-
I have cuda118, rtx A5000, bitsandbytes==0.39.0
When shifting from fp16 Lora fine-tuning to int8 or fp4 I have immidiate x2 performance drop
Is it an expected behaviour or I have a problem on my s…
-
The example in transformer_asr.py is really interesting, but it only shows how to train the model. Can someone give an example of inference is done with the model to get TTS transcripts? Just callin…
-
Hi @OlivierDehaene and @Narsil ,
Thank you for maintaining this great repo. Quick question, technically the optimized kernels in GPTQ and BNB 4bits only support batch size 1. I am a bit confused …
-
In the `_generate_sequence` of ppo_trainer.py, we have
```
with torch.no_grad():
seq = self.actor_model.module.generate(prompts,
…
-
### Describe the bug
UCX deployed using the Intel Daos fileysystem. Upon running 'daos pool quqery tank' I get the following stack trace.
### Steps to Reproduce
- daos pool query tank
### …
-
**Describe the bug**
Applying Elixir's `String.replace/4` to some specific concatenations strings causes segmentation faults.
**To Reproduce**
Following Dockerfile reproduces the segmentation fau…
-
Another wish list request:
Is it possible for the segmenting code to recognize that it is reading a zero filled segment and accelerate the rate at which it is processed? When segmenting with 8 KiB …
xcfmc updated
6 months ago
-
Do you have a good sense of how much RAM and disk CryptOpt should use? I haphazardly attempted a week-long CryptOpt run and saw it reach memory exhaustion on day 2. Here's an output from the rare inst…
-
Recently [we found out](https://docs.aws.amazon.com/AmazonCloudWatch/latest/APIReference/API_PutMetricData.html) that `PutMetricData` endpoint supports gzipped request bodies.
So, to be able to mak…
-
I have the sequencing data in Nanopore of a small amplicon (230 bp) of the HLA-C gene. One of the challenges I face is the low quality of the sequences obtained, as well as the fragmentation of the se…