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

[tf.lite] Use in-process conversion when the new converter is used

Browse Source 
Out-of-process conversion was a workaround for the legacy converter,
which would generally crash the process when conversion failed. However,
out-of-process conversion also adds a good deal of complexity, so avoid
it when using the new conversion backend.

PiperOrigin-RevId: 312142994
Change-Id: I7ddc83df99ccf24be6e15f46d6a116dce8321933
This commit is contained in:
Jared Duke 2020-05-18 13:27:55 -07:00 committed by TensorFlower Gardener
parent 9282621aa9
commit 869920697b
1 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
tensorflow/lite/python/convert.py
View File

@ -169,9 +169,10 @@ def toco_convert_protos(model_flags_str,
RuntimeError: When conversion fails, an exception is raised with the error
message embedded.
"""
# TODO(aselle): When toco does not use fatal errors for failure, we can
# switch this on.
if not _toco_from_proto_bin:
# Historically, TOCO conversion failures would trigger a crash, so we would
# attempt to run the converter out-of-process. The MLIR conversion pipeline
# surfaces errors instead, and can be safely run in-process.
if enable_mlir_converter or not _toco_from_proto_bin:
try:
model_str = wrap_toco.wrapped_toco_convert(model_flags_str,
toco_flags_str, input_data_str,