Skip to content

Latest commit

 

History

History
20 lines (17 loc) · 655 Bytes

ROADMAP.md

File metadata and controls

20 lines (17 loc) · 655 Bytes

Roadmap

A rough roadmap for implementing new features. All is subject to change at a moment's notice.

Launch

  • Training using pytorch-lightning, with suppport for fp16 and Colab TPUs.
  • Training a GPT-2 model from scratch w/ parametricized context window sizes and parameters
  • PyTorch support for training/generating
  • Generation from Transformer's native generate() function
  • Actual documentation
    • Examples
      • Training on a CPU
      • Training on a GPU
      • Training on multiple GPU (4x T4)
      • Training on a TPU
      • Cross-Training on Multiple Datasets
      • Generate on a CPU
      • Generate on a GPU
    • API docs for all classes