daeing
daeing
Can this code satisfied different size of image use your pretrained model? I found you use pretrained model from deit, and resize each image to 224 * 224? So can...
Where is your dataset from? You collected it by yourself or generate it use some way?
The dataset link is not available now. Would you mind share your downloaded data? Many thanks!
loss = criterion(preds, text, tuple(preds_size), length) File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl return forward_call(*input, **kwargs) File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/loss.py", line 1714, in forward return F.ctc_loss(log_probs, targets, input_lengths, target_lengths, self.blank, self.reduction, File "/opt/conda/lib/python3.8/site-packages/torch/nn/functional.py",...
what is the compute capability of 3090ti?
I noticed that the way of pre-training is to use mask language modeling in your article. Is there any plan to expose this part of the code? Or is there...
loss nan
My graphics card does not have 32G memory. After I changed the batch size to a smaller one, the loss appeared nan, and I couldn't get the best model. Do...
After downloading the dataset, I tried run gen_hybrid_data.py to generate the tree txt according to the train_labels.txt. It shows the error below: unknown word before { train_476 1 unknown word...
您好，请问您的后处理流程的代码可以开源参考一下嘛？您说你没有使用这个方式，是因为这个参数的调节会导致有的图像效果好有的图像效果差是嘛？
感谢作者非常棒的项目，并且开源出来供大家学习交流！ 有一点问题想和您讨论一下： 关于trimap的生成方式，您在这个项目和matte anything项目中好像有点不一样(https://github.com/hustvl/Matte-Anything/blob/7743e36c55f40902461c2854c97d3aea44725c9e/matte_anything.py#L78C1-L86C18)。想请问一下，如果想使用您这个项目，我已知mask的情况下，用matte anything的方式生成mask是不是也是可行的呢？