Prune Callback

Use the pruner in fastai Callback system

Overview

The PruneCallback integrates structured pruning into the fastai training loop. Unlike sparsification (which zeros weights), pruning physically removes network structures (filters, channels) to reduce model size and computation.

Key Differences from SparsifyCallback: - Removes structures entirely (not just zeros) - Uses torch-pruning library for dependency handling - Supports various pruning criteria and schedules

Parameters:

pruning_ratio: Target ratio of parameters to remove (0-1 or 0-100). Values >1 are treated as percentages.
schedule: When to prune (from fasterai.core.schedule). Controls how pruning progresses over training.
context: 'local' (per-layer pruning) or 'global' (across entire model).
criteria: How to select what to prune (from fasterai.core.criteria).

Usage Example

from fasterai.prune.prune_callback import PruneCallback
from fasterai.core.schedule import agp
from fasterai.core.criteria import large_final

# Prune 30% of parameters using automated gradual pruning schedule
cb = PruneCallback(
    pruning_ratio=30,        # Remove 30% of parameters
    schedule=agp,            # Gradual pruning (cubic decay)
    context='global',        # Prune globally across all layers
    criteria=large_final     # Keep weights with largest magnitude
)

learn.fit(10, cbs=[cb])

Overview

Usage Example

See Also