Python Deep Learning: Neural Networks to Transformers
Deep learning powers the systems behind image recognition, natural language processing, speech synthesis, and generative AI. Python dominates this space through TensorFlow and PyTorch, which together account for the vast majority of deep learning research and production deployments.
This learning path covers neural network architectures from the ground up: convolutional networks for images, recurrent networks for sequences, transformers for language, and GANs for generation. Each tutorial includes implementation code with the major frameworks.
Tutorials marked with the cert badge include a final exam that awards a certificate of completion you can download and share.
Python Deep Learning
Foundations of deep learning: neurons, layers, activation functions, backpropagation, and training loops.
Python Convolutional Neural Networks (CNN)
CNN architecture, convolution operations, pooling, and building image classifiers.
Python Recurrent Neural Networks and LSTMs
RNNs for sequence data, vanishing gradients, LSTM cells, and text/time-series applications.
Python Transformers
Transformer architecture, self-attention, positional encoding, and using pre-trained models with Hugging Face.
Python Generative Adversarial Networks (GANs)
GAN architecture, training dynamics, mode collapse, and generating synthetic data.
TensorFlow Python Guide
TensorFlow 2.x with Keras: model building, training, evaluation, and the tf.data pipeline.
PyTorch Python Guide
PyTorch fundamentals: tensors, autograd, nn.Module, training loops, and GPU acceleration.
Model Serving and Batching Strategies
Serving deep learning models in production with batching, caching, and scaling patterns.
Serve a Machine Learning Model with FastAPI
Building an inference API for deep learning models with FastAPI.