Update README.md
Browse files
README.md
CHANGED
|
@@ -22,7 +22,7 @@ This model is the ONNX version of [https://huggingface.co/SamLowe/roberta-base-g
|
|
| 22 |
|
| 23 |
`onnx/model.onnx` is the full precision ONNX version
|
| 24 |
|
| 25 |
-
- that has identical accuracy/metrics to the original
|
| 26 |
- and has the same model size (499MB)
|
| 27 |
- is faster than inference than normal Transformers, particularly for smaller batch sizes
|
| 28 |
- in my tests about 2x to 3x as fast for a batch size of 1 on a 8 core 11th gen i7 CPU using ONNXRuntime
|
|
|
|
| 22 |
|
| 23 |
`onnx/model.onnx` is the full precision ONNX version
|
| 24 |
|
| 25 |
+
- that has identical accuracy/metrics to the original Transformers model
|
| 26 |
- and has the same model size (499MB)
|
| 27 |
- is faster than inference than normal Transformers, particularly for smaller batch sizes
|
| 28 |
- in my tests about 2x to 3x as fast for a batch size of 1 on a 8 core 11th gen i7 CPU using ONNXRuntime
|