onnx2keras

onnx2keras copied to clipboard

Hi, I'm trying to export a model using a Softsign activation and it's not working: ``` File "...miniconda3/envs/py39/lib/python3.9/site-packages/onnx2keras/converter.py", line 175, in onnx_to_keras AVAILABLE_CONVERTERS[node_type]( KeyError: 'Abs' ``` Could you add support...

peterukk

Hello friends, I tried to use a simple [onnx mnist model ](https://github.com/onnx/models/blob/master/vision/classification/mnist/model/mnist-1.onnx), but got an error: IndexError: list index (1) out of range. The code: `import onnx` `from onnx2keras import...

enuzhin

I am trying to convert onnx model to keras. I am getting the following error  Any clue how to solve the following error?

sangjun04

Here is a simple steps to get tensorflow (keras) model from onnx and it throws this error: `ValueError: Input 0 is incompatible with layer model: expected shape=(None, 3, 60, 60),...

seyeeet

- Add `Resize` key to AVAILABLE_CONVERTERS as `convert_upsample` function - Add additional Clip logic in case if `min` and `max` params are contained in available layers, not in `node.attributes` -...

MaximKuklin

Currently change_ordering=True doesn't conver global average pool. With the addition of argument axis to the target function target_func will fall under this condition https://github.com/nerox8664/onnx2keras/blob/b5d77c0920eeefef86ea309d7776452a85ec454b/onnx2keras/converter.py#L242 And later conversion here. Sorry didn't...

Vozf

The purpose of this PR is to convert Keras Lambda layers in Keras specific layers when converting softmax We are converting here Lambda layer that runs a softmax computation in...

doronHarnoy

This PR is fixing a naming problem that is causing the call of `Keras` layers to fail. In `TensorFlow.Keras` the char `:` is not supported for `name_scope` hence using it...

doronHarnoy

Add new operators

1

This PR adds support for the Sign and LogSoftmax operators. Limited tested on a couple of models.

teobaluta

ibadr