mindspore-lab/mindnlp
Easy-to-use and high-performance NLP and LLM framework based on MindSpore, compatible with models and datasets of 🤗Huggingface.
PythonApache-2.0
Issues
- 1
weight_norm 将已经删除的weight权重又重新添加回去
#1111 opened by ultranationalism - 1
测试失败,提示runtimeerror,详细错误在文档里面
#1215 opened by ilovexsir - 2
适配GLM4-9b,提供一个推理代码可以吗?
#1163 opened by jesuswa - 1
调用Accumulator报RuntimeError
#1130 opened by FLoutione - 1
OpenELMForCausalLM推理报错
#1191 opened by Ethan-Chen-plus - 1
Bloom相关接口无法导入
#1192 opened by Ethan-Chen-plus - 0
qwen2速度慢
#1168 opened by AnitaSherry - 4
无法跑通qwen1.5
#1152 opened by AnitaSherry - 1
- 0
- 0
Add adapter过程中peft_config缺少属性
#1128 opened by hubosynapse - 5
Add ChatPDF demo
#973 opened by lvyufeng - 3
- 1
- 0
希望适配支持 EMU2多模态模型
#1090 opened by dongdongqiang2018 - 0
MindNLP介绍PPT
#1054 opened by lvyufeng - 0
- 0
tensor索引错误
#1082 opened by QAQusers - 4
应用chatglm3-cli_demo.py直接拉取报错,麻烦协助看下
#1021 opened by maninmancod - 0
add an example of starcoder with llm-vscode
#1035 opened by lvyufeng - 1
Bert-LSTM-CRF 例子,提示ValueError: For 'MatMul' the input dimensions must be equal, but got 'x1_col': 768 and 'x2_row': 3072.
#986 opened by NoComments - 0
PEFT中的add_adapter接口
#1058 opened by hubosynapse - 1
希望可以提供prompt tuning的代码样例
#1028 opened by beepkh - 1
执行ut test_modeling_graph_chatglm.py反序列化报错
#1065 opened by Jia-Xueyu - 2
StarCoder2-2b pipeline生成时报错'AttributeError: 'tokenizers.Tokenizer' object has no attribute 'get_added_tokens_decoder'
#1074 opened by xing-yiren - 1
mindnlp不支持stream_generate,影响使用,麻烦开发支持,谢谢
#1023 opened by rchjia - 7
whisper-tiny无法输出简中
#1062 opened by yang-441 - 0
希望可以提供COT推理的代码样例
#1029 opened by beepkh - 2
whisper推理速度问题
#1060 opened by yang-441 - 0
add core module to design a New Tensor
#1053 opened by lvyufeng - 0
支持类bitsandbytes的量化能力
#1052 opened by lvyufeng - 0
支持huggingface、modelscope、openmind、wisemodel等平台的模型上传
#1051 opened by lvyufeng - 0
- 0
- 2
希望Trainer支持data_collator
#1020 opened by PhilipGAQ - 1
希望可以提供Mixtral 8x7B模型适配版本
#1024 opened by gccyya - 2
使用mindnlp套件本地快速拉起推理失败,麻烦协助
#1018 opened by maninmancod - 0
- 0
segformer 适配
#1005 opened by dongdongqiang2018 - 0
When using the pythia-410m-deduped test model with GPT NeoX, there is a weight disappearance issue when using the float16 parameter
#1002 opened by FlySky-z - 0
mindspore框架同步问题
#1006 opened by cqulilujia - 2
如何通过MindSpore_lite来使用mindnlp
#1004 opened by yang-441 - 1
关于embedding模型的调用示例
#989 opened by zhangheyi-1 - 0
GPT2 inference performance need to be boosted, if we want to beat pytorch
#990 opened by WilliamLiuAtCPC - 3
Bug in examples/text_generation
#981 opened by WilliamLiuAtCPC - 0
Support qwen2 moe
#970 opened by lvyufeng - 0
Add RLHF demo
#975 opened by lvyufeng - 0
Add llama index(sql) demo
#974 opened by lvyufeng - 0
support zero offload
#972 opened by lvyufeng - 0
Support dbrx
#971 opened by lvyufeng