Hugging Face's logo
Victor Sanh's picture
205 59 166

Victor Sanh PRO

VictorSanh
mervenoyan's profile picture fdaudens's profile picture lhoestq's profile picture
·
SanhEstPasMoi
VictorSanh

AI & ML interests

None yet

Articles

Organizations

Hugging Face's profile picture BigScience Workshop's profile picture Big Science - Modeling Metadata's profile picture HuggingFaceM4's profile picture BigCode's profile picture Benchmark Cluster's profile picture Hugging Face H4's profile picture Hugging Face Smol Cluster's profile picture Hugging Face TB Research's profile picture ZeroGPU Explorers's profile picture Pixel Parsing's profile picture Hugging Face Assignments's profile picture Social Post Explorers's profile picture DistilBERT community's profile picture Hacks/Hackers AI sandbox's profile picture

VictorSanh's activity

New activity in HuggingFaceM4/idefics2-8b 3 days ago

How does the attention_mask contribute to the projector preformance?

12
#45 opened 23 days ago by lucasjin

How to modify the weights of the LLM section

2
#62 opened 3 days ago by cookey39
New activity in HuggingFaceM4/idefics2-8b-chatty 3 days ago

fix conversation typo

3
#6 opened 3 days ago by gautierdag
New activity in HuggingFaceM4/siglip-so400m-14-980-flash-attn2-navit 3 days ago

When will transformers' s siglip supports FlashAtten?

2
#5 opened 3 days ago by lucasjin
New activity in HuggingFaceM4/idefics2-8b 5 days ago

How is the image resolution expanded in a vision encoder?

2
#57 opened 10 days ago by efei
New activity in HuggingFaceM4/idefics2-8b 6 days ago

OSError: HuggingFaceM4/idefics2-8b does not appear to have a file named config.json.

2
#60 opened 8 days ago by zzxslp
New activity in HuggingFaceM4/idefics2-8b 8 days ago

Bug in attention mask

1
#58 opened 10 days ago by lucasjin
New activity in HuggingFaceM4/siglip-so400m-14-980-flash-attn2-navit 9 days ago

What processor to use?

9
#4 opened 30 days ago by floschne
New activity in HuggingFaceM4/idefics2-8b 9 days ago

Provided tuning script maybe error

1
#59 opened 10 days ago by efei
New activity in HuggingFaceM4/idefics2-8b 11 days ago

shape mismatch: value tensor of shape [2320] cannot be broadcast to indexing result of shape [2262]

4
#52 opened 17 days ago by yeargun
commented a paper 12 days ago
New activity in HuggingFaceM4/idefics2-8b 13 days ago

Idefics2-pretraining

4
#54 opened 15 days ago by orrzohar
New activity in HuggingFaceM4/idefics2_playground 13 days ago

🚩 Report

1
#9 opened 15 days ago by Idrissou01
New activity in HuggingFaceM4/idefics2-8b-chatty 13 days ago

How to merge an adapter to the base model

1
#4 opened 14 days ago by Ali-C137
New activity in HuggingFaceM4/WebSight 13 days ago

Concerns About Dataset Quality

1
#11 opened 14 days ago by miroslavjaros
New activity in HuggingFaceM4/Caltech-101 15 days ago

PermissionError: [WinError 32] The process cannot access the file because it is being used by another process

1
#3 opened 15 days ago by Chaoses-Ib
New activity in HuggingFaceM4/idefics2-8b 15 days ago

Dedicated Inference Endpoints for Idefics2-8b

8
#32 opened about 1 month ago by zesquirrelnator

After fine tuning, there is a problem for using it.

4
#50 opened 18 days ago by SalmanFaroz

OutOfMemoryError: CUDA out of memory. Tried to allocate 112.00 MiB. GPU

2
#51 opened 17 days ago by mdeniz1
New activity in HuggingFaceM4/idefics2_playground 15 days ago

getting error when uploading json file

1
#7 opened 19 days ago by mahen23

GPU Only?

