Skip to content
This repository has been archived by the owner on Oct 13, 2021. It is now read-only.

Releases: onnx/keras-onnx

v1.7.0

08 Jun 06:40
babb949
Compare
Choose a tag to compare

The major update includes:

  1. supports tensorflow 2.x
  2. support ONNX 1.7
  3. support most of huggingface/transformers
  4. improve the RNN model conversion
  5. pytest to be the default unit test
  6. enable flake8 style check
    ......

Details:
Support tf.nn.leaky_relu and fix advanced_activations (#514)
Add both tf.nn.X and tf.compat.v1.nn.X to activation_map (#513)
Handle different input cases for tf.nn.relu6 support (#510)
Support tf.nn.relu6 (#506)
Run flake8 for keras2onnx dir, add flake8 to UT (#504)
Add time_major handling for bidirectional lstms (#498)
try to enable flake8 checker. (#503)
Support Max/Min opset 12 (#500)
Support onnx 1.7 and ort 1.3 in UT and nightly build. (#496)
Fix shrink_mask for dynamic input shape (#499)
Add transformer TFXLNet to nightly build (#495)
Add support for tf.nn.depth_to_space lambda (#492)
add tf.where and logical ops supports. (#490)
Support dynamic end for tf.strided_slice conversion (#491)
in tf2.x, the tf.keras will be the default model format (#486)
Support tf.ArgMax/Min and Add UT for tf.Einsum (#488)
Support tensorflow 2.2 (#484)
Adjust input output sizes when any dim is None (#480)
Fix GPT2 UT for transformers==2.8.0 (#478)
Fix GPT2 output order mismatch (#476)
Handle conv layer spec when input0 is type SpaceToBatchND (#475)
improve the converter debugging (#466)
Add UT for PushTranspose optimizer Unsqueeze case (#472)
Unit test for TopK (#470)
Use greater/less_equal from onnxconverter_common (#469)
Increase error bar for test_efn (#468)
Enable test for efficientNet, fix coverage pytest (#464)
Update the behavior on custom op. (#459)
Handle mask-rcnn conversion for ort 1.2 (#452)
Handle time_major in lstm and remove reshape from embedding (#457)
support the random generator ops and fix the issues on tf.op (#453)
Convert tf EinSum, OneHot, LogicalAnd/Not etc (#449)
Better conversion for the subclassing model and code reformat. (#446)
Enable transformers in nightly build (#444)
Add conversion for FloorDiv, ZerosLike and Fix bug in slice (#439)
Add outputs to jupyter notebook EfficientNet (#445)
Fix some tf2.x conversion bugs. (#443)
Jupyter notebook for EfficientNet (#442)
Parametrizing RNN tests (#441)
Use layer_info.inputs as inputs for efficientNet (#438)
unittest -> pytest (#425)
Fix the depthwise conv_2d output issue. (#437)
Support tf.Cumsum conversion (#435)
Upgrade the converter to 1.7.0, along with the onnxconverter-common (#430)
Update unit test and coverage condition (#428)
Fix _create_keras_nodelist for test_rnn_state_passing (#427)
Support RNN for tf2 and tf.keras (#422)
patch for ir_version with onnx 1.7 packages. (#423)
Support masking for tf2 and tf.keras (#421)
Support initial states for Bidirectional RNN (#417)
Handle TimeDistributed layer for tf2 and tf.keras (#420)
Ping keras-segmentation==0.2.0 (#415)
Update README.md
Update README.md
Bidirectional GRU and SimpleRNN support (#413)
support tensorflow 2.2 and some fixing related to subclassed. (#414)
Fix LSTM layer conversion in tf 2.x (#412)
Add unit test for conv+batch fusion (#411)
add conv-1d keras layer spec. (#410)
Deleted the unused tf2onnx code. (#408)
Refactor RNN parameter extracting (#405)
Convert tf.add_n using keras _builtin (#406)
fix the conv/bn issue on NCHW tf.keras (#404)
Bidirectional Masking support (#400)
Relax vgg-seg error bound in nightly build (#402)
Disable vgg16 in tfv2 nightly build (#399)
fixing the conv auto-pads (#397)
Masking RNN with zeros input (#386)
Add InceptionV3 in tf2.x test. (#396)
Add vgg16 and nasnet to tfv2 application (#395)
Add DepthwiseConv2d to subclassed model and efficient-net test cases (#394)
Add tf2 to nightly build; add tf.square conversion (#393)
Fix get_attr string issue in convert_tf_depthwise (#391)
Custom Masking value (#389)
support the swish activation layer. (#390)
Enable more layer converters for the subclassing model. (#383)
Fixing pip install path in README (#388)
Add conv_add to unit test with constant input (#381)
Revert "add batchnormalization layer. (#380)"
add batchnormalization layer. (#380)
Pin onnxruntime for build in use of onnx < 1.6 (#377)
support the tf2.x variable in this converter. (#376)
test (#374)
conv-transpose layer conversion in the subclassing mode. (#372)
Add tf.cast to test_tf_slice in UT (#370)
Update README.md
Update README.md
Cast input argument of tf.slice to int32 (#369)

