Ivelin Ivanov's picture

6 6 21

Ivelin Ivanov PRO

ivelin

·

AI & ML interests

computer vision, vision-language models, multi modal transformers

Recent Activity

updated a model 17 days ago

ivelin/SmolVLM-Instruct-vqav2

liked a Space 21 days ago

donb-hf/eduscape

reacted to merve's post with 😎 25 days ago

small but mighty 🔥 you can fine-tune SmolVLM on an L4 with batch size of 4 and it will only take 16.4 GB VRAM 🫰🏻 also with gradient accumulation simulated batch size is 16 ✨ I made a notebook that includes all the goodies: QLoRA, gradient accumulation, gradient checkpointing with explanations on how they work 💝 https://github.com/huggingface/smollm/blob/main/finetuning/Smol_VLM_FT.ipynb

View all activity

Organizations

ivelin's activity

New activity in OFA-Sys/OFA-Generic_Interface almost 2 years ago

Transfer Learn Visual Grounding to UI RefExp

#5 opened almost 2 years ago by

New activity in OFA-Sys/ofa-base-caption-fairseq-version almost 2 years ago

Farseq -> Transformers conversion

#1 opened over 2 years ago by

New activity in ivelin/ui-refexp almost 2 years ago

Space Gradio build issue

#2 opened almost 2 years ago by

Apply for community grant: Company project

#1 opened almost 2 years ago by

New activity in ncoop57/rico_captions almost 2 years ago

Can we test RicoSCA with OFA and LayotMLv3

#2 opened about 2 years ago by

New activity in ncoop57/rico_captions about 2 years ago

Credit original RicoSCA source

#1 opened about 2 years ago by