requirements.txt

# Base
ipython>=5.5.0
ipywidgets>=7.6.5
tqdm>=4.62.0
numpy>=1.21, <1.27
pandas>=2.2.0 #latest sktime makes tests fail if pandas 2.1
jinja2>=3
scipy>=1.6.1,<=1.11.4 #to fix later (https://github.com/mljar/mljar-supervised/issues/691)
joblib>=1.2.0,<1.4  # joblib<1.2.0 is vulnerable to Arbitrary Code Execution (https://github.com/advisories/GHSA-6hrg-qmvc-2xh8)  # joblib 1.4.0 breaks PyCaret
scikit-learn>1.4.0
pyod>=1.1.3 #1.1.3 to support scikit-learn 1.4 (https://github.com/yzhao062/pyod/issues/542)
imbalanced-learn>=0.12.0 # >=0.12.0 to support scikit-learn 1.4
category-encoders>=2.4.0
lightgbm>=3.0.0
numba>=0.55.0
requests>=2.27.1  # Required by pycaret.datasets
psutil>=5.9.0
markupsafe>=2.0.1  # Fixes Google Colab issue
importlib_metadata>=4.12.0
nbformat>=4.2.0
cloudpickle
deprecation>=2.1.0
xxhash
wurlitzer; platform_system != 'Windows'
trio>=0.22.0,<0.25.0 #fixes problems about third-party packages, remove after httpcore 1.05
setuptools; python_version>='3.12'

# Plotting
matplotlib<3.8.0 #stem(..., use_line_collection=False) is no longer supported. 
mljar-scikit-plot #this is a updated version of original scikit-plot
# yellowbrick<=1.5, have conflict with recent matplotlib by use_line_collection error
# should use yellowbrick>1.5 when fixed (https://github.com/DistrictDataLabs/yellowbrick/issues/1312)
yellowbrick>=1.4
plotly>=5.14.0
kaleido>=0.2.1
schemdraw==0.15  # 0.16 only supports Python >3.8
plotly-resampler>=0.8.3.1

# Time-series
statsmodels>=0.12.1
sktime #sktime adds support for scikit-learn 1.5 in version sktime 0.31
tbats>=1.1.3
pmdarima>=2.0.4