trentuna/candle-annotator
8
0
Fork 0
Code Issues Pull requests Releases Packages Activity Actions
93 commits 1 branch 0 tags 2 MiB
Commit graph

13 commits

Author SHA1 Message Date
Marko Djordjevic
aa81d4f3d0 fix(ml): complete ML pipeline fixes and setup 
- Fix CCI indicator to use HLC prices instead of close only
- Parse datetime column when loading enriched CSV
- Strip timezone from annotation timestamps
- Fix TA-Lib pattern names (CDL3WHITESOLDIERS, CDL3BLACKCROWS)
- Exclude programmatic label columns from training features
- Fix classification report to handle missing classes
- Update MLflow tracking to use localhost:5000
- Grant PostgreSQL permissions to ml_user

Pipeline now runs successfully end-to-end:
- Feature engineering: 2543 rows, 31 columns
- Annotation ingestion: 286 samples
- Training: 89.47% test accuracy with Random Forest
 2026-02-15 21:29:54 +01:00
Marko Djordjevic
ceb4103ec4 fix(ml): parse datetime column and fix TA-Lib pattern names 
- Add parse_dates parameter when loading enriched CSV
- Strip timezone from annotation timestamps to match data
- Fix pattern names: CDLTHREEWHITESOLDIERS -> CDL3WHITESOLDIERS
- Fix pattern names: CDLTHREEBLACKCROWS -> CDL3BLACKCROWS
 2026-02-15 21:13:20 +01:00
Marko Djordjevic
2b86524436 fix(ml): correct CCI indicator signature to use HLC prices 2026-02-15 21:09:47 +01:00
Marko Djordjevic
63486bc7b5 fix(ml): add CCI to hlc_indicators list 
CCI (Commodity Channel Index) requires high, low, and close prices
 2026-02-15 21:08:20 +01:00
Marko Djordjevic
57240d4eea fix(scripts): add created_at timestamps to annotation import 
Set created_at field for both span_label_types and span_annotations
to satisfy NOT NULL constraint
 2026-02-15 19:36:55 +01:00
Marko Djordjevic
a68a681c9b fix(ml): handle date strings in TA-Lib annotation generator 
- Convert date strings to Unix timestamps in load_ohlcv()
- Fix duplicate pattern names (CDL3WHITESOLDIERS/CDL3BLACKCROWS)
- Ensure time column is always integer type
 2026-02-15 19:30:38 +01:00
Marko Djordjevic
847ff67986 feat(ml): add TA-Lib annotation generation and import workflow 
Add complete workflow for using TA-Lib to bootstrap training data:

- generate_talib_annotations.py: Python script to run TA-Lib CDL* functions
  and output span annotations in UI-compatible format
- import_talib_annotations.ts: TypeScript script to import generated
  annotations into the UI database with auto-label-type creation
- npm script 'import-annotations' for easy execution
- TALIB_WORKFLOW.md: Comprehensive guide covering the full cycle:
  * Generate patterns with TA-Lib
  * Import into UI
  * Review and edit in browser
  * Export and train model
  * Compare predictions with TA-Lib detections
  * Iterate for improvement

This enables the intended workflow: use TA-Lib for initial annotations,
manually refine them, then train a model that learns from corrections.
 2026-02-15 19:18:28 +01:00
Marko Djordjevic
3a83fd38e9 feat(ml): implement FastAPI inference service with model loading, preprocessing, and prediction endpoints 2026-02-15 14:29:07 +01:00
Marko Djordjevic
f4c0f9a836 feat(ml): implement training stage with MLflow tracking and model wrappers 
- Create RandomForestModel and XGBoostModel wrappers with class weight support
- Implement temporal and random train/val/test splitting
- Add MLflow experiment tracking with full parameter and metric logging
- Create evaluation module for confusion matrix, feature importance, and classification reports
- Implement model training with sklearn/xgboost flavor logging and optional registry registration
- Store training run metadata in PostgreSQL
- Wire training stage into pipeline.py orchestrator
- Support both RandomForest and XGBoost models with configurable hyperparameters
 2026-02-15 14:22:19 +01:00
Marko Djordjevic
16763b967e feat(ml): implement annotation ingestion with windowed/BIO encoding and TA-Lib patterns 2026-02-15 12:28:58 +01:00
Marko Djordjevic
fd29ab91e0 feat(ml): implement feature engineering pipeline 
- Create pipeline.py with CLI argument parsing for running stages
- Implement TA-Lib indicator computation with multi-output support
- Add candle feature extraction (body_size, wicks, ratios, etc.)
- Create custom feature loader with dynamic module import
- Wire all feature engineering stages with NaN handling
- Tasks completed: 2.2, 2.3, 3.1, 3.2, 3.3, 3.4, 3.5
 2026-02-15 12:22:59 +01:00
Marko Djordjevic
ea339a54a7 feat(ml): add database schema, config parser, and DVC setup 
- Initialize DVC with local storage backend (task 1.6)
- Create PostgreSQL schema for training_runs table (task 1.7)
- Add SQLAlchemy database connection setup (task 1.8)
- Create Pydantic config models for pipeline.yaml (task 2.1)
- Add migration runner for database setup
- Fix pyproject.toml package discovery config
 2026-02-15 12:08:53 +01:00
Marko Djordjevic
1a653c5866 feat: add ML service scaffolding with Python FastAPI, Docker, and MLflow setup 2026-02-15 11:58:31 +01:00