transformers
Run model as compressed/uncompressed mode
#34719
Merged
Overview
Commits
37
Changes
View On GitHub
Commits
  • draft, run model as compreszed/uncompressed mode
    George committed 1 year ago
  • draft
    George committed 1 year ago
  • run run_compressed=False
    George committed 1 year ago
  • Merge branch 'main' into compressed-tensors/run_compressed
    GitHub committed 1 year ago
  • run_compressed as attr
    George committed 1 year ago
  • Merge branch 'compressed-tensors/run_compressed' of github.com:neuralmagic/upstream-transformers into compressed-tensors/run_compressed
    George committed 1 year ago
  • set run_compressed=False using quantization_config
    George committed 1 year ago
  • remove redundant line
    George committed 1 year ago
  • make is_qat_trainable dependent on run_compressed status
    George committed 1 year ago
  • add tests
    George Ohashi committed 1 year ago
  • Merge branch 'main' into compressed-tensors/run_compressed
    GitHub committed 1 year ago
  • lint
    George Ohashi committed 1 year ago
  • Merge branch 'compressed-tensors/run_compressed' of github.com:neuralmagic/upstream-transformers into compressed-tensors/run_compressed
    George Ohashi committed 1 year ago
  • full in docstring
    George Ohashi committed 1 year ago
  • add decompress
    George Ohashi committed 1 year ago
  • comments
    George Ohashi committed 1 year ago
  • Merge branch 'main' into compressed-tensors/run_compressed
    GitHub committed 1 year ago
  • decompress if model is compresssed and not run_compressed
    George Ohashi committed 1 year ago
  • Merge branch 'main' into compressed-tensors/run_compressed
    GitHub committed 1 year ago
  • apply_quant_config logic fix -- populate statedict properly
    George Ohashi committed 1 year ago
  • comments
    George Ohashi committed 1 year ago
  • remove non compressed model
    George Ohashi committed 1 year ago
  • make is_compressed as property
    George Ohashi committed 1 year ago
  • cosmetic
    George Ohashi committed 1 year ago
  • Merge branch 'main' into compressed-tensors/run_compressed
    GitHub committed 1 year ago
  • Merge branch 'main' into compressed-tensors/run_compressed
    GitHub committed 1 year ago
  • run apply_quant_config for non-compressed models -- popualte scales and zeropoints
    George Ohashi committed 1 year ago
  • Merge branch 'compressed-tensors/run_compressed' of github.com:neuralmagic/upstream-transformers into compressed-tensors/run_compressed
    George Ohashi committed 1 year ago
  • add pahtway for decompressing sparse models
    George Ohashi committed 1 year ago
  • typo on is_quantization_compressed
    George Ohashi committed 1 year ago
  • Merge branch 'main' into compressed-tensors/run_compressed
    GitHub committed 1 year ago
  • lint
    George Ohashi committed 1 year ago
  • Merge branch 'compressed-tensors/run_compressed' of github.com:neuralmagic/upstream-transformers into compressed-tensors/run_compressed
    George Ohashi committed 1 year ago
  • fix typo
    George Ohashi committed 1 year ago
  • Merge branch 'main' into compressed-tensors/run_compressed
    GitHub committed 1 year ago
  • Merge branch 'main' into compressed-tensors/run_compressed
    GitHub committed 1 year ago
  • Merge branch 'main' into compressed-tensors/run_compressed
    GitHub committed 1 year ago
FAQ Terms Privacy Refunds Impressum
Loading