hop_length = 256
win_length = 2048
max_win = 3
mu = 4
Gamma = 1
f_novfn = lambda x, sr: get_superflux_novfn(x, sr, hop_length, win_length, max_win, mu, Gamma)
results = evaluate_tempos(f_novfn, get_acf_dft_tempo, hop_length, 44100)
ipd.HTML(results.to_html(escape=False, float_format='%.2f'))

12 / 20

	names	Ground-Truth Tempos	Estimated Tempos	Close Enough
0	train1.wav	[64.5, 129.5]	258.40	False
1	train2.wav	[83.5, 167.5]	333.42	False
2	train3.wav	[76.5, 153.0]	102.34	False
3	train4.wav	[42.0, 126.0]	126.05	True
4	train5.wav	[68.5, 205.5]	68.00	True
5	train6.wav	[41.0, 82.0]	164.06	False
6	train7.wav	[56.5, 113.5]	112.35	True
7	train8.wav	[74.0, 148.0]	295.31	False
8	train9.wav	[64.5, 129.0]	258.40	False
9	train10.wav	[61.0, 122.5]	61.52	True
10	train11.wav	[70.0, 140.0]	139.67	True
11	train12.wav	[27.0, 54.0]	53.83	True
12	train13.wav	[90.0, 180.0]	181.33	True
13	train14.wav	[65.0, 130.0]	130.83	True
14	train15.wav	[62.0, 186.0]	184.57	True
15	train16.wav	[45.0, 90.5]	181.33	False
16	train17.wav	[45.5, 91.5]	93.96	True
17	train18.wav	[61.0, 121.5]	15.13	False
18	train19.wav	[93.5, 188.0]	187.93	True
19	train20.wav	[115.5, 220.5]	219.91	True