12fdd18223
Deleted: - DQN::compute_loss_internal (280 lines) — old Candle forward+loss - DQN::train_step (55 lines) — old Candle training step - DQN::compute_gradients (47 lines) — old gradient accumulation - ComputeLossResult struct — only used by deleted functions - RegimeConditionalDQN::train_step (65 lines) — old dispatch - RegimeConditionalDQN::train_step_gpu_regime (100 lines) — old GPU path - RegimeConditionalDQN::compute_gradients_gpu (130 lines) — old regime gradients - RegimeConditionalDQN::compute_gradients (92 lines) — old dispatch - DQNAgentType::train_step dispatch — dead - DQNAgentType::compute_gradients dispatch — dead - GpuDqnTrainer::upload_batch (71 lines) — old CPU→GPU upload - train_step.rs (500 lines) — entire module including ensure_fused_ctx - dqn_benchmark.rs — used old train_step - examples.rs — used old train_step - validation/adapters.rs (289 lines) — used old train_step - dqn/trainable_adapter.rs — used old train_step - gpu_smoketest.rs — tested old train_step - Gradient accumulation path in training_loop.rs (144 lines) - IQN d_h_s2().clone() → raw pointer (zero alloc) - Causal intervention format! string alloc removed - Dead HER relabel functions (320 lines) Kept: - ensure_fused_ctx logic inlined into training_loop.rs - set_noise_sigma_scale re-added to RegimeConditionalDQN Fixed: - GpuReplayBuffer max_batch_size wired from batch_size parameter (was hardcoded 1024, blocking batch_size=8192) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
1.6 KiB
1.6 KiB
Claude Code Hooks for claude-flow
Purpose
Automatically coordinate, format, and learn from Claude Code operations using hooks.
Available Hooks
Pre-Operation Hooks
- pre-edit: Validate and assign agents before file modifications
- pre-bash: Check command safety and resource requirements
- pre-task: Auto-spawn agents for complex tasks
Post-Operation Hooks
- post-edit: Auto-format code and train neural patterns
- post-bash: Log execution and update metrics
- post-search: Cache results and improve search patterns
MCP Integration Hooks
- mcp-initialized: Persist swarm configuration
- agent-spawned: Update agent roster
- task-orchestrated: Monitor task progress
- neural-trained: Save pattern improvements
Session Hooks
- notify: Custom notifications with swarm status
- session-end: Generate summary and save state
- session-restore: Load previous session state
Configuration
Hooks are configured in .claude/settings.json:
{
"hooks": {
"PreToolUse": [
{
"matcher": "^(Write|Edit|MultiEdit)$",
"hooks": [{
"type": "command",
"command": "npx claude-flow hook pre-edit --file '${tool.params.file_path}'"
}]
}
]
}
}
Benefits
- 🤖 Automatic agent assignment based on file type
- 🎨 Consistent code formatting
- 🧠 Continuous neural pattern improvement
- 💾 Cross-session memory persistence
- 📊 Performance metrics tracking