Commit Graph

113 Commits

Author SHA1 Message Date
robcaulk
05ed1b544f Working base for reinforcement learning model 2022-08-24 13:00:40 +02:00
robcaulk
96d8882f1e Plug mem leak, add training timer 2022-08-22 13:30:30 +02:00
robcaulk
8961b8d560 merge in inference timer and historic predictions handling improvements. 2022-08-14 20:31:15 +02:00
robcaulk
ad846cdb76 fix lock bug, update docstring 2022-08-14 20:24:29 +02:00
robcaulk
58de20af0f make BaseClassifierModel. Add predict_proba to lightgbm 2022-08-13 20:07:31 +02:00
robcaulk
31be707cc8 clean up code, add docstrings 2022-08-13 19:40:24 +02:00
robcaulk
3e38c1b0bd take dynamic sized tail off historic_predictions as return dataframe to strategy. 2022-08-13 19:40:24 +02:00
robcaulk
b1b76a2dbe debug classifier with predict proba 2022-08-13 19:40:24 +02:00
robcaulk
23cc21ce59 add predict_proba to base classifier, improve historic predictions handling 2022-08-13 19:40:24 +02:00
robcaulk
e7de812948 Allow user to user pair_dict for persistent storage of custom data 2022-08-09 16:03:10 +02:00
robcaulk
aef086b02e Improved dict typing, timeframe parser, collect dates associated with training data points 2022-08-09 15:30:25 +02:00
robcaulk
ea64f43e52 bring back auto DF resizing for okx 2022-08-08 01:13:13 +02:00
Matthias
47a30047eb Fix typo 2022-08-06 15:03:10 +02:00
Robert Caulk
c172ce1011 improve flexibility of user defined prediction dataframe 2022-08-06 13:51:19 +02:00
Robert Caulk
07763d0d4f add classifier, improve model naming scheme 2022-08-06 08:33:55 +02:00
Robert Caulk
ce8fbbf743 ensure loading historical df matches frequi indices 2022-08-06 07:25:59 +02:00
Robert Caulk
a3799c4d5d start frequi with historical data if available 2022-08-05 18:27:47 +02:00
Robert Caulk
51a6b4289f improve DBSCAN performance for subsequent trainings 2022-08-04 17:41:58 +02:00
robcaulk
95d3009a95 give user ability to analyze live trade dataframe inside custom prediction model. Add documentation to explain new functionality 2022-08-02 20:14:02 +02:00
robcaulk
946d4c7cfc fix trailing whitespace for flake8 2022-07-31 18:39:46 +02:00
robcaulk
4e68626bcb ensure convolutional window is prepended for frequi consistency 2022-07-31 17:51:19 +02:00
robcaulk
c84d54b35e Fix typing issue, avoid using .get() when unnecessary, convert to fstrings 2022-07-29 08:12:50 +02:00
Matthias
a2a0d35a24 Update missing typehints 2022-07-28 07:07:40 +02:00
Matthias
f31106dc61 Minor comment fixes 2022-07-27 07:27:24 +02:00
Matthias
bc760b7eb2 Simplify small segment in datadrawer 2022-07-26 19:41:49 +02:00
robcaulk
ad25a4cb56 reduce number of pair_dict lookups, remove coin_first 2022-07-26 15:58:40 +02:00
robcaulk
fb4e8430cd isort auto import sorting 2022-07-26 10:51:39 +02:00
robcaulk
e213d0ad55 isolate data_drawer functions from data_kitchen, accommodate tests, add new test 2022-07-26 10:24:14 +02:00
Robert Caulk
c9d46a5237 finish bringing follow_mode up to date 2022-07-25 09:24:40 +02:00
Robert Caulk
ab587747fb first fix for follower path bug 2022-07-24 23:32:24 +02:00
Matthias
70b7a254af Update some areas to use default docstring formatting 2022-07-24 16:51:48 +02:00
robcaulk
50d630a155 remove unnecessary comments from data_drawer.py 2022-07-23 13:35:44 +02:00
robcaulk
c2d6a0e891 add record of contribution to doc and source 2022-07-23 13:04:06 +02:00
robcaulk
40f00196eb use cloudpickle in place of pickle. define Paths once in data_drawer. 2022-07-22 17:37:51 +02:00
robcaulk
e7337728bf add separator in folder name just incase an asset ends in an integer 2022-07-21 11:25:28 +02:00
Robert Caulk
