explosion/curated-transformers issues and pull requests

#98 - Fix import in quantize-transformer CLI

Pull Request - State: closed - Opened by danieldk almost 2 years ago
Labels: type/bug

#97 - Set version to 0.0.3

Pull Request - State: closed - Opened by danieldk almost 2 years ago

#96 - Rename `models.transformer_model` to `models.architectures`

Pull Request - State: closed - Opened by danieldk almost 2 years ago

#95 - Add `init.py` to the `cli` module

Pull Request - State: closed - Opened by danieldk almost 2 years ago

#94 - Prefix encoders with `curated_encoder`

Pull Request - State: closed - Opened by danieldk almost 2 years ago
Labels: type/feature, feat/model

#93 - Public API fixes

Pull Request - State: closed - Opened by shadeMe almost 2 years ago

#92 - Move listeners and transformer pipe

Pull Request - State: closed - Opened by danieldk almost 2 years ago

#91 - Another attempt to include test files

Pull Request - State: closed - Opened by danieldk almost 2 years ago

#90 - Include package files (such as test data)

Pull Request - State: closed - Opened by danieldk almost 2 years ago

#89 - Bump version to 0.0.2

Pull Request - State: closed - Opened by danieldk almost 2 years ago

#88 - Bump version to 0.0.2

Pull Request - State: closed - Opened by danieldk almost 2 years ago - 3 comments

#87 - Increase spacy lowerbound to `v4.0.0.dev0 `

Pull Request - State: closed - Opened by shadeMe almost 2 years ago

#86 - Fix MyPy errors and run in CI

Pull Request - State: closed - Opened by danieldk almost 2 years ago
Labels: bug

#85 - Batch the spans that are passed to the torch transformer model

Pull Request - State: closed - Opened by shadeMe almost 2 years ago
Labels: enhancement

#84 - Distillation: Make `TransformerDistiller` a no-op in regular pipelines

Pull Request - State: closed - Opened by shadeMe almost 2 years ago - 2 comments
Labels: bug

#83 - Distillation: Add `mean` and `squared_l2_norm` normalization for hidden layers

Pull Request - State: open - Opened by shadeMe almost 2 years ago - 1 comment
Labels: enhancement

#82 - Add curated-transformers.TransformerLayersListener.v1

Pull Request - State: closed - Opened by danieldk almost 2 years ago

#81 - `ScalarWeightListener`: Batch documents in forward and backward passes

Pull Request - State: closed - Opened by shadeMe almost 2 years ago

#80 - `ScalarWeightListener`: Batch documents in forward and backward passes

Pull Request - State: closed - Opened by shadeMe almost 2 years ago
Labels: enhancement

#79 - Increase GPU throughput during gradient accumulation

Pull Request - State: closed - Opened by shadeMe almost 2 years ago
Labels: enhancement

#78 - Add pull request template

Pull Request - State: closed - Opened by shadeMe almost 2 years ago

#77 - Speed up pooling by concatenating all transformer outputs

Pull Request - State: closed - Opened by danieldk almost 2 years ago - 1 comment
Labels: enhancement

#76 - Replace TorchScript API with Thinc calls

Pull Request - State: closed - Opened by shadeMe almost 2 years ago

#75 - Normalize identifiers that use `size` and `width` interchangeably

Pull Request - State: closed - Opened by shadeMe almost 2 years ago

#74 - Azure CI pipelines with huggingface model caching

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

#73 - Add `mypy` to requirements

Pull Request - State: closed - Opened by shadeMe about 2 years ago

#72 - Replace GitHub actions CI by Azure pipelines

Pull Request - State: closed - Opened by danieldk about 2 years ago - 3 comments

#71 - Update GitHub URLs to Explosion org

Pull Request - State: closed - Opened by danieldk about 2 years ago

#70 - update readme to point to explosion

Pull Request - State: closed - Opened by svlandeg about 2 years ago - 2 comments

#69 - Transformer: do not update when frozen

Pull Request - State: closed - Opened by danieldk about 2 years ago

#68 - README: update list of supported model types

Pull Request - State: closed - Opened by danieldk about 2 years ago

#67 - Add doc strings to public entry points and their associated classes

Pull Request - State: closed - Opened by shadeMe about 2 years ago

#66 - Fix slow tests

Pull Request - State: closed - Opened by danieldk about 2 years ago

#65 - Move all Torch modules to a subpackage

Pull Request - State: closed - Opened by danieldk about 2 years ago

#64 - Remove duplicate code

Pull Request - State: closed - Opened by shadeMe about 2 years ago

#63 - Clarify layer weighting listener error message

Pull Request - State: closed - Opened by shadeMe about 2 years ago

#62 - Add layer weighting `spacy project` config

Pull Request - State: closed - Opened by shadeMe about 2 years ago - 2 comments

#61 - `mypy` fixes

Pull Request - State: closed - Opened by shadeMe about 2 years ago

#60 - Move embedding projection to BertEmbeddings

Pull Request - State: closed - Opened by danieldk about 2 years ago

#59 - Add error stubs for listener replacement functionality

Pull Request - State: closed - Opened by shadeMe about 2 years ago

#58 - Remove 🤗

Pull Request - State: closed - Opened by shadeMe about 2 years ago

#57 - Compressor app

Pull Request - State: closed - Opened by kadarakos about 2 years ago - 1 comment

#56 - Remove unused `hf_model_name`

Pull Request - State: closed - Opened by danieldk about 2 years ago

#55 - Implement BERT layer using `torch.nn.TransformerEncoderLayer`

Pull Request - State: closed - Opened by danieldk about 2 years ago - 1 comment

#54 - Add missing type annotations

Pull Request - State: closed - Opened by shadeMe about 2 years ago

#53 - Move listeners to their own module

Pull Request - State: closed - Opened by shadeMe about 2 years ago

#52 - `Transformer`: `all_layer_outputs` defaults to `False`

Pull Request - State: closed - Opened by shadeMe about 2 years ago

#49 - Add `gradual_transformer_unfreezing` callback

Pull Request - State: closed - Opened by shadeMe about 2 years ago - 2 comments

#48 - Attention mask handling improvements

Pull Request - State: closed - Opened by danieldk about 2 years ago

#43 - Add `build_camembert_transformer_model_v1` for CamemBERT

Pull Request - State: closed - Opened by danieldk about 2 years ago

GitHub / explosion/curated-transformers issues and pull requests