Files
foxhunt/scripts/dashboard_monitor.sh
jgrusewski 35feadf55e 🚀 Wave 160 Phase 6: CUDA Mandatory + TDD Testing + TFT Complete (21 Agents)
## Major Achievements

### 1. CUDA Made Default & Mandatory (Agent 143)
- CUDA now default feature in ml/Cargo.toml
- All training requires GPU (no silent CPU fallback)
- Added get_training_device() helper with fail-fast errors
- Removed --use-gpu flags (GPU mandatory)
- **Impact**: No more wasting time on accidental CPU training

### 2. TFT Training COMPLETE (Agent 144)
-  Training completed successfully in 7.6 minutes
-  Early stopping at epoch 100/200 (best val loss: 0.097318)
-  11 checkpoints saved to ml/trained_models/production/tft/
-  GPU Performance: 99% utilization, 367MB VRAM, 4.4s/epoch
-  10x speedup vs CPU (4.4s vs 43-55s per epoch)
- **Status**: PRODUCTION READY

### 3. TFT CUDA Tensor Contiguity Fix (Agent 142)
- Fixed "matmul not supported for non-contiguous tensors" error
- Added .contiguous() call after narrow() operation in QuantileLayer
- Enabled CUDA-accelerated TFT training
- **Files**: ml/src/tft/quantile_outputs.rs

### 4. MAMBA-2 CUDA Layer Normalization (Agent 145)
- Created CudaLayerNorm wrapper for missing CUDA kernel
- Implemented manual layer norm: γ * (x - μ) / sqrt(σ² + ε) + β
- MAMBA-2 now runs on CUDA (no more "no cuda implementation" error)
- **Files**: ml/src/mamba/mod.rs

### 5. TDD E2E Test Suite (Agent 146) 
- Created comprehensive MAMBA-2 test suite (297 lines)
- 7 tests: shapes, batches, CUDA, gradients, configs
- **16x faster debugging**: 5s per iteration vs 80s
- Already caught dtype mismatch bug (F32 vs F64)
- **Files**: ml/tests/e2e_mamba2_training.rs

## Agent Summary (Agents 126-146)

### Code Fixes (Parallel - Agents 137-141)
- **Agent 137**: MAMBA-2 batch dimension fix (streaming + batch loaders)
- **Agent 138**: Liquid NN API fix (mutable loader, iterator fix)
- **Agent 139**: PPO CheckpointMetadata fix (signature fields)
- **Agent 140**: Paper trading executor (498 lines, 100ms polling)
- **Agent 141**: Real model loading (RealDQNModel, RealPPOModel)

### Infrastructure (Agents 143-146)
- **Agent 143**: CUDA mandatory (Cargo.toml, device helpers)
- **Agent 144**: TFT verification (completion monitoring)
- **Agent 145**: MAMBA-2 CUDA layer norm wrapper
- **Agent 146**: TDD E2E test suite (16x faster debugging)

## Files Modified

### Core ML Infrastructure
- ml/Cargo.toml: Added default = ["minimal-inference", "cuda"]
- ml/src/lib.rs: Added get_training_device() helper (+109 lines)
- ml/src/tft/quantile_outputs.rs: Fixed tensor contiguity
- ml/src/mamba/mod.rs: Added CudaLayerNorm wrapper (+41 lines)

### Training Scripts
- ml/examples/train_tft_dbn.rs: Removed --use-gpu flag
- ml/examples/train_ppo.rs: Removed --use-gpu flag
- ml/examples/train_mamba2_dbn.rs: Forced CUDA-only mode
- ml/examples/train_liquid_dbn.rs: Fixed API usage

### Data Loaders
- ml/src/data_loaders/dbn_sequence_loader.rs: Fixed batch dimensions
- ml/src/data_loaders/streaming_dbn_loader.rs: Fixed batch dimensions

### Trading Service
- services/trading_service/src/paper_trading_executor.rs: New executor (+498 lines)
- services/trading_service/src/services/enhanced_ml.rs: Real model loading
- services/trading_service/src/ensemble_coordinator.rs: Integration

