Embeddings from BERT for original tokens #148
Comments
|
Hi, you should read the discussion in #64. I left this issue open for reference on these questions. |
ocavue
pushed a commit
to ocavue/transformers
that referenced
this issue
Sep 13, 2023
…loses huggingface#148) (huggingface#227) * Define custom CLIP ONNX configs * Update conversion script * Support specifying custom model file name * Use int64 for CLIP input ids * Add support for CLIP text and vision models * Fix JSDoc * Add docs for `CLIPTextModelWithProjection` * Add docs for `CLIPVisionModelWithProjection` * Add unit test for CLIP text models * Add unit test for CLIP vision models * Set resize precision to 3 decimal places * Fix `RawImage.save()` function * Throw error when reading image and status != 200 * Create basic semantic image search application * Separate out components * Add `update-database` script * Update transformers.js version
ZYC-ModelCloud
pushed a commit
to ZYC-ModelCloud/transformers
that referenced
this issue
Nov 14, 2024
* Update run_tests.yml * Update ip * clear cache * fix test not run * fix no tests * Update run_tests.yml
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I am trying out the
extract_features.pyexample program. I noticed that a sentence gets split into tokens and the embeddings are generated. For example, if you had the sentence “Definitely not”, and the corresponding workpieces can be [“Def”, “##in”, “##ite”, “##ly”, “not”]. It then generates the embeddings for these tokens.My question is how do I train an NER system on CoNLL dataset?
I want to extract embeddings for original tokens for training an NER with a neural architecture. If you have come across any resource that gives a clear explanation on how to carry this out, post it here.
The text was updated successfully, but these errors were encountered: