Use main_process_first in the examples (#581)

examples/by_feature/checkpointing.py

@@ -72,11 +72,13 @@ def tokenize_function(examples):
         return outputs
 
     # Apply the method we just defined to all the examples in all the splits of the dataset
-    tokenized_datasets = datasets.map(
-        tokenize_function,
-        batched=True,
-        remove_columns=["idx", "sentence1", "sentence2"],
-    )
+    # starting with the main process first:
+    with accelerator.main_process_first():
+        tokenized_datasets = datasets.map(
+            tokenize_function,
+            batched=True,
+            remove_columns=["idx", "sentence1", "sentence2"],
+        )
 
     # We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
     # transformers library

examples/by_feature/cross_validation.py

@@ -92,11 +92,13 @@ def tokenize_function(examples):
         return outputs
 
     # Apply the method we just defined to all the examples in all the splits of the dataset
-    tokenized_datasets = datasets.map(
-        tokenize_function,
-        batched=True,
-        remove_columns=["idx", "sentence1", "sentence2"],
-    )
+    # starting with the main process first:
+    with accelerator.main_process_first():
+        tokenized_datasets = datasets.map(
+            tokenize_function,
+            batched=True,
+            remove_columns=["idx", "sentence1", "sentence2"],
+        )
 
     # We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
     # transformers library

examples/by_feature/fsdp_with_peak_mem_tracking.py

@@ -127,11 +127,13 @@ def tokenize_function(examples):
         return outputs
 
     # Apply the method we just defined to all the examples in all the splits of the dataset
-    tokenized_datasets = datasets.map(
-        tokenize_function,
-        batched=True,
-        remove_columns=["idx", "sentence1", "sentence2"],
-    )
+    # starting with the main process first:
+    with accelerator.main_process_first():
+        tokenized_datasets = datasets.map(
+            tokenize_function,
+            batched=True,
+            remove_columns=["idx", "sentence1", "sentence2"],
+        )
 
     # We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
     # transformers library

examples/by_feature/gradient_accumulation.py

@@ -67,11 +67,13 @@ def tokenize_function(examples):
         return outputs
 
     # Apply the method we just defined to all the examples in all the splits of the dataset
-    tokenized_datasets = datasets.map(
-        tokenize_function,
-        batched=True,
-        remove_columns=["idx", "sentence1", "sentence2"],
-    )
+    # starting with the main process first:
+    with accelerator.main_process_first():
+        tokenized_datasets = datasets.map(
+            tokenize_function,
+            batched=True,
+            remove_columns=["idx", "sentence1", "sentence2"],
+        )
 
     # We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
     # transformers library

examples/by_feature/memory.py

@@ -72,11 +72,13 @@ def tokenize_function(examples):
         return outputs
 
     # Apply the method we just defined to all the examples in all the splits of the dataset
-    tokenized_datasets = datasets.map(
-        tokenize_function,
-        batched=True,
-        remove_columns=["idx", "sentence1", "sentence2"],
-    )
+    # starting with the main process first:
+    with accelerator.main_process_first():
+        tokenized_datasets = datasets.map(
+            tokenize_function,
+            batched=True,
+            remove_columns=["idx", "sentence1", "sentence2"],
+        )
 
     # We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
     # transformers library

examples/by_feature/multi_process_metrics.py

@@ -74,11 +74,13 @@ def tokenize_function(examples):
         return outputs
 
     # Apply the method we just defined to all the examples in all the splits of the dataset
-    tokenized_datasets = datasets.map(
-        tokenize_function,
-        batched=True,
-        remove_columns=["idx", "sentence1", "sentence2"],
-    )
+    # starting with the main process first:
+    with accelerator.main_process_first():
+        tokenized_datasets = datasets.map(
+            tokenize_function,
+            batched=True,
+            remove_columns=["idx", "sentence1", "sentence2"],
+        )
 
     # We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
     # transformers library

examples/by_feature/tracking.py

@@ -72,11 +72,13 @@ def tokenize_function(examples):
         return outputs
 
     # Apply the method we just defined to all the examples in all the splits of the dataset
-    tokenized_datasets = datasets.map(
-        tokenize_function,
-        batched=True,
-        remove_columns=["idx", "sentence1", "sentence2"],
-    )
+    # starting with the main process first:
+    with accelerator.main_process_first():
+        tokenized_datasets = datasets.map(
+            tokenize_function,
+            batched=True,
+            remove_columns=["idx", "sentence1", "sentence2"],
+        )
 
     # We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
     # transformers library

examples/complete_nlp_example.py

@@ -91,11 +91,13 @@ def tokenize_function(examples):
         return outputs
 
     # Apply the method we just defined to all the examples in all the splits of the dataset
-    tokenized_datasets = datasets.map(
-        tokenize_function,
-        batched=True,
-        remove_columns=["idx", "sentence1", "sentence2"],
-    )
+    # starting with the main process first:
+    with accelerator.main_process_first():
+        tokenized_datasets = datasets.map(
+            tokenize_function,
+            batched=True,
+            remove_columns=["idx", "sentence1", "sentence2"],
+        )
 
     # We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
     # transformers library

examples/nlp_example.py

@@ -65,11 +65,13 @@ def tokenize_function(examples):
         return outputs
 
     # Apply the method we just defined to all the examples in all the splits of the dataset
-    tokenized_datasets = datasets.map(
-        tokenize_function,
-        batched=True,
-        remove_columns=["idx", "sentence1", "sentence2"],
-    )
+    # starting with the main process first:
+    with accelerator.main_process_first():
+        tokenized_datasets = datasets.map(
+            tokenize_function,
+            batched=True,
+            remove_columns=["idx", "sentence1", "sentence2"],
+        )
 
     # We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
     # transformers library