Régis Pierrard's picture

Régis Pierrard

regisss

·

regisss

AI & ML interests

None yet

Recent Activity

updated a dataset about 1 month ago

regisss/benchmarks

Reacted to onekq's post with 🔥 about 1 month ago

I'm now working on finetuning of coding models. If you are GPU-hungry like me, you will find quantized models very helpful. But quantization for finetuning and inference are different and incompatible. So I made two collections here. Inference (GGUF, via Ollama, CPU is enough) https://huggingface.co/collections/onekq-ai/ollama-ready-coding-models-67118c3cfa1af2cf04a926d6 Finetuning (Bitsandbytes, QLora, GPU is needed) https://huggingface.co/collections/onekq-ai/qlora-ready-coding-models-67118771ce001b8f4cf946b2 For quantization, the inference models are far more popular on HF than finetuning models. I use https://huggingface.co/QuantFactory to generate inference models (GGUF), and there are a few other choices. But there hasn't been such a service for finetuning models. DIY isn't too hard though. I made a few myself and you can find the script in the model cards. If the original model is small enough, you can even do it on a free T4 (available via Google Colab). If you know a (small) coding model worthy of quantization, please let me know and I'd love to add it to the collections.

posted an update about 1 month ago

Interested in performing inference with an ONNX model?⚡️ The Optimum docs about model inference with ONNX Runtime is now much clearer and simpler! You want to deploy your favorite model on the hub but you don't know how to export it to the ONNX format? You can do it in one line of code as follows: ```py from optimum.onnxruntime import ORTModelForSequenceClassification # Load the model from the hub and export it to the ONNX format model_id = "distilbert-base-uncased-finetuned-sst-2-english" model = ORTModelForSequenceClassification.from_pretrained(model_id, export=True) ``` Check out the whole guide 👉 https://huggingface.co/docs/optimum/onnxruntime/usage_guides/models

View all activity

Articles

Organizing a Privacy-preserving Hackathon

Accelerating Vision-Language Models: BridgeTower on Habana Gaudi2

Fast Inference on Large Language Models: BLOOMZ on Habana Gaudi2 Accelerator

Faster Training and Inference: Habana Gaudi®2 vs Nvidia A100 80GB

Organizations

regisss's activity

upvoted an article about 1 month ago

Article

Organizing a Privacy-preserving Hackathon

By

•

Oct 17

• 8

upvoted an article 7 months ago

Article

Energy Scores for AI Models

By

•

May 9

• 30