tensorflow-onnx

tensorflow-onnx copied to clipboard

Fix catch_errors flag and check axes of expand_dims in einsum optimizer

1

acmore

tf2onnx.convert has no attribute from_saved_model

3

I am trying to convert my custom objectdetection model .pb to .onnx model here is the code snippet ``` import tf2onnx # Replace with your paths saved_model_path = '/path/to/saved_model' onnx_export_path...

marghoobtarar

Issue loading onnx model after conversion

1

**Describe the bug** [saved_model.zip](https://github.com/onnx/tensorflow-onnx/files/11893732/saved_model.zip) I am currently trying to convert and run inference on the attached model, it was made using tensorflow 1 and at first the conversion seems to...

carl-krikorian

Input tensor to ExpandDims is lost when ExpandDims is in first layer of the input model.

1

I came across this issue while attempting to convert Meta's Wav2Letter model to .onnx format. From some preliminary investigation, it seems that in cases where a model has the ExpandDims...

ArmDude

tf2onnx.convert.from_graph_def argument support.

1

# Ask a Question ### Question https://github.com/onnx/tensorflow-onnx/blob/8f8d49a88656a334ec1f4c000552cbdce211a8ac/tf2onnx/convert.py#L588C12-L588C12 Is there any reason why this function doesn't support `optimizer=` argument? ### Further information - Is this issue related to a specific model?...

PeiyanGong

Cannot convert saved model that uses keras Preprocessing layer

5

Conversion failed from saved model to onnx if the model use `tf.keras.layers.IntegerLookup` and `tf.keras.layers.Discretization`. ### System information - Docker container: `nvcr.io/nvidia/tensorflow:22.12-tf2-py3` - OS Platform and Distribution (e.g., Linux Ubuntu 20.04)...

tuanavu

UnicodeDecodeError: 'ascii' codec can't decode byte 0xff in position 1024: ordinal not in range(128)

2

**Describe the bug** I'm seeing a `UnicodeDecodeError: 'ascii' codec can't decode byte 0xff in position 1024: ordinal not in range(128)`. The code in tf_utils.py (https://github.com/onnx/tensorflow-onnx/blob/main/tf2onnx/tf_utils.py#L57) seems to mark this as...

mon95

Is it possible to add a decode_predictions layer to converted Keras models?

1

# Ask a Question In the example notebook for converting keras CV models (https://github.com/onnx/tensorflow-onnx/blob/main/tutorials/keras-resnet50.ipynb) the keras method decode_predictions is used to translate the outputs of both the keras model and...

george-synx

LayerNormalization is not show as LayerNormalization in onnx graph

2

# Ask a Question ### Question The LayerNormalization was translate to some operators and BatchNormalization, when I use opset=17。 I hope it could be a unique operator rather than a...

Azulita0317

Support Depthwise Conv3D

1

### Describe the feature request Right now, TF2ONNX only supports Depthwise Conv2D. Conv3D has different DataFormat from 2D

ShuaiShao93