microsoft/Olive

Issues

• Early stopping in Hugging Face models

  #859 opened 8 months ago
  7

• assert conversion_footprint and optimizer_footprint with Stable Diffusion conversion

  #855 opened a year ago
  8

• Is this pass flow possible for Stable Diffusion?: OrtTransformersOptimization → IncDynamicQuantization or IncStaticQuantization

  #852 opened a year ago
  46

• ImportError: cannot import name 'load_model' from 'onnxruntime.quantization.quant_utils'

  #849 opened a year ago
  13

• Validation errors for HfConfig

  #847 opened a year ago
  5

• AssertionError - assert conversion_footprint and optimizer_footprint when converting a model with size>512

  #842 opened a year ago
  2

• [Bug]: stable diffusion performance degrade

  #834 opened a year ago
  2

• [Bug]: Merged ONNX model exceeds 2GB, the model will not be checked without `save_path` given.

  #831 opened 8 months ago
  5

• Converted Whisper do not way to return timestamps

  #829 opened a year ago
  5

• Unable to convert openai/whisper-large-v3

  #823 opened a year ago
  3

• Whisper does not converted using onnxruntime-directml

  #813 opened a year ago
  3

• No safety checker in SDXL

  #804 opened a year ago
  3

• Optimized model slower than original one on CUDAExecutionProvider

  #787 opened a year ago
  12

• [FR]: Pass func_kwargs to _evaluate_custom

  #783 opened a year ago
  1

• [Bug]: AssertionError: No valid accelerator specified for target system.

  #782 opened a year ago
  4

• [Bug]: local_resource_path.save_to_dir(save_dir, save_name, overwrite) failed

  #777 opened a year ago
  9

• How do I create a SDXL Olive model that is optimized for various Execution providers - DirectML, CPU, GPU

  #776 opened a year ago
  8

• [Bug]: assert conversion_footprint and optimizer_footprint for LCM SDXL model

  #772 opened a year ago
  13

• Olive requires an old version of torchmetrics ~= 0.10.0

  #768 opened a year ago
  2

• [Bug]: Black images generated with SDXL olive model when using Image-to-Image. Text to Image works fine!

  #764 opened a year ago
  11

• [Bug]: Stable Diffusion conversion - TypeError: 'NoneType' object is not callable

  #757 opened a year ago
  1

• [Bug]: Converting Facebook MusicGen model to Onnx - RuntimeError: Expected tensor for argument #1 'indices' to have one of the following scalar types: Long, Int; but got torch.FloatTensor instead

  #756 opened a year ago
  6

• [Bug]: whisper-medium process convertation finished with "Failed to run Olive on cpu-cpu: Error parsing message"

  #755 opened a year ago
  6

• [Bug]: ModuleNotFoundError: No module named 'olive'

  #754 opened a year ago
  10

• Whisper prompting support

  #753 opened 9 months ago
  7

• [Bug]: Nodes in a graph must be topologically sorted

  #740 opened a year ago
  11

• [Bug]: Whisper optimization example: ModuleNotFoundError: No module named 'olive.workflows'

  #737 opened a year ago
  2

• [Bug]: MIT/ast-finetuned-speech-commands-v2 leads to KeyError in evaluation?

  #735 opened 8 months ago
  6

• [Bug]: Error Node (BeamSearch_node) has input size 12 not in range [min=5, max=10]

  #734 opened a year ago
  4

• [Bug]: During Inference with ORTStableDiffusionXLPipeline - [ONNXRuntimeError] : 9 : NOT_IMPLEMENTED : Could not find an implementation for GroupNorm(1) node with name 'GroupNorm_22'

  #733 opened a year ago
  8

• [FR]: Support Stable Diffusion models with LCM

  #730 opened a year ago
  7

• [FR]: Hope Olive supports Pydantic version 2.0 or higher

  #709 opened a year ago
  4

• Add ONNXRuntimePackages builds for Java

  #697 opened a year ago
  5

• [FR]: Any roadmap for arm architecture

  #687 opened a year ago
  1

• [Bug]: TypeError: __init__() got an unexpected keyword argument 'io_config'

  #686 opened a year ago
  9

• [FR]: Whisper int4 support

  #678 opened 8 months ago
  15

• [FR]: About path.

  #675 opened a year ago
  2

• [Bug]: Running Olive with ROCMExecutionProvider.

  #667 opened a year ago
  8

• CUDAExecutionProvider is not in the session providers

  #656 opened a year ago
  3

• [Bug]: ImportError: cannot import name 'load_model' from ' 6800xt model

  #652 opened a year ago
  2

• [FR]: Enable optimization by onnxruntime on stable_diffusion.py

  #645 opened a year ago
  9

• [FR]: quantize a fine-tuned model

  #640 opened a year ago
  7

• [Bug]: Vitis quantization doesn't work with ORT 1.16.

  #629 opened a year ago
  0

• model_type field required (type=value_error.missing)

  #628 opened a year ago
  5

• [Bug]: Vitis AI quantizer with symmetric uint8 activation wrongfully hard-codes `zero_point=0`

  #623 opened 8 months ago
  6

• [Bug]: `ActivationSymmetric` default is not enforced in Vitis AI quantizer

  #622 opened a year ago
  6

• [Bug]: ort-nightly-gpu not installing

  #611 opened a year ago
  4

• [Question] Whisper GPU mobile NNAPI

  #578 opened a year ago
  1

• [Bug]: No model is selected. Skip packaging output artifacts. whisper

  #575 opened a year ago
  8

• [Bug]: OnnxQuantization

  #573 opened a year ago
  8