Contributions
Our community contributors in this release include @cjermain, @sonu1-p, @CNugteren , @buddhapuneeth and etc. Thanks a lot for their effort to make this converter better.

v1.6.1

03 Apr 18:52
Compare
Choose a tag to compare

A patch release for the issue if there is a new onnx release.

v1.6.5

30 Jan 21:32
5df283e
Compare
Choose a tag to compare

The major update for this release is to support the tf.keras in tensorflow 2.0/2.1, which enable some popular models conversion. like huggingface/transformers.

v1.6.0

30 Oct 23:10
dbe7911
Compare
Choose a tag to compare

Major updates:

  1. Support opset 11 (onnx 1.6.0)
  2. Support causal padding for Conv1D
  3. Convert Keras Flatten Layer to ONNX Flatten Op
  4. Support layers of GaussianNoise, GaussianDropout, AlphaDropout

v1.5.2

27 Sep 23:02
28dd0b6
Compare
Choose a tag to compare

Major update:

  1. Improve submodel and shared model conversion to handle more challenging cases.
  2. Fix and validate several object detection, LSTM, GAN models correctly, and add them to nightly build.
  3. Enable tf direction conversion, add command line support.
  4. Add keras making layer, fix time_distributed layer, LSTM, dot, etc.

v1.5.1

26 Jul 21:23
ffba640
Compare
Choose a tag to compare

Major update:

  1. Work with all tf.keras from multiple tensorflow version, and any bug fixed.
  2. Support ONNX symbolic name constraint.
  3. Better support Keras model layer conversion.
  4. support MaskRCNN and Yolo3 which be run with ONNXRuntime.
  5. Verify model conversion for more categories, such as Speech and GAN
  6. Fixed LSTM/BLSTM conversion bugs.

v1.5.0

10 Jun 23:12
9bb6773
Compare
Choose a tag to compare

keras2onnx version 1.5.0 is now available! This version features ONNX Opset 10 support, compatibility with conversion of state-of-the-art object detection models (YoloV3), and increased test coverage.

How do I use the latest keras2onnx package?

pip install keras2onnx --upgrade
python -c "import keras2onnx"

Note: keras2onnx has been tested with Python 3.5, 3.6, and 3.7. It does not currently support Python 2.x.

Highlights since the last release

  • Updating package version to 1.5.0 (#113)
  • Add OnnxOperatorBuilder (#112)
  • Handle multiple dimensions case for BatchNormalization (#110, #106, #104)
  • Improving test coverage + documentation (#109, #107, #100, #99, #79, #72, #70)
  • Enable the dynamic batch size for the converted model (#93)
  • Bug fixes / Conversion Updates
    • Bug fix in forward and bi-directional LSTM in handling bias (#103)
    • Clean up StridedSlice opset 9 (#73)
    • Fix softmax activation in conv and dense (#69)
    • Address interpolation attribute for Keras Upsampling2D (#101)
    • Fix bidirectional output_seq=False case (#85)
  • CI Build Updates
    • keras-onnx CI support for ONNX 1.5, Python 3.7 (#78)
    • Add CI build for keras applications (#66)
    • Work around conda permission issue in linux (#102)
  • Opset 10 updates
    • Updating ThresholdedRelu and onnxconverter-common package (#96)
    • Convert Upsample using Resize op (#59)
    • Update StridedSlice (#61)