### Tests
- ml/tests/e2e_mamba2_training.rs: New TDD test suite (+297 lines)

### Trainers
- ml/src/trainers/tft.rs: Fixed CheckpointMetadata signature fields

## Performance Metrics

### TFT Training
- Duration: 7.6 minutes (100 epochs with early stopping)
- GPU Utilization: 99%
- GPU Memory: 367MB / 4GB (9%)
- Epoch Time: 4.4 seconds (vs 43-55s on CPU)
- Speedup: 10x vs CPU
- Status:  PRODUCTION READY

### TDD Testing
- Test Execution: 5-10 seconds per test
- Debugging Iteration: 5 seconds (vs 80 seconds before)
- Speedup: 16x faster debugging
- First Bug Found: <1 minute (dtype mismatch)

## Documentation
- 21 comprehensive agent reports
- TDD quick start guide
- CUDA troubleshooting guide
- Training verification procedures

## Next Steps
1. Fix MAMBA-2 dtype mismatch (F32→F64) - 2 minutes
2. Run MAMBA-2 tests until passing - 5-10 minutes
3. Launch full MAMBA-2 training - 200 epochs
4. Launch Liquid NN training

## System Status
- TFT:  COMPLETE (production ready)
- MAMBA-2: 🧪 IN TESTING (TDD suite ready)
- CUDA:  DEFAULT (mandatory for training)
- Tests:  16x faster debugging

