End of training
Browse files- README.md +10 -8
- all_results.json +11 -11
- eval_results.json +7 -7
- logs/events.out.tfevents.1686033598.serv-3317.3377907.8 +3 -0
- train_results.json +4 -4
- trainer_state.json +77 -77
README.md
CHANGED
@@ -1,4 +1,6 @@
|
|
1 |
---
|
|
|
|
|
2 |
tags:
|
3 |
- generated_from_trainer
|
4 |
datasets:
|
@@ -13,7 +15,7 @@ model-index:
|
|
13 |
name: Text Classification
|
14 |
type: text-classification
|
15 |
dataset:
|
16 |
-
name:
|
17 |
type: glue
|
18 |
config: mrpc
|
19 |
split: validation
|
@@ -21,10 +23,10 @@ model-index:
|
|
21 |
metrics:
|
22 |
- name: Accuracy
|
23 |
type: accuracy
|
24 |
-
value: 0.
|
25 |
- name: F1
|
26 |
type: f1
|
27 |
-
value: 0.
|
28 |
---
|
29 |
|
30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -32,12 +34,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
32 |
|
33 |
# hBERTv1_new_pretrain_w_init__mrpc
|
34 |
|
35 |
-
This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the
|
36 |
It achieves the following results on the evaluation set:
|
37 |
-
- Loss:
|
38 |
-
- Accuracy: 0.
|
39 |
-
- F1: 0.
|
40 |
-
- Combined Score: 0.
|
41 |
|
42 |
## Model description
|
43 |
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
|
|
15 |
name: Text Classification
|
16 |
type: text-classification
|
17 |
dataset:
|
18 |
+
name: GLUE MRPC
|
19 |
type: glue
|
20 |
config: mrpc
|
21 |
split: validation
|
|
|
23 |
metrics:
|
24 |
- name: Accuracy
|
25 |
type: accuracy
|
26 |
+
value: 0.6862745098039216
|
27 |
- name: F1
|
28 |
type: f1
|
29 |
+
value: 0.7894736842105262
|
30 |
---
|
31 |
|
32 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
34 |
|
35 |
# hBERTv1_new_pretrain_w_init__mrpc
|
36 |
|
37 |
+
This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the GLUE MRPC dataset.
|
38 |
It achieves the following results on the evaluation set:
|
39 |
+
- Loss: 0.6082
|
40 |
+
- Accuracy: 0.6863
|
41 |
+
- F1: 0.7895
|
42 |
+
- Combined Score: 0.7379
|
43 |
|
44 |
## Model description
|
45 |
|
all_results.json
CHANGED
@@ -1,16 +1,16 @@
|
|
1 |
{
|
2 |
"epoch": 9.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_combined_score": 0.
|
5 |
-
"eval_f1": 0.
|
6 |
-
"eval_loss": 0.
|
7 |
-
"eval_runtime": 0.
|
8 |
"eval_samples": 408,
|
9 |
-
"eval_samples_per_second":
|
10 |
-
"eval_steps_per_second": 5.
|
11 |
-
"train_loss": 0.
|
12 |
-
"train_runtime": 225.
|
13 |
"train_samples": 3668,
|
14 |
-
"train_samples_per_second":
|
15 |
-
"train_steps_per_second": 6.
|
16 |
}
|
|
|
1 |
{
|
2 |
"epoch": 9.0,
|
3 |
+
"eval_accuracy": 0.6862745098039216,
|
4 |
+
"eval_combined_score": 0.737874097007224,
|
5 |
+
"eval_f1": 0.7894736842105262,
|
6 |
+
"eval_loss": 0.6082243323326111,
|
7 |
+
"eval_runtime": 0.748,
|
8 |
"eval_samples": 408,
|
9 |
+
"eval_samples_per_second": 545.43,
|
10 |
+
"eval_steps_per_second": 5.347,
|
11 |
+
"train_loss": 0.46000806216535894,
|
12 |
+
"train_runtime": 225.7005,
|
13 |
"train_samples": 3668,
|
14 |
+
"train_samples_per_second": 812.581,
|
15 |
+
"train_steps_per_second": 6.424
|
16 |
}
|
eval_results.json
CHANGED
@@ -1,11 +1,11 @@
|
|
1 |
{
|
2 |
"epoch": 9.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_combined_score": 0.
