pad-sequences

Pad variable length sequences with multiples features.

Installation via pip

The pad-sequences git repo is available as PyPi package

pip install "pad-sequences>=0.6.0"

Usage

from pad_sequences import pad_sequences_multi
import tensorflow as tf
# import torch

seq = []
seq.append([[1, 1, 1], [2, 2, 2]])
seq.append([[1, 1, 1], [2, 2, 2], [3, 3, 3], [4, 4, 4]])

n_samples = len(seq)
n_features = len(seq[0][0])
n_timesteps = 3

# for input sequences
padded = pad_sequences_multi(seq, padding='pre', value=0, 
                             truncating='pre', maxlen=n_timesteps)
# for output sequences
# padded = pad_sequences_multi(seq, padding='post', value=0, 
#                              truncating='post', maxlen=n_timesteps)

X = tf.reshape(padded, [n_samples, n_timesteps, n_features])
# X = torch.reshape(torch.tensor(padded), [n_samples, n_timesteps, n_features])

Check the examples folder for notebooks.

Appendix

Install a virtual environment

python3 -m venv .venv
source .venv/bin/activate
pip3 install --upgrade pip
pip3 install -r requirements-dev.txt
pip3 install -r requirements-demo.txt

(If your git repo is stored in a folder with whitespaces, then don't use the subfolder .venv. Use an absolute path without whitespaces.)

Other python commands

• Jupyter for the examples: jupyter lab
• Check syntax: flake8 --ignore=F401 --exclude=$(grep -v '^#' .gitignore | xargs | sed -e 's/ /,/g')
• Run Unit Tests: pytest

Publish

pandoc README.md --from markdown --to rst -s -o README.rst
python setup.py sdist 
twine upload -r pypi dist/*

Clean up

find . -type f -name "*.pyc" | xargs rm
find . -type d -name "__pycache__" | xargs rm -r
rm -r .pytest_cache
rm -r .venv

Debugging

• Notebooks to profile python code are in the profile folder