henryscheible commited on
Commit
4060dd0
·
1 Parent(s): eb5b8c9

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -150
README.md CHANGED
@@ -2,26 +2,9 @@
2
  license: mit
3
  tags:
4
  - generated_from_trainer
5
- datasets:
6
- - stereoset
7
- metrics:
8
- - accuracy
9
  model-index:
10
  - name: gpt2_stereoset_classifieronly
11
- results:
12
- - task:
13
- name: Text Classification
14
- type: text-classification
15
- dataset:
16
- name: stereoset
17
- type: stereoset
18
- config: intersentence
19
- split: validation
20
- args: intersentence
21
- metrics:
22
- - name: Accuracy
23
- type: accuracy
24
- value: 0.5023547880690737
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,14 +12,7 @@ should probably proofread and complete it, then remove this comment. -->
29
 
30
  # gpt2_stereoset_classifieronly
31
 
32
- This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the stereoset dataset.
33
- It achieves the following results on the evaluation set:
34
- - Loss: 0.7039
35
- - Accuracy: 0.5024
36
- - Tp: 0.2732
37
- - Tn: 0.2292
38
- - Fp: 0.2857
39
- - Fn: 0.2119
40
 
41
  ## Model description
42
 
@@ -55,7 +31,7 @@ More information needed
55
  ### Training hyperparameters
56
 
57
  The following hyperparameters were used during training:
58
- - learning_rate: 0.0001
59
  - train_batch_size: 64
60
  - eval_batch_size: 64
61
  - seed: 42
@@ -63,129 +39,6 @@ The following hyperparameters were used during training:
63
  - lr_scheduler_type: linear
64
  - num_epochs: 50
65
 
66
- ### Training results
67
-
68
- | Training Loss | Epoch | Step | Validation Loss | Accuracy | Tp | Tn | Fp | Fn |
69
- |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|
70
- | 0.7235 | 0.43 | 20 | 0.7240 | 0.5133 | 0.3163 | 0.1970 | 0.3179 | 0.1688 |
71
- | 0.7247 | 0.85 | 40 | 0.7156 | 0.5204 | 0.2386 | 0.2818 | 0.2331 | 0.2465 |
72
- | 0.7237 | 1.28 | 60 | 0.7167 | 0.5133 | 0.2841 | 0.2292 | 0.2857 | 0.2009 |
73
- | 0.7396 | 1.7 | 80 | 0.7158 | 0.5126 | 0.2896 | 0.2229 | 0.2920 | 0.1954 |
74
- | 0.7144 | 2.13 | 100 | 0.7120 | 0.5267 | 0.2504 | 0.2763 | 0.2386 | 0.2347 |
75
- | 0.7122 | 2.55 | 120 | 0.7108 | 0.5275 | 0.2394 | 0.2881 | 0.2268 | 0.2457 |
76
- | 0.7356 | 2.98 | 140 | 0.7109 | 0.5212 | 0.2567 | 0.2645 | 0.2504 | 0.2284 |
77
- | 0.7232 | 3.4 | 160 | 0.7097 | 0.5228 | 0.1962 | 0.3265 | 0.1884 | 0.2889 |
78
- | 0.6975 | 3.83 | 180 | 0.7105 | 0.5204 | 0.2700 | 0.2504 | 0.2645 | 0.2151 |
79
- | 0.7329 | 4.26 | 200 | 0.7092 | 0.5212 | 0.2410 | 0.2802 | 0.2347 | 0.2441 |
80
- | 0.7158 | 4.68 | 220 | 0.7101 | 0.5110 | 0.2786 | 0.2323 | 0.2826 | 0.2064 |
81
- | 0.7136 | 5.11 | 240 | 0.7086 | 0.5181 | 0.2504 | 0.2677 | 0.2473 | 0.2347 |
82
- | 0.7172 | 5.53 | 260 | 0.7086 | 0.5212 | 0.1648 | 0.3564 | 0.1586 | 0.3203 |
83
- | 0.7328 | 5.96 | 280 | 0.7135 | 0.4874 | 0.3320 | 0.1554 | 0.3595 | 0.1531 |
84
- | 0.7275 | 6.38 | 300 | 0.7096 | 0.5024 | 0.2975 | 0.2049 | 0.3100 | 0.1876 |
85
- | 0.7186 | 6.81 | 320 | 0.7075 | 0.5212 | 0.1672 | 0.3540 | 0.1609 | 0.3179 |
86
- | 0.718 | 7.23 | 340 | 0.7063 | 0.5102 | 0.2339 | 0.2763 | 0.2386 | 0.2512 |
87
- | 0.7102 | 7.66 | 360 | 0.7057 | 0.5126 | 0.2284 | 0.2841 | 0.2308 | 0.2567 |
88
- | 0.7186 | 8.09 | 380 | 0.7091 | 0.4953 | 0.3053 | 0.1900 | 0.3250 | 0.1797 |
89
- | 0.7119 | 8.51 | 400 | 0.7066 | 0.5031 | 0.2567 | 0.2465 | 0.2684 | 0.2284 |
90
- | 0.711 | 8.94 | 420 | 0.7059 | 0.5078 | 0.2418 | 0.2661 | 0.2488 | 0.2433 |
91
- | 0.7183 | 9.36 | 440 | 0.7053 | 0.5141 | 0.2190 | 0.2951 | 0.2198 | 0.2661 |
92
- | 0.7126 | 9.79 | 460 | 0.7051 | 0.5149 | 0.2190 | 0.2959 | 0.2190 | 0.2661 |
93
- | 0.7131 | 10.21 | 480 | 0.7060 | 0.5024 | 0.2645 | 0.2378 | 0.2771 | 0.2206 |
94
- | 0.7037 | 10.64 | 500 | 0.7049 | 0.5 | 0.2535 | 0.2465 | 0.2684 | 0.2316 |
95
- | 0.6955 | 11.06 | 520 | 0.7063 | 0.4890 | 0.2786 | 0.2104 | 0.3046 | 0.2064 |
96
- | 0.7125 | 11.49 | 540 | 0.7058 | 0.4898 | 0.2739 | 0.2159 | 0.2991 | 0.2111 |
97
- | 0.6999 | 11.91 | 560 | 0.7039 | 0.5196 | 0.1915 | 0.3281 | 0.1868 | 0.2936 |
98
- | 0.6898 | 12.34 | 580 | 0.7046 | 0.5063 | 0.2410 | 0.2653 | 0.2496 | 0.2441 |
99
- | 0.7017 | 12.77 | 600 | 0.7061 | 0.4898 | 0.2669 | 0.2229 | 0.2920 | 0.2182 |
100
- | 0.7106 | 13.19 | 620 | 0.7051 | 0.5078 | 0.2473 | 0.2606 | 0.2543 | 0.2378 |
101
- | 0.6795 | 13.62 | 640 | 0.7087 | 0.4929 | 0.3171 | 0.1758 | 0.3391 | 0.1680 |
102
- | 0.7016 | 14.04 | 660 | 0.7044 | 0.5118 | 0.2025 | 0.3093 | 0.2057 | 0.2826 |
103
- | 0.7009 | 14.47 | 680 | 0.7069 | 0.4914 | 0.2975 | 0.1939 | 0.3210 | 0.1876 |
104
- | 0.6967 | 14.89 | 700 | 0.7044 | 0.5047 | 0.2504 | 0.2543 | 0.2606 | 0.2347 |
105
- | 0.7006 | 15.32 | 720 | 0.7046 | 0.5016 | 0.2488 | 0.2527 | 0.2622 | 0.2363 |
106
- | 0.7049 | 15.74 | 740 | 0.7053 | 0.4984 | 0.2653 | 0.2331 | 0.2818 | 0.2198 |
107
- | 0.6988 | 16.17 | 760 | 0.7037 | 0.5086 | 0.2214 | 0.2873 | 0.2276 | 0.2637 |
108
- | 0.717 | 16.6 | 780 | 0.7045 | 0.5149 | 0.1578 | 0.3571 | 0.1578 | 0.3273 |
109
- | 0.6905 | 17.02 | 800 | 0.7116 | 0.4890 | 0.3619 | 0.1272 | 0.3878 | 0.1232 |
110
- | 0.701 | 17.45 | 820 | 0.7041 | 0.5024 | 0.2582 | 0.2441 | 0.2708 | 0.2268 |
111
- | 0.6868 | 17.87 | 840 | 0.7043 | 0.5039 | 0.2575 | 0.2465 | 0.2684 | 0.2276 |
112
- | 0.6866 | 18.3 | 860 | 0.7041 | 0.5055 | 0.2527 | 0.2527 | 0.2622 | 0.2323 |
113
- | 0.6969 | 18.72 | 880 | 0.7058 | 0.5008 | 0.2967 | 0.2041 | 0.3108 | 0.1884 |
114
- | 0.6895 | 19.15 | 900 | 0.7041 | 0.5016 | 0.2559 | 0.2457 | 0.2692 | 0.2292 |
115
- | 0.7065 | 19.57 | 920 | 0.7033 | 0.5071 | 0.2072 | 0.2998 | 0.2151 | 0.2779 |
116
- | 0.6806 | 20.0 | 940 | 0.7058 | 0.4969 | 0.3046 | 0.1923 | 0.3226 | 0.1805 |
117
- | 0.7007 | 20.43 | 960 | 0.7033 | 0.5016 | 0.2363 | 0.2653 | 0.2496 | 0.2488 |
118
- | 0.7004 | 20.85 | 980 | 0.7066 | 0.4906 | 0.3148 | 0.1758 | 0.3391 | 0.1703 |
119
- | 0.7054 | 21.28 | 1000 | 0.7059 | 0.4898 | 0.2959 | 0.1939 | 0.3210 | 0.1892 |
120
- | 0.6933 | 21.7 | 1020 | 0.7034 | 0.5016 | 0.2496 | 0.2520 | 0.2630 | 0.2355 |
121
- | 0.694 | 22.13 | 1040 | 0.7065 | 0.4922 | 0.3171 | 0.1750 | 0.3399 | 0.1680 |
122
- | 0.696 | 22.55 | 1060 | 0.7034 | 0.4961 | 0.2457 | 0.2504 | 0.2645 | 0.2394 |
123
- | 0.6904 | 22.98 | 1080 | 0.7039 | 0.5008 | 0.2653 | 0.2355 | 0.2794 | 0.2198 |
124
- | 0.7095 | 23.4 | 1100 | 0.7044 | 0.4984 | 0.2763 | 0.2221 | 0.2928 | 0.2088 |
125
- | 0.7079 | 23.83 | 1120 | 0.7043 | 0.5008 | 0.2802 | 0.2206 | 0.2943 | 0.2049 |
126
- | 0.6991 | 24.26 | 1140 | 0.7053 | 0.4992 | 0.3069 | 0.1923 | 0.3226 | 0.1782 |
127
- | 0.7001 | 24.68 | 1160 | 0.7032 | 0.4992 | 0.2512 | 0.2480 | 0.2669 | 0.2339 |
128
- | 0.6855 | 25.11 | 1180 | 0.7032 | 0.5016 | 0.2457 | 0.2559 | 0.2590 | 0.2394 |
129
- | 0.6921 | 25.53 | 1200 | 0.7038 | 0.5 | 0.2700 | 0.2300 | 0.2849 | 0.2151 |
130
- | 0.7019 | 25.96 | 1220 | 0.7028 | 0.5016 | 0.2284 | 0.2732 | 0.2418 | 0.2567 |
131
- | 0.707 | 26.38 | 1240 | 0.7036 | 0.4984 | 0.2645 | 0.2339 | 0.2810 | 0.2206 |
132
- | 0.694 | 26.81 | 1260 | 0.7026 | 0.5063 | 0.2394 | 0.2669 | 0.2480 | 0.2457 |
133
