# TensorFlow

TensorFlow is an open source software library for machine learning in various kinds of perceptual and language understanding tasks. It is open source and self-hosted, installable on Docker Hub and Proxmox.

Canonical page: https://ourlocalhost.com/app/6203/tensorflow

## Description

TensorFlow is an open source software library for machine learning in various kinds of perceptual and language understanding tasks.

## Platform availability

Available on 1 of 10 platforms: Proxmox.

## Health score

82/100 (Excellent), based on 3 of 4 factors.
- Maintained: 100 (last commit 2 days ago)
- Popular: 100 (201,650 GitHub stars)
- Easy to install: 25 (0 install platforms, install notes, Docker image)

## Links

- GitHub: https://github.com/tensorflow/tensorflow
- Docker Hub: https://hub.docker.com/r/rocker/tensorflow
- Proxmox: https://community-scripts.github.io/ProxmoxVE/scripts/tensorflow
- Official website: https://tensorflow.org
- Documentation: https://tensorflow.org/install

## Common questions

### How can I run TensorFlow with GPU acceleration?

The current release installed with pip supports CUDA-enabled GPU cards on Ubuntu and Windows. For other devices such as DirectX and macOS Metal, use TensorFlow device plugins.

### What extra is needed to load models or data from Google Cloud Storage?

Install TensorFlow with the GCS filesystem extra, such as tensorflow[gcs-filesystem]. Without that extra, GCS:// file paths will not work because TensorFlow cannot access Cloud Storage.

### How do I update TensorFlow to the latest version?

Add the --upgrade flag to the pip install commands for tensorflow or tensorflow-cpu. Nightly builds are also available as tf-nightly and tf-nightly-cpu on PyPI for testing.

### What kinds of TensorFlow problems should I troubleshoot?

Common issues include installation errors, GPU acceleration failures, model training problems, memory bottlenecks, and version compatibility challenges.

### Where can I look for documented build and installation errors?

TensorFlow's install error page links error messages to a solution or discussion. It notes that TensorFlow uses GitHub issues, Stack Overflow, and the TensorFlow Forum to track and discuss build and installation problems.

---
Source: Our Local Host (https://ourlocalhost.com/app/6203/tensorflow), the self-hosted and open-source app directory.