921f3899f0 revert pickle reading for historic predictions 2022-07-17 16:06:36 +02:00
Robert Caulk
41eeb99177 load pickle file for writing 2022-07-17 10:05:21 +02:00
Robert Caulk
9d184586f1 fix bug in historic prediction saving 2022-07-16 21:16:59 +02:00
Wagner Costa Santos
ca2029a46b fix issue when bot restarts with same identifier, does not load prediction history 2022-07-14 18:55:24 -03:00
Robert Caulk
fea63fba12 Fix saving/loading historic predictions 2022-07-12 10:12:50 +02:00
Robert Caulk
8ce6b18318 start collecting indefinite history of predictions. Allow user to generate statistics on these predictions. Direct FreqAI to save these to disk and reload them if available. 2022-07-11 22:01:48 +02:00
robcaulk
ffb39a5029 black formatting on freqai files 2022-07-03 10:59:38 +02:00
robcaulk
106131ff0f Rehaul organization of return values 2022-07-02 18:09:38 +02:00
robcaulk
93e1410ed9 first step toward cleaning output and enabling multimodel training per pair 2022-07-01 14:00:30 +02:00
robcaulk
051b99791d reduce unnecessary verbosity, fix error on first training sweep, add LightGBMPredictionModel 2022-06-26 19:04:23 +02:00
robcaulk
f631ae911b add model expiration feature, fix bug in DI return values 2022-06-17 14:55:40 +02:00
robcaulk
c5de0c49e4 first functional scanning commit 2022-06-16 00:24:18 +02:00
robcaulk
cab8f517b4 add lock to datadrawer 2022-06-07 01:07:30 +02:00
Robert Caulk
2451ed8c88
Quick bug fix 2022-06-06 15:11:54 -06:00
robcaulk
16b4a5b71f rehaul of backend data management - increasing performance by holding history in memory, reducing load on the ratelimit by only pinging exchange once per candle. Improve code readability. 2022-06-03 15:19:46 +02:00
robcaulk
0b0dd8dd80 Force high priority pair to be trained before anything else. 2022-06-01 07:55:05 +02:00
robcaulk
bac4ced382 Ensure follower predictions are persistent and uniquely stored 2022-05-31 14:35:04 +02:00
robcaulk
0306f5ca13 Add autopurge feature so that FreqAI cleans up after itself when it no longer needs old models on disk 2022-05-31 11:58:21 +02:00
robcaulk
45f4f0f603 ensure follower sends back null arrays in case leader hasnt created a model file yet 2022-05-31 01:48:48 +02:00
robcaulk
606f18e5c1 Add follow_mode feature so that secondary bots can be launched with the same identifier and load models trained by the leader 2022-05-30 21:35:48 +02:00
robcaulk
5b4c649d43 detect variable sized dataframes coming from strat, adjust our stored/returned data accordingly 2022-05-30 13:55:46 +02:00
robcaulk
e229902381 fix bug in previous commit 2022-05-30 12:48:22 +02:00
robcaulk
a20651efd8 Increase performance by only predicting on most recent candle instead of full strat provided dataframe. Collect predictions and store them so that we can feed true predictions back to strategy (so that frequi isnt updating historic predictions based on newly trained models). 2022-05-30 11:37:05 +02:00
robcaulk
83dd453723 catch errors occuring on background thread, and make sure to keep the ball rolling. Improve pair retraining queue. 2022-05-28 18:26:19 +02:00
robcaulk
58b5abbaa6 improve multithreaded training queue system 2022-05-24 15:28:38 +02:00
robcaulk
255d35976e add priority metadata to pairs to avoid a sync of train time + train period 2022-05-24 12:58:53 +02:00
robcaulk
059c285425 paying closer attention to managing live retraining on separate thread without affecting prediction of other coins on master thread 2022-05-24 12:01:01 +02:00
robcaulk
b0d2d13eb1 improve data persistence/mapping for live/dry. This accommodates quick reloads after crash and handles multi-pair cleanly 2022-05-23 21:05:05 +02:00