nhanv commited on
Commit
dd20389
·
1 Parent(s): fdad744

upload model

Browse files
README.md ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ tags:
4
+ - generated_from_trainer
5
+ metrics:
6
+ - precision
7
+ - recall
8
+ - f1
9
+ - accuracy
10
+ model-index:
11
+ - name: reco-ner
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # reco-ner
19
+
20
+ This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on an unknown dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 0.0668
23
+ - Precision: 0.8125
24
+ - Recall: 0.8790
25
+ - F1: 0.8444
26
+ - Accuracy: 0.9819
27
+
28
+ ## Model description
29
+
30
+ More information needed
31
+
32
+ ## Intended uses & limitations
33
+
34
+ More information needed
35
+
36
+ ## Training and evaluation data
37
+
38
+ More information needed
39
+
40
+ ## Training procedure
41
+
42
+ ### Training hyperparameters
43
+
44
+ The following hyperparameters were used during training:
45
+ - learning_rate: 5e-05
46
+ - train_batch_size: 16
47
+ - eval_batch_size: 4
48
+ - seed: 42
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: linear
51
+ - num_epochs: 10.0
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
56
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
57
+ | 0.4516 | 1.0 | 626 | 0.4047 | 0.4332 | 0.4564 | 0.4445 | 0.8980 |
58
+ | 0.3677 | 2.0 | 1252 | 0.2774 | 0.4918 | 0.5731 | 0.5293 | 0.9193 |
59
+ | 0.2892 | 3.0 | 1878 | 0.2133 | 0.6139 | 0.6581 | 0.6353 | 0.9384 |
60
+ | 0.2736 | 4.0 | 2504 | 0.1772 | 0.6248 | 0.6854 | 0.6537 | 0.9488 |
61
+ | 0.221 | 5.0 | 3130 | 0.1503 | 0.6295 | 0.7328 | 0.6772 | 0.9560 |
62
+ | 0.1569 | 6.0 | 3756 | 0.1283 | 0.6821 | 0.8108 | 0.7409 | 0.9623 |
63
+ | 0.1534 | 7.0 | 4382 | 0.0995 | 0.7412 | 0.8119 | 0.7749 | 0.9708 |
64
+ | 0.089 | 8.0 | 5008 | 0.0846 | 0.7695 | 0.8353 | 0.8010 | 0.9760 |
65
+ | 0.0923 | 9.0 | 5634 | 0.0743 | 0.7881 | 0.8740 | 0.8289 | 0.9789 |
66
+ | 0.0711 | 10.0 | 6260 | 0.0668 | 0.8125 | 0.8790 | 0.8444 | 0.9819 |
67
+
68
+
69
+ ### Framework versions
70
+
71
+ - Transformers 4.22.2
72
+ - Pytorch 1.12.1+cu113
73
+ - Datasets 2.4.0
74
+ - Tokenizers 0.12.1
added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "[MASK]": 128000
3
+ }
all_results.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.9819289686983922,
4
+ "eval_f1": 0.8444211629125196,
5
+ "eval_loss": 0.06683139503002167,
6
+ "eval_precision": 0.8125,
7
+ "eval_recall": 0.8789531079607416,
8
+ "eval_runtime": 68.0971,
9
+ "eval_samples": 10007,
10
+ "eval_samples_per_second": 146.952,
11
+ "eval_steps_per_second": 36.742,
12
+ "train_loss": 0.217802461039144,
13
+ "train_runtime": 2284.8138,
14
+ "train_samples": 10007,
15
+ "train_samples_per_second": 43.798,
16
+ "train_steps_per_second": 2.74
17
+ }
config.json ADDED
@@ -0,0 +1,118 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/deberta-v3-base",
3
+ "architectures": [
4
+ "DebertaV2ForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "finetuning_task": "ner",
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "B-Company",
13
+ "1": "B-Date",
14
+ "2": "B-Degree",
15
+ "3": "B-Education",
16
+ "4": "B-Email",
17
+ "5": "B-Experiments",
18
+ "6": "B-Gender",
19
+ "7": "B-Institution",
20
+ "8": "B-Job",
21
+ "9": "B-Language",
22
+ "10": "B-Location",
23
+ "11": "B-Name",
24
+ "12": "B-Overall",