|
5 |
-
"eval_f1": 0.
|
6 |
-
"eval_loss": 0.
|
7 |
-
"eval_runtime": 0.
|
8 |
"eval_samples": 408,
|
9 |
-
"eval_samples_per_second":
|
10 |
-
"eval_steps_per_second": 5.
|
11 |
}
|
|
|
1 |
{
|
2 |
"epoch": 9.0,
|
3 |
+
"eval_accuracy": 0.6862745098039216,
|
4 |
+
"eval_combined_score": 0.737874097007224,
|
5 |
+
"eval_f1": 0.7894736842105262,
|
6 |
+
"eval_loss": 0.6082243323326111,
|
7 |
+
"eval_runtime": 0.748,
|
8 |
"eval_samples": 408,
|
9 |
+
"eval_samples_per_second": 545.43,
|
10 |
+
"eval_steps_per_second": 5.347
|
11 |
}
|
logs/events.out.tfevents.1686033598.serv-3317.3377907.8
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:52302955d2af1b6fc5043f428ac37b205e9677853456c337d771e77f98b25884
|
3 |
+
size 467
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 9.0,
|
3 |
-
"train_loss": 0.
|
4 |
-
"train_runtime": 225.
|
5 |
"train_samples": 3668,
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 6.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 9.0,
|
3 |
+
"train_loss": 0.46000806216535894,
|
4 |
+
"train_runtime": 225.7005,
|
5 |
"train_samples": 3668,
|
6 |
+
"train_samples_per_second": 812.581,
|
7 |
+
"train_steps_per_second": 6.424
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
"best_model_checkpoint": "hBERTv1_new_pretrain_w_init__mrpc/checkpoint-116",
|
4 |
"epoch": 9.0,
|
5 |
"global_step": 261,
|
@@ -9,8 +9,8 @@
|
|
9 |
"log_history": [
|
10 |
{
|
11 |
"epoch": 1.0,
|
12 |
-
"learning_rate":
|
13 |
-
"loss":
|
14 |
"step": 29
|
15 |
},
|
16 |
{
|
@@ -18,16 +18,16 @@
|
|
18 |
"eval_accuracy": 0.6838235294117647,
|
19 |
"eval_combined_score": 0.7480253018237863,
|
20 |
"eval_f1": 0.8122270742358079,
|
21 |
-
"eval_loss": 0.
|
22 |
-
"eval_runtime": 0.
|
23 |
-
"eval_samples_per_second":
|
24 |
-
"eval_steps_per_second": 5.
|
25 |
"step": 29
|
26 |
},
|
27 |
{
|
28 |
"epoch": 2.0,
|
29 |
-
"learning_rate":
|
30 |
-
"loss": 0.
|
31 |
"step": 58
|
32 |
},
|
33 |
{
|
@@ -35,16 +35,16 @@
|
|
35 |
"eval_accuracy": 0.6838235294117647,
|
36 |
"eval_combined_score": 0.7480253018237863,
|
37 |
"eval_f1": 0.8122270742358079,
|
38 |
-
"eval_loss": 0.
|
39 |
-
"eval_runtime": 0.
|
40 |
-
"eval_samples_per_second":
|
41 |
-
"eval_steps_per_second": 5.
|
42 |
"step": 58
|
43 |
},
|
44 |
{
|
45 |
"epoch": 3.0,
|
46 |
-
"learning_rate":
|
47 |
-
"loss": 0.
|
48 |
"step": 87
|
49 |
},
|
50 |
{
|
@@ -52,122 +52,122 @@
|
|
52 |
"eval_accuracy": 0.6838235294117647,
|
53 |
"eval_combined_score": 0.7480253018237863,
|
54 |
"eval_f1": 0.8122270742358079,
|
55 |
-
"eval_loss": 0.
|
56 |
-
"eval_runtime": 0.
|
57 |
-
"eval_samples_per_second":
|
58 |
-
"eval_steps_per_second": 5.
