TFLearn is an open-source deep learning library that provides a higher-level, Keras-style API on top of TensorFlow, designed to make building neural networks faster and more approachable. Hosted on GitHub, it has accumulated more than 9,500 stars and, while its last major activity was in 2024, it remains a widely referenced teaching and prototyping library in the TensorFlow ecosystem.
TFLearn is a modular and transparent deep learning library built on top of TensorFlow. It was designed to provide a higher-level API in order to facilitate and speed up experimentation, while remaining fully transparent and compatible with raw TensorFlow code. Instead of wrestling with graph boilerplate, you compose models from built-in layers, regularizers, optimizers, and metrics, and can still drop down to native TensorFlow tensors whenever you need full control. The project includes tutorials and examples covering the common workloads: convolutional networks for images, recurrent networks for sequences, and generative models. Because everything resolves to standard TensorFlow operations, models trained with TFLearn can be served and extended with the regular TensorFlow toolchain.
TFLearn is completely free and open source. There is no paid edition or hosted service: you install the library, run it against your own TensorFlow installation, and pay only for your own compute. Documentation at tflearn.org is free as well.
TFLearn suits students, educators, and practitioners who want to learn or prototype deep learning in the TensorFlow world without low-level friction. It is a good fit for coursework, quick experiments, and legacy TensorFlow projects that value readable, compact model definitions over the newest framework features.