Hugging Face's logo

thrunlab
/
t5-large_cola_sp0_ar0

Text Classification
Transformers
PyTorch
t5
Generated from Trainer
Eval Results (legacy)
Model card Files
xet
 Community
1
lukeleeai commited on
Commit
80bc2e6
·
1 Parent(s): d88dc0a

End of training

Browse files
Files changed (2) hide show
  1. README.md +68 -71
  2. pytorch_model.bin +1 -1
README.md CHANGED
@@ -22,7 +22,7 @@ model-index:
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
- value: 0.873046875
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [t5-large](https://huggingface.co/t5-large) on the glue dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 0.4531
36
- - Accuracy: 0.8730
37
 
38
  ## Model description
39
 
@@ -59,79 +59,76 @@ The following hyperparameters were used during training:
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
  - lr_scheduler_type: linear
61
  - lr_scheduler_warmup_steps: 20
62
- - num_epochs: 4
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
67
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
68
- | 0.6179 | 0.05 | 25 | 0.7262 | 0.6913 |
69
- | 0.5561 | 0.11 | 50 | 0.5504 | 0.6961 |
70
- | 0.5041 | 0.16 | 75 | 0.5027 | 0.7996 |
71
- | 0.454 | 0.21 | 100 | 0.5353 | 0.7881 |
72
- | 0.3981 | 0.27 | 125 | 0.5393 | 0.8063 |
73
- | 0.4205 | 0.32 | 150 | 0.5582 | 0.7987 |
74
- | 0.4305 | 0.37 | 175 | 0.4384 | 0.8073 |
75
- | 0.4041 | 0.42 | 200 | 0.5255 | 0.8140 |
76
- | 0.4319 | 0.48 | 225 | 0.5297 | 0.8102 |
77
- | 0.3674 | 0.53 | 250 | 0.4788 | 0.8226 |
78
- | 0.4013 | 0.58 | 275 | 0.5254 | 0.8150 |
79
- | 0.4003 | 0.64 | 300 | 0.4980 | 0.8044 |
80
- | 0.3794 | 0.69 | 325 | 0.4988 | 0.8188 |
81
- | 0.4002 | 0.74 | 350 | 0.4188 | 0.8274 |
82
- | 0.3793 | 0.8 | 375 | 0.4412 | 0.8188 |
83
- | 0.3502 | 0.85 | 400 | 0.4862 | 0.8178 |
84
- | 0.3563 | 0.9 | 425 | 0.4951 | 0.8082 |
85
- | 0.3962 | 0.96 | 450 | 0.5133 | 0.8073 |
86
- | 0.3474 | 1.01 | 475 | 0.4908 | 0.8265 |
87
- | 0.22 | 1.06 | 500 | 0.6653 | 0.8198 |
88
- | 0.2745 | 1.11 | 525 | 0.4787 | 0.8360 |
89
- | 0.2181 | 1.17 | 550 | 0.6265 | 0.8245 |
90
- | 0.283 | 1.22 | 575 | 0.6629 | 0.8265 |
91
