About the default config

[tcmyxc]

I try your code in your repo default config, but I get the acc in cifar10-lt-ir-100 is 87.00%

Is it the software version?

[ynu-yangpeng]

use python main.py --dataset cifar10 -a resnet32 --num_classes 10 --imbanlance_rate 0.01 --beta 0.5 --lr 0.01 --epochs 200 -b 64 --momentum 0.9 --weight_decay 5e-3 --resample_weighting 0.0 --label_weighting 1.2 --contrast_weight 1

[tcmyxc]

For other imbanlance rates, does the config need to be modified?

For example, is the weight of contrast learning loss still 1?

[tcmyxc]

in test file, the func eval_training is not correct, you should consider batch size.

you can see my code:

        model.eval()
        size = len(val_loader.dataset)
        correct = 0
        for i,  (inputs, labels) in enumerate(val_loader):
            inputs, labels = inputs.cuda(), labels.cuda()
            with torch.no_grad():
                logits = model(inputs,train=False)
                correct += (logits.argmax(1) == labels).type(torch.float).sum().item()

                top1, top5 = accuracy(logits.data, labels.data, topk=(1,5))
            output = 'Test:  ' + str(i) +' Prec@1:  ' + str(top1.item())
            print(output, end="\r")

        correct /= size
        return correct

[yc-cui]

Have you run it? Does the modified eval_training have a big impact?

[tcmyxc]

Have you run it? Does the modified eval_training have a big impact?

The experimental results are similar to those provided by the repo, but the script for evaluating weights afterwards is incorrect

[ynu-yangpeng]

Please set the batch size to 64 for both training and testing.

[tcmyxc]

Please set the batch size to 64 for both training and testing.

There is really a problem with your test weight code, you can check. For the same checkpoint file, the results of your test code are different every time.

测试脚本里面，你每次都是把准确率存起来，然后求和，然后除以列表长度，这是不对的。你应该把每个小批量预测正确的数量存起来，在最后求和（或者一直累计），之后再除以数据集样本总量。你的训练脚本就没有这个问题。一个固定的模型权重文件，对应的ACC已经确定了，不可能随着你验证的时候设置的批量大小变化，也不会随着时间变化。

the train code is right.

[ynu-yangpeng]

Thank you for your question. We have updated the code.