|
59 |
"step": 87
|
60 |
},
|
61 |
{
|
62 |
"epoch": 4.0,
|
63 |
-
"learning_rate":
|
64 |
-
"loss": 0.
|
65 |
"step": 116
|
66 |
},
|
67 |
{
|
68 |
"epoch": 4.0,
|
69 |
-
"eval_accuracy": 0.
|
70 |
-
"eval_combined_score": 0.
|
71 |
-
"eval_f1": 0.
|
72 |
-
"eval_loss": 0.
|
73 |
-
"eval_runtime": 0.
|
74 |
-
"eval_samples_per_second":
|
75 |
-
"eval_steps_per_second": 5.
|
76 |
"step": 116
|
77 |
},
|
78 |
{
|
79 |
"epoch": 5.0,
|
80 |
-
"learning_rate":
|
81 |
-
"loss": 0.
|
82 |
"step": 145
|
83 |
},
|
84 |
{
|
85 |
"epoch": 5.0,
|
86 |
-
"eval_accuracy": 0.
|
87 |
-
"eval_combined_score": 0.
|
88 |
-
"eval_f1": 0.
|
89 |
-
"eval_loss": 0.
|
90 |
-
"eval_runtime": 0.
|
91 |
-
"eval_samples_per_second": 549.
|
92 |
-
"eval_steps_per_second": 5.
|
93 |
"step": 145
|
94 |
},
|
95 |
{
|
96 |
"epoch": 6.0,
|
97 |
-
"learning_rate":
|
98 |
-
"loss": 0.
|
99 |
"step": 174
|
100 |
},
|
101 |
{
|
102 |
"epoch": 6.0,
|
103 |
-
"eval_accuracy": 0.
|
104 |
-
"eval_combined_score": 0.
|
105 |
-
"eval_f1": 0.
|
106 |
-
"eval_loss": 0.
|
107 |
-
"eval_runtime": 0.
|
108 |
-
"eval_samples_per_second": 550.
|
109 |
-
"eval_steps_per_second": 5.
|
110 |
"step": 174
|
111 |
},
|
112 |
{
|
113 |
"epoch": 7.0,
|
114 |
-
"learning_rate":
|
115 |
-
"loss": 0.
|
116 |
"step": 203
|
117 |
},
|
118 |
{
|
119 |
"epoch": 7.0,
|
120 |
-
"eval_accuracy": 0.
|
121 |
-
"eval_combined_score": 0.
|
122 |
-
"eval_f1": 0.
|
123 |
-
"eval_loss":
|
124 |
-
"eval_runtime": 0.
|
125 |
-
"eval_samples_per_second":
|
126 |
-
"eval_steps_per_second": 5.
|
127 |
"step": 203
|
128 |
},
|
129 |
{
|
130 |
"epoch": 8.0,
|
131 |
-
"learning_rate":
|
132 |
-
"loss": 0.
|
133 |
"step": 232
|
134 |
},
|
135 |
{
|
136 |
"epoch": 8.0,
|
137 |
-
"eval_accuracy": 0.
|
138 |
-
"eval_combined_score": 0.
|
139 |
-
"eval_f1": 0.
|
140 |
-
"eval_loss":
|
141 |
-
"eval_runtime": 0.
|
142 |
-
"eval_samples_per_second":
|
143 |
-
"eval_steps_per_second": 5.
|
144 |
"step": 232
|
145 |
},
|
146 |
{
|
147 |
"epoch": 9.0,
|
148 |
-
"learning_rate":
|
149 |
-
"loss": 0.
|
150 |
"step": 261
|
151 |
},
|
152 |
{
|
153 |
"epoch": 9.0,
|
154 |
-
"eval_accuracy": 0.
|
155 |
-
"eval_combined_score": 0.
|
156 |
-
"eval_f1": 0.
|
157 |
-
"eval_loss":
|
158 |
-
"eval_runtime": 0.
|
159 |
-
"eval_samples_per_second":
|
160 |
-
"eval_steps_per_second": 5.
