gitadmin/foxhunt
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
Files
b10cdebfb17201c1395615d7be63ebbea720623e
foxhunt/crates/ml/src
History
jgrusewski b10cdebfb1 fix(ml): wire OFI features into GPU batch state construction 
The GPU path (build_batch_states / build_state_tensor) only concatenated
market[40] + portfolio[3] = 43 dims, while state_dim was set to 51 when
OFI was enabled. This caused shape mismatch [128,43] vs [51,1024].

- Add ofi_features: Option<Tensor> field to DqnGpuData
- Add upload_ofi() method for MBP-10 order book features (8 dims/bar)
- Concatenate OFI in build_batch_states: [count,40]+[count,3]+[count,8]=[count,51]
- Concatenate OFI in build_state_tensor: [1,40]+[1,3]+[1,8]=[1,51]
- Wire trainer to call upload_ofi() after GPU data upload
- Fix CPU path to zero-pad regime_features when OFI enabled but data missing

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-07 20:16:32 +01:00
..
asset_selection
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
backtesting
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
benchmark
fix(dqn,ppo): prevent 45-action collapse with entropy, exploration, and monitoring fixes
2026-03-03 02:04:22 +01:00
checkpoint
cleanup: delete 92 scattered .md files and .serena artifacts
2026-03-01 22:42:36 +01:00
common
feat(ml): wire PPO to 45 factored actions via FactoredAction
2026-03-02 15:41:23 +01:00
config
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
cuda_common
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
cuda_pipeline
fix(ml): wire OFI features into GPU batch state construction
2026-03-07 20:16:32 +01:00
data_loaders
feat(ml): real trade data pipeline for VPIN/Kyle's Lambda + offline RL
2026-03-07 19:14:46 +01:00
data_pipeline
refactor: delete 22 dead examples, 4 CSVs, consolidate data to test_data/
2026-02-26 01:24:02 +01:00
data_validation
refactor(ml): rename real_data_loader to data_loader
2026-03-02 15:41:23 +01:00
deployment
docs(ml): clarify edge-case comments from final review
2026-03-03 09:26:53 +01:00
diffusion
fix(ml): add ensure_training_dtype boundary casts to all model forward methods
2026-03-03 17:42:41 +01:00
dqn
feat(ml): real trade data pipeline for VPIN/Kyle's Lambda + offline RL
2026-03-07 19:14:46 +01:00
ensemble
feat(ml): real trade data pipeline for VPIN/Kyle's Lambda + offline RL
2026-03-07 19:14:46 +01:00
evaluation
fix(ml): improve DQN backtest Sharpe — greedy eval, exposure-scaled reward, cost alignment
2026-03-06 20:25:42 +01:00
explainability
feat(ml): BF16 VarBuilder, checkpoints, and training tensors for PPO
2026-03-03 16:16:07 +01:00
features
feat(ml): real trade data pipeline for VPIN/Kyle's Lambda + offline RL
2026-03-07 19:14:46 +01:00
flash_attention
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
gpu
feat(ml): re-enable OFI features from MBP-10 order book data
2026-03-07 13:00:07 +01:00
gpu_benchmarks
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
hyperopt
feat(ml): wire OFI data dirs into hyperopt DQN adapter
2026-03-07 19:36:42 +01:00
integration
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
kan
fix(ml): add ensure_training_dtype boundary casts to all model forward methods
2026-03-03 17:42:41 +01:00
labeling
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
liquid
fix(ml): add ensure_training_dtype boundary casts to all model forward methods
2026-03-03 17:42:41 +01:00
mamba
fix(ml): add ensure_training_dtype boundary casts to all model forward methods
2026-03-03 17:42:41 +01:00
memory_optimization
feat(ml): comprehensive GPU saturation audit — 58 fixes across all 10 models
2026-03-02 15:41:23 +01:00
metrics
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
microstructure
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
model_registry
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
observability
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
optimizers
feat(training): move metrics to common::metrics, fix all clippy errors
2026-03-01 15:56:43 +01:00
paper_trading
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
ppo
refactor(ml): eliminate f64↔f32 round-trips and dead deps across ML crate
