        # Update mosaic border
        # b = int(random.uniform(0.25 * imgsz, 0.75 * imgsz + gs) // gs * gs)
        # dataset.mosaic_border = [b - imgsz, -b]  # height, width borders

        mloss = torch.zeros(4, device=device)  # mean losses
        if rank != -1:
            dataloader.sampler.set_epoch(epoch)
        pbar = enumerate(dataloader)
        logger.info(('\n' + '%10s' * 8) % ('Epoch', 'gpu_mem', 'diff', 'score.0', 'score.1', 'total', 'labels', 'img_size'))
        if rank in [-1, 0]:
            pbar = tqdm(pbar, total=nb)  # progress bar
        optimizer.zero_grad()
        for i, (imgs, targets, paths, _) in pbar:  # batch -------------------------------------------------------------
            ni = i + nb * epoch  # number integrated batches (since train start)
            imgs = imgs.to(device, non_blocking=True).float() / 255.0  # uint8 to float32, 0-255 to 0.0-1.0

            # Warmup
            if ni <= nw:
                xi = [0, nw]  # x interp
                # model.gr = np.interp(ni, xi, [0.0, 1.0])  # iou loss ratio (obj_loss = 1.0 or iou)
                accumulate = max(1, np.interp(ni, xi, [1, nbs / total_batch_size]).round())
                for j, x in enumerate(optimizer.param_groups):
                    # bias lr falls from 0.1 to lr0, all other lrs rise from 0.0 to lr0
                    x['lr'] = np.interp(ni, xi, [hyp['warmup_bias_lr'] if j == 2 else 0.0, x['initial_lr'] * lf(epoch)])
                    if 'momentum' in x:
                        x['momentum'] = np.interp(ni, xi, [hyp['warmup_momentum'], hyp['momentum']])

            # Multi-scale
            if opt.multi_scale:
                sz = random.randrange(imgsz * 0.5, imgsz * 1.5 + gs) // gs * gs  # size
                sf = sz / max(imgs.shape[2:])  # scale factor
                if sf != 1:
                    ns = [math.ceil(x * sf / gs) * gs for x in imgs.shape[2:]]  # new shape (stretched to gs-multiple)
                    imgs = F.interpolate(imgs, size=ns, mode='bilinear', align_corners=False)

            # Forward
            with amp.autocast(enabled=cuda):
                pred = model(imgs)  # forward
                loss, loss_items = compute_loss_ota_dy(pred, targets.to(device), imgs)  # loss scaled by batch_size
                if rank != -1:
                    loss *= opt.world_size  # gradient averaged between devices in DDP mode
                if opt.quad:
                    loss *= 4.

            # Backward
            scaler.scale(loss).backward()

            # Optimize
            if ni % accumulate == 0:
                scaler.step(optimizer)  # optimizer.step
                scaler.update()
                optimizer.zero_grad()
                if ema:
                    ema.update(model)

            # Print
            if rank in [-1, 0]:
                mloss = (mloss * i + loss_items) / (i + 1)  # update mean losses
                mem = '%.3gG' % (torch.cuda.memory_reserved() / 1E9 if torch.cuda.is_available() else 0)  # (GB)
                s = ('%10s' * 2 + '%10.4g' * 6) % (
                    '%g/%g' % (epoch, epochs - 1), mem, *mloss, targets.shape[0], imgs.shape[-1])
                pbar.set_description(s)

                # Plot
                if plots and ni < 10:
                    f = save_dir / f'train_batch{ni}.jpg'  # filename
                    Thread(target=plot_images, args=(imgs, targets, paths, f), daemon=True).start()
                    # if tb_writer:
                    #     tb_writer.add_image(f, result, dataformats='HWC', global_step=epoch)
                    #     tb_writer.add_graph(torch.jit.trace(model, imgs, strict=False), [])  # add model graph
                elif plots and ni == 10 and wandb_logger.wandb:
                    wandb_logger.log({"Mosaics": [wandb_logger.wandb.Image(str(x), caption=x.name) for x in
                                                  save_dir.glob('train*.jpg') if x.exists()]})

            # end batch ------------------------------------------------------------------------------------------------
        # end epoch ----------------------------------------------------------------------------------------------------