2
#8 opened 18 days ago by Joshagibby
New activity in HuggingFaceM4/idefics2-8b-chatty 25 days ago

[Question] question about hyperparameter

1
#1 opened 25 days ago by Lala-chick
New activity in HuggingFaceM4/idefics2_playground 25 days ago

css fix for Duplicate button spacing

2
#4 opened 26 days ago by ysharma
New activity in HuggingFaceM4/idefics2-8b-base 27 days ago

red-teaming learnings in prep of `idefics2-8b-chatty` release
#4 opened about 1 month ago by VictorSanh
New activity in HuggingFaceM4/idefics2-8b 29 days ago

Setting compute_metrics in Trainer() leads to AttributeError

3
#38 opened about 1 month ago by Eyel

Fine-tuned idefics for inference?

3
#37 opened about 1 month ago by baldesco

How can I deploy idefics2-8b with TensorRT + Triton?

9
#31 opened about 1 month ago by catworld1212
New activity in HuggingFaceM4/idefics2-8b about 1 month ago

How many images can it take as an input at a time?

1
#39 opened about 1 month ago by jysung

As of April 18th, 2024, Idefics2 is integrated into the newest pypi release of Transformers
#21 opened about 1 month ago by VictorSanh

NameError: name 'awq_ext' is not defined

2
#29 opened about 1 month ago by abhijeet06793

Text model not being loaded with Flash Attention 2

4
#27 opened about 1 month ago by starzmustdie

Quantized version?

1
#24 opened about 1 month ago by MichielBontenbal

Is TGI possible?

7
#3 opened about 2 months ago by RonanMcGovern
New activity in HuggingFaceM4/idefics2-8b-AWQ about 1 month ago

Add chat template to processor_config.json

1
#2 opened about 1 month ago by Rocketknight1
New activity in HuggingFaceM4/idefics2-8b about 1 month ago

Add chat template to processor_config.json

2
#36 opened about 1 month ago by Rocketknight1

Update README.md

1
#35 opened about 1 month ago by SalmanFaroz

Multi-gpu fine-tuning

16
#30 opened about 1 month ago by matbee
New activity in HuggingFaceM4/idefics2_playground about 1 month ago

Getting image_flag error

1
#2 opened about 1 month ago by meg
New activity in HuggingFaceM4/idefics2-8b about 1 month ago

What is the minimum number of examples to fine-tune IDEFICS-2 to get the expected output?

16
#14 opened about 2 months ago by catworld1212

how to use idefics2 with different parameter value (defined in the demo)

1
#28 opened about 1 month ago by karambos

Error running idefics2-8b-AWQ

23
#7 opened about 2 months ago by oliverguhr

load_image(URL) is invalid

6
#17 opened about 2 months ago by micaelcarvalho

Fine Tuning with QLora Overfitting

1
#26 opened about 1 month ago by RicoRausch

Fine-tuning with LoRA?

2
#25 opened about 1 month ago by PatchouliPatch
New activity in HuggingFaceM4/the_cauldron about 1 month ago

Model Architectures For Training

1
#5 opened about 1 month ago by Taylor658
New activity in HuggingFaceM4/idefics2-8b about 1 month ago

CUDA out of memory on A100 with 40GB

7
#8 opened about 2 months ago by SkalskiP

Cuda OOM by simply doing a forward pass on an A6000 (48GB VRAM)

10
#11 opened about 2 months ago by starzmustdie

About the token / per image in report

2
#13 opened about 2 months ago by lucasjin

OCR on dense images by increasing resampler_n_latents

3
#22 opened about 1 month ago by tonibert

Model is incompatible with Inference Endpoints

2
#23 opened about 1 month ago by sebbyjp

KeyError: 'idefics2'

2
#6 opened about 2 months ago by AlexWuKing

Which version of transformers?

3
#2 opened about 2 months ago by harpreetsahota

ValueError: The checkpoint you are trying to load has model type `idefics2

2
#9 opened about 2 months ago by karambos

Memory optimization for long sequence with many small images: reduce resampler_n_latents

4
#18 opened about 2 months ago by ch272h
New activity in HuggingFaceM4/idefics2-8b-AWQ about 2 months ago

OOM at inferencing

2
#1 opened about 2 months ago by abalogh
New activity in HuggingFaceM4/idefics2-8b about 2 months ago

Model download resume problem

4
#16 opened about 2 months ago by karambos

model parameters (temperature, top_p, ...)

1
#15 opened about 2 months ago by Nitsugua

Where is source code?

1
#12 opened about 2 months ago by lucasjin
commented a paper about 1 year ago