Hugging Face's logo

YCheetax
/
esm2_t12_35M_UR50D-finetuned-localization

Text Classification
Transformers
PyTorch
TensorBoard
esm
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
YCheetax commited on
Commit
bbf4ef5
•
1 Parent(s): 63c87f2

Model save

Browse files
Files changed (1) hide show
  1. README.md +8 -105
README.md CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [facebook/esm2_t12_35M_UR50D](https://huggingface.co/facebook/esm2_t12_35M_UR50D) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 4.3200
21
- - F1: 0.0208
22
 
23
  ## Model description
24
 
@@ -47,112 +47,15 @@ The following hyperparameters were used during training:
47
  - total_eval_batch_size: 24
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
- - num_epochs: 100
51
 
52
  ### Training results
53
 
54
- | Training Loss | Epoch | Step | Validation Loss | F1 |
55
- |:-------------:|:-----:|:----:|:---------------:|:------:|
56
- | No log | 1.0 | 5 | 9.0954 | 0.0 |
57
- | No log | 2.0 | 10 | 9.0396 | 0.0 |
58
- | No log | 3.0 | 15 | 8.9842 | 0.0180 |
59
- | No log | 4.0 | 20 | 8.9289 | 0.0208 |
60
- | No log | 5.0 | 25 | 8.8736 | 0.0208 |
61
- | No log | 6.0 | 30 | 8.8179 | 0.0208 |
62
- | No log | 7.0 | 35 | 8.7617 | 0.0208 |
63
- | No log | 8.0 | 40 | 8.7048 | 0.0208 |
64
- | No log | 9.0 | 45 | 8.6472 | 0.0208 |
65
- | No log | 10.0 | 50 | 8.5888 | 0.0208 |
66
- | No log | 11.0 | 55 | 8.5294 | 0.0208 |
67
- | No log | 12.0 | 60 | 8.4687 | 0.0208 |
68
- | No log | 13.0 | 65 | 8.4071 | 0.0208 |
69
- | No log | 14.0 | 70 | 8.3442 | 0.0208 |
70
- | No log | 15.0 | 75 | 8.2803 | 0.0208 |
71
- | No log | 16.0 | 80 | 8.2152 | 0.0208 |
72
- | No log | 17.0 | 85 | 8.1491 | 0.0208 |
73
- | No log | 18.0 | 90 | 8.0819 | 0.0208 |
74
- | No log | 19.0 | 95 | 8.0136 | 0.0208 |
75
- | No log | 20.0 | 100 | 7.9443 | 0.0208 |
76
- | No log | 21.0 | 105 | 7.8741 | 0.0208 |
77
- | No log | 22.0 | 110 | 7.8029 | 0.0208 |
78
- | No log | 23.0 | 115 | 7.7308 | 0.0208 |
79
- | No log | 24.0 | 120 | 7.6576 | 0.0208 |
80
- | No log | 25.0 | 125 | 7.5834 | 0.0208 |
81
- | No log | 26.0 | 130 | 7.5083 | 0.0208 |
82
- | No log | 27.0 | 135 | 7.4330 | 0.0208 |
83
- | No log | 28.0 | 140 | 7.3573 | 0.0208 |
84
- | No log | 29.0 | 145 | 7.2820 | 0.0208 |
85
- | No log | 30.0 | 150 | 7.2068 | 0.0208 |
86
- | No log | 31.0 | 155 | 7.1315 | 0.0208 |
87
- | No log | 32.0 | 160 | 7.0560 | 0.0208 |
88
- | No log | 33.0 | 165 | 6.9807 | 0.0208 |
89
- | No log | 34.0 | 170 | 6.9055 | 0.0208 |
90
- | No log | 35.0 | 175 | 6.8307 | 0.0208 |
91
- | No log | 36.0 | 180 | 6.7565 | 0.0208 |
92
- | No log | 37.0 | 185 | 6.6831 | 0.0208 |
93
- | No log | 38.0 | 190 | 6.6100 | 0.0208 |
94
- | No log | 39.0 | 195 | 6.5380 | 0.0208 |
95
- | No log | 40.0 | 200 | 6.4667 | 0.0208 |
96
- | No log | 41.0 | 205 | 6.3962 | 0.0208 |
97
- | No log | 42.0 | 210 | 6.3268 | 0.0208 |
98
- | No log | 43.0 | 215 | 6.2585 | 0.0208 |
99
- | No log | 44.0 | 220 | 6.1912 | 0.0208 |
100
- | No log | 45.0 | 225 | 6.1249 | 0.0208 |
101
- | No log | 46.0 | 230 | 6.0597 | 0.0208 |
102
- | No log | 47.0 | 235 | 5.9954 | 0.0208 |
103
- | No log | 48.0 | 240 | 5.9325 | 0.0208 |
104
- | No log | 49.0 | 245 | 5.8706 | 0.0208 |
105
