tensorflow-onnx
tensorflow-onnx copied to clipboard
onnx
Convert TensorFlow, Keras, Tensorflow.js and Tflite models to ONNX
# Summary - current tf2onnx doesn't support to convert without transpose NHWC to NCHW in split node. - and the most of functions are not consider when operator has multiple...
Fix Edge Case: Shared Layer Output Between Model Output and Internal Layers # 🎯 Summary This PR addresses a specific edge case that occurs when converting TensorFlow models to ONNX....
**Describe the bug** ``` poetry add tf2onnx Using version ^1.16.1 for tf2onnx Updating dependencies Resolving dependencies... (0.4s) Because no versions of tf2onnx match >1.16.1,=3.20,=1.16.1,=3.20,=5.28.0) and no versions of tensorflow match...
I got "StatefulPartitionedCall is not supported" when trying to convert a model created with keras and Conv2D layers inside. The onnx model is created, but can't be run because of...
**Describe the bug** `tf.clip_by_value` gets converted to a sequence of `min`->`max` instead of `clip`. **System information** - OS Platform and Distribution (e.g., Linux Ubuntu 18.04*): Ubuntu 24.04 - TensorFlow Version:...
**Describe the bug** I have inference function and params of a jax model and convert it to a tf saved model. I encounter a issue when i convert the saved...
Before submitting your request, please review past submissions to ensure that it is not a duplicate of a known feature request. ### Describe the feature request We are releasing ONNX...
After conversion, onn.checker throws error "No Op registered for RFFT2D with domain_version of 18" **Code** import tf2onnx tflite_path = "/kaggle/input/birdnet/tensorflow2/default/1/CustomClassifier.tflite" onnx_path = "/kaggle/working/bpi.onnx" model_proto, external_tensor_storage = tf2onnx.convert.from_tflite(tflite_path, opset=18) onnx.checker.check_model(onnx_path, True)...
is it possible to get this working with a dynamic batch size bg shape = (None, const, const, const) ```py patch1 = tf.image.extract_patches(bg, sizes=[1, self.k_size, self.k_size, 1], strides=[1, self.stride, self.stride,...
