End of training
Browse files- README.md +24 -2
- all_results.json +16 -0
- eval_results.json +11 -0
- train_results.json +8 -0
- trainer_state.json +172 -0
README.md
CHANGED
@@ -1,13 +1,30 @@
|
|
1 |
---
|
|
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: bert-base-uncased
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
- glue
|
|
|
|
|
8 |
model-index:
|
9 |
- name: out-glue-stsb
|
10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
---
|
12 |
|
13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -15,7 +32,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# out-glue-stsb
|
17 |
|
18 |
-
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the
|
|
|
|
|
|
|
|
|
|
|
19 |
|
20 |
## Model description
|
21 |
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
license: apache-2.0
|
5 |
base_model: bert-base-uncased
|
6 |
tags:
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
- glue
|
10 |
+
metrics:
|
11 |
+
- spearmanr
|
12 |
model-index:
|
13 |
- name: out-glue-stsb
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
name: Text Classification
|
17 |
+
type: text-classification
|
18 |
+
dataset:
|
19 |
+
name: GLUE STSB
|
20 |
+
type: glue
|
21 |
+
config: stsb
|
22 |
+
split: validation
|
23 |
+
args: stsb
|
24 |
+
metrics:
|
25 |
+
- name: Spearmanr
|
26 |
+
type: spearmanr
|
27 |
+
value: 0.8807188703976201
|
28 |
---
|
29 |
|
30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
32 |
|
33 |
# out-glue-stsb
|
34 |
|
35 |
+
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the GLUE STSB dataset.
|
36 |
+
It achieves the following results on the evaluation set:
|
37 |
+
- Loss: 0.5336
|
38 |
+
- Pearson: 0.8853
|
39 |
+
- Spearmanr: 0.8807
|
40 |
+
- Combined Score: 0.8830
|
41 |
|
42 |
## Model description
|
43 |
|
all_results.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 20.0,
|
3 |
+
"eval_combined_score": 0.8830287553865886,
|
4 |
+
"eval_loss": 0.533576250076294,
|
5 |
+
"eval_pearson": 0.8853386403755572,
|
6 |
+
"eval_runtime": 1.1414,
|
7 |
+
"eval_samples": 1500,
|
8 |
+
"eval_samples_per_second": 1314.172,
|
9 |
+
"eval_spearmanr": 0.8807188703976201,
|
10 |
+
"eval_steps_per_second": 5.257,
|
11 |
+
"train_loss": 0.4903079410394033,
|
12 |
+
"train_runtime": 237.6265,
|
13 |
+
"train_samples": 5749,
|
14 |
+
"train_samples_per_second": 483.869,
|
15 |
+
"train_steps_per_second": 2.525
|
16 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 20.0,
|
3 |
+
"eval_combined_score": 0.8830287553865886,
|
4 |
+
"eval_loss": 0.533576250076294,
|
5 |
+
"eval_pearson": 0.8853386403755572,
|
6 |
+
"eval_runtime": 1.1414,
|
7 |
+
"eval_samples": 1500,
|
8 |
+
"eval_samples_per_second": 1314.172,
|
9 |
+
"eval_spearmanr": 0.8807188703976201,
|
10 |
+
"eval_steps_per_second": 5.257
|
11 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 20.0,
|
3 |
+
"train_loss": 0.4903079410394033,
|
4 |
+
"train_runtime": 237.6265,
|
5 |
+
"train_samples": 5749,
|
6 |
+
"train_samples_per_second": 483.869,
|
7 |
+
"train_steps_per_second": 2.525
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,172 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 20.0,
|
5 |
+
"eval_steps": 2.0,
|
6 |
+
"global_step": 600,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.83,
|
13 |
+
"learning_rate": 1.926666666666667e-05,
|
14 |
+
"loss": 5.4856,
|
15 |
+
"step": 25
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 1.67,
|
19 |
+
"learning_rate": 1.8433333333333335e-05,
|
20 |
+
"loss": 1.108,
|
21 |
+
"step": 50
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 2.5,
|
25 |
+
"learning_rate": 1.76e-05,
|
26 |
+
"loss": 0.6811,
|
27 |
+
"step": 75
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 3.33,
|
31 |
+
"learning_rate": 1.676666666666667e-05,
