wenmeng zhou comments

Results 74 comments of


                                            wenmeng zhou

Englishfriend.ly

image file is broken, please upload it again

在用pyinstaller打包代码报错，但是验证内置模型就没问题KeyError: 'speech_dfsmn_kws_char_farfield is not in the pipelines registry group keyword-spotting. Please make sure the correct version of ModelScope library is used.'

```shell 2024-04-29 01:14:24,335 - modelscope - INFO - AST-Scanning the path "C:\Users\Administrator\AppData\Local\Temp_MEI43202\modelscope" with the following sub folders ['models', 'metrics', 'pipelines', 'preprocessors', 'trainers', 'msdatasets', 'exporters'] 2024-04-29 01:14:24,335 - modelscope - INFO...

关于使用ModelScope离线启动已下载过的模型时遇到的问题

model_id直接换成本地cache目录

RuntimeError: Failed to import modelscope.models.audio.tts.sambert_hifi because of the following error (look up to see its traceback): No module named 'kantts'

try to execute `pip install modelscope[audio]`

官方文档环境安装能优化下吗？

你好，直接docker run 拉起进行在里面运行代码就可以。开源社区的文档更多的面向开发者，一些最基础的信息如果我们没有描述清楚可以借助搜索引擎查询即可示例命令 ```shell docker run --gpus='"device=0"' registry.cn-beijing.aliyuncs.com/modelscope-repo/modelscope:ubuntu22.04-cuda11.8.0-py310-torch2.1.0-tf2.14.0-1.10.0 bash ```

如果一个模型仓库里面只改了一个小文件，是不是整个仓库都重新下载？

目前只会增量更新有更新的文件，如果所有文件都下载是非预期行为，可以把tokenizer.json删除后再执行下，看下是否能复现。版本问题是由于模型官方提供者没有打版本导致，我们沟通下

Tools/method getting the pre-trained tensorflow weight

`8-bit quantization` support

8 bit is a well trade-off between speed and accuracy in practice, so 8bit support is strongly required to add to vllm

如何设定推理时的线程数？

现在的问题是pipeline占用cpu核数过多，导致程序运行缓慢？

PunctuationProcessingPipeline: cannot import name 'check_argument_types' from 'typeguard'

can you provide the original error log and traceback？