un-certainty

**安装方式**：https://github.com/PaddlePaddle/PaddleHub/blob/release/v2.1/docs/docs_ch/get_start/linux_quickstart.md **问题**：耗时过久（非网络问题，依赖包太多，opencv,fonttools下载了一堆版本），且最终报错`ResolutionTooDeep` **错误**： ``` Traceback (most recent call last): File "/data/envs/conda/envs/py38/lib/python3.8/site-packages/pip/_internal/cli/base_command.py", line 173, i n _main status = self.run(options, args) File "/data/envs/conda/envs/py38/lib/python3.8/site-packages/pip/_internal/cli/req_command.py", line 203, in wrapper return func(self, options, args)...

How to achieve faster convergence speed?

1

Hi, Thank you for the simple yet elegant work! I wonder if you have encountered any convergence issue? I am finetuning a [XGLM(4.5B)](https://huggingface.co/facebook/xglm-4.5B) for a text generation task, with only...

Hi GLM team, It will be very helpful if you can provide a script to export the model into ONNX format.

Download link is broken

3

Hi @chujiezheng , The link is broken: https://1drv.ms/f/s!Aky8v8NZbQx1qjlNdQb7Tb9vLSKc 

Support for XGLM: How to achieve faster inference speed?

5

## Describe a requested feature Thanks for releasing this great library! I am currently working on deploying [facebook/xglm-7.5B](https://huggingface.co/facebook/xglm-7.5B), which is currently not supported by parallelformers. [POLICY.md](https://github.com/tunib-ai/parallelformers/blob/main/POLICY.md) provides a comprehensive guide...

Hello, I am reaching out for clarification on the continuous batching mechanism within the VLLM worker, as described in the [VLLM Integration README](https://github.com/lm-sys/FastChat/blob/main/docs/vllm_integration.md). The documentation states that the system "offers...

I tried deploying the `Baichuan2-13B-Chat` model to [TGI](https://huggingface.co/docs/text-generation-inference/en/quicktour). However the performance drops with the same checkpoint. What's the cause and how can I resolve this? ``` Query: 世界第二高峰是哪里？ HuggingFace：乔戈里峰（Qogir Shi），海拔8611米，位于巴基斯坦和中国边境，是世界第二高峰。...