serii_zad1

'p_' + str(p) for p in percentiles_list]
    if iqr and 25 in percentiles_list and 75 in percentiles_list:
        names.append('iqr')
    if amplitude and 1 in percentiles_list and 99 in percentiles_list:
        names.append('perc_amp')
    if len(x) == 0:
        values = [0 for i in range(len(names))]
        return values, names
    if len(percentiles_list) > 0 and all([0 < q < 100 for q in percentiles_list]):
        values = list(np.percentile(x, percentiles_list))
    else:
        values = []
    if iqr:
        q1 = percentiles_list.index(25)
        q3 = percentiles_list.index(75)
        values.append(values[q3] - values[q1])
    if amplitude and 1 in percentiles_list and 99 in percentiles_list:
        q1 = percentiles_list.index(1)
        q3 = percentiles_list.index(99)
        values.append(values[q3] - values[q1])
        return values, names


def stats_calculate_all(X_data):
    stats_all_names = stats_all_names = ['len', 'min', 'max', 'range', 'mean', 'hmean', 'gmean', 'var', 'std', 'skew',
                                         'kurtosis', 'median', 'mode', 'energy', 'energy_sample', 'snr']

    xnp = np.array(X_data)
    n = len(X_data)
    if n == 0:
        values = [0 for i in range(len(stats_all_names))]
        return values, stats_all_names
    values = []
    vmin = float(min(xnp))
    if vmin < 1:
        offset = 1 + abs(vmin)
    else:
        offset = 0
    vmax = float(max(xnp))
    #vrange = vmax - vmin
    vmean = float(np.mean(xnp))
    vstd = float(np.std(xnp))
    #venergy = float(sum(np.array(xnp) ** 2))
    #venergy_sample = venergy / n
    #snr = 0.0
    #if vstd != 0: snr = vmean / vstd
    values.append(vmin)
    values.append(vmax)
    #values.append(vrange)
    values.append(vmean)
    # values.append(float(sp.hmean(xnp + offset)))
    # values.append(float(sp.gmean(xnp + offset)))
    # values.append(vstd ** 2)
    values.append(vstd)
    # values.append(sp.skew(xnp))
    # values.append(sp.kurtosis(xnp))
    #values.append(np.median(xnp))
    # vmode = sp.mode(xnp)
    # vmode = float(vmode[0][0])
    # values.append(vmode)
    # values.append(venergy)
    # values.append(venergy_sample)
    #values.append(snr)
    return values, stats_all_names


if __name__ == "__main__":
     x = np.array(X_data)

    # vasiot kod tuka
     shift=input()
     w_long=input()
     #print(x.shape)
     #print(x[:10, 0].shape)
     #print(x[:10, 0])
     #print(sorted(x[:10, 0]))
     #print(stats_calculate_all(x[:10, 0]))
     #print(percentiles_all(x[:10, 0], iqr=False, amplitude=False, percentiles_list=[10, 20, 50, 70, 90]))
     #shift = 50
     #w_long=100
     #w_short = 20
     for i in range(w_long, len(X_data), shift):
        row=[]
        for j in range(x.shape[1]):
            #x.shape[1]
            x_winow_long = x[i - w_long:i,j]
            #x_winow_short = x[i - w_short:i, j]
            #p1 = percentiles_all(x_winow_long)
            #p2 = percentiles_all(x_winow_short)
            s1, stat_names = stats_calculate_all(x_winow_long)
            #s2, _ = stats_calculate_all(x_winow_short)
            #s2=round((s1),2)
            #long_MVA = s1[4]
            row+=s1
            #print(s1,row)
            #short_MVA = s2[4]
        row3=[round(r,2) for r in row]
        print(i,row3)