|
161 |
"step": 261
|
162 |
},
|
163 |
{
|
164 |
"epoch": 9.0,
|
165 |
"step": 261,
|
166 |
"total_flos": 4888970768941056.0,
|
167 |
-
"train_loss": 0.
|
168 |
-
"train_runtime": 225.
|
169 |
-
"train_samples_per_second":
|
170 |
-
"train_steps_per_second": 6.
|
171 |
}
|
172 |
],
|
173 |
"max_steps": 1450,
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.6082243323326111,
|
3 |
"best_model_checkpoint": "hBERTv1_new_pretrain_w_init__mrpc/checkpoint-116",
|
4 |
"epoch": 9.0,
|
5 |
"global_step": 261,
|
|
|
9 |
"log_history": [
|
10 |
{
|
11 |
"epoch": 1.0,
|
12 |
+
"learning_rate": 3.9200000000000004e-05,
|
13 |
+
"loss": 0.7111,
|
14 |
"step": 29
|
15 |
},
|
16 |
{
|
|
|
18 |
"eval_accuracy": 0.6838235294117647,
|
19 |
"eval_combined_score": 0.7480253018237863,
|
20 |
"eval_f1": 0.8122270742358079,
|
21 |
+
"eval_loss": 0.6563864350318909,
|
22 |
+
"eval_runtime": 0.7398,
|
23 |
+
"eval_samples_per_second": 551.531,
|
24 |
+
"eval_steps_per_second": 5.407,
|
25 |
"step": 29
|
26 |
},
|
27 |
{
|
28 |
"epoch": 2.0,
|
29 |
+
"learning_rate": 3.8400000000000005e-05,
|
30 |
+
"loss": 0.6641,
|
31 |
"step": 58
|
32 |
},
|
33 |
{
|
|
|
35 |
"eval_accuracy": 0.6838235294117647,
|
36 |
"eval_combined_score": 0.7480253018237863,
|
37 |
"eval_f1": 0.8122270742358079,
|
38 |
+
"eval_loss": 0.6160457730293274,
|
39 |
+
"eval_runtime": 0.7439,
|
40 |
+
"eval_samples_per_second": 548.435,
|
41 |
+
"eval_steps_per_second": 5.377,
|
42 |
"step": 58
|
43 |
},
|
44 |
{
|
45 |
"epoch": 3.0,
|
46 |
+
"learning_rate": 3.76e-05,
|
47 |
+
"loss": 0.6156,
|
48 |
"step": 87
|
49 |
},
|
50 |
{
|
|
|
52 |
"eval_accuracy": 0.6838235294117647,
|
53 |
"eval_combined_score": 0.7480253018237863,
|
54 |
"eval_f1": 0.8122270742358079,
|
55 |
+
"eval_loss": 0.6354035139083862,
|
56 |
+
"eval_runtime": 0.7456,
|
57 |
+
"eval_samples_per_second": 547.23,
|
58 |
+
"eval_steps_per_second": 5.365,
|
59 |
"step": 87
|
60 |
},
|
61 |
{
|
62 |
"epoch": 4.0,
|
63 |
+
"learning_rate": 3.680000000000001e-05,
|
64 |
+
"loss": 0.5817,
|
65 |
"step": 116
|
66 |
},
|
67 |
{
|
68 |
"epoch": 4.0,
|
69 |
+
"eval_accuracy": 0.6862745098039216,
|
70 |
+
"eval_combined_score": 0.737874097007224,
|
71 |
+
"eval_f1": 0.7894736842105262,
|
72 |
+
"eval_loss": 0.6082243323326111,
|
73 |
+
"eval_runtime": 0.7468,
|
74 |
+
"eval_samples_per_second": 546.334,
|
75 |
+
"eval_steps_per_second": 5.356,
|
76 |
"step": 116
|
77 |
},
|
78 |
{
|
79 |
"epoch": 5.0,
|
80 |
+
"learning_rate": 3.6e-05,
|
81 |
+
"loss": 0.5091,
|
82 |
"step": 145
|
83 |
},
|
84 |
{
|
85 |
"epoch": 5.0,
|
86 |
+
"eval_accuracy": 0.5073529411764706,
