1pikachu1111
commited on
Commit
•
19ad58c
1
Parent(s):
024c1ed
update int8 onnx model and readme
Browse filesSigned-off-by: dujun <[email protected]>
- README.md +4 -2
- model.onnx +2 -2
README.md
CHANGED
@@ -63,12 +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 [Alireza1044/albert-base-v2-sst2](https://huggingface.co/Alireza1044/albert-base-v2-sst2).
|
65 |
|
|
|
|
|
66 |
#### Test result
|
67 |
|
68 |
| |INT8|FP32|
|
69 |
|---|:---:|:---:|
|
70 |
-
| **Accuracy (eval-accuracy)** |0.
|
71 |
-
| **Model size (MB)** |
|
72 |
|
73 |
|
74 |
#### Load ONNX model:
|
|
|
63 |
|
64 |
The original fp32 model comes from the fine-tuned model [Alireza1044/albert-base-v2-sst2](https://huggingface.co/Alireza1044/albert-base-v2-sst2).
|
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-accuracy)** |0.9140|0.9232|
|
73 |
+
| **Model size (MB)** |50|45|
|
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:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:64266db101b17dc05cb4fd7b0643d7eda5425f02494feca37c38b2153b3d10fc
|
3 |
+
size 51754627
|