In this tutorial, we will introduce the shape of weight in torch.nn.Linear().
In this tutorial, we will use some examples to show how to use torch.nn.Module.modules().
In this tutorial, we will discuss if we have saved a model object using torch.save(), can we change our model structure before we plan to load this saved model?
This tutorial will introduce you how to fix torch.save() take much disk space when we are using torch.save().
In this tutorial, we will use an example to show you how to split a big wave file to some clips (small wave files) with same length.
Padding method will affect the performance of LLM. There are two padding method: left and right padding.
As to a tokenizer instance, it contains add_special_tokens parameter. In this tutorial, we will introduce what it mean.
Most of LLMs are decoder-only architectures, which means they are not trained to continue from pad tokens. This strategy may cause wrong outputs when batch inference.
In this tutorial, we will introduce load_in_8bit parameter in AutoModelForCausalLM.from_pretrained()
In this tutorial, we will tell you a simple way to find and solve the nan error when training a torch model.