How to properly make the matrix of the parameters combinations

[novoz00]

Hi everyone! It's been a month now I'm giving my first try to serious algotrading. I'm digging vectobt and got stuck on this issue that maybe someone can give some avice on.
I have this strategy on 5m TF, basically relying on a fast and a slow ema, on RSI and on some moving averages of the RSI. Specifically I want to go long when fast_ema is above slow_ema AND when fast_sma_rsi is above slow_ema_rsi AND rsi is oversold... when the opposite I want to go short.
Now... I managed to plot a heatmap but it has a weird diagonal and it's not fully populated. I'm afraid it's not actually backtesting all the combinations of the parameters I want it to it but I can't undestand how and where I'm supposed to make the product of the parameters.
Here is the relevant code. Thanks to anyone that want to give me a hint 🖖

import vectorbt as vbt
import numpy as np
import pandas as pd
from numba import njit

@njit
def apply_funch_nb(rsi, rsi_sma, rsi_ema, ema_fast, ema_slow, long_oversold, long_overbought, short_oversold, short_overbought):
    signals = np.where((ema_fast > ema_slow) & (rsi_sma > rsi_ema) & (rsi < long_oversold), 1, 0) #--------------------------------- LONG
    signals = np.where((rsi > long_overbought) & (signals == 0), 2, signals)     #----------------------------------------------- close LONG
    signals = np.where((ema_fast < ema_slow) & (rsi_sma < rsi_ema) & (rsi > short_overbought) & (signals == 0), -1, signals) #--- SHORT
    signals = np.where((rsi < short_oversold) & (signals == 0), -2, signals)     #----------------------------------------------- close SHORT
    return signals

def custom_ind(close, rsi_window, ema_fast_window, ema_slow_window, 
               sma_rsi_window, ema_rsi_window, long_oversold, long_overbought, 
               short_oversold, short_overbought):
    EMA = vbt.IndicatorFactory.from_talib('EMA')
    SMA = vbt.IndicatorFactory.from_talib('SMA')
    RSI = vbt.IndicatorFactory.from_talib('RSI')
    rsi = RSI.run(close, rsi_window).real.to_numpy()
    rsi_sma = SMA.run(rsi, sma_rsi_window).real.to_numpy()
    rsi_ema = EMA.run(rsi, ema_rsi_window).real.to_numpy()
    ema_fast = EMA.run(close, ema_fast_window).real.to_numpy()
    ema_slow = EMA.run(close, ema_slow_window).real.to_numpy()
    return apply_funch_nb(rsi, rsi_sma, rsi_ema, ema_fast, ema_slow, long_oversold, long_overbought, short_oversold, short_overbought)


ind = vbt.IndicatorFactory(
        class_name = "Combination",
        short_name = "comb",
        input_names = ["close"],
        param_names = ["rsi_window", "ema_fast_window", "ema_slow_window", "sma_rsi_window", "ema_rsi_window", 
                       "long_oversold", "long_overbought", "short_oversold", "short_overbought"],
        output_names = ["value"]
        ).from_apply_func(
                custom_ind,
                close = data['Close'],
                rsi_window = 14,
                ema_fast_window = 20,
                ema_slow_window = 45,
                sma_rsi_window = 9,
                ema_rsi_window = 45,
                long_oversold = 45,
                long_overbought = 80,
                short_oversold = 55,
                short_overbought = 20
                )

res = ind.run(
    close = data['Close'],
    rsi_window =            np.arange(10, 40, step=2, dtype=int),
    ema_fast_window =       np.arange(10, 25, step=1, dtype=int),
    ema_slow_window =       np.arange(30, 60, step=2, dtype=int),
    sma_rsi_window =        np.arange(5, 20,  step=1, dtype=int),
    ema_rsi_window =        np.arange(35, 50, step=1, dtype=int),
    long_oversold =         np.arange(40, 55, step=1, dtype=int),
    long_overbought =       np.arange(70, 85, step=1, dtype=int),
    short_oversold =        np.arange(10, 25, step=1, dtype=int),
    short_overbought =      np.arange(35, 50, step=1, dtype=int)
)

long_entries  = res.value == 1
long_exits    = res.value == 2
short_entries = res.value == -1
short_exits   = res.value == -2

pf = vbt.Portfolio.from_signals(
    close=data['Close'],
    entries=long_entries,
    exits=long_exits,
    short_entries=short_entries,
    short_exits=short_exits,
    fees=0.00055,
    sl_stop=0.0005,
    tp_stop=0.02
)

returns = pf.total_return()
fig = returns.vbt.heatmap(
    x_level = 'comb_sma_rsi_window',
    y_level = 'comb_ema_fast_window',
)
fig.show()