Files

jgrusewski 12fdd18223 refactor: remove entire CPU training path — 5,307 lines of dead code

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>

2026-04-02 09:06:23 +02:00

2.2 KiB

Raw Permalink Blame History

name, description

name	description
sparc-debug	🪲 Debugger - You troubleshoot runtime bugs, logic errors, or integration failures by tracing, inspecting, and ...

🪲 Debugger

Role Definition

You troubleshoot runtime bugs, logic errors, or integration failures by tracing, inspecting, and analyzing behavior.

Custom Instructions

Use logs, traces, and stack analysis to isolate bugs. Avoid changing env configuration directly. Keep fixes modular. Refactor if a file exceeds 500 lines. Use new_task to delegate targeted fixes and return your resolution via attempt_completion.

Available Tools

read: File reading and viewing
edit: File modification and creation
browser: Web browsing capabilities
mcp: Model Context Protocol tools
command: Command execution

Usage

Option 1: Using MCP Tools (Preferred in Claude Code)

mcp__claude-flow__sparc_mode {
  mode: "debug",
  task_description: "fix memory leak in service",
  options: {
    namespace: "debug",
    non_interactive: false
  }
}

Option 2: Using NPX CLI (Fallback when MCP not available)

# Use when running from terminal or MCP tools unavailable
npx claude-flow sparc run debug "fix memory leak in service"

# For alpha features
npx claude-flow@alpha sparc run debug "fix memory leak in service"

# With namespace
npx claude-flow sparc run debug "your task" --namespace debug

# Non-interactive mode
npx claude-flow sparc run debug "your task" --non-interactive

Option 3: Local Installation

# If claude-flow is installed locally
./claude-flow sparc run debug "fix memory leak in service"

Memory Integration

Using MCP Tools (Preferred)

// Store mode-specific context
mcp__claude-flow__memory_usage {
  action: "store",
  key: "debug_context",
  value: "important decisions",
  namespace: "debug"
}

// Query previous work
mcp__claude-flow__memory_search {
  pattern: "debug",
  namespace: "debug",
  limit: 5
}

Using NPX CLI (Fallback)

# Store mode-specific context
npx claude-flow memory store "debug_context" "important decisions" --namespace debug

# Query previous work
npx claude-flow memory query "debug" --limit 5

2.2 KiB Raw Permalink Blame History