        # Scheduler
        lr = [x['lr'] for x in optimizer.param_groups]  # for tensorboard
        scheduler.step()

        # DDP process 0 or single-GPU
        if rank in [-1, 0]:
            # mAP
            ema.update_attr(model, include=['yaml', 'nc', 'hyp', 'gr', 'names', 'stride', 'class_weights'])
            final_epoch = epoch + 1 == epochs
            if not opt.notest or final_epoch:  # Calculate mAP
                wandb_logger.current_epoch = epoch + 1
                test_batch_size = 1
                results, maps, times = test.test(data_dict,
                                                 batch_size=test_batch_size,
                                                 imgsz=imgsz_test,
                                                 model=ema.ema,
                                                 single_cls=opt.single_cls,
                                                 dataloader=testloader,
                                                 save_dir=save_dir,
                                                 verbose=nc < 50 and final_epoch,
                                                 plots=plots and final_epoch,
                                                 wandb_logger=wandb_logger,
                                                 compute_loss=compute_loss,
                                                 is_coco=is_coco,
                                                 v5_metric=opt.v5_metric,
                                                 opt=opt)

            # Write
            with open(results_file, 'a') as f:
                f.write(s + '%10.4g' * 7 % results + '\n')  # append metrics, val_loss
            if len(opt.name) and opt.bucket:
                os.system('gsutil cp %s gs://%s/results/results%s.txt' % (results_file, opt.bucket, opt.name))

            # Log
            tags = ['train/box_loss', 'train/obj_loss', 'train/cls_loss',  # train loss
                    'metrics/precision', 'metrics/recall', 'metrics/mAP_0.5', 'metrics/mAP_0.5:0.95',
                    'val/box_loss', 'val/obj_loss', 'val/cls_loss',  # val loss
                    'x/lr0', 'x/lr1', 'x/lr2']  # params
            for x, tag in zip(list(mloss[:-1]) + list(results) + lr, tags):
                if tb_writer:
                    tb_writer.add_scalar(tag, x, epoch)  # tensorboard
                if wandb_logger.wandb:
                    wandb_logger.log({tag: x})  # W&B

            # Update best mAP
            fi = fitness(np.array(results).reshape(1, -1))  # weighted combination of [P, R, mAP@.5, mAP@.5-.95]
            if fi > best_fitness:
                best_fitness = fi
            wandb_logger.end_epoch(best_result=best_fitness == fi)

            # Save model
            if (not opt.nosave) or final_epoch:  # if save
                ckpt = {'epoch': epoch,
                        'best_fitness': best_fitness,
                        'training_results': results_file.read_text(),
                        'model': get_state_dict(deepcopy(model.module if is_parallel(model) else model).half()),
                        'ema': get_state_dict(deepcopy(ema.ema).half()),
                        'updates': ema.updates,
                        'optimizer': optimizer.state_dict(),
                        'wandb_id': wandb_logger.wandb_run.id if wandb_logger.wandb else None}

                # Save last, best and delete
                torch.save(ckpt, last)
                if best_fitness == fi:
                    torch.save(ckpt, best)
                if (best_fitness == fi) and (epoch >= 200):
                    torch.save(ckpt, wdir / 'best_{:03d}.pt'.format(epoch))
                if epoch == 0:
                    torch.save(ckpt, wdir / 'epoch_{:03d}.pt'.format(epoch))
                elif ((epoch+1) % 25) == 0:
                    torch.save(ckpt, wdir / 'epoch_{:03d}.pt'.format(epoch))
                elif epoch >= (epochs-5):
                    torch.save(ckpt, wdir / 'epoch_{:03d}.pt'.format(epoch))
                if wandb_logger.wandb:
                    if ((epoch + 1) % opt.save_period == 0 and not final_epoch) and opt.save_period != -1:
                        wandb_logger.log_model(
                            last.parent, opt, epoch, fi, best_model=best_fitness == fi)
                del ckpt