|
32 |
+
"loss": 0.5356,
|
33 |
+
"step": 100
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 4.17,
|
37 |
+
"learning_rate": 1.5933333333333336e-05,
|
38 |
+
"loss": 0.4742,
|
39 |
+
"step": 125
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 5.0,
|
43 |
+
"learning_rate": 1.5100000000000001e-05,
|
44 |
+
"loss": 0.3909,
|
45 |
+
"step": 150
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 5.83,
|
49 |
+
"learning_rate": 1.4266666666666668e-05,
|
50 |
+
"loss": 0.3342,
|
51 |
+
"step": 175
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 6.67,
|
55 |
+
"learning_rate": 1.3433333333333334e-05,
|
56 |
+
"loss": 0.2914,
|
57 |
+
"step": 200
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 7.5,
|
61 |
+
"learning_rate": 1.2600000000000001e-05,
|
62 |
+
"loss": 0.2638,
|
63 |
+
"step": 225
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 8.33,
|
67 |
+
"learning_rate": 1.1766666666666669e-05,
|
68 |
+
"loss": 0.2275,
|
69 |
+
"step": 250
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 9.17,
|
73 |
+
"learning_rate": 1.0933333333333334e-05,
|
74 |
+
"loss": 0.2238,
|
75 |
+
"step": 275
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 10.0,
|
79 |
+
"learning_rate": 1.0100000000000002e-05,
|
80 |
+
"loss": 0.1909,
|
81 |
+
"step": 300
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 10.83,
|
85 |
+
"learning_rate": 9.266666666666667e-06,
|
86 |
+
"loss": 0.17,
|
87 |
+
"step": 325
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 11.67,
|
91 |
+
"learning_rate": 8.433333333333334e-06,
|
92 |
+
"loss": 0.1637,
|
93 |
+
"step": 350
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 12.5,
|
97 |
+
"learning_rate": 7.600000000000001e-06,
|
98 |
+
"loss": 0.1513,
|
99 |
+
"step": 375
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 13.33,
|
103 |
+
"learning_rate": 6.7666666666666665e-06,
|
104 |
+
"loss": 0.1415,
|
105 |
+
"step": 400
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 14.17,
|
109 |
+
"learning_rate": 5.933333333333335e-06,
|
110 |
+
"loss": 0.1404,
|
111 |
+
"step": 425
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 15.0,
|
115 |
+
"learning_rate": 5.1e-06,
|
116 |
+
"loss": 0.1244,
|
117 |
+
"step": 450
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 15.83,
|
121 |
+
"learning_rate": 4.266666666666668e-06,
|
122 |
+
"loss": 0.1235,
|
123 |
+
"step": 475
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 16.67,
|
127 |
+
"learning_rate": 3.4333333333333336e-06,
|
128 |
+
"loss": 0.115,
|
129 |
+
"step": 500
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 17.5,
|
133 |
+
"learning_rate": 2.6e-06,
|
134 |
+
"loss": 0.1095,
|
135 |
+
"step": 525
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 18.33,
|
139 |
+
"learning_rate": 1.7666666666666668e-06,
|
140 |
+
"loss": 0.1088,
|
141 |
+
"step": 550
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 19.17,
|
145 |
+
"learning_rate": 9.333333333333334e-07,
|
146 |
+
"loss": 0.106,
|
147 |
+
"step": 575
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 20.0,
|
151 |
+
"learning_rate": 1.0000000000000001e-07,
|
152 |
+
"loss": 0.1063,
|
153 |
+
"step": 600
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 20.0,
|
157 |
+
"step": 600,
|
158 |
+
"total_flos": 7563059380055040.0,
|
159 |
+
"train_loss": 0.4903079410394033,
|
160 |
+
"train_runtime": 237.6265,
|
161 |
+
"train_samples_per_second": 483.869,
|
162 |
+
"train_steps_per_second": 2.525
|
163 |
+
}
|
164 |
+
],
|
165 |
+
"logging_steps": 25,
|
166 |
+
"max_steps": 600,
|
167 |
+
"num_train_epochs": 20,
|
168 |
+
"save_steps": 500,
|
169 |
+
"total_flos": 7563059380055040.0,
|
170 |
+
"trial_name": null,
|
171 |
+
"trial_params": null
|
172 |
+
}
|