Yoni Gozlan

[WIP] Add OmDet-Turbo

4

# What does this PR do? This PR adds support for [OmDet-Turbo](https://github.com/om-ai-lab/OmDet), an open-vocabulary detection model from Om Research Lab. ## Who can review? @amyeroberts @qubvel

Batched MultiHeadAttention

4

Hi! Yoni from Hugging Face again. I'm opening a separate issue because there seems to be a potentially important problem in the model's encoder. https://github.com/om-ai-lab/OmDet/blob/542ce974ee22e16f9e532500e3f84e4702c03abf/omdet/omdet_v2_turbo/ela_encoder.py#L27 Shouldn't this MultiHeadAttention be initialized...

Integrating OmDet Turbo in Transformers 🤗

14

Hi Om people! I am an MLE at Hugging Face, and given the popularity and performance of your model, we wanted to see if you would be interested in working...

Uniformize kwargs for LLaVa processor and update docs

1

# What does this PR do? Adds uniformized processors kwargs following https://github.com/huggingface/transformers/issues/31911 for LLaVa ## Before submitting - [ ] This PR fixes a typo or improves the docs (you...

Fix continue_final_message for image-text-to-text chat templates

1

# What does this PR do? The `content` field for an image-text-to-text model is a list, which is not currently taken into account when `continue_final_message` is set to True in...

# What does this PR do? Currently, the modeling file of Paligemma uses `from ...utils import logging`, and the processing file uses `import logging`. That means if we want to...

Add image text to text pipeline

6

# What does this PR do? Add image-text-to-text pipeline! A split of this PR with only model-specific pre and post processing is available [here](https://github.com/huggingface/transformers/pull/32471), in order to reduce the loc...

Add Idefics2/3 and SmolVLM Fast image processors + improvements for fast image processors

2

# What does this PR do? Several things added to this PR: - Idefics2/3 + smolvlm fast image processors. Cc @andimarafioti :) - Improvements in the base fast image processors...

Improve BatchFeature: stack list and nested lists of torch tensors

2

I have been wanting to change that for a while, it shouldn't be a breaking change, but align what we support in `BatchFeature` between numpy arrays and torch tensors. The...

🚨Default to fast image processors for all models

7

# What does this PR do? Following the trial testing with Qwen_VL image processors, this extends defaulting to fast image processors even for checkpoints saved with a slow one to...