adzhua

AssertionError: GPT3Trainer: GPT3ForTextGeneration: pipeline_model parallel group is not initialized

2

请问按照https://modelscope.cn/models/damo/nlp_gpt3_text-generation_1.3B/summary继续训练gpt3报如下错误： AssertionError: GPT3Trainer: GPT3ForTextGeneration: pipeline_model parallel group is not initialized是什么原因啦？是因为某个参数设置错误吗？

RuntimeError: DistributedGPT3Pipeline: Error building extension 'scaled_upper_triang_masked_softmax_cuda'

6

请问调用pipe = pipeline(Tasks.text_generation, model='damo/nlp_gpt3_text-generation_1.3B')时候报： RuntimeError: DistributedGPT3Pipeline: Error building extension 'scaled_upper_triang_masked_softmax_cuda' 这是什么原因啦

使用glm-10b-chinese调用generate方法有时时会出错

1

有时候调用generate会报错RuntimeError: The size of tensor a (529) must match the size of tensor b (528) at non-singleton dimension 1 请问这是什么原因？

相似度结果

3

你好，我试下了两个相同的文本计算的相似度不为1，这是什么原因啦

训练集数据

4

请问训练结束后，训练集中的问题答案能完全记住不？

[Question]: docker compose failed!

8

### Describe your problem run docker compose up -d, received: WARN[0000] The "MACOS" variable is not set. Defaulting to a blank string. [+] Running 0/30 ⠧ ragflow [⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀] Pulling 161.7s...

[Question]: response error

2

### Describe your problem  what is the reason?

[Bug]:top_p is assigned to top_k

2

### Is there an existing issue for the same bug? - [X] I have checked the existing issues. ### Branch name latest ### Commit ID c3b232 ### Other environment information...

### Describe your problem deepdoc/vision/ocr.py load_mode第72-76行，onnxruntime session的代码如下： if False and ort.get_device() == "GPU": sess = ort.InferenceSession( model_file_path, options=options, providers=['CUDAExecutionProvider']) 如果装的是onnxruntime gpu版本的，这样是不是不会用GPU执行？

[Question]: the ragflow minio container keeps restarting

1

### Describe your problem