[kaggle] dataframeの容量を圧縮して、メモリー不足を回避する

ソースコード

def reduce_mem_usage(train_data):
    """ iterate through all the columns of a dataframe and modify the data type
    to reduce memory usage.
    """
    start_mem = train_data.memory_usage().sum() / 1024**2
    print('Memory usage of dataframe is {:.2f} MB'.format(start_mem))
    
    for col in train_data.columns:
        col_type = train_data[col].dtype

        if col_type != object:
            c_min = train_data[col].min()
            c_max = train_data[col].max()
            if str(col_type)[:3] == 'int':
                if c_min > np.iinfo(np.int8).min and c_max < np.iinfo(np.int8).max:
                    train_data[col] = train_data[col].astype(np.int8)
                elif c_min > np.iinfo(np.int16).min and c_max < np.iinfo(np.int16).max:
                    train_data[col] = train_data[col].astype(np.int16)
                elif c_min > np.iinfo(np.int32).min and c_max < np.iinfo(np.int32).max:
                    train_data[col] = train_data[col].astype(np.int32)
                elif c_min > np.iinfo(np.int64).min and c_max < np.iinfo(np.int64).max:
                    train_data[col] = train_data[col].astype(np.int64)  
            else:
                if c_min > np.finfo(np.float16).min and c_max < np.finfo(np.float16).max:
                    train_data[col] = train_data[col].astype(np.float16)
                elif c_min > np.finfo(np.float32).min and c_max < np.finfo(np.float32).max:
                    train_data[col] = train_data[col].astype(np.float32)
                else:
                    train_data[col] = train_data[col].astype(np.float64)
        else:
            train_data[col] = train_data[col].astype('category')

    end_mem = train_data.memory_usage().sum() / 1024**2
    print('Memory usage after optimization is: {:.2f} MB'.format(end_mem))
    print('Decreased by {:.1f}%'.format(100 * (start_mem - end_mem) / start_mem))

    return train_data

呼び出し

df = reduce_mem_usage(df)

出力

Memory usage of dataframe is 109.14 MB
Memory usage after optimization is: 79.94 MB
Decreased by 26.8%

出典

How To Fix Your Notebook tried to allocate more memory than available. It has been restarted.