https://huggingface.co/JustinDu/BARTxiv with ONNX weights to be compatible with Transformers.js.
Steps to get ONNX quantized weights
- Use optimum-cli to convert the PyTorch weights to ONNX format.
optimum-cli export onnx --model JustinDu/BARTxiv BARTxiv_onnx
- Use the
quantizescript in the Transformers.js repo to quantize the ONNX weights.
python -m scripts.quantize --input_folder BARTxiv_onnx --output_folder BARTxiv_onnx/onnx
Move the .onnx files in the BARTxiv_onnx folder to BARTxiv_onnx/onnx (these are the FP32 weights).
- Downloads last month
- 38
Inference API (serverless) does not yet support transformers.js models for this pipeline type.