An open-source machine learning framework for everyone.

TensorFlow™ is an open source software library for high performance numerical computation. Its flexible architecture allows easy deployment of computation across a variety of platforms (CPUs, GPUs, TPUs), and from desktops to clusters of servers to mobile and edge devices. Originally developed by researchers and engineers from the Google Brain team within Google’s AI organization, it comes with strong support for machine learning and deep learning and the flexible numerical computation core is used across many other scientific domains.

lightbulb_outline View ideas list


  • machine learning
  • deep learning
  • python
  • data analysis
  • c/c++


  • Other
  • deep learning
  • machine learning
  • python
  • data
  • data science
comment IRC Channel
email Mailing list
mail_outline Contact email

TensorFlow 2020 Projects

  • Ayushi
    A fast, cross-platform image loader in Swift
    Apps and web pages often contain a lot of images and at times they are remote and contribute highly to data usage, so they require a loader to delay...
  • Roland Gavrilescu
    Flax HOWTOs
    Flax is a high-performance neural network library for JAX designed for flexibility. In order to allow users get started easily, Flax provides a...
  • Vo Van Nghia
    GCS, S3 and Hadoop support for Tensorflow
    Currently, there are no supports for S3 and GCS on Windows. It leads to two problems: The limitation of storage capacity and inconvenient of...
  • Yiming Shi
    Help migrate a TensorFlow library to TensorFlow 2.x
    My interest lies in software engineering (especially debugging) and machine learning applications. I am very interested in the project Help migrate a...
  • Robin Baumann
    Implement Mesh R-CNN in TensorFlow Graphics
    The Mesh-RCNN technique was recently introduced to estimate 3D models from natural 2D images. It uses a combination of an object detection module...
  • Yasaswi
    Implement TF 2.x Model Garden Examples
    The main objective of this project will be to implement various neural network models and to improve the documentation for the TensorFlow User...
  • Daisuke Kasuga
    Implementing “implicit 3D representations” in TensorFlow Graphics
    In recent years, “implicit 3D representations” have been attracting attention as an appropriate 3D expression for deep neural networks. At CVPR 2019,...
  • Vijay Tadikamalla
    Improving TensorFlow Datasets Library
    Everyone wants to focus on improving their model performance, but a significant portion of their time is used up while searching for a proper dataset...
  • Hedron
    Incorporating point-based convolutions in Tensorflow Graphics
    Point-based learning techniques have shown great potential in standard Computer Vision tasks such as 3D scan segmentation or 3D object...
  • Swarup
    Keras RFC Implementation - ImagePipeline based on API is the defacto way of loading and preprocessing data in TensorFlow for training machine learning models on CPUs, GPUs and TPUs. There...
  • Abhineet Choudhary
    Let's not Go Off Tangent - Add Efficient Tangent batching to ForwardAccumulator
    Current native support for forward-mode differentiation calls backward on a function twice. But executing this as a tf.function prevents retracing...
  • Jin Dong
    New Tools for TensorFlow Hub
    This project aims at implementing new tools to support the transfer learning pipelines, ranging from object detection to embedding-based similarity...
  • seungjaeryanlee
    Reinforcement Learning to S4TF
    Reinforcement learning is a class of machine learning methods that uses the reward signal of the environment to infer optimal actions. Reinforcement...
  • Zhibo Zhang
    SOTA ML Paper/Model Implementations
    Neural Tangents (Infinite-width NNs) migration and reconstruction for TensorFlow 2.x, originally based on JAX...
  • Vishal V
    State of the Art TensorFlow 2.x Models
    State-of-the-art Deep Learning models for the TensorFlow Model Garden implemented from the most recent research papers and trained on the datasets...
  • Dejia Xu
    TensorFlow Hub New Models and Demos Application
    I will implement DeepFillv2 and add detailed Colab demos for it. The model will be "fine-tunable" to support the “reusable” characteristic of...
  • Amish Garg
    TensorFlow Lite Flutter Support Suite
    TensorFlow Lite Flutter Support Suite is a cluster of three projects TfLite Flutter Plugin, TfLite Flutter Helper library, Flutter example ML app...