Edit model card

https://huggingface.co/microsoft/Florence-2-large with ONNX weights to be compatible with Transformers.js.

Usage (Transformers.js)

NOTE: Florence-2 support is experimental and requires you to install Transformers.js v3 from source.

If you haven't already, you can install the Transformers.js JavaScript library from GitHub using:

npm install xenova/transformers.js#v3

Example: Perform image captioning with onnx-community/Florence-2-large.

import {
    Florence2ForConditionalGeneration,
    AutoProcessor,
    AutoTokenizer,
    RawImage,
} from '@xenova/transformers';

// Load model, processor, and tokenizer
const model_id = 'onnx-community/Florence-2-large';
const model = await Florence2ForConditionalGeneration.from_pretrained(model_id, {
    dtype: {
        embed_tokens: 'fp16', // or 'fp32'
        vision_encoder: 'fp16', // or 'fp32'
        encoder_model: 'q4',
        decoder_model_merged: 'q4',
    },
});
const processor = await AutoProcessor.from_pretrained(model_id);
const tokenizer = await AutoTokenizer.from_pretrained(model_id);

// Load image and prepare vision inputs
const url = 'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/car.jpg';
const image = await RawImage.fromURL(url);
const vision_inputs = await processor(image);

// Specify task and prepare text inputs
const task = '<MORE_DETAILED_CAPTION>';
const prompts = processor.construct_prompts(task);
const text_inputs = tokenizer(prompts);

// Generate text
const generated_ids = await model.generate({
    ...text_inputs,
    ...vision_inputs,
    max_new_tokens: 256,
});

// Decode generated text
const generated_text = tokenizer.batch_decode(generated_ids, { skip_special_tokens: false })[0];

// Post-process the generated text
const result = processor.post_process_generation(generated_text, task, image.size);
console.log(result);
// { '<MORE_DETAILED_CAPTION>': 'The image shows a vintage Volkswagen Beetle car parked on a cobblestone street in front of a yellow building with two wooden doors. The car is a bright turquoise color and has a classic design with a round body and a sloping roofline. It has two doors on either side of the car, one on the left side and one in the center, with a brown door on the right side. The doors are made of wood and have a rustic, weathered look. The building behind the car is painted in a light yellow color and appears to be old and dilapidated. The sky is blue and there are trees in the background. The image is taken from a low angle, looking up at the car and the building.' }

We also released an online demo, which you can try yourself: https://huggingface.co/spaces/Xenova/florence2-webgpu


Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using 🤗 Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).

Downloads last month
65
Inference API
Inference API (serverless) does not yet support transformers.js models for this pipeline type.

Model tree for onnx-community/Florence-2-large

Quantized
(1)
this model