AlexanderLutsenko/nobuco

Issues

• The Possibility to use older version of Tensorflow such as 2.7

  #69 opened 2 months ago by Fatih-Kiralioglu
  0

• Quantization support?

  #68 opened 2 months ago by happyTonakai
  2

• tensorflowjs woes with saved keras file

  #67 opened 3 months ago by johndpope
  4

• Unimplemented Nodes: roi_align

  #63 opened 4 months ago by hanspres1999
  3

• multihead attention - no converter found

  #65 opened 3 months ago by johndpope
  3

• ModuleNotFoundError: No module named 'keras.src.engine'

  #27 opened 10 months ago by ivanstepanovftw
  7

• TensorFlow 2 PyTorch?

  #12 opened a year ago by innat
  7

• Exception: Unimplemented nodes

  #62 opened 5 months ago by kabyanil
  1

• Convert pretrained .torch model (including weights)

  #34 opened 9 months ago by CarlosNacher
  4

• Is the MultiheadAttention converter outdated?

  #46 opened 8 months ago by crimson206
  3

• speed issue after conversion

  #54 opened 6 months ago by thegodone
  7

• When to worry about Imprecise conversion / Max diff

  #60 opened 5 months ago by chadrockey
  2

• Extremely High GPU VRAM Usage During Conversion

  #61 opened 5 months ago by chadrockey
  2

• Stateful model

  #55 opened 6 months ago by johan-sightic
  2

• issue in tf.equal converter_eq

  #53 opened 6 months ago by thegodone
  2

• Signature output names

  #51 opened 6 months ago by johan-sightic
  3

• FFT/STFT support

  #50 opened 7 months ago by happyTonakai
  1

• Force static shapes through the graph?

  #48 opened 7 months ago by tonystratum
  9

• Modes for torch.nn.functional.pad and tf.pad

  #49 opened 7 months ago by happyTonakai
  2

• Preserve node naming/namespacing

  #47 opened 8 months ago by matemijolovic
  4

• Conversion fails randomly on models with multiple inputs

  #37 opened 8 months ago by mmjamm
  7

• raise Exception(f'Failed conversion: {self.original_node}')

  #28 opened 8 months ago by mr-sarthakgupta
  2

• Identity layer is considered as 'Unimplemented nodes'

  #38 opened 8 months ago by crimson206
  1

• ValueError from adding tf.Tensor and None when exporting model with dynamic shapes

  #36 opened 8 months ago by NathanielB123
  0

• When groups equal input channels then PyTorch Conv2D is mapped to Keras DepthwiseConv2D

  #33 opened 9 months ago by dr-costas
  5

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

  #30 opened 9 months ago by pedrofrodenas
  1

• Failed to conversion a convolution when padding argument is 'valid' or default

  #31 opened 9 months ago by kokeshing
  2

• Troubleshooting Evaluation and Compilation Issues with Keras_yolov5s.h5 Conversion

  #29 opened 9 months ago by bspark7
  0

• I am getting 'Unimplemented nodes' exception

  #20 opened 10 months ago by kiyanisarah
  5

• Debugging tensor shapes when using dynamic axes

  #24 opened a year ago by BenCrulis
  8

• Parameter count mismatch when converting `nn.TransformerEncoderLayer`

  #26 opened 10 months ago by till-m
  6

• Convolution layers: Add support for `same` padding.

  #25 opened 10 months ago by till-m
  2

• Custom Softplus Layer and Einsum Error when load_model with Keras

  #23 opened a year ago by wredan
  7

• Keras symbolic inputs/outputs do not implement `__len__`

  #22 opened a year ago by federicoparra
  8

• Imprecise conversion for custom log_softmax converter

  #21 opened a year ago by coxep
  4

• Tensorflow warns that variables were used in Lambda layers but are not present in tracked objects

  #19 opened a year ago by tcwalther
  2

• unnecessary python 3.9 limitation

  #18 opened a year ago by Heatdh
  1

• Add support for torch.repeat_interleave

  #17 opened a year ago by stakemura
  2

• How to llama 2 conversion

  #16 opened a year ago by yige0617
  1

• LogSoftMax conversion

  #15 opened a year ago by thunder3057
  1

• Custom initializer for tf weights

  #13 opened a year ago by d-a-yakovlev
  6

• Much slower training after convert torch code to tf code

  #14 opened a year ago by chenghuige
  1

• Add support for multidimensional aggregating function

  #11 opened a year ago by alievrusik
  1

• pytorch version?

  #8 opened 2 years ago by mychina75
  1

• Thank you for releasing this wonderful repository.

  #6 opened 2 years ago by PINTO0309
  2

• [Question] Are dynamic axes supported?

  #3 opened 2 years ago by dazzle-me
  2

• TypeError: converter_mean() got an unexpected keyword argument 'keepdims'

  #4 opened 2 years ago by hengck23
  2

• Depthwise convolution conversion fails?

  #2 opened 2 years ago by dazzle-me
  1