feat: enhance training reporting and directory management
- Added functionality to store run name and ID in the training results. - Implemented report generation in the `DataManagerRepository`, including methods to create a run directory and generate comprehensive reports. - Updated `TrainModelResult` to include `run_id` and `run_dir` attributes for better tracking of training sessions.
This commit is contained in:
@@ -248,6 +248,14 @@ class Training(SientiaMonitoring):
|
||||
) as run_info:
|
||||
wrapper.store_model(name=train_params.model_name)
|
||||
|
||||
train_result.run_name = run_info.run_name
|
||||
train_result.run_id = run_info.run_id
|
||||
|
||||
train_result = self.data_manager_repository.generate_report(
|
||||
train_result,
|
||||
metadata=metadata,
|
||||
)
|
||||
|
||||
model_saved = True
|
||||
|
||||
return {
|
||||
|
||||
Reference in New Issue
Block a user