Training step
A training step is an iteration in which a model is updated during training using data and an optimization objective.
Deep Learning with R, Third Edition
A training step is one iteration that evaluates a model on a batch, computes its loss and gradients, and updates its parameters.
- How it is used
- Uses each step to update model weights on a batch within a repeated training loop.
- How deep it goes
- Works through prediction, loss, automatic gradients, and updates across backends, then extends the step to Keras metrics and fit().
- Examples
- Uses MNIST image batches and a full-batch linear classifier on 2,000 synthetic points.
- What you can do after
- Apply it via Keras with TensorFlow, PyTorch, or JAX, including a custom train_step() used by fit().
Read how Deep Learning with R, Third Edition explains training step →
Deep Learning with Python, Third Edition
A single update of a model on one batch, consisting of prediction, loss computation, gradient computation, and weight adjustment.
- How it is used
- Uses a batch-level update as the building block of training loops and custom Keras fit() behavior.
- How deep it goes
- Works through prediction, loss, gradients, and updates, then implements backend-specific steps and custom metric handling.
- Examples
- Uses MNIST image classification and a JAX linear classifier to demonstrate training steps.
- What you can do after
- Apply it via TensorFlow, PyTorch, or JAX, including by overriding Keras train_step().
Read how Deep Learning with Python, Third Edition explains training step →
Local Image Generation
A training step is one iteration in which a training procedure updates a model using a batch of training data.
- How it is used
- Uses step counts to configure Flux LoRA training and choose a saved model version from sample outputs.
- How deep it goes
- Explains step counts procedurally through configuration, checkpoint retention, and visual comparison rather than update internals.
- Examples
- Compares Keats images at 1,250, 3,250, 4,750, 5,500, and 6,000 steps, including photo and cartoon prompts.
- What you can do after
- Operate/configure it by setting a training total, retaining checkpoints, and selecting a step from generated-image tests.
Also covered in
- Sutskever's List
- Time Series Forecasting Using Foundation Models
- Build a DeepSeek Model (From Scratch)In early access (MEAP)
What to know first
- optimizationOptimization is the systematic selection or modification of values, configurations, or implementations to improve a defined objective, often subject to constraints.
- objective functionAn objective function is a mathematical function that assigns values to candidate solutions or parameter settings, which an optimization procedure seeks to minimize or maximize.
- computation graphA computation graph is a directed graph whose nodes represent computational operations or values and whose edges represent dependencies between them.
- fine-tuningFine-tuning is the continued training of a pretrained model on targeted data or objectives to adapt its parameters or behavior to a task, domain, or use.
Related concepts
- fit methodused by · A fit method is an operation that estimates model parameters or learns a model or data transformation from supplied data.
- Keras modelpart of · A Keras model is a trainable object representing a neural-network computation and providing interfaces for training, evaluation, prediction, and saving.
- epochcontrasts with · An epoch is one complete pass through the training data during model training.
- stepkind of · A step is a discrete, ordered unit of work or action within a process, workflow, procedure, or task.
- batch sizeconfigured by · Batch size is the number of items grouped for processing together in a single batch or computational operation.
Questions about training step
What is training step?
A training step is an iteration in which a model is updated during training using data and an optimization objective.
Which Manning books explain training step?
Deep Learning with R, Third Edition, Deep Learning with Python, Third Edition and Local Image Generation.
What should I know before learning training step?
Manning books that teach training step usually introduce optimization, objective function, computation graph and fine-tuning first.
What is the difference between training step and epoch?
A training step is an iteration in which a model is updated during training using data and an optimization objective. An epoch is one complete pass through the training data during model training.
What is training step a kind of?
Training step is a kind of gradient-based optimization. A method that uses derivatives of a function with respect to variables to guide optimization or analyze how outputs respond to changes in those variables.
What is training step part of?
Training step is part of Keras model. A Keras model is a trainable object representing a neural-network computation and providing interfaces for training, evaluation, prediction, and saving.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.


