Easy to use BiLSTM+CRF sequence tagging for text.
Project description
This repo is a fork of https://github.com/Hironsan/anago
Updated with support for tf 2.0 and newer keras versions.
This repo is a fork of https://github.com/Hironsan/anago and is modified to suit my needs. The module's name is changed so that it doesn't conflict with the installation of the original repo and so that I can push my changes to PyPi.
Unless you know why are you using this, don't use this module. Use the excellent https://github.com/Hironsan/anago
The purpose of creating a fork (and a separate module) instead of using the excellent anago is as follows.
- Maintain compatibility with newer tf and keras versions.
- Anago's installation comes with ELMo, allennlp and some other stuff, which are not needed for the intended use cases of this repo (mainly DeepSegment).
- It's not easy to include a git repo link as a requirement in setup.py while publishing to PyPi. So, this module is renamed to seqtag-keras.
**Full credit and thanks for the original implementation of this repo goes to https://github.com/Hironsan/ and https://github.com/chakki-works/ . **
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file seqtag_keras-1.0.6.tar.gz
.
File metadata
- Download URL: seqtag_keras-1.0.6.tar.gz
- Upload date:
- Size: 18.8 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.22.0 setuptools/49.6.0 requests-toolbelt/0.9.1 tqdm/4.40.0 CPython/3.6.8
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 8a52766f288c206788cc0c5f22dfdedf523995ce86c4c0afa0ba508e2fe26f34 |
|
MD5 | fdcdb71ce719e5ffaca93433842d2cdf |
|
BLAKE2b-256 | db743d33bbe04671a5adc8b4d29bd3b81c624903f2d77cb03c13381d2f0e171c |
File details
Details for the file seqtag_keras-1.0.6-py2.py3-none-any.whl
.
File metadata
- Download URL: seqtag_keras-1.0.6-py2.py3-none-any.whl
- Upload date:
- Size: 21.3 kB
- Tags: Python 2, Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.22.0 setuptools/49.6.0 requests-toolbelt/0.9.1 tqdm/4.40.0 CPython/3.6.8
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 3ca75daa613a19befdca377fa9ecdece1597c2efe5c22b5da30de7d41092df1f |
|
MD5 | 2518b27d259096326e9e4c702147d52a |
|
BLAKE2b-256 | 101ad4ad57102dcbde20c682d0ebf793b768ceedd413be549ba7ed4489d28dc5 |