huggingface--transformers/docs/source/en/model_doc/timm_wrapper.md

TimmWrapper

Overview

Helper class to enable loading timm models to be used with the transformers library and its autoclasses.

from urllib.request import urlopen

import torch
from PIL import Image

from transformers import AutoImageProcessor, AutoModelForImageClassification


# Load image
image = Image.open(urlopen(
    'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))

# Load model and image processor
checkpoint = "timm/resnet50.a1_in1k"
image_processor = AutoImageProcessor.from_pretrained(checkpoint)
model = AutoModelForImageClassification.from_pretrained(checkpoint).eval( device_map="auto")

# Preprocess image
inputs = image_processor(image)

# Forward pass
with torch.no_grad():
    logits = model(**inputs).logits

# Get top 5 predictions
top5_probabilities, top5_class_indices = torch.topk(logits.softmax(dim=1) * 100, k=5)

Resources

A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with TimmWrapper.

• Collection of Example Notebook 🌎

Tip

For a more detailed overview please read the official blog post on the timm integration.

TimmWrapperConfig

autodoc TimmWrapperConfig

TimmWrapperImageProcessor

autodoc TimmWrapperImageProcessor - preprocess

TimmWrapperModel

autodoc TimmWrapperModel - forward

TimmWrapperForImageClassification

autodoc TimmWrapperForImageClassification - forward