NSS / src /entity /config_entity.py
Kshitijk20's picture
Clean deployment without binary files
a21e473
from datetime import datetime
import os
from src.constant import training_pipeline
class TrainingPipelineConfig:
def __init__(self, timestamp = datetime.now()):
timestamp = timestamp.strftime("%m_%d_%Y_%H_%M_%S")
self.pipeline_name = training_pipeline.PIPELINE_NAME
self.artifact_name = training_pipeline.ARTIFACT_DIR
self.artifact_dir = os.path.join(self.artifact_name, timestamp)
self.model_dir=os.path.join("final_model")
self.timestamp:str = timestamp
class Data_ingestion_config:
def __init__(self, training_pipeline_config:TrainingPipelineConfig):
self.data_ingestion_dir = os.path.join(training_pipeline_config.artifact_dir,training_pipeline.DATA_INGESTION_DIR_NAME)
self.feature_store_file_path = os.path.join(self.data_ingestion_dir, training_pipeline.DATA_INGESTION_FEATURE_STORE_DIR, training_pipeline.FILE_NAME)
self.train_file_path = os.path.join(self.data_ingestion_dir, training_pipeline.DATA_INGESTION_INGESTED_DIR, training_pipeline.TRAIN_FILE_NAME)
self.test_file_path = os.path.join(self.data_ingestion_dir, training_pipeline.DATA_INGESTION_INGESTED_DIR, training_pipeline.TEST_FILE_NAME)
self.database_name = training_pipeline.DATA_INGESTION_DATBASE_NANE
self.collection_name = training_pipeline.DATA_INGESTION_COLLECTION_NAME
self.train_test_split_ratio = training_pipeline.DATA_INGESTION_TRAIN_TEST_SPLIT_RATION
class Data_validation_config:
def __init__(self,training_pipeline_config: TrainingPipelineConfig):
self.data_validation_dir:str = os.path.join(training_pipeline_config.artifact_dir, training_pipeline.DATA_VALIDATION_DIR_NAMR)
self.valid_data_dir:str = os.path.join(self.data_validation_dir, training_pipeline.DATA_VALIDATION_VALID_DIR)
self.invalid_data_dir:str = os.path.join(self.data_validation_dir, training_pipeline.DATA_VALIDATION_INVALID_DIR)
self.valid_train_file_path:str = os.path.join(self.valid_data_dir, training_pipeline.TRAIN_FILE_NAME)
self.valid_test_file_path:str = os.path.join(self.valid_data_dir, training_pipeline.TEST_FILE_NAME)
self.invalid_train_file_path:str = os.path.join(self.invalid_data_dir, training_pipeline.TRAIN_FILE_NAME)
self.invalid_test_file_path:str = os.path.join(self.invalid_data_dir, training_pipeline.TEST_FILE_NAME)
self.drift_report_file_path:str = os.path.join(self.data_validation_dir, training_pipeline.DATA_VALIDATION_DRIFT_REPORT_DIR, training_pipeline.DATA_VALIDATION_DRIFT_REPORT_FILE_NAME)
class Data_transformation_config:
def __init__(self, training_pipeline_config: TrainingPipelineConfig):
self.data_transformation_dir:str = os.path.join(training_pipeline_config.artifact_dir, training_pipeline.DATA_TRANSFORMATION_DIR_NAME)
self.transformed_train_file_path: str = os.path.join( self.data_transformation_dir,training_pipeline.DATA_TRANSFORMATION_TRANSFORMED_DIR_NAME,
training_pipeline.TRAIN_FILE_NAME.replace("csv", "npy"),)
self.transformed_test_file_path: str = os.path.join(self.data_transformation_dir, training_pipeline.DATA_TRANSFORMATION_TRANSFORMED_DIR_NAME,
training_pipeline.TEST_FILE_NAME.replace("csv", "npy"),
)
self.transformed_object_file_path: str = os.path.join( self.data_transformation_dir, training_pipeline.DATA_TRANSFORMATION_TRANSFORMED_OBJECT_DIR,
training_pipeline.PREPROCESSING_OBJECT_FILE_NAME)
class Model_trainer_config:
def __init__(self, training_pipeline_config: TrainingPipelineConfig):
self.model_trainer_dir:str = os.path.join(
training_pipeline_config.artifact_dir, training_pipeline.MODEL_TRAINER_DIR_NAME
)
self.trained_model_file_path:str = os.path.join(
self.model_trainer_dir, training_pipeline.MODEL_TRAINER_MODEL_DIR, training_pipeline.MODEL_TRAINER_MODEL_NAME
)
self.expected_accuracy:float = training_pipeline.MODEL_TRAINER_EXPECTED_SCORE
self.overfitting_underfitting_threshold = training_pipeline.MODEL_TRAINER_OVERFITTING_UNDERFITTING_THRESHOLD