paying closer attention to managing live retraining on separate thread without affecting prediction of other coins on master thread
This commit is contained in:
@@ -24,6 +24,7 @@ class FreqaiDataDrawer:
|
||||
self.pair_dict: Dict[str, Any] = {}
|
||||
# dictionary holding all actively inferenced models in memory given a model filename
|
||||
self.model_dictionary: Dict[str, Any] = {}
|
||||
self.pair_data_dict: Dict[str, Any] = {}
|
||||
self.full_path = full_path
|
||||
self.load_drawer_from_disk()
|
||||
|
||||
|
Reference in New Issue
Block a user