WanBenLe

I mean I've already done this, but I don't know if you'd like me to mention this part of the code to linarmodels according to the github process. 

I was using IVGMM before and found that there is no support for multi-IV and panel GMM, yesterday I checked the source code and found that there is no support...

One of my ideas is whether it is possible to pass in panel data through the pannel data object, but I would first make this part of the code run...

I found some documentation that may help to systematically update the work involved: http://scholar.harvard.edu/files/wirev_092218-_corrected_0.pdf www.cambridge.org/core/journals/econometric-theory/article/abs/large-system-of-seemingly-unrelated-regressions-a-penalized-quasimaximum-likelihood-estimation-perspective/049B50430D3563728D69E541D5BEFE37 We can also check the results from Stata's ivreg2 module. http://fmwww.bc.edu/RePEc/bocode/i/ivreg2.html I may follow up...

+1 for this feature, datadreamer seems couldn't Improve inference speed(with model copy)) of a single model on multiple GPUs

I have fixed the issue of llava-v1.5 in the latest transformers version and supported llava-v1.6 (LLavaNext), can I create a PR for these? with /root/autodl-tmp/wsl/AutoAWQ-main/awq/modules/fused/model.py and others ``` if input_ids...

> Hi @WanBenLe, would you mind sharing the exact script for quantization to get `llava-v1.6-34b-hf-awq`? @casper-hansen I'm also wondering if the PR will be merged soon For the unmerged version(AutoAWQ==0.25),...

> 你好，请问autoawq 0.2.5支持llava 1.5吗，能给一下示例代码吗，要求的最低transformers版本是什么？ 你要不直接用example的官方示例+我的那个代码试试?应该能跑起来. 下面是两个链接,一个是原始的llava-v1.5pr https://github.com/casper-hansen/AutoAWQ/pull/250 一个是新的等待merged的pr https://github.com/casper-hansen/AutoAWQ/pull/471

@1SingleFeng 你可以看一下是不是你的model还是数据没有to(cuda)之类的,如果你不打算用cuda可以设置os的cuda设备为空

AWQ is a weight quantization-only optimization based on the activation of each layer of the model, so it does not perform high GPU resource calculations ('model.quantize' in your code, but...