* refactor(Evaluate): print out accuracy, f1, etc. for the final & meta predictions, separated out evaluation step
* fix(Linter): ran
* fix(Tests): syntax change
* fix(Inference): runs now again
* fix(Linter): ran
* feat(Labeling): purge overlapping events, sort dataframe at loading time
* fix(Linter): ran
* refactor(Labeling): moved purge_overlapping_events one abstraction level higher
* fix(Data): renamed class
* fix(Data): corrected parameter name
* fix(Config): parameters
* fix(Data): fixed path
* fix(Data): uncommented required code
* feat(EventFilters): use vol based CUSUM
* fix(Config): only retrain every 2000 samples
* fix(Config): filter out even more events
* fix(Inference): added remove_overlapping_events
* refactor(Types): simplified type hierarchy
* feat(Project): use 5 minute data, running training in parallel, sped up cusum filter by 10x with numba
* fix(WalkForward): inference mini-batch parallelization
* fix(WalkForward): don't use the parallel version of any of the functions
* feat(CI): download the data required
* fix(Project): 5min_crypto folder added
* fix(Evaluate): make sure we have numerical stability in returns
* feat(Models): use SKLearn models directly to enable composability
* feat(Inference): batched inference now working, added forecasting_horizon
* fix(Inference): works again
* fix(Inference)
* chore(Models): remove unused Ensemble model
* fix(Labeller): don't just forward shift returns, also take the sum of the data happened until then
* Update test.yml
* fix(MetaLabeling): previously misinterpreted meta-labeling, now also multiplying base model's prediction with the meta model's prediction
* fix(Evaluate): print results
* fix(Evaluate): make sure we have numerical stability in returns
* fix(Inference): only output and print stats in training mode
* fix(Evaluate): don't add miniscule amount to result
* refactor(Training): added InferenceResult & TrainedModel types
* refactor(Pipeline): introduced TrainingOutcome, BetSizingWithMetaOutcome, etc.
* fix(Pipeline): getting it to compile
* refactor(WalkForward): separate preprocessing step
* feat(Pipeline): separate out transformations processing step
* refactor(Pipeline): use the Directional model terminology, put bet_sizing into pipeline instead of hiding it in a step
* refactor(WalkForward): moved functions to separate folder
* fix(WalkForward): use sparse array to store models, process transformations in parallel (lot faster)
* fix(Tests): and evaluation
* fix(Tests): for realz
* fix(Inference): preloading everything now, renamed primary models to directional models
* fix(BetSizing): was running transformations on the wrong data, oops
* fix(BetSizing): concatenated on the wrong axis accidentally
* fix(Reporting): able to use the new Stats type
* fix(BetSizing): renamed int column names
* fix(Portfolio): name the column properly
* fix(Reporting): rename the correct Series, lol
* fix(Inference): walk_forwad_inference() can deal with models not being aligned with the starting index
* fix(WalkForward): accidentally using the wrong index
* fix(WalkForward): use the correct indicies to fetch last model/transformations
* fix(CI): changed the name of the results
* refactor(Config): use a Config object instead of dictionary of dictionaries!
* fix(Config): use default_ensemble_config
* fix(Portfolio): fixed portfolio construction
* refactor(Project): removed regression method (we can still use regression models, but we'll need map them to classes later)
* fix(Training): removed mistakenly left in `method` parameter
* fix, feat: Fixed inference processing data. Add transformation attribute.
* feat: Added transformations step, refractored the loop to make more sense (divided the train and inference loop).
* feat: Truncated models over time and transformations over time. Fixed some typing aswell.
* fix: Fixed a number of out of array problems.
* feat: Inference now works!
* fix(Steps): runtime error not checking for None
* fix(Steps): preloaded transformers are not optional anymore, sped up training by temporary increasing the retrain_every
* fix(CI): disable ray memory monitoring
* refactor(Inference): removed truncate_models and replaced it with filling X with NaN until inference should start
* feat(Inference): added index_from parameter
* fix(Tests): walk_forward test
* refactor(Pipeline): only predict one asset
* refactor(Inference): removed select_models step, inference code moved to run_inference.py so it matches convention (similar to run_pipeline.py)
* fix(Evaluation): adjust transaction costs
* fix(Config): adjusted retrain_every
Co-authored-by: Daniel Szemerey <szemereydaniel@gmail.com>
Co-authored-by: Mark Aron Szulyovszky <mark.szulyovszky@gmail.com>
* feat(Transformations): removed feature-selection pre-processing step completely
* fix(Core): removed unnecessary `original_X`
* fix(Transformations): use the X_expanding_window to transform subsequent data
* fix(RFE): should check for model correctly
* fix(Config): only re-train the model every 40 timestamp
* fix(MetaLabeling): pass in the correct X to meta-labeling step
* fix(Transformation): PCA should at least keep as many features as sliding_window_size
* feat(Transformations): cache transformations across the same asset
* fix(Tests): missing preloaded_transformations arg
* chore(Config): got rid of unnecessary 'classification_models' and 'regression_models' dictionary keys
* feat: Added optional import of models.
* fix: Models weren't wrapped into abstract class, fixed it.
* chore: Deleted leftover comments.
* fix: Same merge commit as on remote.
* fix: System wasn't putting in RF because there was no differentiation between RF as regressor and RF as classificator.
* fix(Models): use the XGBoostModel wrapper
Co-authored-by: Daniel Szemerey <szemereydaniel@gmail.com>
Co-authored-by: Mark Aron Szulyovszky <mark.szulyovszky@gmail.com>
* feat: Basic pipeline extended.
