A bridge course for people who already know PyTorch. Every lesson maps a concept you learned in the PyTorch course to its Keras equivalent — tensors, Sequential/Functional models, the compile/fit loop, tf.data, callbacks, GradientTape, and saving/serving — so you get productive in TensorFlow fast, without relearning the fundamentals.
Before you start
This course assumes the Deep Learning with PyTorch course — it teaches only what’s different in Keras, not deep learning from scratch. You need Python 3.10+ and tensorflow (which bundles Keras). A CPU install works for everything here.
Setup & Tensors
You already know deep learning from PyTorch — this course maps it to Keras. Start with the two frameworks side by side, install TensorFlow, and meet tf.Tensor.
Building Models
nn.Module has two Keras counterparts: the Sequential API for simple stacks and the Functional API for anything with branches. Map your PyTorch model-building instincts across.
compile() & fit()
This is the biggest difference from PyTorch. Instead of writing the training loop by hand, you configure it with compile() and run it with fit() — convenience for control.
Data Pipelines with tf.data
tf.data.Dataset is Keras’s answer to Dataset + DataLoader — batching, shuffling, and prefetching, but as a chained, lazy pipeline. Feed it straight into fit().
Callbacks & Custom Training
The techniques you coded by hand in PyTorch — early stopping, checkpoints, LR schedules — are built-in Keras callbacks. And when fit() isn’t enough, GradientTape gives you the manual loop back.
Saving & Serving
Ship the trained model. Save in the Keras or SavedModel format, reload for inference, and export to TensorFlow Serving or TFLite — the TF equivalents of TorchScript/ONNX.