AlexanderLutsenko/nobuco issues and pull requests

#68 - Quantization support?

Issue - State: closed - Opened by happyTonakai 4 months ago - 2 comments

#67 - tensorflowjs woes with saved keras file

Issue - State: closed - Opened by johndpope 4 months ago - 4 comments

#66 - Update README.md

Pull Request - State: open - Opened by johndpope 4 months ago - 1 comment

#65 - multihead attention - no converter found

Issue - State: closed - Opened by johndpope 4 months ago - 3 comments

#64 - color traceback on error messages / upgraded progressbar

Pull Request - State: open - Opened by johndpope 4 months ago - 3 comments

#63 - Unimplemented Nodes: roi_align

Issue - State: closed - Opened by hanspres1999 6 months ago - 3 comments

#62 - Exception: Unimplemented nodes

Issue - State: open - Opened by kabyanil 6 months ago - 1 comment

#61 - Extremely High GPU VRAM Usage During Conversion

Issue - State: closed - Opened by chadrockey 6 months ago - 2 comments

#60 - When to worry about Imprecise conversion / Max diff

Issue - State: closed - Opened by chadrockey 6 months ago - 2 comments

#59 - Update linear.py

Pull Request - State: closed - Opened by kukugpt 7 months ago

#58 - Update misc.py

Pull Request - State: closed - Opened by kukugpt 7 months ago - 1 comment

#57 - Extended return

Pull Request - State: open - Opened by crimson206 7 months ago - 6 comments

#56 - ValidationResult Typehinting

Pull Request - State: closed - Opened by crimson206 7 months ago

#55 - Stateful model

Issue - State: closed - Opened by johan-sightic 7 months ago - 2 comments

#54 - speed issue after conversion

Issue - State: open - Opened by thegodone 7 months ago - 7 comments

#53 - issue in tf.equal converter_eq

Issue - State: closed - Opened by thegodone 7 months ago - 2 comments

#51 - Signature output names

Issue - State: closed - Opened by johan-sightic 7 months ago - 3 comments

#50 - FFT/STFT support

Issue - State: open - Opened by happyTonakai 8 months ago - 1 comment
Labels: help wanted

#49 - Modes for torch.nn.functional.pad and tf.pad

Issue - State: closed - Opened by happyTonakai 8 months ago - 2 comments

#48 - Force static shapes through the graph?

Issue - State: closed - Opened by tonystratum 8 months ago - 9 comments

#47 - Preserve node naming/namespacing

Issue - State: open - Opened by matemijolovic 9 months ago - 4 comments
Labels: enhancement

#46 - Is the MultiheadAttention converter outdated?

Issue - State: closed - Opened by crimson206 9 months ago - 3 comments

#45 - Added python default .gitignore

Pull Request - State: closed - Opened by crimson206 9 months ago - 1 comment

#44 - Hotfix gitignore

Pull Request - State: closed - Opened by crimson206 9 months ago

#43 - Added getattribute converter

Pull Request - State: closed - Opened by crimson206 9 months ago - 1 comment

#42 - Added unittest setup with a unittest

Pull Request - State: open - Opened by crimson206 9 months ago

#41 - Added math-sign converter

Pull Request - State: closed - Opened by crimson206 9 months ago

#38 - Identity layer is considered as 'Unimplemented nodes'

Issue - State: closed - Opened by crimson206 9 months ago - 1 comment

#37 - Conversion fails randomly on models with multiple inputs

Issue - State: closed - Opened by mmjamm 9 months ago - 7 comments
Labels: bug, windows

#36 - ValueError from adding tf.Tensor and None when exporting model with dynamic shapes

Issue - State: open - Opened by NathanielB123 10 months ago
Labels: bug

#34 - Convert pretrained .torch model (including weights)

Issue - State: open - Opened by CarlosNacher 10 months ago - 4 comments

#33 - When groups equal input channels then PyTorch Conv2D is mapped to Keras DepthwiseConv2D

Issue - State: closed - Opened by dr-costas 10 months ago - 5 comments

#32 - FIX #31: failed to conversion a convolution when padding argument is 'valid' or default

