update int8 onnx model and readme

Signed-off-by: dujun <[email protected]>

README.md

@@ -63,14 +63,14 @@ This is an INT8 ONNX model quantized with [Intel® Neural Compressor](https://gi
 
 The original fp32 model comes from the fine-tuned model [electra-small-discriminator-mrpc](https://huggingface.co/Intel/electra-small-discriminator-mrpc).
 
-The calibration dataloader is the eval dataloader. The default calibration sampling size 100 isn't divisible exactly by batch size 8. So the real sampling size is 104.
+The calibration dataloader is the eval dataloader. The calibration sampling size is 100.
 
 #### Test result
 
 |   |INT8|FP32|
 |---|:---:|:---:|
-| **Accuracy (eval-f1)** |0.8993|0.8983|
-| **Model size (MB)**  |32|52|
+| **Accuracy (eval-f1)** |0.8957|0.8983|
+| **Model size (MB)**  |16|52|
 
 
 #### Load ONNX model:

model.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9aaab0ea74e1aba289dae90f053c4d7dbdb9ebc577100b77cdd8736cee3f8683
-size 32868991
+oid sha256:11550dc03b7a071d26cd3b5191c9df3a83113227bb82a6e78f28e6c2ca59e6b3
+size 16314577