Install Teradata Package for Python on IBM PowerPC - Install Teradata Package for Python on IBM PowerPC - Teradata Package for Python

Teradata® Package for Python User Guide
Deployment
VantageCloud
VantageCore
Edition
VMware
Enterprise
IntelliFlex
Product
Teradata Package for Python
Release Number
20.00
Published
March 2025
ft:locale
en-US
ft:lastEdition
2026-02-20
dita:mapPath
nvi1706202040305.ditamap
dita:ditavalPath
plt1683835213376.ditaval
dita:id
rkb1531260709148
Product Category
Teradata Vantage
Prerequisite
  • Python 3.9 or higher
  • conda-forge channel provided by conda package manager - Required for installing platform-dependent packages like pandas, numpy, scipy, scikit-learn, lightgbm, and matplotlib on PowerPC architecture.

    These packages contain compiled C/C++/Fortran extensions that need pre-built binary wheels. Conda-forge provides PowerPC-compatible binaries, while PyPI lacks PowerPC wheels for most scientific packages, forcing source compilation that fails with GNU Compiler Collection (GCC) 8.5.0 internal compiler errors on PowerPC systems.

Important: Do NOT install the scientific dependencies using pip; compiler errors will be returned.
Complete the following steps to install Teradata Package for Python on IBM PowerPC. See Common Issues and Solutions with Installing Teradata Package for Python on IMB PowerPC to address any issues during the installation process.
  1. Install Conda (if not already installed).
    1. Execute the following to download Miniconda for PowerPC.
      cd /tmp
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-ppc64le.sh
bash Miniconda3-latest-Linux-ppc64le.sh -b -p $HOME/miniconda3
    2. Add conda to PATH.
      export PATH="$HOME/miniconda3/bin:$PATH"
echo 'export PATH="$HOME/miniconda3/bin:$PATH"' >> ~/.bashrc
source ~/.bashrc
  2. Create a new Conda environment for teradataml.
    conda create -n teradataml python=3.9 -y
conda activate teradataml
  3. Install scientific dependencies using conda-forge in the following order.
    Important: You must install the packages in the order specified:
    • 'numpy' is the foundation for all scientific computing.
    • 'pandas' depends on numpy.
    • 'scikit-learn' depends on numpy and scipy (auto-installed).
    • 'lightgbm' depends on numpy and scikit-learn.
    • 'matplotlib' can depend on numpy for certain operations.
    conda install -c conda-forge numpy -y
conda install -c conda-forge pandas -y
conda install -c conda-forge scikit-learn -y
conda install -c conda-forge lightgbm -y
conda install -c conda-forge matplotlib -y
  4. Verify the scientific stack installation.
    python -c "
import numpy as np
import pandas as pd
import sklearn
import lightgbm as lgb
import matplotlib.pyplot as plt
print(f'✅ NumPy: {np.__version__}')
print(f'✅ Pandas: {pd.__version__}')
print(f'✅ Scikit-learn: {sklearn.__version__}')
print(f'✅ LightGBM: {lgb.__version__}')
print(f'✅ Matplotlib: {plt.matplotlib.__version__}')
print('🎉 All scientific dependencies installed successfully!')
"
  5. Install teradataml using pip.
    pip install teradataml --no-cache-dir
  6. Verify the installation completed successfully.
    python -c "
import teradataml as tdml
print(f'✅ TeradataML: {tdml.__version__}')
print('🎉 TeradataML installation complete!')
"
PostrequisiteReview the following topics for more information: