unilm | scale Self-supervised Pre | Natural Language Processing library

by microsoft Python Version: s2s-ft.v0.3 License: MIT

X-Ray Key Features Code Snippets(3)Community Discussions(1)Vulnerabilities Install Support

kandi X-RAY | unilm Summary

unilm is a Python library typically used in Artificial Intelligence, Natural Language Processing, Deep Learning, Pytorch, Bert, Transformer applications. unilm has no bugs, it has no vulnerabilities, it has a Permissive License and it has medium support. However unilm build file is not available. You can download it from GitHub.

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

Support

Quality

Security

License

Reuse

Support

unilm has a medium active ecosystem.

It has 12771 star(s) with 1868 fork(s). There are 260 watchers for this library.

It had no major release in the last 12 months.

There are 337 open issues and 650 have been closed. On average issues are closed in 7 days. There are 27 open pull requests and 0 closed requests.

It has a neutral sentiment in the developer community.

The latest version of unilm is s2s-ft.v0.3

Quality

unilm has 0 bugs and 0 code smells.

Security

unilm has no vulnerabilities reported, and its dependent libraries have no vulnerabilities reported.

unilm code analysis shows 0 unresolved vulnerabilities.

There are 0 security hotspots that need review.

License

unilm is licensed under the MIT License. This license is Permissive.

Permissive licenses have the least restrictions, and you can use them in most projects.

Reuse

unilm releases are available to install and integrate.

unilm has no build file. You will be need to create the build yourself to build the component from source.

Top functions reviewed by kandi - BETA

kandi has reviewed unilm and discovered the below as its top functions. This is intended to give you an instant insight into unilm implemented functionality, and help decide if they suit your requirements.

Generate a sequence generator .
Converts the examples to a dataset .
Performs a beam search .
Generates a sequence of sequences from the model .
Generate nbest and re - processes .
Computes predictions for all the training features .
Write predictions .
Prepare the model for decoding .
Write out predictions to be extended .
Computes the predictions for the predictions .

Get all kandi verified functions for this library.

unilm Key Features

No Key Features are available at this moment for unilm.

unilm Examples and Code Snippets

UniLM模型放在了另一个仓库：https://github.com/zhangwanyu2020/UniLM

Python

Lines of Code : 5

License : No License

Copy

 （2）将文本转换成bert所需的数据格式：tokens\segment_id\mask_id
 
 （3）download bert预训练模型，加载到我的模型中
 
 （4）在bert模型后面，加一个简单的全连接层，进行预测输出

Pretrained-Unilm-Chinese,Usage,pretrain

Python

Lines of Code : 5

License : No License

Copy

python run_pretraining_google.py \
  --bert_config_file=$BERT_BASE_DIR/bert_config.json \
  --init_checkpoint=$BERT_BASE_DIR/bert_model.ckpt \
  --input_file=$DATA_BASE_DIR/wiki_sent_pair.txt \
  --output_dir=$OUT_PUT_BASE_DIR/checkpoint

Pretrained-Unilm-Chinese,Usage,fine-tune

Python

Lines of Code : 1

License : No License

Copy

python task_summary.py

Community Discussions

Trending Discussions on unilm

torch.nn.CrossEntropyLoss().ignore_index is crashing when importing transfomers library

QUESTION

torch.nn.CrossEntropyLoss().ignore_index is crashing when importing transfomers library

Asked 2021-Jan-28 at 09:25

I am using layoutlm github which require python 3.6, transformer 2.9.0. I created an conda env:

...

ANSWER

Answered 2021-Jan-28 at 09:25

It seems something was broken on layoutlm with pytorch 1.4 related issue. Switching to pytorch 1.6 fix the issue with the core dump, and the layoutlm code run without any modification.

Source https://stackoverflow.com/questions/65582498

Community Discussions, Code Snippets contain sources that include Stack Exchange Network

Vulnerabilities

No vulnerabilities reported

Install unilm

You can download it from GitHub.
You can use unilm like any standard Python library. You will need to make sure that you have a development environment consisting of a Python distribution including header files, a compiler, pip, and git installed. Make sure that your pip, setuptools, and wheel are up to date. When using pip it is generally recommended to install packages in a virtual environment to avoid changes to the system.

Support

For help or issues using the pre-trained models, please submit a GitHub issue. For other communications, please contact Furu Wei (fuwei@microsoft.com).

Find more information at: