text-dataset-distillation/get_plot.py at master · Harshp1802/text-dataset-distillation · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
import matplotlib.pyplot as plt

# set up matplotlib plot style
plt.style.use('ggplot')
# decrease font size
plt.rcParams.update({'font.size': 7})

vanilla = [17.5967,
53.7366,
51.1595,
55.0001,
54.1534,
52.1684,
57.2255,
54.0067,
58.4196,
54.4811,
56.4682,
55.8747,
54.1779,
53.2649,
52.8565,
55.0286,
52.0682,
53.8368,
53.3479,
52.9864,
56.0524,
54.5246,
54.9186,
55.3477,
56.0842,
55.0182,
]

skiplookup = [16.3981,
52.5351,
52.1985,
54.9383,
51.811,
50.3123,
53.9874,
53.531,
53.5545,
50.9516,
55.3949,
54.2889,
53.6612,
52.9043,
51.5983,
50.3625,
50.424,
54.7583,
48.2386,
51.7404,
56.356,
52.3364,
53.5734,
53.7937,
54.7532,
53.2813,
]

vocab_gumbel = [27.3376,
34.8215,
37.673,
30.3174,
34.8948,
33.0873,
36.7162,
34.4061,
39.3859,
36.9784,
33.509,
33.7688,
27.9969,
35.6353,
38.3488,
37.2604,
34.1665,
36.0762,
28.4481,
28.5977,
39.6989,
38.4121,
35.5306,
39.5799,
31.3746,
38.8703,
]

vocab_softmax = [29.0373,
36.1231,
30.5822,
43.9765,
41.9127,
38.4912,
46.3447,
44.5014,
44.2985,
36.5913,
35.8083,
42.3114,
42.5383,
36.3701,
31.2442,
38.6972,
40.4692,
45.6989,
41.051,
43.5506,
38.9295,
40.7971,
40.2469,
46.1467,
45.9028,
41.192,
]

x_axis_val = list(range(0, 251, 10))

#plot lines
plt.plot(x_axis_val, vanilla, label='vanilla')
plt.plot(x_axis_val, skiplookup, label='skiplookup')
plt.plot(x_axis_val, vocab_softmax, label='vocab_softmax')
plt.plot(x_axis_val, vocab_gumbel, label='vocab_gumbel')
plt.legend()
# x axis label
plt.xlabel('Epochs')
# y axis label
plt.ylabel('F1 (macro))')
# plot title
plt.title('F1 vs epochs for all approaches with 10 distilled images per class')
# save plot
plt.savefig('plot.png', bbox_inches='tight', dpi=300)