-
imagenet对我来说太大了，我想用自己的图片来进行训练，我是不是直接把八万张图片丢到train文件夹里面就可以了呢，请问一张a100够吗，我需要调整哪些别的超参数呢，谢谢您的工作，祝您一切顺利！！！
-
您好！
我想问一下：
Q1: 在模型训练的时候，模型最终是几个输出？是 **BatchFormer前的classifier** + **BatchFormer后的classifier** 这两个输出呢？然后分别与ground truth计算呢？
Q2: 代码里面，如下所示，这行代码表示的是**BatchFormer后的classifier**的输出呢？还是对模型两个classifier…
-
Hello,
How does your pipeline work for patch classification? After extracting the features from the patches, for classification what code needs to be executed? TIA.
-
Hi,
Thank you for your valuable work. Could you please provide how long have you trained your model (moco v3)?
To be specific, how many epochs (for your currently released model) and how long does…
-
I met an issue training resnet-50 with moco-v3. Under the distributed training setting with 16 V100 GPUs (each process only has one gpu, batch size 4096), I can get the training loss at about 27.2 in …
-
As there is no code related to figure7 interpolation in the GitHub repository.
-
Dear author
I have reproduced your code using 64 V100 GPUs. Every setting is the same as paper (batch size 4096), The end-to-end finetuning is almost the same as paper. However, the linear pro…
-
-
Hello, I was wondering whether there is a documentation for training the model from scratch on any type of dataset. Thanks!
-
There's small nuances in how the dynamo runners benchmark models that can make certain torchbench models fail
Some models might be explicitly skipped, others might fail because of some dtype conve…