search

lucidrains / vit-pytorch

Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch
MIT License
20.42k stars 3.03k forks source link

where is the train.py? tks #185

Open henbucuoshanghai opened 2 years ago

jieruyao49 commented 1 year ago

Have you solved this problem? I have the same problem

michaelkoepf commented 1 year ago

There is neither a training script nor are there pre-trained models. This is a repository that provides implementations of different ViT architectures. If you need an example how to train a model from scratch, check out the provided notebook. Basically, can also use any other tutorial that provides information on training classifiers in PyTorch (e.g., this one) and just swap the classifier (however, keep in mind that depending on your data, additionally modifications might be necessary).