About Time Series Forecasting

Demonstrate common approaches for Time Series forecasting task.

Setup

OS

Ubuntu 18.04.6 LTS

01-TimeSeries-Forecasting.ipynb

Environment

conda create --name timeseries python==3.8.3 --yes

conda activate timeseries

conda install notebook ipykernel --yes

ipython kernel install --name timeseries --display-name "Python 3.8.3 (timeseries)" --user

pip packages

pip install requests==2.27.1

pip install plotly==5.6.0

pip install statsmodels==0.13.2

pip install featuretools==1.6.0

conda packages

conda install pandas=1.4.1

conda install chardet=4.0.0

conda install scikit-learn=1.0.2

conda install matplotlib=3.5.1

02-TimeSeries-Forecasting-PyTorch-Lightning.ipynb

Environment

conda create --name timeseriespytorchlight python==3.8.12 --yes

conda activate timeseriespytorchlight

conda install notebook ipykernel --yes

ipython kernel install --name timeseriespytorchlight --display-name "Python 3.8.12 (timeseriespytorchlight)" --user

pip packages

pip install pyarrow==7.0.0

conda packages

conda install -c conda-forge ipywidgets

conda install pandas=1.4.1

conda install scikit-learn=1.0.2

conda install matplotlib=3.5.1

conda install pytorch=1.10.2 -c pytorch

conda install pytorch-forecasting=0.9.2 -c conda-forge