Hugging Face's logo

pig4431
/
TweetEval_ELECTRA_5E

Text Classification
Transformers
PyTorch
electra
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
pig4431 commited on
Commit
86490a1
•
1 Parent(s): 8d891aa

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +196 -0
README.md ADDED
@@ -0,0 +1,196 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - tweet_eval
7
+ metrics:
8
+ - accuracy
9
+ model-index:
10
+ - name: TweetEval_ELECTRA_5E
11
+ results:
12
+ - task:
13
+ name: Text Classification
14
+ type: text-classification
15
+ dataset:
16
+ name: tweet_eval
17
+ type: tweet_eval
18
+ config: sentiment
19
+ split: train
20
+ args: sentiment
21
+ metrics:
22
+ - name: Accuracy
23
+ type: accuracy
24
+ value: 0.9066666666666666
25
+ ---
26
+
27
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
28
+ should probably proofread and complete it, then remove this comment. -->
29
+
30
+ # TweetEval_ELECTRA_5E
31
+
32
+ This model is a fine-tuned version of [google/electra-small-discriminator](https://huggingface.co/google/electra-small-discriminator) on the tweet_eval dataset.
33
+ It achieves the following results on the evaluation set:
34
+ - Loss: 0.2935
35
+ - Accuracy: 0.9067
36
+
37
+ ## Model description
38
+
39
+ More information needed
40
+
41
+ ## Intended uses & limitations
42
+
43
+ More information needed
44
+
45
+ ## Training and evaluation data
46
+
47
+ More information needed
48
+
49
+ ## Training procedure
50
+
51
+ ### Training hyperparameters
52
+
53
+ The following hyperparameters were used during training:
54
+ - learning_rate: 1e-05
55
+ - train_batch_size: 16
56
+ - eval_batch_size: 8
57
+ - seed: 42
58
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
59
+ - lr_scheduler_type: linear
60
+ - num_epochs: 5
61
+
62
+ ### Training results
63
+
64
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
65
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
66
+ | 0.6466 | 0.04 | 50 | 0.6006 | 0.7333 |
67
+ | 0.5974 | 0.08 | 100 | 0.5769 | 0.7333 |
68
+ | 0.5884 | 0.12 | 150 | 0.5486 | 0.7333 |
69
+ | 0.5601 | 0.16 | 200 | 0.4799 | 0.76 |
70
+ | 0.5125 | 0.2 | 250 | 0.4380 | 0.8533 |
71
+ | 0.4603 | 0.24 | 300 | 0.4169 | 0.84 |
72
+ | 0.4353 | 0.28 | 350 | 0.3775 | 0.86 |
73
+ | 0.4498 | 0.32 | 400 | 0.3460 | 0.9 |
74
+ | 0.4014 | 0.37 | 450 | 0.3812 | 0.8467 |
75
+ | 0.4072 | 0.41 | 500 | 0.3383 | 0.88 |
76
+ | 0.3891 | 0.45 | 550 | 0.3377 | 0.88 |
77
+ | 0.3482 | 0.49 | 600 | 0.3289 | 0.8933 |
78
+ | 0.3705 | 0.53 | 650 | 0.3162 | 0.8933 |
79
+ | 0.3249 | 0.57 | 700 | 0.2967 | 0.9 |
80
+ | 0.332 | 0.61 | 750 | 0.2925 | 0.8867 |
81
+ | 0.3166 | 0.65 | 800 | 0.2916 | 0.9067 |
