how to use pretrained model in simplest way?

[Uljibuh]

Before raising a question, you may need to check the following listed items.

Checklist

1. I have searched related issues but cannot get the expected help.
2. I have read the FAQ documentation but cannot get the expected help.

[Uljibuh]

i have triained the video swin transformer on a private dataset, the dataset is 3D medical image, around 400 data in training set, the traning result is not good, sometime the model does not converges at all. i am wondering how to utilize the pretranined model for my own dataset in simplest way? my data input size is (1, 128, 128, 50), (channel, height, width, depth), i am quite new to vision transormer, and i am not familar all configuration files, btw, i am working in kaggle environment.

i have run these codes, and have installed them all. wondering how to use pretrained model and then trian the model on my own dataset?

!pip install git+https://github.com/open-mmlab/mim.git

!git clone https://github.com/open-mmlab/mmaction2.git

pip install -r requirements/build.txt

thank you in advance.