Releases: chengchingwen/Transformers.jl
Releases · chengchingwen/Transformers.jl
v0.2.0
Transformers v0.2.0
Closed issues:
- Redesign
AbstractHGFConfig
loading with JSON3.jl (#10) - About choice of Tokenizers (#14)
- Creating a new Stacks.jl library (#63)
- Support for HuggingFace bert-base-portuguese-cased (#78)
- scibert models missing loading_method? (#91)
- Transformer tutorial is outdated (#98)
- Transformers.jl failing to load in Julia 1.8 on Mac M1 (#122)
- Bug(?) while displaying a huggingface model (#123)
Merged pull requests:
- 0.2 (#126) (@chengchingwen)
v0.1.25
v0.1.24
v0.1.23
Transformers v0.1.23
Merged pull requests:
- Handle post-processor from fast tokenizer and add some tokenizers (#115) (@chengchingwen)
v0.1.22
Transformers v0.1.22
Closed issues:
- Tokenizer not support yet | Example GPT2_TextGeneration (#112)
Merged pull requests:
- Refactor hgf tokenizer code (#114) (@chengchingwen)
v0.1.21
Transformers v0.1.21
Closed issues:
- MbedTLS error code -76: NET - Reading information from the socket failed (#107)
- Evaluation into the closed module
Transformers
breaks incremental compilation (enable_gpu) (#108)
Merged pull requests:
- gpt / gpt2 textencoder and hgf gpt2 tokenizer (#111) (@chengchingwen)
v0.1.20
v0.1.19
Transformers v0.1.19
Closed issues:
- BERT CoLA Example Question (#102)
Merged pull requests:
- Adapting Huggingfaceapi (#103) (@chengchingwen)
- Use Base.libblas_name instead of LinearAlgebra.BLAS.libblas. (#104) (@jwkvam)
v0.1.18
Transformers v0.1.18
Closed issues:
Merged pull requests:
- addresses issue #92 (#93) (@Alexander-Barth)