|
87 |
+
"eval_combined_score": 0.5115077958894401,
|
88 |
+
"eval_f1": 0.5156626506024096,
|
89 |
+
"eval_loss": 0.7811743021011353,
|
90 |
+
"eval_runtime": 0.7431,
|
91 |
+
"eval_samples_per_second": 549.025,
|
92 |
+
"eval_steps_per_second": 5.383,
|
93 |
"step": 145
|
94 |
},
|
95 |
{
|
96 |
"epoch": 6.0,
|
97 |
+
"learning_rate": 3.52e-05,
|
98 |
+
"loss": 0.3973,
|
99 |
"step": 174
|
100 |
},
|
101 |
{
|
102 |
"epoch": 6.0,
|
103 |
+
"eval_accuracy": 0.6544117647058824,
|
104 |
+
"eval_combined_score": 0.705444224321853,
|
105 |
+
"eval_f1": 0.7564766839378239,
|
106 |
+
"eval_loss": 0.7949338555335999,
|
107 |
+
"eval_runtime": 0.7416,
|
108 |
+
"eval_samples_per_second": 550.17,
|
109 |
+
"eval_steps_per_second": 5.394,
|
110 |
"step": 174
|
111 |
},
|
112 |
{
|
113 |
"epoch": 7.0,
|
114 |
+
"learning_rate": 3.44e-05,
|
115 |
+
"loss": 0.2966,
|
116 |
"step": 203
|
117 |
},
|
118 |
{
|
119 |
"epoch": 7.0,
|
120 |
+
"eval_accuracy": 0.6078431372549019,
|
121 |
+
"eval_combined_score": 0.6482795452811474,
|
122 |
+
"eval_f1": 0.688715953307393,
|
123 |
+
"eval_loss": 1.0388320684432983,
|
124 |
+
"eval_runtime": 0.7411,
|
125 |
+
"eval_samples_per_second": 550.501,
|
126 |
+
"eval_steps_per_second": 5.397,
|
127 |
"step": 203
|
128 |
},
|
129 |
{
|
130 |
"epoch": 8.0,
|
131 |
+
"learning_rate": 3.3600000000000004e-05,
|
132 |
+
"loss": 0.2024,
|
133 |
"step": 232
|
134 |
},
|
135 |
{
|
136 |
"epoch": 8.0,
|
137 |
+
"eval_accuracy": 0.6200980392156863,
|
138 |
+
"eval_combined_score": 0.6662642329659136,
|
139 |
+
"eval_f1": 0.7124304267161411,
|
140 |
+
"eval_loss": 1.0064674615859985,
|
141 |
+
"eval_runtime": 0.7424,
|
142 |
+
"eval_samples_per_second": 549.6,
|
143 |
+
"eval_steps_per_second": 5.388,
|
144 |
"step": 232
|
145 |
},
|
146 |
{
|
147 |
"epoch": 9.0,
|
148 |
+
"learning_rate": 3.28e-05,
|
149 |
+
"loss": 0.1621,
|
150 |
"step": 261
|
151 |
},
|
152 |
{
|
153 |
"epoch": 9.0,
|
154 |
+
"eval_accuracy": 0.5735294117647058,
|
155 |
+
"eval_combined_score": 0.6155048633626679,
|
156 |
+
"eval_f1": 0.65748031496063,
|
157 |
+
"eval_loss": 1.307615876197815,
|
158 |
+
"eval_runtime": 0.7454,
|
159 |
+
"eval_samples_per_second": 547.342,
|
160 |
+
"eval_steps_per_second": 5.366,
|
161 |
"step": 261
|
162 |
},
|
163 |
{
|
164 |
"epoch": 9.0,
|
165 |
"step": 261,
|
166 |
"total_flos": 4888970768941056.0,
|
167 |
+
"train_loss": 0.46000806216535894,
|
168 |
+
"train_runtime": 225.7005,
|
169 |
+
"train_samples_per_second": 812.581,
|
170 |
+
"train_steps_per_second": 6.424
|
171 |
}
|
172 |
],
|
173 |
"max_steps": 1450,
|