Video Classification
Transformers
PyTorch
Safetensors
English
xclip
feature-extraction
vision
Eval Results (legacy)
Instructions to use aurelio-ai/xclip-base-patch16-zero-shot with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use aurelio-ai/xclip-base-patch16-zero-shot with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("video-classification", model="aurelio-ai/xclip-base-patch16-zero-shot")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("aurelio-ai/xclip-base-patch16-zero-shot") model = AutoModel.from_pretrained("aurelio-ai/xclip-base-patch16-zero-shot", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
-
reverted back
Verified signature0a6e420 -
updated hugginface hub
Verified signaturea979a85 -
updating transformer version
Verified signatureb21cd05 -
update hugginface hub
Verified signatureOn huggingface-hub==0.17.3 getting ImportError: cannot import name 'HF_HOME' from 'huggingface_hub.constants' (/usr/local/lib/python3.10/dist-packages/huggingface_hub/constants.py)582f445
-
Added a variable to handle OCR extraction frame rate
8ed3901 -
Instead of OCR every frame, do it every 6 video frames
32565de -
Instead of OCR every frame, doit every 8 video frames
1dfe4c5
-
Changed back to set, with jsondumps for serializing
374a5b3 -
Changed set for list to be json serializable
f33eeea -
Added easyocr for videoframes
0dbfcb8
-
Removed some loggings
cd14c77 -
Added back some logging messages
83cd9c3 -
Added back some logging messages
41bc78a -
Limited batch size to 4
281ccac -
Returned to ThreadPoolExecutor
54d363d -
Changed ThreadPoolExecutor for ProcessPoolExecutor
7c13868 -
Changed OpenCV2 for decord in extracting videoframes
39adffa -
Changed OpenCV2 for decord in extracting videoframes
ccc4e99
-
Added timeit loggings
ade4ac9 -
Added timeit loggings
ab10ce8 -
Modified the squeeze() application to embeddings
dae7ab5 -
Added normalization using torch and shape check
073334d -
Added normalization using torch
0f7ecda -
Removed normalization of embeddings
a0a72c0 -
Removed troch.stack()
c45d9fe -
Modified embed_frames_with_xclip_processing
cbcc67d
-
Added more logging messages
b26ff1b -
Added logging messages
cec58ea -
Initial commit of XCLIP video processing endpoint only video frames
e8faf30
-
Duplicate from microsoft/xclip-base-patch16-zero-shot
Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>e30ecf9