1pikachu1111 commited on
Commit
5bea157
1 Parent(s): c731696

update int8 onnx model and readme

Browse files

Signed-off-by: dujun <jun.du@intel.com>

Files changed (2) hide show
  1. README.md +3 -3
  2. model.onnx +2 -2
README.md CHANGED
@@ -63,14 +63,14 @@ This is an INT8 ONNX model quantized with [Intel® Neural Compressor](https://gi
63
 
64
  The original fp32 model comes from the fine-tuned model [xlnet-base-cased-mrpc](https://huggingface.co/Intel/xlnet-base-cased-mrpc).
65
 
66
- 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.
67
 
68
  #### Test result
69
 
70
  | |INT8|FP32|
71
  |---|:---:|:---:|
72
- | **Accuracy (eval-f1)** |0.8935|0.8986|
73
- | **Model size (MB)** |286|448|
74
 
75
 
76
  #### Load ONNX model:
 
63
 
64
  The original fp32 model comes from the fine-tuned model [xlnet-base-cased-mrpc](https://huggingface.co/Intel/xlnet-base-cased-mrpc).
65
 
66
+ The calibration dataloader is the eval dataloader. The calibration sampling size is 100.
67
 
68
  #### Test result
69
 
70
  | |INT8|FP32|
71
  |---|:---:|:---:|
72
+ | **Accuracy (eval-f1)** |0.8974|0.8986|
73
+ | **Model size (MB)** |226|448|
74
 
75
 
76
  #### Load ONNX model:
model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5c71602889b26bf3079ca71f4bb721481d24d6f082d9edada98d3dfe41e9454d
3
- size 299662965
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7299d27322dd1ca9fc9b3a98a064eddbc590a2bbf1d254df79e89463432113fd
3
+ size 236931543