OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Tensorpack

Tensorpack is a Python framework for training neural networks with graph-mode TensorFlow, offering multi-GPU and distributed training strategies, Python data-loading utilities, and example workflows for research tasks.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Tensorpack — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Tensorpack when a project needs a Python training interface around graph-mode TensorFlow, especially for data loading, multi-GPU or distributed training, or adapting its research examples.

Who it is for

  • Python developers building TensorFlow neural-network training workflows.
  • Researchers adapting reproducible model implementations or creating custom data-loading pipelines.

Documented uses

  • Build neural-network training workflows with TensorFlow.
  • Use data-parallel multi-GPU or distributed training strategies.
  • Create Python data-processing pipelines with tensorpack.dataflow.
  • Study or adapt example scripts for computer vision, reinforcement learning, speech recognition, and language modeling.
Before you choose

Requirements & limits

The project is publicly available as an Apache-2.0-licensed repository. The README lists Python 3.3 or later and TensorFlow 1.5 or later for framework use; TensorFlow is not required for tensorpack.dataflow alone.

  • Tensorpack is based on graph-mode TensorFlow; with TensorFlow 2 it uses TF1 compatibility mode.
  • The README notes that some examples have not been migrated to support TensorFlow 2.
  • The project describes itself as not yet stable and recommends pinning the exact Tensorpack version used as a dependency.
  • OpenCV Python bindings are optional but required for many features.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Tensorpack from its GitHub repository

Install Tensorpack using the installation method and prerequisites listed in its README.

Before you start

  • Python 3.3 or newer.
  • TensorFlow 1.5 or newer, unless using only tensorpack.dataflow.
  • Python bindings for OpenCV are optional, but required by many features.

Steps

  1. Install the listed prerequisites for the features you intend to use.

  2. Run the command below.

    Documented command
    pip install --upgrade git+https://github.com/tensorpack/tensorpack.git

  3. For a user-local installation, add --user to the pip command.

  4. Pin the exact Tensorpack version used in your project dependencies, as the README says Tensorpack is not yet stable.

Expected outcome

Tensorpack is installed from the GitHub repository using the README's pip method.

Guide limitations

  • When using TensorFlow 2, Tensorpack uses TensorFlow's TF1 compatibility mode; some repository examples have not been migrated to support TensorFlow 2.
  • TensorFlow is not required when using tensorpack.dataflow alone as a data-processing library.
  • The README describes Tensorpack as not yet stable.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
tensorpack/tensorpack · Official homepage title · retained in its source language
How we review sourcesSuggest a correction