A few things learned, and shared.
32 sessions
A hands-on workshop on representation learning for computer vision using Keras and TensorFlow. Attendees learn how pretrained backbones, embeddings, and transfer learning turn raw pixels into powerful, reusable feature representations for downstream…
An introduction to representation learning for computer vision: how deep neural networks learn compact, transferable feature embeddings from images. Covers feature extraction, transfer learning, and why high-quality representations are the foundation of…
The opening session of the ML Ascent series on pretraining deep learning models for computer vision. Explores how large-scale pretraining and transfer learning with TensorFlow and Keras give vision models a head…
Explore the frontier of multimodal AI with contrastive learning for visual-language alignment. This talk unpacks how models like CLIP learn a shared image-text embedding space through contrastive objectives, powering zero-shot classification, semantic…
Master parameter-efficient fine-tuning of Large Language Models with LoRA (Low-Rank Adaptation). Learn how LoRA adapts LLMs to custom tasks using a fraction of the compute and memory, plus practical guidance on training…
A deep dive into the Transformer architecture that powers modern generative AI. Covers self-attention, multi-head attention, positional encoding, and the encoder-decoder design behind today's Large Language Models, explaining why Transformers revolutionized NLP…
A practical guide to Generative Adversarial Networks (GANs). Learn how the generator and discriminator compete during training, common failure modes like mode collapse, and hands-on techniques for training stable GANs to synthesize…
Go beyond Keras Model.fit and take full control of model training. This session covers custom training loops with tf.GradientTape, custom losses and metrics, and advanced TensorFlow workflows for when the high-level API…
A practical session on deploying machine learning models to production. Covers model serialization, serving with TensorFlow, building API endpoints, optimizing for inference, and the MLOps practices that move a trained model from…
A hands-on workshop on deep computer vision for object detection. Learn how detectors locate and classify multiple objects in an image, the role of bounding boxes and anchors, and how to implement…
A practical deep-dive into image segmentation with deep learning. Covers semantic and instance segmentation, encoder-decoder architectures such as U-Net, and how to train pixel-level computer-vision models for applications like medical imaging and…
The foundations of deep computer vision for image classification. Learn how convolutional neural networks (CNNs) extract features from images, how to build and train classifiers with TensorFlow and Keras, and best practices…
At BlaBlaConf, this talk explores the art of MLOps in the TensorFlow ecosystem: building production-grade machine learning pipelines with TFX, data validation, model training, deployment, and monitoring. A practical look at taking…
An accessible introduction to JAX and Flax for high-performance machine learning. Covers JAX's NumPy-style API, automatic differentiation, just-in-time (JIT) compilation, and vectorization, plus building and training neural networks with the Flax library.
An accessible introduction to Reinforcement Learning (RL). Covers agents, environments, rewards, policies, and value functions, explaining how RL algorithms learn optimal behavior through trial and error, illustrated with intuitive examples and real-world…
Presented at DevFest, this talk covers the art of MLOps in the TensorFlow ecosystem: designing reproducible ML pipelines, automating training and deployment, and monitoring models in production with TensorFlow Extended (TFX) for…
An approachable introduction to Machine Learning. This ML 101 session demystifies core concepts, supervised versus unsupervised learning, training and evaluation, features and models, giving beginners a clear mental map of how machines…
A hands-on workshop building a brain tumor classifier with TensorFlow. Participants work through loading medical MRI imagery, preprocessing, and training a convolutional neural network to classify tumors, a practical introduction to deep…
A practical talk on remote sensing and machine learning: classifying land cover from multispectral Sentinel-2 satellite imagery using Google Earth Engine and TensorFlow. Covers working with geospatial data and spectral bands to…
Part of the 30DaysofML #MLReady series: build your first image classifier with a dense neural network in TensorFlow. A beginner-friendly walkthrough of loading image data, designing a fully-connected network, and training and…
Part of the 30DaysofML #MLReady series: time series forecasting with TensorFlow. Learn how to frame forecasting problems, prepare sequential data with windowing, and build neural networks that predict future values from historical…
Part of the 30DaysofML #MLReady series: detecting anomalies in time-series data with TensorFlow. Covers autoencoder-based anomaly detection, reconstruction error, and how to flag unusual patterns in sequential data for monitoring and predictive…
Part of the 30DaysofML #MLReady series: an introduction to Natural Language Processing (NLP). Covers text preprocessing, tokenization, and turning raw text into machine-readable features, the foundation for building NLP models with TensorFlow.
Part of the 30DaysofML #MLReady series: understanding word embeddings in NLP. Learn how words become dense vectors that capture semantic meaning, how embeddings are trained, and how to use them as input…
Part of the 30DaysofML #MLReady series: hands-on sentiment analysis with TensorFlow. Build a model that classifies text as positive or negative, covering text vectorization, word embeddings, and recurrent and dense architectures for…
Part of the 30DaysofML #MLStart series: an introduction to Exploratory Data Analysis (EDA). Learn how to inspect, visualize, and understand datasets, spotting patterns, outliers, and relationships, before building any machine learning model.
Part of the 30DaysofML #MLStart series: AI 101, a beginner-friendly introduction to Artificial Intelligence. Covers what AI is, how it differs from machine learning and deep learning, and the everyday applications already…
The kickoff session of the 30DaysofML program, a community learning journey into machine learning. This opening talk introduces the roadmap, goals, and the hands-on path participants follow from AI basics to building…
A follow-up session on personal branding for developers and tech professionals. Practical guidance on building an authentic online presence, growing a community, sharing your work, and turning your expertise into real opportunities…
A talk on personal branding for students and developers in tech. Learn how to craft your professional identity, build a portfolio and online presence, and communicate your skills to stand out to…
An approachable introduction to Machine Learning. This Machine Learning 101 session covers the core ideas behind how machines learn from data, supervised and unsupervised learning, models, training, and evaluation, designed to make…
A beginner-friendly introduction to Artificial Intelligence. This AI 101 talk explains what AI really is, the key concepts behind machine learning and deep learning, and how AI is transforming industries, perfect for…
