Split model overhead into two constants: MODEL_OVERHEAD_MB (pure
model weights for batch-size capping) and TRIAL_VRAM_MB (total
per-trial VRAM for concurrent hyperopt planning). DQN trials
empirically consume ~7 GB each on L40S (model + GPU replay buffer +
experience collector + CUDA allocations + fragmentation), not the
200 MB previously estimated. This caused plan_hyperopt to compute
128 concurrent trials instead of the actual 5, inflating PSO
particles from 20→128 and total trials from 20→384 via .max()
instead of .min(), guaranteeing a 4h timeout kill.
Fix auto-scaling to: (1) match particles to GPU concurrency for
maximum hardware utilization on any node, (2) cap particles at
max_trials to never inflate the trial budget, (3) never auto-inflate
the total trial count.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The auto-detect heuristic used cpus/2 ("smart cap"), designed for
CPU-bound workloads. DQN/PPO trials are GPU-bound — each rayon
thread submits CUDA kernels and waits on cudaDeviceSynchronize(),
using minimal CPU. cpus-1 is the correct cap.
On L40S-1-48G (8 vCPU): 3 threads → 7 threads (2.3× more trials).
Also bumps CI CPU limit 7500m→8000m to expose all 8 cores.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The batch span processor needs a tokio runtime for gRPC transport and
periodic flush. Async services already have one via #[tokio::main], but
sync training binaries (hyperopt, train, evaluate) don't.
Previous approach (making binaries async with #[tokio::main]) caused
"Cannot start a runtime from within a runtime" panics because the ML
crate's internal code creates its own tokio runtimes for block_on().
New approach: build_otel_tracer() detects runtime context via
Handle::try_current(). If absent, it creates a dedicated 1-worker
multi-thread runtime stored in a process-lifetime OnceLock. The worker
thread actively polls the OTLP batch export task.
Reverts training binaries to sync fn main() so internal runtime creation
(hyperopt adapters, DQN/PPO trainers) continues working as before.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
All 6 training binaries (hyperopt_baseline_rl, hyperopt_baseline_supervised,
train_baseline_rl, train_baseline_supervised, evaluate_baseline,
evaluate_supervised) used sync fn main() but the OTLP batch exporter
requires a tokio runtime (tonic/hyper-util gRPC transport). This caused
an immediate panic on CI when OTEL_EXPORTER_OTLP_ENDPOINT was set.
Fix: #[tokio::main(flavor = "current_thread")] on all 6 binaries.
Also fix pre-existing clippy warnings (shadow, let_underscore_must_use,
doc_markdown, cognitive_complexity, integer_division, unsafe_code).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Both DQN and PPO eval paths used old 3-action index matching (0=Buy,
1=Sell, 2=Hold). Now uses FactoredAction.target_exposure() for
exposure-weighted returns and order-type-specific transaction costs.
PPO path had .to_int() which doesn't exist on FactoredAction.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- DQN PER: defer td_errors to_vec1() after loss.to_scalar() — piggyback on
existing pipeline flush instead of forcing premature GPU→CPU stall
- PPO trajectories: capacity-hint Vec allocations, extend_flat_states methods,
states_flat field on TrajectoryBatch for zero-copy GPU upload
- TGGN validate(): batch N per-sample losses on GPU → single to_scalar() sync
(was N GPU→CPU syncs)
- Liquid backward(): batch grad-norm per-param sqr().sum_all() on GPU → single
to_scalar() sync (was N GPU→CPU syncs per optimizer step)
- Liquid validate(): same N→1 GPU sync reduction as TGGN
- DQN trainer: restore EpochPrefetcher/DoubleBufferedLoader API (wrongly deleted)
- train_baseline_rl: wire DoubleBuffer GPU pre-upload — after CPU prefetch
completes, immediately upload next fold to GPU via DqnGpuData::upload() so
next fold starts with data already resident on GPU
2478 tests pass, 0 clippy warnings, 0 compile errors.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Emit set_eval_metrics (directional_accuracy, sharpe, profit_factor, return)
per fold for supervised model evaluation. Start metrics server on :9094.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Emit set_eval_metrics (win_rate, sharpe, profit_factor, return) per fold
for DQN/PPO evaluation. Start metrics server on :9094. Feeds training
cockpit Grafana dashboard eval panels.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace tracing_subscriber::fmt() with init_observability() which adds
JSON structured logging + optional OTLP export to Tempo. When
OTEL_EXPORTER_OTLP_ENDPOINT env var is unset, falls back to fmt-only.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Emit set_epoch, set_epoch_loss, set_validation_loss, set_iteration_seconds
after each DQN/PPO fold completes. Feeds training cockpit Grafana dashboard.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Previously train_baseline_rl.rs only passed commission (tx_cost_bps) to
DQN/PPO trainers, ignoring bid-ask spread slippage. Now computes per-fold
average spread via spread_cost_bps() (same as evaluate_baseline) and passes
total cost (commission + spread) to both trainers.
