Downloads · 30 days
10
3% of all-time downloads
juliensimon/distilbert-amazon-shoe-reviews-onnx-quantized
distilbert-amazon-shoe-reviews-onnx-quantized is a text classification model from juliensimon. Use it when you need a label for a piece of text. It is set up for transformers.
Dynamically quantized ONNX export of juliensimon/distilbert-amazon-shoe-reviews for maximum CPU inference speed using Optimum.
Downloads · 30 days
10
3% of all-time downloads
All-time downloads
381
Public
Repo size
345 MB
Likes
0
Public
Click a slice to open those files.
.onnx172 MB · 99%
From the Hugging Face model README
Dynamically quantized ONNX export of juliensimon/distilbert-amazon-shoe-reviews for maximum CPU inference speed using Optimum.
Video walkthrough: Accelerate Transformer inference on CPU with Optimum and ONNX
| Detail | Value |
|---|---|
| Base model | juliensimon/distilbert-amazon-shoe-reviews |
| Task | Star rating prediction (1–5 stars) from shoe reviews |
| Format | ONNX with dynamic quantization (INT8 weights) |
from optimum.onnxruntime import ORTModelForSequenceClassification
from transformers import AutoTokenizer, pipeline
model = ORTModelForSequenceClassification.from_pretrained("juliensimon/distilbert-amazon-shoe-reviews-onnx-quantized")
tokenizer = AutoTokenizer.from_pretrained("juliensimon/distilbert-amazon-shoe-reviews-onnx-quantized")
classifier = pipeline("text-classification", model=model, tokenizer=tokenizer)
classifier("These shoes are incredibly comfortable!")