- | 0.6928 | 27.23 | 1280 | 0.7040 | 0.5063 | 0.2834 | 0.2229 | 0.2920 | 0.2017 |
134
- | 0.6922 | 27.66 | 1300 | 0.7046 | 0.5086 | 0.2991 | 0.2096 | 0.3053 | 0.1860 |
135
- | 0.7018 | 28.09 | 1320 | 0.7035 | 0.5039 | 0.2677 | 0.2363 | 0.2786 | 0.2174 |
136
- | 0.7018 | 28.51 | 1340 | 0.7042 | 0.5055 | 0.2841 | 0.2214 | 0.2936 | 0.2009 |
137
- | 0.7004 | 28.94 | 1360 | 0.7026 | 0.5031 | 0.2292 | 0.2739 | 0.2410 | 0.2559 |
138
- | 0.7006 | 29.36 | 1380 | 0.7032 | 0.5008 | 0.2527 | 0.2480 | 0.2669 | 0.2323 |
139
- | 0.6912 | 29.79 | 1400 | 0.7043 | 0.5031 | 0.2802 | 0.2229 | 0.2920 | 0.2049 |
140
- | 0.7098 | 30.21 | 1420 | 0.7042 | 0.5071 | 0.2920 | 0.2151 | 0.2998 | 0.1931 |
141
- | 0.6909 | 30.64 | 1440 | 0.7026 | 0.5055 | 0.2473 | 0.2582 | 0.2567 | 0.2378 |
142
- | 0.6822 | 31.06 | 1460 | 0.7025 | 0.5071 | 0.2410 | 0.2661 | 0.2488 | 0.2441 |
143
- | 0.6888 | 31.49 | 1480 | 0.7031 | 0.5024 | 0.2637 | 0.2386 | 0.2763 | 0.2214 |
144
- | 0.6931 | 31.91 | 1500 | 0.7022 | 0.5024 | 0.2237 | 0.2786 | 0.2363 | 0.2614 |
145
- | 0.6885 | 32.34 | 1520 | 0.7050 | 0.5031 | 0.3116 | 0.1915 | 0.3234 | 0.1735 |
146
- | 0.6965 | 32.77 | 1540 | 0.7035 | 0.5047 | 0.2622 | 0.2425 | 0.2724 | 0.2229 |
147
- | 0.7038 | 33.19 | 1560 | 0.7029 | 0.5078 | 0.2512 | 0.2567 | 0.2582 | 0.2339 |
148
- | 0.6867 | 33.62 | 1580 | 0.7026 | 0.5047 | 0.2370 | 0.2677 | 0.2473 | 0.2480 |
149
- | 0.6855 | 34.04 | 1600 | 0.7044 | 0.5031 | 0.3022 | 0.2009 | 0.3140 | 0.1829 |
150
- | 0.6921 | 34.47 | 1620 | 0.7060 | 0.4976 | 0.3242 | 0.1735 | 0.3414 | 0.1609 |
151
- | 0.693 | 34.89 | 1640 | 0.7033 | 0.5126 | 0.2763 | 0.2363 | 0.2786 | 0.2088 |
152
- | 0.6838 | 35.32 | 1660 | 0.7034 | 0.5071 | 0.2684 | 0.2386 | 0.2763 | 0.2166 |
153
- | 0.6931 | 35.74 | 1680 | 0.7031 | 0.5055 | 0.2630 | 0.2425 | 0.2724 | 0.2221 |
154
- | 0.6885 | 36.17 | 1700 | 0.7030 | 0.5055 | 0.2614 | 0.2441 | 0.2708 | 0.2237 |
155
- | 0.6956 | 36.6 | 1720 | 0.7034 | 0.5118 | 0.2786 | 0.2331 | 0.2818 | 0.2064 |
156
- | 0.6829 | 37.02 | 1740 | 0.7031 | 0.5102 | 0.2637 | 0.2465 | 0.2684 | 0.2214 |
157
- | 0.7025 | 37.45 | 1760 | 0.7030 | 0.5063 | 0.2582 | 0.2480 | 0.2669 | 0.2268 |
158
- | 0.6988 | 37.87 | 1780 | 0.7038 | 0.5063 | 0.2857 | 0.2206 | 0.2943 | 0.1994 |
159
- | 0.6868 | 38.3 | 1800 | 0.7032 | 0.5078 | 0.2637 | 0.2441 | 0.2708 | 0.2214 |
160
- | 0.6911 | 38.72 | 1820 | 0.7035 | 0.5055 | 0.2700 | 0.2355 | 0.2794 | 0.2151 |
161
- | 0.6978 | 39.15 | 1840 | 0.7040 | 0.5063 | 0.2873 | 0.2190 | 0.2959 | 0.1978 |
162
- | 0.6945 | 39.57 | 1860 | 0.7036 | 0.5047 | 0.2732 | 0.2316 | 0.2834 | 0.2119 |
163
- | 0.6906 | 40.0 | 1880 | 0.7035 | 0.5071 | 0.2669 | 0.2402 | 0.2747 | 0.2182 |
164
- | 0.6822 | 40.43 | 1900 | 0.7035 | 0.5078 | 0.2637 | 0.2441 | 0.2708 | 0.2214 |
