任务池

共 7 条，当前显示第 1–7 条（第 1/1 页）

• Implement `generate` (inference) for torch exported text-generation models

  huggingface/transformers·#32504·Python·★ 166457·774 天未动·0 条评论

  Feature requestGenerationExecuTorch

  75

  可认领
• Caching Past Key values of any length for Vision LLM's

  huggingface/transformers·#31096·Python·★ 166457·845 天未动·2 条评论

  Feature requestGeneration

  74

  可认领
• Adaptive Decoding Support

  huggingface/transformers·#31250·Python·★ 166457·838 天未动·1 条评论

  Feature requestGeneration

  74

  可认领
• Pass input_ids to _update_model_kwargs_for_generation

  huggingface/transformers·#33548·Python·★ 166457·732 天未动·1 条评论

  Feature requestGeneration

  74

  可认领
• Tuning generation_config in Trainer hyperparameter_search (Optuna backend)

  huggingface/transformers·#33755·Python·★ 166457·724 天未动·1 条评论

  trainerFeature requestGeneration

  74

  可认领
• Add support for prefix_allowed_tokens_fn to maintain a state throughout decoding

  huggingface/transformers·#28935·Python·★ 166457·951 天未动·4 条评论

  Feature requestGeneration

  73

  可认领
• TextStreamer: An option to print / put every single token instead of whole words

  huggingface/transformers·#28849·Python·★ 166457·888 天未动·3 条评论

  Feature requestGeneration

  71

  可认领