openvinotoolkit/nncf

Issues

• untimeError: CUDA is available for PyTorch, but NNcF could not compile GPu quan tization extensions.

  #3033 opened a month ago by huangqiu15444
  3

• PyTorch 2.5.0

  #3036 opened a month ago by alexsu52
  0

• [TorchFX] Torch FX/PyTorch 2 Export Quantization

  #2766 opened 5 months ago by alexsu52
  4

• Object Detection Quantization fails with OpenVINO 2024.3

  #2889 opened 3 months ago by fmishra747
  2

• How to use custom dataloder in QAT?

  #2982 opened a month ago by huangqiu15444
  5

• [Good First Issue][NNCF]: Dump actual_subset_size to ov.Model

  #2562 opened a month ago by l-bat
  16

• [Good First Issue][NNCF][FBC/BC] Add a case with depthwise/transpose convolutions to the template test

  #2916 opened a month ago by daniil-lyakhov
  15

• [Torch FX] INT4 data-free weights compression

  #3005 opened a month ago by alexsu52
  0

• [Good First Issue][NNCF]: [TorchFX] Test MinMax algorithm

  #2872 opened 2 months ago by daniil-lyakhov
  10

• The question about function create_compressed_model()：RuntimeError: CUDA error: device-side assert triggered

  #2688 opened 2 months ago by zbnlala
  5

• [TorchFX] Add edges shape to `test_models.py`

  #2904 opened 2 months ago by daniil-lyakhov
  0

• [TorchFX] Support NNCFGraph.is_shared method

  #2938 opened 3 months ago by daniil-lyakhov
  1

• KeyError: 'bf16'

  #2980 opened 2 months ago by hezhiqian01
  5

• [Good First Issue][NNCF][TorchFX]: Test model transformer

  #2775 opened 2 months ago by daniil-lyakhov
  16

• [Good First Issue][NNCF]: [TorchFX] Test PTQ MinMax parameters

  #2873 opened 2 months ago by daniil-lyakhov
  15

• [Torch FX] Support for ExportedProgram from torch.export.export in Weights Compression

  #2987 opened 2 months ago by anzr299
  1

• [Good First Issue][NNCF]: [TorchFX] Test PTQ quantizer config

  #2874 opened 2 months ago by daniil-lyakhov
  8

• Type mismatch in onnx quantize_with_accuracy_control validation_fn

  #2957 opened 2 months ago by w4-jinhyeonkim
  4

• NNCF can not compile model: Default output not supported (opset1::VariadicSplit)

  #2943 opened 2 months ago by cansik
  12

• [TorchFX] Use FakeTensors in meta instead of actual tensors

  #2934 opened 3 months ago by daniil-lyakhov
  0

• Accuracy Aware Quantization- Object Segmentation fails for OpenVINO 2024.3 and ultralytics_8.2.74 with yolov8-seg model and coco128-seg dataset

  #2907 opened 3 months ago by fmishra747
  2

• [TorchFX][Optimization] Run eliminate dead code once for constant correction commands

  #2929 opened 3 months ago by daniil-lyakhov
  0

• [TorchFX] Stop using PT commands and migrate to FX commands

  #2928 opened 3 months ago by daniil-lyakhov
  0

• [TorchFX] Use node names vs node map in the Model Transformer

  #2906 opened 3 months ago by daniil-lyakhov
  0

• Failed to export NNCF QAT model for Ultralytics YOLOv8

  #2846 opened 4 months ago by quinnZE
  5

• [Good First Issue][NNCF][TorchFX]: Test statistic aggregator

  #2777 opened 4 months ago by daniil-lyakhov
  5

• [TorchFX] TorchFX metatypes support

  #2778 opened 4 months ago by daniil-lyakhov
  2

• [NNCF][TorchFX]: Test Fast Bias Correction algorithm

  #2809 opened 4 months ago by daniil-lyakhov
  2

• nncf + ultralytics yolov8 training-time compression

  #2486 opened 9 months ago by SofyaLL
  10

• test_autoq_precision_init Fails due to connection timeout

  #2843 opened 4 months ago by anzr299
  1

• DeformConv2d Cannot be Quantized

  #2794 opened 5 months ago by anzr299
  1

• When will there be a c++ interface?

  #2753 opened 5 months ago by qq978358810
  2

• ov.compile_model(quantized_model, device_name="GPU") is not working

  #2773 opened 5 months ago by paguilomanas
  5

• Very inconsistent quantization results on Tensorflow

  #2743 opened 5 months ago by apoorvu-sharechat
  10

• [DISCUSSION]: Metrics for NNCF

  #2763 opened 5 months ago by AbhijithGanesh
  1

• IndexError: list index out of range When I try to quantize llama models using OVQuantizer

  #2755 opened 5 months ago by Alwahsh
  2

• AttributeError: 'list' object has no attribute 'keys' when executing yolov8_quantize_with_accuracy_control example

  #2669 opened 6 months ago by openvino-book
  4

• [Good First Issue][NNCF]: Remove compress_to_fp16=False from examples

  #2668 opened 6 months ago by alexsu52
  3

• [Good First Issue][NNCF]: fix invalid error reporting in JSON schema

  #2570 opened 8 months ago by ljaljushkin
  19

• [Good First Issue][NNCF]: dump the ignored scope more gracefully

  #2564 opened 6 months ago by ljaljushkin
  4

• [Good First Issue][NNCF]: Add INT8 weight compression conformance test for Tinyllama-1.1b PyTorch model

  #2527 opened 7 months ago by alexsu52
  19

• [Good First Issue][NNCF]: Optimize memory footprint by removing redundant collected statistics

  #2557 opened 8 months ago by kshpv
  8

• Ultralytics yolov8 QAT example

  #2489 opened 9 months ago by alexsu52
  1

• PTQ of Fast R-CNN crashes in PyTorch backend

  #2568 opened 8 months ago by alexsu52
  1

• [Good First Issue][NNCF]: check number of u8, u4 constants in weight compression tests

  #2565 opened 7 months ago by ljaljushkin
  10

• Why doesn't the size and precision of the model change after INT4 quantization?

  #2553 opened 8 months ago by xingfenghaizeiwang
  2

• [Good First Issue][NNCF]: Add tests for torch device utils

  #2579 opened 8 months ago by daniil-lyakhov
  5

• [Good First Issue][NNCF]: Fixing NNCFGraph export for visualization in Netron

  #2552 opened 8 months ago by andrey-churkin
  6

• Thanks to our Contributors

  #2499 opened 9 months ago by alexsu52
  1

• Compressed models that call torch.is_floating_point() during inference are traced with runtime error.

  #2432 opened 10 months ago by alexsu52
  0