165
- | 0.6933 | 40.85 | 1920 | 0.7042 | 0.5071 | 0.2873 | 0.2198 | 0.2951 | 0.1978 |
166
- | 0.6963 | 41.28 | 1940 | 0.7040 | 0.5047 | 0.2810 | 0.2237 | 0.2912 | 0.2041 |
167
- | 0.6994 | 41.7 | 1960 | 0.7038 | 0.5024 | 0.2724 | 0.2300 | 0.2849 | 0.2127 |
168
- | 0.6836 | 42.13 | 1980 | 0.7037 | 0.5016 | 0.2661 | 0.2355 | 0.2794 | 0.2190 |
169
- | 0.6882 | 42.55 | 2000 | 0.7034 | 0.5 | 0.2527 | 0.2473 | 0.2677 | 0.2323 |
170
- | 0.6953 | 42.98 | 2020 | 0.7036 | 0.5 | 0.2669 | 0.2331 | 0.2818 | 0.2182 |
171
- | 0.7027 | 43.4 | 2040 | 0.7038 | 0.5039 | 0.2732 | 0.2308 | 0.2841 | 0.2119 |
172
- | 0.6951 | 43.83 | 2060 | 0.7037 | 0.5024 | 0.2684 | 0.2339 | 0.2810 | 0.2166 |
173
- | 0.6938 | 44.26 | 2080 | 0.7040 | 0.5047 | 0.2802 | 0.2245 | 0.2904 | 0.2049 |
174
- | 0.6809 | 44.68 | 2100 | 0.7040 | 0.5071 | 0.2810 | 0.2261 | 0.2889 | 0.2041 |
175
- | 0.6886 | 45.11 | 2120 | 0.7036 | 0.5031 | 0.2637 | 0.2394 | 0.2755 | 0.2214 |
176
- | 0.6947 | 45.53 | 2140 | 0.7035 | 0.5039 | 0.2606 | 0.2433 | 0.2716 | 0.2245 |
177
- | 0.694 | 45.96 | 2160 | 0.7035 | 0.5039 | 0.2614 | 0.2425 | 0.2724 | 0.2237 |
178
- | 0.6888 | 46.38 | 2180 | 0.7040 | 0.5071 | 0.2794 | 0.2276 | 0.2873 | 0.2057 |
179
- | 0.6893 | 46.81 | 2200 | 0.7036 | 0.5024 | 0.2637 | 0.2386 | 0.2763 | 0.2214 |
180
- | 0.6873 | 47.23 | 2220 | 0.7038 | 0.5039 | 0.2732 | 0.2308 | 0.2841 | 0.2119 |
181
- | 0.6974 | 47.66 | 2240 | 0.7040 | 0.5078 | 0.2802 | 0.2276 | 0.2873 | 0.2049 |
182
- | 0.6853 | 48.09 | 2260 | 0.7038 | 0.5039 | 0.2732 | 0.2308 | 0.2841 | 0.2119 |
183
- | 0.6974 | 48.51 | 2280 | 0.7038 | 0.5031 | 0.2724 | 0.2308 | 0.2841 | 0.2127 |
184
- | 0.6833 | 48.94 | 2300 | 0.7038 | 0.5039 | 0.2732 | 0.2308 | 0.2841 | 0.2119 |
185
- | 0.692 | 49.36 | 2320 | 0.7038 | 0.5031 | 0.2732 | 0.2300 | 0.2849 | 0.2119 |
186
- | 0.6771 | 49.79 | 2340 | 0.7039 | 0.5024 | 0.2732 | 0.2292 | 0.2857 | 0.2119 |
187
-
188
-
189
  ### Framework versions
190
 
191
  - Transformers 4.26.1
 
2
  license: mit
3
  tags:
4
  - generated_from_trainer
 
 
 
 
5
  model-index:
6
  - name: gpt2_stereoset_classifieronly
7
+ results: []
 
 
 
 
 
 
 
 
 
 
 
 
 
8
  ---
9
 
10
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
12
 
13
  # gpt2_stereoset_classifieronly
14
 
15
+ This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
 
 
 
 
 
 
 
16
 
17
  ## Model description
18
 
 
31
  ### Training hyperparameters
32
 
33
  The following hyperparameters were used during training:
34
+ - learning_rate: 0.0005
35
  - train_batch_size: 64
36
  - eval_batch_size: 64
37
  - seed: 42
 
39
  - lr_scheduler_type: linear
40
  - num_epochs: 50
41
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
42
  ### Framework versions
43
 
44
  - Transformers 4.26.1