You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
zyw_hw a99445a1e8 pipeline llama2 tokenizer bug fix 6 months ago
..
__init__.py add blip2 2-stage inference support 8 months ago
base_pipeline.py batch predict for pipeline 6 months ago
build_pipeline.py 修改build_optim和注释问题 1 year ago
fill_mask_pipeline.py API整改:auto_class、pipeline 7 months ago
image_classification_pipeline.py API整改:auto_class、pipeline 7 months ago
image_to_text_generation_pipeline.py fix multi_image infer with pipeline in blip2 doc and improve prompt and template 7 months ago
masked_image_modeling_pipeline.py 添加分布式推理 10 months ago
pipeline.py pipeline llama2 tokenizer bug fix 6 months ago
question_answering_pipeline.py 添加分布式推理 10 months ago
segment_anything_pipeline.py 新增SAM模型推理 8 months ago
text_classification_pipeline.py Wrapper Pipeline部分API说明文档整改 7 months ago
text_generation_pipeline.py batch predict for pipeline 6 months ago
token_classification_pipeline.py Wrapper Pipeline部分API说明文档整改 7 months ago
translation_pipeline.py Wrapper Pipeline部分API说明文档整改 7 months ago
zero_shot_image_classification_pipeline.py Wrapper Pipeline部分API说明文档整改 7 months ago

MindSpore Transformers套件的目标是构建一个大模型训练、推理、部署的全流程套件: 提供业内主流的Transformer类预训练模型, 涵盖丰富的并行特性。 期望帮助用户轻松的实现大模型训练。 文档:https://mindformers.readthedocs.io/zh-cn/latest/

Jupyter Notebook Python Markdown Shell