- | 0.311 | 1.27 | 600 | 0.4652 | 0.8399 |
92
- | 0.2169 | 1.33 | 625 | 0.5805 | 0.8370 |
93
- | 0.2764 | 1.38 | 650 | 0.5420 | 0.8322 |
94
- | 0.2186 | 1.43 | 675 | 0.5668 | 0.8322 |
95
- | 0.3297 | 1.49 | 700 | 0.6130 | 0.8332 |
96
- | 0.2419 | 1.54 | 725 | 0.4862 | 0.8332 |
97
- | 0.2713 | 1.59 | 750 | 0.4718 | 0.8255 |
98
- | 0.2681 | 1.65 | 775 | 0.4710 | 0.8322 |
99
- | 0.231 | 1.7 | 800 | 0.6525 | 0.8121 |
100
- | 0.2286 | 1.75 | 825 | 0.5772 | 0.8236 |
101
- | 0.2965 | 1.8 | 850 | 0.5093 | 0.8303 |
102
- | 0.2791 | 1.86 | 875 | 0.5118 | 0.8303 |
103
- | 0.276 | 1.91 | 900 | 0.5549 | 0.8313 |
104
- | 0.2555 | 1.96 | 925 | 0.5499 | 0.8255 |
105
- | 0.2361 | 2.02 | 950 | 0.6162 | 0.8245 |
106
- | 0.1706 | 2.07 | 975 | 0.5747 | 0.8380 |
107
- | 0.1566 | 2.12 | 1000 | 1.1363 | 0.8265 |
108
- | 0.2285 | 2.18 | 1025 | 0.9228 | 0.8341 |
109
- | 0.1501 | 2.23 | 1050 | 0.7669 | 0.8284 |
110
- | 0.1484 | 2.28 | 1075 | 0.8961 | 0.8380 |
111
- | 0.1639 | 2.34 | 1100 | 1.1919 | 0.8303 |
112
- | 0.1021 | 2.39 | 1125 | 3.1020 | 0.8351 |
113
- | 0.1543 | 2.44 | 1150 | 3.1121 | 0.8332 |
114
- | 0.208 | 2.49 | 1175 | 0.8105 | 0.8265 |
115
- | 0.1513 | 2.55 | 1200 | 0.6624 | 0.8360 |
116
- | 0.1866 | 2.6 | 1225 | 0.5803 | 0.8255 |
117
- | 0.1504 | 2.65 | 1250 | 0.7874 | 0.8341 |
118
- | 0.1929 | 2.71 | 1275 | 0.6037 | 0.8293 |
119
- | 0.1972 | 2.76 | 1300 | 0.8352 | 0.8217 |
120
- | 0.1037 | 2.81 | 1325 | 1.9576 | 0.8293 |
121
- | 0.1279 | 2.87 | 1350 | 3.1590 | 0.8303 |
122
- | 0.2085 | 2.92 | 1375 | 4.3899 | 0.8284 |
123
- | 0.191 | 2.97 | 1400 | 3.4493 | 0.8341 |
124
- | 0.1469 | 3.03 | 1425 | 2.4196 | 0.8303 |
125
- | 0.1009 | 3.08 | 1450 | 2.8744 | 0.8360 |
126
- | 0.1243 | 3.13 | 1475 | 3.1569 | 0.8380 |
127
- | 0.0695 | 3.18 | 1500 | 4.7343 | 0.8389 |
128
- | 0.3294 | 3.24 | 1525 | 5.9461 | 0.8370 |
129
- | 0.35 | 3.29 | 1550 | 6.4079 | 0.8341 |
130
- | 0.1144 | 3.34 | 1575 | 5.5799 | 0.8313 |
131
- | 0.1151 | 3.4 | 1600 | 5.1090 | 0.8332 |
132
- | 0.1197 | 3.45 | 1625 | 5.1001 | 0.8341 |
133
- | 0.6171 | 3.5 | 1650 | 4.7312 | 0.8360 |
134
- | 0.1307 | 3.56 | 1675 | 4.3671 | 0.8360 |
135
 