82
+ | 0.334 | 0.69 | 850 | 0.3083 | 0.8667 |
83
+ | 0.3039 | 0.73 | 900 | 0.2966 | 0.8867 |
84
+ | 0.3066 | 0.77 | 950 | 0.3054 | 0.88 |
85
+ | 0.3238 | 0.81 | 1000 | 0.3060 | 0.88 |
86
+ | 0.308 | 0.85 | 1050 | 0.3103 | 0.88 |
87
+ | 0.2889 | 0.89 | 1100 | 0.2922 | 0.88 |
88
+ | 0.2773 | 0.93 | 1150 | 0.2986 | 0.8933 |
89
+ | 0.3078 | 0.97 | 1200 | 0.2852 | 0.8933 |
90
+ | 0.2529 | 1.01 | 1250 | 0.2957 | 0.8933 |
91
+ | 0.2968 | 1.06 | 1300 | 0.2893 | 0.8867 |
92
+ | 0.2536 | 1.1 | 1350 | 0.2902 | 0.88 |
93
+ | 0.2836 | 1.14 | 1400 | 0.3085 | 0.88 |
94
+ | 0.3066 | 1.18 | 1450 | 0.2909 | 0.88 |
95
+ | 0.28 | 1.22 | 1500 | 0.2953 | 0.8867 |
96
+ | 0.2549 | 1.26 | 1550 | 0.3019 | 0.8867 |
97
+ | 0.2974 | 1.3 | 1600 | 0.2796 | 0.88 |
98
+ | 0.2808 | 1.34 | 1650 | 0.2762 | 0.9 |
99
+ | 0.2548 | 1.38 | 1700 | 0.2808 | 0.9 |
100
+ | 0.2879 | 1.42 | 1750 | 0.2819 | 0.8933 |
101
+ | 0.2583 | 1.46 | 1800 | 0.2904 | 0.88 |
102
+ | 0.2387 | 1.5 | 1850 | 0.3016 | 0.8733 |
103
+ | 0.2574 | 1.54 | 1900 | 0.2981 | 0.8933 |
104
+ | 0.2589 | 1.58 | 1950 | 0.2907 | 0.8933 |
105
+ | 0.2436 | 1.62 | 2000 | 0.2926 | 0.8867 |
106
+ | 0.2606 | 1.66 | 2050 | 0.2807 | 0.8933 |
107
+ | 0.2841 | 1.7 | 2100 | 0.2805 | 0.9 |
108
+ | 0.2497 | 1.75 | 2150 | 0.2765 | 0.8867 |
109
+ | 0.2866 | 1.79 | 2200 | 0.2821 | 0.9 |
110
+ | 0.2614 | 1.83 | 2250 | 0.2759 | 0.8867 |
111
+ | 0.2605 | 1.87 | 2300 | 0.2704 | 0.8933 |
112
+ | 0.2365 | 1.91 | 2350 | 0.2623 | 0.9 |
113
+ | 0.2274 | 1.95 | 2400 | 0.2651 | 0.8933 |
114
+ | 0.2564 | 1.99 | 2450 | 0.2664 | 0.9 |
115
+ | 0.2481 | 2.03 | 2500 | 0.2706 | 0.9 |
116
+ | 0.2382 | 2.07 | 2550 | 0.2819 | 0.8933 |
117
+ | 0.2351 | 2.11 | 2600 | 0.2848 | 0.9 |
118
+ | 0.18 | 2.15 | 2650 | 0.2881 | 0.8933 |
119
+ | 0.2343 | 2.19 | 2700 | 0.2983 | 0.9 |
120
+ | 0.2043 | 2.23 | 2750 | 0.2908 | 0.8933 |
121
+ | 0.2272 | 2.27 | 2800 | 0.3000 | 0.8867 |
122
+ | 0.246 | 2.31 | 2850 | 0.3136 | 0.8867 |
123
+ | 0.2577 | 2.35 | 2900 | 0.3126 | 0.88 |
124
+ | 0.2316 | 2.39 | 2950 | 0.2803 | 0.8933 |
125
+ | 0.2156 | 2.44 | 3000 | 0.2737 | 0.9067 |
126
+ | 0.223 | 2.48 | 3050 | 0.2883 | 0.8933 |
127
+ | 0.2215 | 2.52 | 3100 | 0.2660 | 0.8867 |
128
+ | 0.2488 | 2.56 | 3150 | 0.2551 | 0.9 |
129
+ | 0.2095 | 2.6 | 3200 | 0.2645 | 0.9 |
130
+ | 0.2247 | 2.64 | 3250 | 0.2751 | 0.8933 |
131
+ | 0.2292 | 2.68 | 3300 | 0.2851 | 0.8867 |
132
+ | 0.237 | 2.72 | 3350 | 0.2824 | 0.8867 |
133
+ | 0.2086 | 2.76 | 3400 | 0.2805 | 0.8867 |
134
+ | 0.2063 | 2.8 | 3450 | 0.2771 | 0.9 |
135
+ | 0.2015 | 2.84 | 3500 | 0.2981 | 0.8933 |
136
+ | 0.2036 | 2.88 | 3550 | 0.2937 | 0.8933 |
137