Removes #[allow(dead_code)] — function is now used by all 4 example binaries.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace raw DQN::new() + manual training loop in the walk-forward
training binary with DQNTrainer, which automatically activates:
- Mixed precision (BF16/F16 auto-detected from GPU)
- Dynamic batch sizing (AutoBatchSizer + HardwareBudget)
- Gradient accumulation
- Full Rainbow DQN (PER, dueling, C51, noisy nets, n-step)
- Regime-conditional Q-networks
- Portfolio tracking, Kelly sizing, entropy regularization
The walk-forward fold structure (data loading, feature extraction,
window generation, normalization) stays in the binary — only per-fold
training delegates to DQNTrainer::train_with_preloaded_data().
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Quality audit found 2 dead code paths in the GPU optimization commit:
1. Data caching: preload_data() was defined on all 10 hyperopt adapters
but never called. Now wired in both hyperopt binaries (RL + supervised)
before the trial loop. Each model preloads training data once into
Arc<Vec<...>>, eliminating per-trial disk I/O.
2. PPO mixed precision: config.mixed_precision was stored but never used
in forward passes. Added forward_mixed() to PolicyNetwork and
ValueNetwork (same BF16/FP16 pattern as DQN's NetworkLayers). Stored
on network structs and auto-applied via forward(). Wired in
PPO::with_device() for MLP networks.
Also fixes missing mixed_precision field in 2 test files and
trading_service PPOConfig literal.
5 files changed, +152/-20. 2418 tests pass, workspace compiles clean.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Wire BF16/FP16 mixed precision end-to-end for DQN and PPO with auto-detection
from GPU name (Ampere+ → BF16, Volta/Turing → FP16). Add hidden_dim_base to
hyperopt and wire through training/eval binaries. Reduce GPU sync points: make
DQN NaN checks periodic (every 100 steps), replace PPO GAE GPU round-trip with
pure CPU implementation. Cache training data across hyperopt trials for all 10
models via Arc. Batch DQN experience storage (128x fewer lock acquisitions).
Correct VRAM constants and batch bounds for all 9 supervised model adapters.
28 files changed, +1207/-208 lines. 2418 tests pass.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Move Prometheus training metrics from example-local baseline_common/
to common::metrics::{server,training_metrics} following the existing
grpc_metrics.rs pattern. Fix 29 let_underscore_must_use clippy errors
in push_metrics.rs, 3 shadow lint errors in training binaries, and
demote gradient clipping log from warn to debug.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add metrics server lifecycle (init, port 9094, active_workers) to both
hyperopt_baseline_rl and hyperopt_baseline_supervised. All 18 metrics
are registered and exposed; inner PSO trial loops can be instrumented
incrementally.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Create shared baseline_common/metrics.rs module that registers all 18
dashboard-expected metrics (11 gauges, 5 counters, 2 histograms) and
spawns a lightweight HTTP metrics server on port 9094.
Instrument train_baseline_supervised and train_baseline_rl with:
- Epoch progress, training/validation loss gauges
- Checkpoint save timing, size, and failure counters
- NaN/gradient explosion detection counters
- Data loading latency histograms
- Active workers lifecycle (1 on start, 0 on exit)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Remove artificial swarm_size cap from plan_hyperopt() that limited
concurrent trials to n_particles (default 20). Now concurrency is
purely VRAM-driven with a hardware cap of 128 threads.
optimize_parallel() auto-scales n_particles to match GPU budget:
- L4 24GB: ~65 concurrent DQN trials (was 20)
- H100 80GB: 128 concurrent DQN trials (was 20)
- CPU/small GPU: falls back to configured n_particles
max_trials scales proportionally to ensure 3+ PSO iterations
for convergence with larger swarms.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Auto-detect now considers both CPU count and GPU VRAM when choosing
concurrent trial count. Prevents OOM on smaller GPUs.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The previous "auto-detect" logic forced CPU which was wrong — GPU is
faster for forward/backward even with parallel trials (DQN/PPO use
<350MB of 24GB VRAM across 7 threads).
Changes:
- Remove --device flag, always require CUDA GPU
- Add DQNTrainer::new_with_device() to share CUDA context across trials
- Propagate hyperopt device to internal DQN trainer (was ignoring it)
- PPO/DQN adapters error on missing GPU instead of silent CPU fallback
- Downgrade batch-size clamping from warn to debug
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
DQN/PPO networks are tiny (3 layers × 128 neurons). Running parallel
hyperopt on GPU wastes cores because CUDA context serializes across
threads — 5 trials on L4 only used 2000m of 6000m requested CPU.