136
 
137
  ### Framework versions
 
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
+ value: 0.880859375
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  This model is a fine-tuned version of [t5-large](https://huggingface.co/t5-large) on the glue dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.4179
36
+ - Accuracy: 0.8809
37
 
38
  ## Model description
39
 
 
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
  - lr_scheduler_type: linear
61
  - lr_scheduler_warmup_steps: 20
62
+ - num_epochs: 6
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
67
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
68
+ | 0.5885 | 0.05 | 25 | 0.6751 | 0.6913 |
69
+ | 0.5475 | 0.11 | 50 | 0.5338 | 0.6913 |
70
+ | 0.5122 | 0.16 | 75 | 0.4847 | 0.7919 |
71
+ | 0.4486 | 0.21 | 100 | 0.5089 | 0.7996 |
72
+ | 0.4087 | 0.27 | 125 | 0.5139 | 0.8063 |
73
+ | 0.4022 | 0.32 | 150 | 0.5188 | 0.8035 |
74
+ | 0.4245 | 0.37 | 175 | 0.5196 | 0.7987 |
75
+ | 0.4298 | 0.42 | 200 | 0.6226 | 0.8006 |
76
+ | 0.4326 | 0.48 | 225 | 0.6169 | 0.8015 |
77
+ | 0.4321 | 0.53 | 250 | 0.6173 | 0.7987 |
78
+ | 0.4288 | 0.58 | 275 | 0.4786 | 0.8102 |
79
+ | 0.3914 | 0.64 | 300 | 0.5147 | 0.8054 |
80
+ | 0.3519 | 0.69 | 325 | 0.5691 | 0.8150 |
81
+ | 0.4036 | 0.74 | 350 | 0.4560 | 0.8236 |
82
+ | 0.3706 | 0.8 | 375 | 0.4640 | 0.8245 |
83
+ | 0.3584 | 0.85 | 400 | 0.4605 | 0.8207 |
84
+ | 0.3539 | 0.9 | 425 | 0.4932 | 0.8217 |
85
+ | 0.3982 | 0.96 | 450 | 0.5397 | 0.8073 |
86
+ | 0.3352 | 1.01 | 475 | 0.5490 | 0.8150 |
87
+ | 0.2631 | 1.06 | 500 | 0.9244 | 0.8121 |
88
+ | 0.2992 | 1.11 | 525 | 0.5666 | 0.8169 |
89
+ | 0.2308 | 1.17 | 550 | 0.7285 | 0.8178 |
90
+ | 0.2893 | 1.22 | 575 | 0.6907 | 0.8198 |
91
+ | 0.2809 | 1.27 | 600 | 0.4998 | 0.8140 |
92
+ | 0.2469 | 1.33 | 625 | 0.7260 | 0.8236 |
93
+ | 0.331 | 1.38 | 650 | 0.5812 | 0.8293 |
94
+ | 0.286 | 1.43 | 675 | 0.5102 | 0.8360 |
95
+ | 0.347 | 1.49 | 700 | 0.5696 | 0.8255 |
96
+ | 0.2971 | 1.54 | 725 | 0.4114 | 0.8380 |
97
+ | 0.3048 | 1.59 | 750 | 0.5466 | 0.8169 |
98
+ | 0.3168 | 1.65 | 775 | 0.4787 | 0.8274 |
99
+ | 0.2247 | 1.7 | 800 | 0.7926 | 0.8063 |
100
+ | 0.2666 | 1.75 | 825 | 0.5763 | 0.8274 |
101
+ | 0.2856 | 1.8 | 850 | 0.5131 | 0.8303 |
102
+ | 0.2967 | 1.86 | 875 | 0.4970 | 0.8293 |
103
+ | 0.296 | 1.91 | 900 | 0.5532 | 0.8293 |
104
+ | 0.2828 | 1.96 | 925 | 0.4777 | 0.8274 |
105
+ | 0.2708 | 2.02 | 950 | 0.5433 | 0.8351 |
106
+ | 0.1406 | 2.07 | 975 | 0.6351 | 0.8351 |
107
+ | 0.2046 | 2.12 | 1000 | 0.6058 | 0.8332 |
108
+ | 0.2227 | 2.18 | 1025 | 0.5616 | 0.8408 |
109
+ | 0.1551 | 2.23 | 1050 | 1.0299 | 0.8360 |
110
+ | 0.1465 | 2.28 | 1075 | 0.7842 | 0.8380 |
111
+ | 0.2171 | 2.34 | 1100 | 0.6329 | 0.8437 |
112
+ | 0.1588 | 2.39 | 1125 | 0.7575 | 0.8418 |
113
+ | 0.4245 | 2.44 | 1150 | 0.7603 | 0.8351 |
114
+ | 0.2124 | 2.49 | 1175 | 0.5838 | 0.8447 |
115
+ | 0.2333 | 2.55 | 1200 | 0.4896 | 0.8418 |
116
+ | 0.1943 | 2.6 | 1225 | 0.6343 | 0.8332 |
117
+ | 0.1961 | 2.65 | 1250 | 0.6343 | 0.8284 |
118
+ | 0.1981 | 2.71 | 1275 | 0.6145 | 0.8332 |
119
+ | 0.2151 | 2.76 | 1300 | 0.6335 | 0.8360 |
120
+ | 0.1634 | 2.81 | 1325 | 1.1357 | 0.8399 |
121
+ | 0.1526 | 2.87 | 1350 | 1.0044 | 0.8293 |
122
+ | 0.2096 | 2.92 | 1375 | 0.7761 | 0.8360 |
123
+ | 0.2135 | 2.97 | 1400 | 0.9338 | 0.8351 |
124
+ | 0.155 | 3.03 | 1425 | 3.3297 | 0.8360 |
125
+ | 0.3667 | 3.08 | 1450 | 4.0564 | 0.8370 |
126
+ | 0.5925 | 3.13 | 1475 | 6.7411 | 0.8408 |
127
+ | 0.5866 | 3.18 | 1500 | 7.1940 | 0.8399 |
128
+ | 0.3812 | 3.24 | 1525 | 7.0097 | 0.8351 |
129
+ | 0.1041 | 3.29 | 1550 | 7.0157 | 0.8351 |
130
+ | 0.3451 | 3.34 | 1575 | 6.2653 | 0.8418 |
131
+ | 0.1121 | 3.4 | 1600 | 4.2608 | 0.8485 |
 
 
 
132
 
133
 
134
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:67a3b69134ba8a843ab5ab0ccda32d261b70a5f61028492d3962615a90803a37
3
  size 2955086993
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f79c9ef5ec15efa0c84e8d8739597da4e56fd674206e30cdb7cdadf935470a00
3
  size 2955086993