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-10-14 23:13:34 +02:00

137 lines
5.1 KiB
Bash
Executable File

#!/bin/bash
# Hyperparameter Tuning Dashboard Monitor
# Real-time status dashboard for all 5 model tuning jobs
clear
echo "=================================================================="
echo " HYPERPARAMETER TUNING DASHBOARD"
echo "=================================================================="
echo "Updated: $(date)"
echo ""
# Check GPU status
echo "--- GPU STATUS ---"
nvidia-smi --query-gpu=index,name,utilization.gpu,memory.used,memory.total,temperature.gpu --format=csv | tail -n +2
echo ""
# Check DQN status
echo "--- DQN TUNING STATUS ---"
if ps -p 3911478 > /dev/null 2>&1; then
DQN_TRIALS=$(grep -c "Trial .* completed" /tmp/tuning_run.log 2>/dev/null || echo "0")
DQN_RUNTIME=$(ps -p 3911478 -o etime --no-headers)
echo "Status: RUNNING (PID 3911478)"
echo "Runtime: $DQN_RUNTIME"
echo "Progress: $DQN_TRIALS/50 trials ($(echo "scale=1; $DQN_TRIALS*100/50" | bc)%)"
LAST_TRIAL=$(grep "Trial .* completed" /tmp/tuning_run.log | tail -1)
echo "Last: $LAST_TRIAL"
else
DQN_TRIALS=$(grep -c "Trial .* completed" /tmp/tuning_run.log 2>/dev/null || echo "0")
echo "Status: COMPLETE"
echo "Completed: $DQN_TRIALS/50 trials"
fi
echo ""
# Check PPO status
echo "--- PPO TUNING STATUS ---"
PPO_PID=$(cat /tmp/ppo_tuning.pid 2>/dev/null)
if [ -n "$PPO_PID" ] && ps -p $PPO_PID > /dev/null 2>&1; then
PPO_TRIALS=$(grep -c "Trial .* completed" /tmp/ppo_tuning_run.log 2>/dev/null || echo "0")
PPO_RUNTIME=$(ps -p $PPO_PID -o etime --no-headers)
echo "Status: RUNNING (PID $PPO_PID)"
echo "Runtime: $PPO_RUNTIME"
echo "Progress: $PPO_TRIALS/50 trials ($(echo "scale=1; $PPO_TRIALS*100/50" | bc)%)"
else
if [ -f /tmp/ppo_tuning_run.log ]; then
PPO_TRIALS=$(grep -c "Trial .* completed" /tmp/ppo_tuning_run.log 2>/dev/null || echo "0")
if [ "$PPO_TRIALS" -gt 0 ]; then
echo "Status: COMPLETE"
echo "Completed: $PPO_TRIALS/50 trials"
else
echo "Status: NOT STARTED"
fi
else
echo "Status: NOT STARTED"
fi
fi
echo ""
# Check TFT status
echo "--- TFT TUNING STATUS ---"
TFT_PID=$(cat /tmp/tft_tuning.pid 2>/dev/null)
if [ -n "$TFT_PID" ] && ps -p $TFT_PID > /dev/null 2>&1; then
TFT_TRIALS=$(grep -c "Trial .* completed" /tmp/tft_tuning_run.log 2>/dev/null || echo "0")
TFT_RUNTIME=$(ps -p $TFT_PID -o etime --no-headers)
echo "Status: RUNNING (PID $TFT_PID)"
echo "Runtime: $TFT_RUNTIME"
echo "Progress: $TFT_TRIALS/50 trials ($(echo "scale=1; $TFT_TRIALS*100/50" | bc)%)"
else
if [ -f /tmp/tft_tuning_run.log ]; then
TFT_TRIALS=$(grep -c "Trial .* completed" /tmp/tft_tuning_run.log 2>/dev/null || echo "0")
if [ "$TFT_TRIALS" -gt 0 ]; then
echo "Status: COMPLETE"
echo "Completed: $TFT_TRIALS/50 trials"
else
echo "Status: NOT STARTED"
fi
else
echo "Status: NOT STARTED"
fi
fi
echo ""
# Check MAMBA-2 status
echo "--- MAMBA-2 TUNING STATUS ---"
MAMBA2_PID=$(cat /tmp/mamba2_tuning.pid 2>/dev/null)
if [ -n "$MAMBA2_PID" ] && ps -p $MAMBA2_PID > /dev/null 2>&1; then
MAMBA2_TRIALS=$(grep -c "Trial .* completed" /tmp/mamba2_tuning_run.log 2>/dev/null || echo "0")
MAMBA2_RUNTIME=$(ps -p $MAMBA2_PID -o etime --no-headers)
echo "Status: RUNNING (PID $MAMBA2_PID)"
echo "Runtime: $MAMBA2_RUNTIME"
echo "Progress: $MAMBA2_TRIALS/50 trials ($(echo "scale=1; $MAMBA2_TRIALS*100/50" | bc)%)"
else
if [ -f /tmp/mamba2_tuning_run.log ]; then
MAMBA2_TRIALS=$(grep -c "Trial .* completed" /tmp/mamba2_tuning_run.log 2>/dev/null || echo "0")
if [ "$MAMBA2_TRIALS" -gt 0 ]; then
echo "Status: COMPLETE"
echo "Completed: $MAMBA2_TRIALS/50 trials"
else
echo "Status: NOT STARTED"
fi
else
echo "Status: NOT STARTED"
fi
fi
echo ""
# Check Liquid status
echo "--- LIQUID TUNING STATUS ---"
LIQUID_PID=$(cat /tmp/liquid_tuning.pid 2>/dev/null)
if [ -n "$LIQUID_PID" ] && ps -p $LIQUID_PID > /dev/null 2>&1; then
LIQUID_TRIALS=$(grep -c "Trial .* completed" /tmp/liquid_tuning_run.log 2>/dev/null || echo "0")
LIQUID_RUNTIME=$(ps -p $LIQUID_PID -o etime --no-headers)
echo "Status: RUNNING (PID $LIQUID_PID)"
echo "Runtime: $LIQUID_RUNTIME"
echo "Progress: $LIQUID_TRIALS/50 trials ($(echo "scale=1; $LIQUID_TRIALS*100/50" | bc)%)"
else
if [ -f /tmp/liquid_tuning_run.log ]; then
LIQUID_TRIALS=$(grep -c "Trial .* completed" /tmp/liquid_tuning_run.log 2>/dev/null || echo "0")
if [ "$LIQUID_TRIALS" -gt 0 ]; then
echo "Status: COMPLETE"
echo "Completed: $LIQUID_TRIALS/50 trials"
else
echo "Status: NOT STARTED"
fi
else
echo "Status: NOT STARTED"
fi
fi
echo ""
echo "=================================================================="
echo "Commands:"
echo " Watch dashboard: watch -n 30 /home/jgrusewski/Work/foxhunt/scripts/dashboard_monitor.sh"
echo " View DQN log: tail -f /tmp/tuning_run.log"
echo " View PPO log: tail -f /tmp/ppo_tuning_run.log"
echo " View status: cat /tmp/tuning_pipeline_status.txt"
echo "=================================================================="