* feat: Added conversion of model list to existing structure (model_name, model_in_time). Fixed loading of previous models and dicts.
* fix: Had an unfinished function.
* fix: Inference wasn't getting model_over_time. Now transformations are not getting it either yet.
Co-authored-by: Daniel Szemerey <szemereydaniel@gmail.com>
* feat(Transformations): added Transformations abstraction & handling in walk_forward_train() & inference()
* fix(WalkForward): use Dataframes to call Transformation.fit_transform()
* feat(WalkForward): restored option for models to recieve unscaled data
* fix(Transformations): output DataFrame as expected
* fix(Tests): missing new property
* refr: Took out main primary and secondary loops and data processing.
* feat: Tidied the code up.
* feat: Saving models and results now works in a type safe way.
* fix: There was error in the saving function.
* chore: Took out some remaining comments.
* fix: Fixed the previous data checking process.
* feat: Fixed model selection method. I will continue the inference after we merged.
Co-authored-by: Daniel Szemerey <szemereydaniel@gmail.com>
* refactor(WalkForward): separate train / test functions (draft) to potentially help with inference later
* fix(Training): use the new separate train / test functions
* feat(Training): return and pass in scalers that are necessary for inference
* fix(Project): runtime errors
* fix(WalkForward): use the correct `train_from` value
* fix(Tests): for new walk_forward functions()
* refactor(WalkForward): rename `walk_forward_test()` to `walk_forward_inference()`
* feat: Basic scaffolding up for inference process after training.
* feat: Saving and loading models works. Inference works nearly.
* feat: Added inference pipeline.
* feat: Saving model now accoring to date and time; loading models now selects from latest file. Fixed the creation of dictionary of models.
* feat: Added lightweight asset config, but full pipeline.
* feat: Added new naming for dictionary.
* fix: Fixed dictionary naming convention.
* fix: Fixed naming again, now the model structure is good
* fix: Changed the output path and the return values from run_pipeline.
* feat: Added function to make sure folder exists for output models.
Co-authored-by: Daniel Szemerey <szemereydaniel@gmail.com>
* feat(Sweep): try to filter out some not great models
* fix(Sweep): yaml
* fix(Sweep): yaml
* fix(Config): remove some models that do not perform well
* feat(Reporting): added vectorbt-based backtest
* fix(Reporting): added transaction costs
* feat(Reporting): added ability to rebalance only every n days
* feat(Dependencies): added pytorch
* fix(Dependencies): added pytorch-lightning
* feat(CI): added portfolio reporting step
* feat(Reporting): save weights as well
* fix(Reporting): start with less cash
* fix(FeatureExtractor): use a rolling z-score instead of StandardScaler with unavoidable lookahead bias
* chore(Archive): removed archived models
* fix(FeatureExtractors): syntax
* fix(FeatureExtractors): mistake with expanding window
* fix(Evaluation): correlation test should work on a per asset level, not per model level
* fix(Evaluation): correlations series initalized correctly
* fix(Reporting): don't name the run after the incorrectly supposed model_type
* fix(Reporting): put back send_report_to_wandb() into its original place
* fix(CI): sending reports again in comment
* refactor(Naming): use `primary_models` & `meta_labeling_models`
* refactor(Naming): using primary * meta_labeling across config and in pipeline
* feat(Pipeline): added back Ensemble models
* fix(Pipeline): compiler error
* fix(Config): typo
* chore(Pipeline): removed unused averaging step
* revert the changes in discretizing
* chore(Pipeline): remove sharpe improvement logging
* fix(Pipeline): ensemble predictions should be a pd.Series instead of a DataFrame
* fix(Pipeline): discard unnecessary ensemble_probabilities
* fix(Pipeline): fixes regarding various meta-labeling ensemble bugs
* fix(Reporting): use the new naming convention
* fix(Reporting): use the right variable
* feat(Sweep): new sweep for ensemble models
* fix(Sweep): config reference
* fix(Config): simplified dev config
* fix(Models): use the faster LR model
* fix(Models): use LGBM in the meta-labeling model for speed
* fix(Selection): always use the first model for feature selection, commented out caching from select_features() as it's close to redundant in terms of speed
* feat(Models): added lightGBM, moved other models to separate files
* feat(Models): added non-working statsmodel wrapper
* fix(Models): added work-in-progress comment to StatsModels
* feat(DataLoader): added load_only_returns() method
* feat(Portfolio): load predictions
* feat(Portfolio): normalize weights
* feat(Portfolio): started integrating with portfoliobt
* feat(Portfolio): include fees in the portfolio construction
* feat(Portfolio): demo of pyportfolioopt
* feat(Portfolio): get efficient frontier calculation to work
* feat(Portfolio): add a few strategies to create weights
* chore(Dependencies): remove pyportfolioopt for now
* fix(Dependencies): try to install all dependencies with pip
* fix(Dependencies): indentation
* fix(Dependencies): corrected pytorch module name
* fix(Dependencies): try to have as many modules installed by conda for the sake of sanity?
* fix(Dependencies): put fracdiff into pip modules
* fix(Dependencies): revert to using pip almost exclusively
* feat(Portfolio): added alphalens
* fix(Portfolio): got limited weights working
* feat(Portfolio): trying to get alphalens to work
* feat(Portfolio): alphalens working
* fix(Dependencies): removed vectorbt
* fix(Dependencies): use alphalens-reloaded
* fix(Dependencies): added conda source for alphalens-reloaded
* refactor(Portfolio): removed traces of vectorbt
* feat(Reporting): factor reporting done
* feat(Portfolio): added pyfolio reporting (fails bc alphalens is not working properly lol)
Mark Aron Szulyovszky