Pull Request - State: closed - Opened by kokeshing 10 months ago - 1 comment

#31 - Failed to conversion a convolution when padding argument is 'valid' or default

Issue - State: closed - Opened by kokeshing 10 months ago - 2 comments

#30 - nn.Conv2d and F.conv2d with groups == input_channels (DepthWise) generates PartitionedCall in tensorflow frozen_graph

Issue - State: closed - Opened by pedrofrodenas 11 months ago - 1 comment

#29 - Troubleshooting Evaluation and Compilation Issues with Keras_yolov5s.h5 Conversion

Issue - State: open - Opened by bspark7 11 months ago

#28 - raise Exception(f'Failed conversion: {self.original_node}')

Issue - State: closed - Opened by mr-sarthakgupta 11 months ago - 2 comments

#27 - ModuleNotFoundError: No module named 'keras.src.engine'

Issue - State: open - Opened by ivanstepanovftw 11 months ago - 7 comments
Labels: Keras 3

#26 - Parameter count mismatch when converting `nn.TransformerEncoderLayer`

Issue - State: closed - Opened by till-m 12 months ago - 6 comments
Labels: bug

#25 - Convolution layers: Add support for `same` padding.

Issue - State: closed - Opened by till-m 12 months ago - 2 comments

#24 - Debugging tensor shapes when using dynamic axes

Issue - State: open - Opened by BenCrulis 12 months ago - 8 comments
Labels: enhancement

#23 - Custom Softplus Layer and Einsum Error when load_model with Keras

Issue - State: closed - Opened by wredan 12 months ago - 7 comments

#22 - Keras symbolic inputs/outputs do not implement `len`

Issue - State: closed - Opened by federicoparra about 1 year ago - 8 comments

#21 - Imprecise conversion for custom log_softmax converter

Issue - State: closed - Opened by coxep about 1 year ago - 4 comments

#20 - I am getting 'Unimplemented nodes' exception

Issue - State: closed - Opened by kiyanisarah about 1 year ago - 5 comments

#19 - Tensorflow warns that variables were used in Lambda layers but are not present in tracked objects

Issue - State: closed - Opened by tcwalther about 1 year ago - 2 comments

#18 - unnecessary python 3.9 limitation

Issue - State: closed - Opened by Heatdh about 1 year ago - 1 comment

#17 - Add support for torch.repeat_interleave

Issue - State: closed - Opened by stakemura about 1 year ago - 2 comments

#16 - How to llama 2 conversion

Issue - State: open - Opened by yige0617 over 1 year ago - 1 comment
Labels: TODO

#15 - LogSoftMax conversion

Issue - State: closed - Opened by thunder3057 over 1 year ago - 1 comment

#14 - Much slower training after convert torch code to tf code

Issue - State: closed - Opened by chenghuige over 1 year ago - 1 comment

#13 - Custom initializer for tf weights

Issue - State: closed - Opened by d-a-yakovlev over 1 year ago - 6 comments

#12 - TensorFlow 2 PyTorch?

Issue - State: closed - Opened by innat over 1 year ago - 7 comments

#11 - Add support for multidimensional aggregating function

Issue - State: closed - Opened by alievrusik over 1 year ago - 1 comment

#10 - Added nn.SiLU activation

Pull Request - State: closed - Opened by ffs333 over 1 year ago

#9 - Added nn.GELU activation

Pull Request - State: closed - Opened by ffs333 over 1 year ago - 3 comments

#8 - pytorch version?

Issue - State: closed - Opened by mychina75 almost 2 years ago - 1 comment

#7 - new implementation of prelu

Pull Request - State: closed - Opened by On-JungWoan almost 2 years ago - 1 comment

#6 - Thank you for releasing this wonderful repository.

Issue - State: closed - Opened by PINTO0309 almost 2 years ago - 2 comments

#5 - add sin, cos ops

Pull Request - State: closed - Opened by On-JungWoan almost 2 years ago - 7 comments

GitHub / AlexanderLutsenko/nobuco issues and pull requests