Carlo Moro

I'm also facing this issue and have not been able to overcome it yet Its usefulness decreases if we can't apply it on another test set @kirisakow your answer is...

@anishafluffy I have been able to overcome this issue. After inspecting mca.py in the repo, I've noticed that one of the first things is the one hot encoding with pd.get_dummies(X)...

I realized that I have 2 seconds of latency on amd cpus (ryzen 1700, ryzen 3500u), but close to zero delay on intel based systems, even low end ones.

> @cnmoro And the rest of the hardware/OS is identical? I have a hard time thinking of something where the CPU model could cause this. For the ryzen 1700 yes,...

thank you for the comment ! I'll try your solution and post the results later

I have the same error

I managed to solve it (but I had to downgrade to python 3.9, it was the only one that successfully installed lxml) 1 - clone https://github.com/aws/aws-lambda-base-images/tree/python3.9 2 - change the...

Same error here. Traceback (most recent call last): File "/usr/local/lib/python3.10/dist-packages/qdrant_client/http/api_client.py", line 95, in send_inner response = self._client.send(request) File "/usr/local/lib/python3.10/dist-packages/httpx/_client.py", line 901, in send response = self._send_handling_auth( File "/usr/local/lib/python3.10/dist-packages/httpx/_client.py", line 929,...

I have noticed the same issue when trying to compress portuguese text (I used the Vicuna model, which works very well for this language). Also, it appears that the GPTQ...

there is a memory leak of sorts. after every 20~50 translations or so, you must do a gc.collect() to avoid it.