Edit model card

INT8 T5 small finetuned on CNN-News

Post-training dynamic quantization

This is an INT8 PyTorch model quantized with huggingface/optimum-intel through the usage of Intel® Neural Compressor.

The original fp32 model comes from the fine-tuned model shivaniNK8/t5-small-finetuned-cnn-news.

The calibration dataloader is the train dataloader. The default calibration sampling size 100 isn't divisible exactly by batch size 8, so the real sampling size is 104.

The linear modules lm.head, fall back to fp32 for less than 1% relative accuracy loss.

Evaluation result

Accuracy (eval-rouge1) 38.9981 39.2142
Model size 154M 242M

Load with optimum:

from optimum.intel.neural_compressor.quantization import IncQuantizedModelForSeq2SeqLM
int8_model = IncQuantizedModelForSeq2SeqLM.from_pretrained(
Downloads last month
Hosted inference API
This model can be loaded on the Inference API on-demand.