Changes:
- Add --device flag to hyperopt_baseline_rl (auto/cpu/cuda)
- Auto mode forces CPU for parallel runs (no CUDA contention)
- CPU mode uses all available cores (no 2-core reserve)
- Add with_device() builder to DQN/PPO hyperopt trainers
- Downgrade "portfolio value <= 0" and GPU utilization warnings
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Enable concurrent trial evaluation for DQN/PPO hyperparameter
optimization via clone-per-particle pattern — each PSO particle
clones the trainer and trains independently, replacing the previous
Arc<Mutex> serialization bottleneck. On L4 (8 vCPU) this yields
~4-5x throughput improvement.
Changes:
- DQNTrainer/PPOTrainer: Clone with Arc<AtomicUsize> trial counter
- DQNTrainer: replace unsafe mutable aliasing with Arc<Mutex> for
best_trial tracking
- ArgminOptimizer: add optimize_parallel() with ParallelObjectiveFunction
and scoped-thread LHS evaluation
- CLI: --parallel 0 (auto-detect CPUs-2), --initial-capital 35000,
--tx-cost-bps 0.1 (IBKR ES all-in)
- CI: both hyperopt jobs use --parallel 0 + IBKR ES cost defaults
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Extend hyperopt infrastructure to support all 10 ML models (DQN, PPO +
8 supervised). Previously only TFT and Mamba2 had hyperopt trainers.
- Add HyperparameterOptimizable impl for Liquid, TGGN, TLOB, KAN, xLSTM, Diffusion
- Create shared_data.rs with common data prep utilities (build_flat_pairs,
build_sequence_pairs, write_trial_result_json)
- Extend hyperopt_baseline_supervised binary to dispatch all 8 models
(individual, "both" for tft+mamba2, "all" for all 8)
- Add CI jobs: 7 train-validate + 10 hyperopt jobs for all models
- Fix DiffusionMetrics NaN default, XLSTMMetrics serde, safe indexing
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
evaluate_supervised looked for checkpoints at {models_dir}/{model}_fold{N}_best
but training saves to {models_dir}/{model}/{model}_fold{N}_best (model subdirectory).
Also saves NormStats JSON per fold during training so evaluation uses
training-time normalization instead of computing from test data (data leakage).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New evaluate_supervised binary runs walk-forward inference on supervised
model checkpoints (TFT, Mamba2, etc.), converts directional predictions
to trading signals, and computes Sharpe/MaxDD/WinRate/DirAccuracy.
CI changes:
- train-validate-dqn → train-validate-rl (trains+evals DQN+PPO)
- train-validate-tft now runs evaluate_supervised after training
- web/api fallback rules added to train-validate and deploy stages
- evaluate_supervised added to compile-services and Docker images
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add dimension validation in DQN, PPO, Mamba2, TGGN, TLOB, Liquid,
KAN, xLSTM, Diffusion constructors (fail-fast on zero-dim inputs
that would cause CUDA_ERROR_INVALID_VALUE at runtime)
- Add num_unknown_features > 0 guard to TFT (temporal input required)
- Fix 12 dead-code/unused warnings in test compilation
- Remove opt-level=3 and codegen-units=1 from target rustflags
(was forcing O3 + single-thread codegen on dev/test builds)
- Remove hardcoded jobs=16 cap (cargo now auto-detects CPU count)
- Switch linker to clang+lld (2-5x faster linking)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The training binary creates flat [batch, 51] feature vectors (one per
bar), but TFTConfig::default() had sequence_length=50, making the
adapter expect [batch, 51*50=2550]. Set sequence_length=1 and
prediction_horizon=1 to match the actual point-wise sample format.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
TFT create_model used TFTConfig::default() values for num_known_features(10)
and num_unknown_features(210) totaling 220, but input_dim was 51 from the
feature extractor. Set both explicitly: known=0, unknown=feature_dim.
S3 uploader now uses path-style requests (required for Scaleway S3) and
explicitly passes AWS credentials from env vars instead of relying on the
instance metadata credential provider (unavailable on Kapsule).
Also fix runner tags lost during session (kapsule, rust, docker restored).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
TFT create_model used TFTConfig::default() values for num_known_features(10)
and num_unknown_features(210) totaling 220, but input_dim was 51 from the
feature extractor. Set both explicitly: known=0, unknown=feature_dim.
S3 uploader now uses path-style requests (required for Scaleway S3) and
explicitly passes AWS credentials from env vars instead of relying on the
instance metadata credential provider (unavailable on Kapsule).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Extend ml_training_service to dispatch GPU training jobs as K8s batch/v1
Jobs, collect results via a Rust sidecar uploader, and support model
promotion with operator approval via fxt CLI.
