Batch Size Experiment Game

Test small and large batches, observe noisy gradients, compare convergence, manage memory, and discover efficient training choices through interactive machine learning experiments and challenges.

Training laboratory

Mission and challenge objectives

Choose a controlled experiment, predict the result, then evaluate the measured trade-offs.

Score 0
Ready for an experiment
No active training run
Experiment setup

Configure the dataset, model, and training process

Keep most settings unchanged when comparing batch sizes fairly.

Game and challenge

Dataset controls

Model controls

Batch settings

Recommendation considers data, model, memory, and noise.

Optimisation controls

Training controls

Forecast before training

Make performance predictions

Predictions are scored when the experiment finishes.

Live training

Canvas batch-processing simulation

Samples enter batches, gradients update the model, and memory pressure changes with batch size.

Dots: samples Box: active batch Wave: gradient update Blocks: memory
Epoch progress0 / 60
Epoch0
Train loss
Validation loss
Validation accuracy
Gradient variance
Memory
Updates0
Samples / second
Elapsed time0.0s
Stability
Generalisation gap
Learning rate
Plotly dashboards

Live curves and experiment comparisons

Zoom, pan, hover, and download each interactive graph.

Experiment history

Rank, pin, duplicate, and compare completed runs

Results are stored in this browser and remain available after reloading.

Rank Experiment Batch Optimiser Rate Epochs Time Train loss Validation loss Accuracy Stability Memory Gap Score Actions
Run an experiment to create the first comparison record.
Interpretation

Automatic findings and learning feedback

The explanation updates after every completed run.

Start an experiment to receive a batch-size interpretation.

Small batches

They update weights frequently. Their gradients are noisy. Noise can sometimes escape weak solutions.

Large batches

They process samples efficiently. They need more memory. Their gradients are usually smoother.

Fair comparisons

Keep data, model, seed, and optimiser fixed. Change one setting. Compare several repeated runs.

Learning-rate interaction

Larger batches may tolerate larger rates. Excessive rates still diverge. Scaling rules are starting points.

Convergence

Fast epochs do not guarantee fewer epochs. Count updates too. Check validation performance before deciding.

Generalisation

Smooth training can still overfit. Compare validation loss. Prefer balanced performance over one metric.

Progress

Achievements

Complete varied experiments to unlock learning milestones.

Guide

How to use the game

  1. Select a challenge and configure one batch size.
  2. Make predictions before starting the training simulation.
  3. Observe sample batches, gradient movement, and memory blocks.
  4. Study the Plotly curves after every simulated epoch.
  5. Repeat with another batch size while keeping other settings fixed.
  6. Rank results using speed, stability, convergence, accuracy, and memory.

Frequently asked questions

Is the training real?

The game uses a deterministic educational simulation. It models common trends without training a production model.

Why can results vary?

Noise, seed, optimiser, learning rate, and model complexity influence every simulated trajectory.

What is full-batch training?

One update uses the complete training set. It is smooth but can be slow and memory intensive.

What is convergence?

Convergence means loss improves slowly and approaches a stable level or chosen target.

Which batch size is best?

No universal winner exists. Hardware, dataset size, model architecture, and accuracy goals determine the best choice.

Why measure throughput?

Throughput shows processed samples per second. It separates computational speed from epoch count.

Related Calculators

Train the Tiny ModelHyperparameter Tuning RaceEpoch ControlModel Selection TournamentBias-Variance Balancing GameRegularisation DefenderEarly Stopping Challenge

Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.