diff --git a/src/aiunn/finetune.py b/src/aiunn/finetune.py index 6ce67b4..0226db4 100644 --- a/src/aiunn/finetune.py +++ b/src/aiunn/finetune.py @@ -15,7 +15,7 @@ from aiia import AIIA, AIIAConfig, AIIABase, AIIABaseShared, AIIAmoe, AIIAchunke class aiuNNDataset(torch.utils.data.Dataset): def __init__(self, parquet_path): # Read the Parquet file - self.df = pd.read_parquet(parquet_path) + self.df = pd.read_parquet(parquet_path).head(2500) # Data augmentation pipeline without Resize as it's redundant self.augmentation = Compose([