- K8s dispatcher creates Jobs on gpu-training pool with native sidecar
- training_uploader crate: watches DONE/FAILED marker, uploads to S3,
reports completion via ReportJobCompletion gRPC
- PromotionManager compares metrics, queues better models for approval
- 4 new proto RPCs: ReportJobCompletion, ListPendingPromotions,
ApprovePromotion, RejectPromotion
- fxt commands: train start, model list/approve/reject
- Training binaries write DONE/FAILED markers + metrics.json
- Dockerfile, K8s job template, and CI pipeline updated
- StartTraining gracefully falls back to in-process when outside K8s
- 27 new tests (16 service + 11 promotion), 141 total service tests pass
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds a parent/child GitLab CI pipeline for ML model training:
- Generator script produces per-model hyperopt/train/evaluate jobs
- Parent pipeline (.gitlab-ci-training.yml) with manual trigger
- NFS-backed ReadWriteMany PVC for shared training outputs
- Hyperopt params wired into training binaries (DQN, PPO, TFT, Mamba2)
- Shared DBN loader eliminates duplicate code across hyperopt adapters
- Supervised hyperopt unified to DBN data (was parquet-only)
Pipeline: hyperopt (4 models) → train (10 models) → evaluate ensemble
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Delete 22 dead/placeholder/broken example files (-3,489 lines code)
- Delete 4 tracked CSV files (-1.1M lines, were accidentally committed)
- Move baseline training data default from data/cache/ to test_data/
- Update 5 unified binary defaults, gitignore, k8s upload comment, docs
- Consolidate all training data under test_data/futures-baseline/
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Split the build pipeline: one compile-services job builds all 8 service
binaries with PVC-backed sccache, saves as artifacts. Then 9 Kaniko jobs
just package pre-built binaries into slim runtime images (~30s each).
Before: 9 parallel Kaniko jobs each doing full cargo build --release
(~20min each, no sccache, 9x duplicated dep compilation)
After: 1 compile job with sccache (~5min cached) + 9 package jobs (~30s)
- Add compile stage between test and build
- Add Dockerfile.runtime (minimal debian + pre-built binary)
- Add Dockerfile.web-gateway-runtime (Node dashboard + pre-built binary)
- Keep Dockerfile.training via Kaniko (needs CUDA dev image for H100)
- Remove all SCCACHE_BUCKET build-args from service builds
- Use dir:// context for Kaniko (only sends build-out/ dir, not full repo)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Mass cleanup of crates/ml/:
- Delete 121 dead/broken/superseded example files (-44,098 lines)
- Delete 6 QAT test files (-4,201 lines) — QAT is disabled at runtime
(trainer.rs falls back to FP32 with warning)
- Delete 2 stale markdown files in examples/
- Delete orphaned src/bin/train_tft.rs (unimplemented stub)
- Clean up Cargo.toml: remove stale [[example]] entries, add missing ones
- Fix stale binary references in log_size_test.rs
- Add infra consistency test (35 checks across Dockerfile/train.sh/Cargo.toml)
Remaining examples (7): train_baseline_rl, train_baseline_supervised,
evaluate_baseline, hyperopt_baseline_rl, hyperopt_baseline_supervised,
download_baseline, cuda_test
All 2390 lib tests pass. All examples compile. 35/35 infra checks pass.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
These two files survived the 20-file consolidation in 022036cb.
Both are now fully superseded:
- train_ppo.rs → train_baseline_rl --model ppo
- train_mamba2.rs → train_baseline_supervised --model mamba2
Also updates entrypoint-generic.sh usage examples to reference
the unified binaries (train_baseline_rl, train_baseline_supervised).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add --cache=true --cache-repo to all 12 Kaniko builds
- Cache Docker layers in Scaleway CR (rg.fr-par.scw.cloud/foxhunt-ci/cache)
- Add Docker Hub auth to devcontainer + infra-runner prepare jobs
- First build populates cache; subsequent builds skip base image pulls
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add tx_cost_bps/tick_size/spread_ticks CLI args to tggn, xlstm, diffusion
- Subtract spread + commission from target returns during data prep
- Delete unused validate_model_parameters from train_mamba2_dbn
- Wire validate_training_batch into mamba2 pre-training validation
All 16 training examples now compile with zero warnings.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add missing Mamba2Config fields (early_stopping_enabled, patience, min_delta, min_epochs)
- Replace unwrap_or on f64 (not Option) with direct field access
- Replace .unwrap() on path.to_str() with safe .ok_or_else()
- Fix DQN closure signature: 3 args (epoch, data, is_final), not 2
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Move 17 library crates into crates/, CLI binary into bin/fxt,
consolidate 10 test crates into testing/, split config crate
from deployment config files.
Root directory reduced from 38+ to ~17 directories.
All Cargo.toml paths and build.rs proto refs updated.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
