Instructions to use Salesforce/blip-image-captioning-large with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Salesforce/blip-image-captioning-large with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "image-to-text" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large")# Load model directly from transformers import AutoProcessor, AutoModelForImageTextToText processor = AutoProcessor.from_pretrained("Salesforce/blip-image-captioning-large") model = AutoModelForImageTextToText.from_pretrained("Salesforce/blip-image-captioning-large") - Notebooks
- Google Colab
- Kaggle
[blip_text_model] num_attention_heads is 8? not 12? [blip_vision_model] eps is 1e-5?
#5
by junnyu - opened
(1) i find https://github.com/salesforce/BLIP/blob/main/configs/med_config.json num_attention_heads is 12, but in this repo num_attention_heads is 8
(2) and blip_vision_model eps shoud be 1e-6, not 1e-5
Hi @junnyu
I think you are right, see related: https://github.com/huggingface/transformers/issues/22625