przybytniowska commited on
Commit
1d5c979
1 Parent(s): e4fb4ba

Training complete

Browse files
Files changed (3) hide show
  1. README.md +54 -0
  2. test_metrics.json +3 -0
  3. train_losses.csv +127 -0
README.md ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model: FacebookAI/roberta-base
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - arrow
8
+ model-index:
9
+ - name: roberta_base_QA_SQUAD_adamw_torch
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # roberta_base_QA_SQUAD_adamw_torch
17
+
18
+ This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on the arrow dataset.
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 2e-05
38
+ - train_batch_size: 32
39
+ - eval_batch_size: 8
40
+ - seed: 42
41
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
+ - lr_scheduler_type: linear
43
+ - num_epochs: 5
44
+
45
+ ### Training results
46
+
47
+
48
+
49
+ ### Framework versions
50
+
51
+ - Transformers 4.34.1
52
+ - Pytorch 2.3.0+cu118
53
+ - Datasets 2.19.0
54
+ - Tokenizers 0.14.1
test_metrics.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "test_accuracy": 0.952
3
+ }
train_losses.csv ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ loss,epoch
2
+ 0.2402,0.04
3
+ 0.1457,0.08
4
+ 0.1223,0.12
5
+ 0.103,0.16
6
+ 0.0809,0.2
7
+ 0.0672,0.24
8
+ 0.0642,0.28
9
+ 0.049,0.32
10
+ 0.0394,0.36
11
+ 0.0378,0.4
12
+ 0.0315,0.44
13
+ 0.0283,0.48
14
+ 0.0203,0.52
15
+ 0.0187,0.56
16
+ 0.023,0.6
17
+ 0.0219,0.64
18
+ 0.0184,0.68
19
+ 0.0206,0.72
20
+ 0.016,0.76
21
+ 0.0102,0.8
22
+ 0.014,0.84
23
+ 0.0113,0.88
24
+ 0.0119,0.92
25
+ 0.0118,0.96
26
+ 0.0125,1.0
27
+ 0.0143,1.04
28
+ 0.0128,1.08
29
+ 0.0132,1.12
30
+ 0.011,1.16
31
+ 0.0094,1.2
32
+ 0.0086,1.24
33
+ 0.0104,1.28
34
+ 0.0082,1.32
35
+ 0.0063,1.36
36
+ 0.0079,1.4
37
+ 0.006,1.44
38
+ 0.0065,1.48
39
+ 0.011,1.52
40
+ 0.0073,1.56
41
+ 0.0053,1.6
42
+ 0.0058,1.64
43
+ 0.006,1.68
44
+ 0.0053,1.72
45
+ 0.0108,1.76
46
+ 0.0092,1.8
47
+ 0.0044,1.84
48
+ 0.0045,1.88
49
+ 0.007,1.92
50
+ 0.0054,1.96
51
+ 0.0037,2.0
52
+ 0.0061,2.04
53
+ 0.0036,2.08
54
+ 0.0036,2.12
55
+ 0.004,2.16
56
+ 0.006,2.2
57
+ 0.0044,2.24
58
+ 0.0046,2.28
59
+ 0.0014,2.32
60
+ 0.0075,2.36
61
+ 0.0036,2.4
62
+ 0.0033,2.44
63
+ 0.003,2.48
64
+ 0.0034,2.52
65
+ 0.0024,2.56
66
+ 0.0023,2.6
67
+ 0.0014,2.64
68
+ 0.0066,2.68
69
+ 0.0034,2.72
70
+ 0.0031,2.76
71
+ 0.0012,2.8
72
+ 0.0029,2.84
73
+ 0.0016,2.88
74
+ 0.0027,2.92
75
+ 0.0012,2.96
76
+ 0.002,3.0
77
+ 0.0024,3.04
78
+ 0.0012,3.08
79
+ 0.0006,3.12
80
+ 0.0032,3.16
81
+ 0.0022,3.2
82
+ 0.0008,3.24
83
+ 0.0021,3.28
84
+ 0.0004,3.32
85
+ 0.0024,3.36
86
+ 0.0001,3.4
87
+ 0.0041,3.44
88
+ 0.0007,3.48
89
+ 0.0003,3.52
90
+ 0.0008,3.56
91
+ 0.0002,3.6
92
+ 0.0031,3.64
93
+ 0.0006,3.68
94
+ 0.0013,3.72
95
+ 0.0005,3.76
96
+ 0.0,3.8
97
+ 0.002,3.84
98
+ 0.0011,3.88
99
+ 0.0,3.92
100
+ 0.0,3.96
101
+ 0.0008,4.0
102
+ 0.0,4.04
103
+ 0.0,4.08
104
+ 0.0,4.12
105
+ 0.0,4.16
106
+ 0.0005,4.2
107
+ 0.0,4.24
108
+ 0.0008,4.28
109
+ 0.002,4.32
110
+ 0.0004,4.36
111
+ 0.0,4.4
112
+ 0.0,4.44
113
+ 0.0003,4.48
114
+ 0.0,4.52
115
+ 0.0,4.56
116
+ 0.0004,4.6
117
+ 0.0,4.64
118
+ 0.0,4.68
119
+ 0.0004,4.72
120
+ 0.0,4.76
121
+ 0.0,4.8
122
+ 0.0,4.84
123
+ 0.0006,4.88
124
+ 0.0006,4.92
125
+ 0.0,4.96
126
+ 0.0,5.0
127
+ 0.012337412198364735,5.0