        # end epoch ----------------------------------------------------------------------------------------------------
    # end training
    if rank in [-1, 0]:
        # Plots
        if plots:
            plot_results(save_dir=save_dir)  # save as results.png
            if wandb_logger.wandb:
                files = ['results.png', 'confusion_matrix.png', *[f'{x}_curve.png' for x in ('F1', 'PR', 'P', 'R')]]
                wandb_logger.log({"Results": [wandb_logger.wandb.Image(str(save_dir / f), caption=f) for f in files
                                              if (save_dir / f).exists()]})
        # Test best.pt
        logger.info('%g epochs completed in %.3f hours.\n' % (epoch - start_epoch + 1, (time.time() - t0) / 3600))
        # Strip optimizers
        final = best if best.exists() else last  # final model
        for f in last, best:
            if f.exists():
                strip_optimizer(f)  # strip optimizers
        if opt.bucket:
            os.system(f'gsutil cp {final} gs://{opt.bucket}/weights')  # upload
        if wandb_logger.wandb:  # Log the stripped model
            wandb_logger.wandb.log_artifact(str(final), type='model',
                                            name='run_' + wandb_logger.wandb_run.id + '_model',
                                            aliases=['last', 'best', 'stripped'])
        wandb_logger.finish_run()
    else:
        dist.destroy_process_group()
    torch.cuda.empty_cache()
    return results


def main(opt):

    # Set DDP variables
    opt.world_size = int(os.environ['WORLD_SIZE']) if 'WORLD_SIZE' in os.environ else 1
    opt.global_rank = int(os.environ['RANK']) if 'RANK' in os.environ else -1
    set_logging(opt.global_rank)

    # Resume
    wandb_run = check_wandb_resume(opt)
    cfg = os.path.join(partinet.DynamicDet.__path__[0], "cfg", f"dy-{opt.backbone_detector}-step2.yaml")
    hyp = os.path.join(partinet.DynamicDet.__path__[0], "hyp", f"hyp.{opt.hyp}.yaml")
    if opt.resume and not wandb_run:  # resume an interrupted run
        ckpt = opt.resume if isinstance(opt.resume, str) else get_latest_run()  # specified or most recent path
        assert os.path.isfile(ckpt), 'ERROR: --resume checkpoint does not exist'
        apriori = opt.global_rank, opt.local_rank
        with open(Path(ckpt).parent.parent / 'opt.yaml') as f:
            opt = argparse.Namespace(**yaml.load(f, Loader=yaml.SafeLoader))  # replace
        cfg, opt.weight, opt.resume = os.path.relpath(Path(ckpt).parent.parent / 'cfg.yaml'), ckpt, True
        opt.batch_size, opt.global_rank, opt.local_rank = opt.total_batch_size, *apriori  # reinstate
        opt.save_dir = os.path.relpath(Path(ckpt).parent.parent)
        logger.info('Resuming training from %s' % ckpt)
    else:
        # opt.hyp = opt.hyp or ('hyp.finetune.yaml' if opt.weights else 'hyp.scratch.yaml')
        opt.data, cfg, hyp = check_file(opt.data), check_file(cfg), check_file(hyp)  # check files
        assert len(cfg), 'cfg must be specified'
        opt.img_size.extend([opt.img_size[-1]] * (2 - len(opt.img_size)))  # extend to 2 sizes (train, test)
        opt.save_dir = increment_path(Path(opt.project) / opt.name, exist_ok=opt.exist_ok)  # increment run

    # DDP mode
    opt.total_batch_size = opt.batch_size
    assert opt.batch_size == 1, 'set batch_size = 1 for dynamic training'

    device = select_device(opt.device, batch_size=opt.batch_size)
    if opt.local_rank != -1:
        assert torch.cuda.device_count() > opt.local_rank
        torch.cuda.set_device(opt.local_rank)
        device = torch.device('cuda', opt.local_rank)
        dist.init_process_group(backend='nccl', init_method='env://')  # distributed backend
        assert opt.batch_size % opt.world_size == 0, '--batch-size must be multiple of CUDA device count'
        opt.batch_size = opt.total_batch_size // opt.world_size

    # Hyperparameters
    with open(hyp) as f:
        hyp = yaml.load(f, Loader=yaml.SafeLoader)  # load hyps

    # Train
    logger.info(opt)
    tb_writer = None  # init loggers
    if opt.global_rank in [-1, 0]:
        prefix = colorstr('tensorboard: ')
        logger.info(f"{prefix}Start with 'tensorboard --logdir {opt.project}', view at http://localhost:6006/")
        tb_writer = SummaryWriter(opt.save_dir)  # Tensorboard
    train(hyp, opt, device, cfg, tb_writer)