+ | 0.247 | 2.92 | 3600 | 0.2985 | 0.8933 |
138
+ | 0.23 | 2.96 | 3650 | 0.2866 | 0.9067 |
139
+ | 0.2625 | 3.0 | 3700 | 0.2836 | 0.9 |
140
+ | 0.2064 | 3.04 | 3750 | 0.2911 | 0.8933 |
141
+ | 0.1867 | 3.08 | 3800 | 0.2868 | 0.8933 |
142
+ | 0.2143 | 3.12 | 3850 | 0.2903 | 0.9 |
143
+ | 0.1993 | 3.17 | 3900 | 0.2987 | 0.8933 |
144
+ | 0.1762 | 3.21 | 3950 | 0.3066 | 0.9067 |
145
+ | 0.1935 | 3.25 | 4000 | 0.3185 | 0.8867 |
146
+ | 0.234 | 3.29 | 4050 | 0.3043 | 0.9067 |
147
+ | 0.195 | 3.33 | 4100 | 0.2905 | 0.9067 |
148
+ | 0.2434 | 3.37 | 4150 | 0.3081 | 0.9 |
149
+ | 0.2168 | 3.41 | 4200 | 0.2919 | 0.9067 |
150
+ | 0.2044 | 3.45 | 4250 | 0.2903 | 0.9 |
151
+ | 0.2419 | 3.49 | 4300 | 0.2955 | 0.8933 |
152
+ | 0.191 | 3.53 | 4350 | 0.2957 | 0.9067 |
153
+ | 0.1927 | 3.57 | 4400 | 0.3075 | 0.8933 |
154
+ | 0.2267 | 3.61 | 4450 | 0.2823 | 0.9067 |
155
+ | 0.1971 | 3.65 | 4500 | 0.2933 | 0.9067 |
156
+ | 0.2164 | 3.69 | 4550 | 0.2910 | 0.9067 |
157
+ | 0.1939 | 3.73 | 4600 | 0.2813 | 0.9067 |
158
+ | 0.1834 | 3.77 | 4650 | 0.2913 | 0.9067 |
159
+ | 0.234 | 3.81 | 4700 | 0.2841 | 0.9067 |
160
+ | 0.2226 | 3.86 | 4750 | 0.2888 | 0.9067 |
161
+ | 0.2176 | 3.9 | 4800 | 0.2902 | 0.9067 |
162
+ | 0.2279 | 3.94 | 4850 | 0.2842 | 0.9067 |
163
+ | 0.1948 | 3.98 | 4900 | 0.2856 | 0.9067 |
164
+ | 0.2044 | 4.02 | 4950 | 0.2845 | 0.9067 |
165
+ | 0.2075 | 4.06 | 5000 | 0.2825 | 0.9067 |
166
+ | 0.1721 | 4.1 | 5050 | 0.2796 | 0.9067 |
167
+ | 0.2206 | 4.14 | 5100 | 0.2752 | 0.9067 |
168
+ | 0.2012 | 4.18 | 5150 | 0.2738 | 0.9067 |
169
+ | 0.1868 | 4.22 | 5200 | 0.2932 | 0.9 |
170
+ | 0.2117 | 4.26 | 5250 | 0.2881 | 0.9 |
171
+ | 0.1946 | 4.3 | 5300 | 0.2985 | 0.9 |
172
+ | 0.2138 | 4.34 | 5350 | 0.3025 | 0.8933 |
173
+ | 0.1841 | 4.38 | 5400 | 0.2906 | 0.9067 |
174
+ | 0.2171 | 4.42 | 5450 | 0.2919 | 0.9067 |
175
+ | 0.2116 | 4.46 | 5500 | 0.2889 | 0.9067 |
176
+ | 0.162 | 4.5 | 5550 | 0.2994 | 0.8933 |
177
+ | 0.1821 | 4.55 | 5600 | 0.2975 | 0.9 |
178
+ | 0.1802 | 4.59 | 5650 | 0.2994 | 0.9 |
179
+ | 0.1619 | 4.63 | 5700 | 0.2978 | 0.9 |
180
+ | 0.1955 | 4.67 | 5750 | 0.2984 | 0.9 |
181
+ | 0.2031 | 4.71 | 5800 | 0.2925 | 0.9067 |
182
+ | 0.1937 | 4.75 | 5850 | 0.2939 | 0.9067 |
183
+ | 0.1799 | 4.79 | 5900 | 0.2955 | 0.9067 |
184
+ | 0.2106 | 4.83 | 5950 | 0.2965 | 0.9067 |
185
+ | 0.196 | 4.87 | 6000 | 0.2954 | 0.9067 |
186
+ | 0.2336 | 4.91 | 6050 | 0.2932 | 0.9067 |
187
+ | 0.1805 | 4.95 | 6100 | 0.2931 | 0.9067 |
188
+ | 0.1877 | 4.99 | 6150 | 0.2935 | 0.9067 |
189
+
190
+
191
+ ### Framework versions
192
+
193
+ - Transformers 4.24.0
194
+ - Pytorch 1.13.0
195
+ - Datasets 2.7.1
196
+ - Tokenizers 0.13.2