PyPI page
Home page
Author:
Your Name
License:
Summary:
To improve EDU segmentation performance using Segbot. As Segbot has an encoder-decoder model architecture, we can replace bidirectional GRU encoder with generative pretraining models such as BART and T5. Evaluate the new model using the RST dataset by using few-shot based settings (e.g. 100 examples) to train the model, instead of using the full dataset.
Latest version:
0.0.115
Required dependencies:
attrs
|
bleach
|
build
|
cachecontrol
|
certifi
|
charset-normalizer
|
cleo
|
click
|
colorama
|
crashtest
|
distlib
|
docutils
|
dulwich
|
filelock
|
fsspec
|
html5lib
|
huggingface-hub
|
idna
|
importlib-metadata
|
installer
|
jinja2
|
joblib
|
jsonschema
|
keyring
|
lockfile
|
markdown-it-py
|
markupsafe
|
mdurl
|
more-itertools
|
mpmath
|
msgpack
|
networkx
|
nltk
|
numpy
|
packaging
|
pexpect
|
pkginfo
|
platformdirs
|
poetry
|
poetry-core
|
poetry-plugin-export
|
ptyprocess
|
pygments
|
pyproject_hooks
|
pyrsistent
|
pywin32-ctypes
|
pyyaml
|
rapidfuzz
|
readme-renderer
|
regex
|
requests
|
requests-toolbelt
|
rfc3986
|
rich
|
shellingham
|
six
|
sympy
|
tokenizers
|
tomlkit
|
torch
|
tqdm
|
transformers
|
trove-classifiers
|
twine
|
typing_extensions
|
urllib3
|
virtualenv
|
webencodings
|
zipp
Downloads last day:
26
Downloads last week:
241
Downloads last month:
637