| base_model: vikp/texify | |
| library_name: transformers.js | |
| pipeline_tag: image-to-text | |
| https://huggingface.co/vikp/texify with ONNX weights to be compatible with Transformers.js. | |
| ## Usage (Transformers.js) | |
| If you haven't already, you can install the [Transformers.js](https://huggingface.co/docs/transformers.js) JavaScript library from [NPM](https://www.npmjs.com/package/@huggingface/transformers) using: | |
| ```bash | |
| npm i @huggingface/transformers | |
| ``` | |
| **Example:** Image-to-text w/ `Xenova/texify`. | |
| ```js | |
| import { pipeline } from '@huggingface/transformers'; | |
| // Create an image-to-text pipeline | |
| const texify = await pipeline('image-to-text', 'Xenova/texify'); | |
| // Generate LaTeX from image | |
| const image = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/latex2.png'; | |
| const latex = await texify(image, { max_new_tokens: 384 }); | |
| console.log(latex); | |
| // [{ generated_text: "$$ |\\ \\frac{1}{x}=\\frac{1}{c}|=|\\ \\frac{c-x}{xc}|=\\frac{1}{|x|}\\cdot\\frac{1}{|c|}\\cdot|x-c|$$\n\nThe factor $$ \\frac{1}{|x|}$$ is not good if its near 0." }] | |
| ``` | |
| | Input image | Visualized output | | |
| |--------|--------| | |
| |  |  | | |
| --- | |
| 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](https://huggingface.co/docs/optimum/index) and structuring your repo like this one (with ONNX weights located in a subfolder named `onnx`). |