Moved use_aug from hyperparameters to separate variable

pipelines/distilbert_trainer.py

@@ -1,4 +1,4 @@
-#usage: python3 -m pipelines.distilbert_trainer --num_epochs 20 --batch_size 8 --num_labels 2 --device 'cuda' --save_path '/tmp' --model_name 'distilbert-base-uncased' --use_aug True
+#usage: python3 -m pipelines.distilbert_trainer --num_epochs 20 --batch_size 8 --num_labels 2 --device 'cuda' --save_path '/tmp' --model_name 'distilbert-base-uncased' --use_aug 'True'
 import sys
 sys.path.append('..')
 
@@ -125,15 +125,15 @@ def data_split(data):
 
     return train, sanity, gold_fraud
 
-def train_model(train_data, hyper_params):
+def train_model(train_data, hyper_params, use_aug=False):
     run = wandb.init(config=hyper_params)
     model = DistilbertModel(**hyper_params)
 
     # #drop train examples with Label=1 and Body less than 4 words
     # train_data = train_data[~((train_data['Label'] == 1) & (train_data['Body'].str.split().str.len() < 4))]
     # train_data = train_data.reset_index(drop=True)
 
-    if hyper_params['use_aug']:
+    if use_aug:
         augmentor = Augmentor()
 
         train_body, train_labels = augmentor(
@@ -256,7 +256,6 @@ def dump_logs_to_wandb(hyper_params, f1_scores, save_path):
         'num_epochs': args.num_epochs,
         'batch_size': args.batch_size,
         'device': args.device,
-        'use_aug': args.use_aug,
     }
 
     # Log in to Weights and Biases
@@ -289,7 +288,7 @@ def dump_logs_to_wandb(hyper_params, f1_scores, save_path):
     train_data, sanity_data, gold_fraud_data = data_split(data)
 
     # Train the model
-    model = train_model(train_data, hyper_params)
+    model = train_model(train_data, hyper_params, use_aug=args.use_aug)
 
     # Test the model
     f1_scores = test_model(train_data, sanity_data, gold_fraud_data, save_path)

pipelines/roberta_trainer.py

@@ -1,4 +1,4 @@
-#usage: python3 -m pipelines.roberta_trainer --num_epochs 20 --batch_size 8 --num_labels 2 --device 'cuda' --save_path '/tmp' --model_name 'roberta-base' --use_aug True
+#usage: python3 -m pipelines.roberta_trainer --num_epochs 20 --batch_size 8 --num_labels 2 --device 'cuda' --save_path '/tmp' --model_name 'roberta-base' --use_aug 'True'
 import sys
 sys.path.append('..')
 
@@ -125,11 +125,11 @@ def data_split(data):
 
     return train, sanity, gold_fraud
 
-def train_model(train_data, hyper_params):
+def train_model(train_data, hyper_params, use_aug=False):
     run = wandb.init(config=hyper_params)
     model = RobertaModel(**hyper_params)
 
-    if hyper_params['use_aug']:
+    if use_aug:
         augmentor = Augmentor()
 
         train_body, train_labels = augmentor(
@@ -252,7 +252,6 @@ def dump_logs_to_wandb(hyper_params, f1_scores, save_path):
         'num_epochs': args.num_epochs,
         'batch_size': args.batch_size,
         'device': args.device,
-        'use_aug': args.use_aug,
     }
 
     # Log in to Weights and Biases
@@ -285,7 +284,7 @@ def dump_logs_to_wandb(hyper_params, f1_scores, save_path):
     train_data, sanity_data, gold_fraud_data = data_split(data)
 
     # Train the model
-    model = train_model(train_data, hyper_params)
+    model = train_model(train_data, hyper_params, use_aug=args.use_aug)
 
     # Test the model
     f1_scores = test_model(train_data, sanity_data, gold_fraud_data, save_path)

pipelines/svm_trainer.py

@@ -1,4 +1,4 @@
-#usage: python3 -m pipelines.svm_trainer --num_labels 2 --C 10 --kernel 'rbf' --save_path '/tmp/model' --use_aug True
+#usage: python3 -m pipelines.svm_trainer --num_labels 2 --C 10 --kernel 'rbf' --save_path '/tmp/model' --use_aug 'True'
 import sys
 sys.path.append('..')
 
@@ -123,15 +123,15 @@ def data_split(data):
 
     return train, sanity, gold_fraud
 
-def train_model(train_data, hyper_params):
+def train_model(train_data, hyper_params, use_aug=False):
     run = wandb.init(config=hyper_params)
     model = SVMModel(**hyper_params)
 
     # #drop train examples with Label=1 and Body less than 4 words
     # train_data = train_data[~((train_data['Label'] == 1) & (train_data['Body'].str.split().str.len() < 4))]
     # train_data = train_data.reset_index(drop=True)
 
-    if hyper_params['use_aug']:
+    if use_aug:
         augmentor = Augmentor()
 
         train_body, train_labels = augmentor(
@@ -240,7 +240,6 @@ def dump_logs_to_wandb(hyper_params, f1_scores, save_path):
         'num_labels': args.num_labels,
         'C': args.C,
         'kernel': args.kernel,
-        'use_aug': args.use_aug,
     }
 
     # Log in to Weights and Biases
@@ -273,7 +272,7 @@ def dump_logs_to_wandb(hyper_params, f1_scores, save_path):
     train_data, sanity_data, gold_fraud_data = data_split(data)
 
     # Train the model
-    model = train_model(train_data, hyper_params)
+    model = train_model(train_data, hyper_params, use_aug=args.use_aug)
 
     # Test the model
     f1_scores = test_and_save_model(train_data, sanity_data, gold_fraud_data, save_path)