WZMIAOMIAO/deep-learning-for-image-processing · error · FileNotFoundError

not found weights file: {}

Error message

not found weights file: {}

What it means

This FileNotFoundError is raised in main() of the efficientnetV2 training script when args.weights points to a path that does not exist. The script only attempts torch.load if os.path.exists(args.weights) is true; otherwise it aborts training early rather than silently starting from random weights.

Source

Thrown at pytorch_classification/Test11_efficientnetV2/train.py:78

                                               collate_fn=train_dataset.collate_fn)

    val_loader = torch.utils.data.DataLoader(val_dataset,
                                             batch_size=batch_size,
                                             shuffle=False,
                                             pin_memory=True,
                                             num_workers=nw,
                                             collate_fn=val_dataset.collate_fn)

    # 如果存在预训练权重则载入
    model = create_model(num_classes=args.num_classes).to(device)
    if args.weights != "":
        if os.path.exists(args.weights):
            weights_dict = torch.load(args.weights, map_location=device)
            load_weights_dict = {k: v for k, v in weights_dict.items()
                                 if model.state_dict()[k].numel() == v.numel()}
            print(model.load_state_dict(load_weights_dict, strict=False))
        else:
            raise FileNotFoundError("not found weights file: {}".format(args.weights))

    # 是否冻结权重
    if args.freeze_layers:
        for name, para in model.named_parameters():
            # 除head外，其他权重全部冻结
            if "head" not in name:
                para.requires_grad_(False)
            else:
                print("training {}".format(name))

    pg = [p for p in model.parameters() if p.requires_grad]
    optimizer = optim.SGD(pg, lr=args.lr, momentum=0.9, weight_decay=1E-4)
    # Scheduler https://arxiv.org/pdf/1812.01187.pdf
    lf = lambda x: ((1 + math.cos(x * math.pi / args.epochs)) / 2) * (1 - args.lrf) + args.lrf  # cosine
    scheduler = lr_scheduler.LambdaLR(optimizer, lr_lambda=lf)

    for epoch in range(args.epochs):
        # train

View on GitHub (pinned to 1ec3fe6f37)

Solutions

1. Check the path exists before running: ls the value you pass to --weights and correct it.
2. Download the pre-trained weights the tutorial expects and pass its exact absolute path to --weights.
3. Run with --weights '' (empty) if you intend to train from scratch, since the script only enters the load branch when args.weights is non-empty/truthy.
4. Use an absolute path to eliminate working-directory ambiguity.

Example fix

// before
python train.py --weights ./efficientnet_v2_rw_s weights.pth
// after
python train.py --weights /abs/path/efficientnet_v2_rw_s-dd5fe8b6.pth

import os
weights = args.weights
if weights and not os.path.isfile(weights):
    raise FileNotFoundError(f"weights not found: {os.path.abspath(weights)}")

try:
    weights_dict = torch.load(args.weights, map_location=device)
except FileNotFoundError as e:
    print(f"WARNING: {e}; training from scratch")
    weights_dict = None

When it happens

Trigger: Running `python train.py --weights path/to/weights.pth` where the file path is wrong, the .pth was never downloaded, or the path is relative to a different working directory than the one the script runs from.

Common situations: Following a tutorial and forgetting to download the official efficientnetv2 pre-trained weights; renaming or moving the weights file after download; running the script from a different cwd so the relative path no longer resolves.

Related errors

• not found weights file: {}
• not found weights file: {}
• not found weights file: {}
• not found weights file: {}
• dataset have {} classes, but input {}

AI-assisted analysis of WZMIAOMIAO/deep-learning-for-image-processing@1ec3fe6f37 (2026-08-30). Data as JSON: /api/errors/36777c42914f2d21. Report an issue: GitHub.