Hello everyone!
Released a blog on the HuggingFace ecosystem where I have shown how to use and customize datasets
and Trainer
for all of your NLP problems.
Contents of the blog:
-
datasets
andTrainer
overview - Using Dynamic Padding and Uniform Length Batching
- Customizing Trainer
- Custom model implementations
- Custom Callbacks
- Weights & Biases integration
- Logging custom things (prediction distribution) to W&B
Tweet highlighting the content of the blog:
Blog link: