experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]

Browse Source 
PiperOrigin-RevId: 252864172
This commit is contained in:
A. Unique TensorFlower 2019-06-12 11:27:30 -07:00 committed by TensorFlower Gardener
parent 365f69743a
commit 0481453417
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
tensorflow/lite/g3doc/performance/post_training_quantization.md
View File

@ -66,7 +66,7 @@ deployment to accelerators that support float. To require the converter to only
output integer operations, one can specify:
```
converter.target_spec.supported_ops = [tf.lite.OpSet.TFLITE_BUILTINS_INT8]
converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]
```
Note: `target_spec.supported_ops` was previously `target_ops` in the Python API.