triple_barrier_method_output_function.py

def triple_barrier_method(df, holding_period=10, upper_lower_multipliers=[2, 1]):
    # Set the close price as a cupy array
    close = cp.array(df['Close'].values, dtype=cp.float64)
    # Set the high price as a cupy array
    high = cp.array(df['High'].values, dtype=cp.float64)
    # Set the low price as a cupy array
    low = cp.array(df['Low'].values, dtype=cp.float64)
    # Set the daily volatility as a cupy array
    daily_volatility = cp.array(df['vol'].values, dtype=cp.float64)
    # Set the barriers empty array as a cupy array
    barriers = np.empty((close.shape[0], 6), dtype=cp.float64)
    # Set the upper and lower multipliers as a cupy array
    upper_lower_multipliers = np.array(upper_lower_multipliers, dtype=cp.float64)

    # Set the block size
    block_size = 256
    # Set the number of blocks
    num_blocks = (close.shape[0] - 1) // block_size + 1

    # Compute the barriers output
    triple_barrier_method_cuda[num_blocks, block_size](close, high, low, daily_volatility, upper_lower_multipliers, holding_period, barriers)

    # Transform the barriers array into a cudf dataframe
    barriers_df = cudf.DataFrame(barriers, columns=['days_passed', 'price', 'vert_barrier', 'top_barrier', 'bottom_barrier', 'out'])

    # Set the barrieres index to 
    barriers_df.set_index(df.index[-len(barriers_df.index):], inplace=True)

    # Transform the dataframe from cudf-based to pandas
    barriers_df = barriers_df.to_pandas()

    # Return the prediction feature
    return barriers_df[['out']]