- | No log | 50.0 | 250 | 5.8102 | 0.0208 |
106
- | No log | 51.0 | 255 | 5.7507 | 0.0208 |
107
- | No log | 52.0 | 260 | 5.6925 | 0.0208 |
108
- | No log | 53.0 | 265 | 5.6356 | 0.0208 |
109
- | No log | 54.0 | 270 | 5.5795 | 0.0208 |
110
- | No log | 55.0 | 275 | 5.5251 | 0.0208 |
111
- | No log | 56.0 | 280 | 5.4714 | 0.0208 |
112
- | No log | 57.0 | 285 | 5.4190 | 0.0208 |
113
- | No log | 58.0 | 290 | 5.3679 | 0.0208 |
114
- | No log | 59.0 | 295 | 5.3180 | 0.0208 |
115
- | No log | 60.0 | 300 | 5.2690 | 0.0208 |
116
- | No log | 61.0 | 305 | 5.2213 | 0.0208 |
117
- | No log | 62.0 | 310 | 5.1749 | 0.0208 |
118
- | No log | 63.0 | 315 | 5.1301 | 0.0208 |
119
- | No log | 64.0 | 320 | 5.0863 | 0.0208 |
120
- | No log | 65.0 | 325 | 5.0441 | 0.0208 |
121
- | No log | 66.0 | 330 | 5.0031 | 0.0208 |
122
- | No log | 67.0 | 335 | 4.9627 | 0.0208 |
123
- | No log | 68.0 | 340 | 4.9235 | 0.0208 |
124
- | No log | 69.0 | 345 | 4.8856 | 0.0208 |
125
- | No log | 70.0 | 350 | 4.8490 | 0.0208 |
126
- | No log | 71.0 | 355 | 4.8138 | 0.0208 |
127
- | No log | 72.0 | 360 | 4.7801 | 0.0208 |
128
- | No log | 73.0 | 365 | 4.7474 | 0.0208 |
129
- | No log | 74.0 | 370 | 4.7158 | 0.0208 |
130
- | No log | 75.0 | 375 | 4.6857 | 0.0208 |
131
- | No log | 76.0 | 380 | 4.6567 | 0.0208 |
132
- | No log | 77.0 | 385 | 4.6289 | 0.0208 |
133
- | No log | 78.0 | 390 | 4.6021 | 0.0208 |
134
- | No log | 79.0 | 395 | 4.5767 | 0.0208 |
135
- | No log | 80.0 | 400 | 4.5527 | 0.0208 |
136
- | No log | 81.0 | 405 | 4.5299 | 0.0208 |
137
- | No log | 82.0 | 410 | 4.5085 | 0.0208 |
138
- | No log | 83.0 | 415 | 4.4883 | 0.0208 |
139
- | No log | 84.0 | 420 | 4.4691 | 0.0208 |
140
- | No log | 85.0 | 425 | 4.4511 | 0.0208 |
141
- | No log | 86.0 | 430 | 4.4344 | 0.0208 |
142
- | No log | 87.0 | 435 | 4.4188 | 0.0208 |
143
- | No log | 88.0 | 440 | 4.4043 | 0.0208 |
144
- | No log | 89.0 | 445 | 4.3909 | 0.0208 |
145
- | No log | 90.0 | 450 | 4.3786 | 0.0208 |
146
- | No log | 91.0 | 455 | 4.3674 | 0.0208 |
147
- | No log | 92.0 | 460 | 4.3574 | 0.0208 |
148
- | No log | 93.0 | 465 | 4.3488 | 0.0208 |
149
- | No log | 94.0 | 470 | 4.3412 | 0.0208 |
150
- | No log | 95.0 | 475 | 4.3348 | 0.0208 |
151
- | No log | 96.0 | 480 | 4.3296 | 0.0208 |
152
- | No log | 97.0 | 485 | 4.3255 | 0.0208 |
153
- | No log | 98.0 | 490 | 4.3226 | 0.0208 |
154
- | No log | 99.0 | 495 | 4.3207 | 0.0208 |
155
- | 6.1728 | 100.0 | 500 | 4.3200 | 0.0208 |
156
 
157
 
158
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [facebook/esm2_t12_35M_UR50D](https://huggingface.co/facebook/esm2_t12_35M_UR50D) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 1.8621
21
+ - F1: 0.0444
22
 
23
  ## Model description
24
 
 
47
  - total_eval_batch_size: 24
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
+ - num_epochs: 3
51
 
52
  ### Training results
53
 
54
+ | Training Loss | Epoch | Step | Validation Loss | F1 |
55
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|
56
+ | 2.2383 | 1.0 | 14099 | 2.5252 | 0.0102 |
57
+ | 1.6872 | 2.0 | 28198 | 2.0152 | 0.0355 |
58
+ | 1.4838 | 3.0 | 42297 | 1.8621 | 0.0444 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
59
 
60
 
61
  ### Framework versions