2026-03-04 18:10:04 +01:00
regime
fix(stubs): implement 15 production stubs, fix routing, delete placeholders
2026-02-28 23:47:31 +01:00
regime_detection
fix(ml): address final review findings — rename, mutability, bounds
2026-03-03 05:23:31 +01:00
registry
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
risk
fix(ml): address final review findings — rename, mutability, bounds
2026-03-03 05:23:31 +01:00
safety
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
security
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
stress_testing
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
tests
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
tft
fix(ml): add ensure_training_dtype boundary casts to all model forward methods
2026-03-03 17:42:41 +01:00
tgnn
fix(ml): add ensure_training_dtype boundary casts to all model forward methods
2026-03-03 17:42:41 +01:00
tlob
fix(ml): add ensure_training_dtype boundary casts to all model forward methods
2026-03-03 17:42:41 +01:00
trainers
fix(ml): wire OFI features into GPU batch state construction
2026-03-07 20:16:32 +01:00
training
feat(ml): comprehensive GPU saturation audit — 58 fixes across all 10 models
2026-03-02 15:41:23 +01:00
transformers
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
types
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
universe
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
validation
fix(ml): use exposure index (0-4) in DQN experience storage and tracking
2026-03-06 16:27:17 +01:00
xlstm
fix(ml): add ensure_training_dtype boundary casts to all model forward methods
2026-03-03 17:42:41 +01:00
.gitignore
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
batch_processing.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
benchmarks.rs
refactor(ml): consolidate duplicate Rainbow DQN agent implementations
2026-03-04 18:27:42 +01:00
bridge.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
cuda_compat.rs
feat(ml): comprehensive GPU saturation audit — 58 fixes across all 10 models
2026-03-02 15:41:23 +01:00
data_loader.rs
fix(ml): BUG #40 — epsilon stuck at 1.0 with noisy nets (100% random actions)
2026-03-04 23:13:25 +01:00
error.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
examples.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
feature_cache.rs
fix(ml): DQN hyperopt overhaul — DSR reward, dead features, C51/noisy/network fixes
2026-03-07 12:08:39 +01:00
gradient_accumulation.rs
fix(ml): implement actual gradient clipping in PPO (was warn-only)
2026-02-27 10:03:42 +01:00
gradient_utils.rs
fix(ml): add ensure_training_dtype boundary casts to all model forward methods
2026-03-03 17:42:41 +01:00
inference_validator.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
inference.rs
fix(ml): remove hidden fake feature importance from inference path
2026-03-02 15:41:23 +01:00
lib.rs
refactor: remove trading_engine dep from ml and risk crates
2026-03-05 23:19:38 +01:00
model_factory.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
model_loader_integration.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
model_registry.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
model.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
performance.rs
fix(ml): ignore perf benchmark tests in debug builds
2026-02-26 18:09:01 +01:00
portfolio_transformer.rs
feat(ml): BF16 VarBuilder, checkpoints, and training tensors for PPO
2026-03-03 16:16:07 +01:00
preprocessing.rs
feat(ml): expand DQN hyperopt search space 27D→31D + GPU-batched eval inference
2026-03-05 20:45:13 +01:00
tensor_ops.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
training_pipeline.rs
feat(ml): comprehensive GPU saturation audit — 58 fixes across all 10 models
2026-03-02 15:41:23 +01:00
training.rs
refactor: delete 4,800 lines of dead ML code and remove MLFeatureExtractor dependency
2026-03-01 18:29:43 +01:00
traits.rs
refactor: restructure repo — crates/, bin/, testing/ layout
2026-02-25 11:56:00 +01:00
walk_forward.rs
feat(ml): real trade data pipeline for VPIN/Kyle's Lambda + offline RL
2026-03-07 19:14:46 +01:00