Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix (deprecated) ONNX exporter to account for new tf2onnx API #15856

Merged
merged 2 commits into from
Feb 28, 2022
Merged

Fix (deprecated) ONNX exporter to account for new tf2onnx API #15856

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
efee3e4
Fix (deprecated) ONNX exporter to account for new tf2onnx API
lewtun Feb 28, 2022
91f0f7c
Fix style
lewtun Feb 28, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 8 additions & 4 deletions src/transformers/convert_graph_to_onnx.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -327,8 +327,8 @@ def convert_tensorflow(nlp: Pipeline, opset: int, output: Path):
try:
import tensorflow as tf

import tf2onnx
from tf2onnx import __version__ as t2ov
from tf2onnx import convert_keras, save_model

print(f"Using framework TensorFlow: {tf.version.VERSION}, tf2onnx: {t2ov}")

Expand All @@ -337,11 +337,15 @@ def convert_tensorflow(nlp: Pipeline, opset: int, output: Path):

# Forward
nlp.model.predict(tokens.data)
onnx_model = convert_keras(nlp.model, nlp.model.name, target_opset=opset)
save_model(onnx_model, output.as_posix())
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

With the new tf2onnx API, we no longer need a separate save model step - this is handled by the output_path arg in tf2onnx.convert.from_keras() below

input_signature = [tf.TensorSpec.from_tensor(tensor, name=key) for key, tensor in tokens.items()]
model_proto, _ = tf2onnx.convert.from_keras(
nlp.model, input_signature, opset=opset, output_path=output.as_posix()
)

except ImportError as e:
raise Exception(f"Cannot import {e.name} required to convert TF model to ONNX. Please install {e.name} first.")
raise Exception(
f"Cannot import {e.name} required to convert TF model to ONNX. Please install {e.name} first. {e}"
)


def convert(
Expand Down