MiniLM
Collection
Fine-tuned version of Microsoft's MiniLM models, trained on the GLUE MRPC dataset. • 3 items • Updated
Text Classification
Transformers
ONNX
English
bert
text-classfication
int8
Intel® Neural Compressor
PostTrainingDynamic
# Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("Intel/MiniLM-L12-H384-uncased-mrpc-int8-dynamic-inc")
model = AutoModelForSequenceClassification.from_pretrained("Intel/MiniLM-L12-H384-uncased-mrpc-int8-dynamic-inc", device_map="auto")Quick Links
INT8 MiniLM-L12-H384-uncased-mrpc
Post-training dynamic quantization
ONNX
This is an INT8 ONNX model quantized with Intel® Neural Compressor.
The original fp32 model comes from the fine-tuned model Intel/MiniLM-L12-H384-uncased-mrpc.
Test result
| INT8 | FP32 | |
|---|---|---|
| Accuracy (eval-f1) | 0.9107 | 0.9097 |
| Model size (MB) | 33 | 128 |
Load ONNX model:
from optimum.onnxruntime import ORTModelForSequenceClassification
model = ORTModelForSequenceClassification.from_pretrained('Intel/MiniLM-L12-H384-uncased-mrpc-int8-dynamic')
- Downloads last month
- 5
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="Intel/MiniLM-L12-H384-uncased-mrpc-int8-dynamic-inc")