added sampling for pretraining data

VahidooX · Nov 19, 2023 · aa1a85d · aa1a85d
1 parent ae8741f
commit aa1a85d
Showing 1 changed file with 2 additions and 0 deletions.
diff --git a/scripts/speechlm_sft/sampling_pretraining_data.py b/scripts/speechlm_sft/sampling_pretraining_data.py
@@ -303,6 +303,8 @@
     dataset_handler = open(dataset_path, 'r', encoding='utf-8')
     dataset_handlers[dataset] = dataset_handler
 
+print("Blend to be used:", DATA_BLEND)
+
 with open(OUTPUT_FILE, 'w', encoding='utf-8') as outf:
     datasets = list(DATA_BLEND.keys())
     weights = list(DATA_BLEND.values())