Update README.md
Browse files
README.md
CHANGED
|
@@ -24,12 +24,8 @@ The calibration dataloader is the train dataloader. The default calibration samp
|
|
| 24 |
|
| 25 |
### Test result
|
| 26 |
|
| 27 |
-
- Batch size = 8
|
| 28 |
-
- [Amazon Web Services](https://aws.amazon.com/) c6i.xlarge (Intel ICE Lake: 4 vCPUs, 8g Memory) instance.
|
| 29 |
-
|
| 30 |
| |INT8|FP32|
|
| 31 |
|---|:---:|:---:|
|
| 32 |
-
| **Throughput (samples/sec)** |47.554|23.046|
|
| 33 |
| **Accuracy (eval-accuracy)** |0.9037|0.9106|
|
| 34 |
| **Model size (MB)** |65|255|
|
| 35 |
|
|
@@ -41,6 +37,3 @@ int8_model = OptimizedModel.from_pretrained(
|
|
| 41 |
'Intel/distilbert-base-uncased-finetuned-sst-2-english-int8-static',
|
| 42 |
)
|
| 43 |
```
|
| 44 |
-
|
| 45 |
-
Notes:
|
| 46 |
-
- The INT8 model has better performance than the FP32 model when the CPU is fully occupied. Otherwise, there will be the illusion that INT8 is inferior to FP32.
|
|
|
|
| 24 |
|
| 25 |
### Test result
|
| 26 |
|
|
|
|
|
|
|
|
|
|
| 27 |
| |INT8|FP32|
|
| 28 |
|---|:---:|:---:|
|
|
|
|
| 29 |
| **Accuracy (eval-accuracy)** |0.9037|0.9106|
|
| 30 |
| **Model size (MB)** |65|255|
|
| 31 |
|
|
|
|
| 37 |
'Intel/distilbert-base-uncased-finetuned-sst-2-english-int8-static',
|
| 38 |
)
|
| 39 |
```
|
|
|
|
|
|
|
|
|