25
+ "13": "B-Phone",
26
+ "14": "B-Position",
27
+ "15": "B-Program",
28
+ "16": "B-Project",
29
+ "17": "B-Skill",
30
+ "18": "B-Training",
31
+ "19": "I-Company",
32
+ "20": "I-Date",
33
+ "21": "I-Degree",
34
+ "22": "I-Education",
35
+ "23": "I-Email",
36
+ "24": "I-Experiments",
37
+ "25": "I-Gender",
38
+ "26": "I-Institution",
39
+ "27": "I-Job",
40
+ "28": "I-Language",
41
+ "29": "I-Location",
42
+ "30": "I-Name",
43
+ "31": "I-Overall",
44
+ "32": "I-Phone",
45
+ "33": "I-Position",
46
+ "34": "I-Program",
47
+ "35": "I-Project",
48
+ "36": "I-Skill",
49
+ "37": "I-Training",
50
+ "38": "O"
51
+ },
52
+ "initializer_range": 0.02,
53
+ "intermediate_size": 3072,
54
+ "label2id": {
55
+ "B-Company": 0,
56
+ "B-Date": 1,
57
+ "B-Degree": 2,
58
+ "B-Education": 3,
59
+ "B-Email": 4,
60
+ "B-Experiments": 5,
61
+ "B-Gender": 6,
62
+ "B-Institution": 7,
63
+ "B-Job": 8,
64
+ "B-Language": 9,
65
+ "B-Location": 10,
66
+ "B-Name": 11,
67
+ "B-Overall": 12,
68
+ "B-Phone": 13,
69
+ "B-Position": 14,
70
+ "B-Program": 15,
71
+ "B-Project": 16,
72
+ "B-Skill": 17,
73
+ "B-Training": 18,
74
+ "I-Company": 19,
75
+ "I-Date": 20,
76
+ "I-Degree": 21,
77
+ "I-Education": 22,
78
+ "I-Email": 23,
79
+ "I-Experiments": 24,
80
+ "I-Gender": 25,
81
+ "I-Institution": 26,
82
+ "I-Job": 27,
83
+ "I-Language": 28,
84
+ "I-Location": 29,
85
+ "I-Name": 30,
86
+ "I-Overall": 31,
87
+ "I-Phone": 32,
88
+ "I-Position": 33,
89
+ "I-Program": 34,
90
+ "I-Project": 35,
91
+ "I-Skill": 36,
92
+ "I-Training": 37,
93
+ "O": 38
94
+ },
95
+ "layer_norm_eps": 1e-07,
96
+ "max_position_embeddings": 512,
97
+ "max_relative_positions": -1,
98
+ "model_type": "deberta-v2",
99
+ "norm_rel_ebd": "layer_norm",
100
+ "num_attention_heads": 12,
101
+ "num_hidden_layers": 12,
102
+ "pad_token_id": 0,
103
+ "pooler_dropout": 0,
104
+ "pooler_hidden_act": "gelu",
105
+ "pooler_hidden_size": 768,
106
+ "pos_att_type": [
107
+ "p2c",
108
+ "c2p"
109
+ ],
110
+ "position_biased_input": false,
111
+ "position_buckets": 256,
112
+ "relative_attention": true,
113
+ "share_att_key": true,
114
+ "torch_dtype": "float32",
115
+ "transformers_version": "4.22.2",
116
+ "type_vocab_size": 0,
117
+ "vocab_size": 128100
118
+ }
eval_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.9819289686983922,
4
+ "eval_f1": 0.8444211629125196,
5
+ "eval_loss": 0.06683139503002167,
6
+ "eval_precision": 0.8125,
7
+ "eval_recall": 0.8789531079607416,
8
+ "eval_runtime": 68.0971,
9
+ "eval_samples": 10007,
10
+ "eval_samples_per_second": 146.952,
11
+ "eval_steps_per_second": 36.742
12
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66f39653aeab7fecd9de7b8637713cc73ba309bfc75eb188bfdabc2aceaba97a
3
+ size 735517743
special_tokens_map.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "[CLS]",
3
+ "cls_token": "[CLS]",
4
+ "eos_token": "[SEP]",
5
+ "mask_token": "[MASK]",
6
+ "pad_token": "[PAD]",
7
+ "sep_token": "[SEP]",
8
+ "unk_token": "[UNK]"
9
+ }
spm.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
3
+ size 2464616
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "[CLS]",
3
+ "cls_token": "[CLS]",
4
+ "do_lower_case": false,
5
+ "eos_token": "[SEP]",
6
+ "mask_token": "[MASK]",
7
+ "name_or_path": "microsoft/deberta-v3-base",
8
+ "pad_token": "[PAD]",
9
+ "sep_token": "[SEP]",
10
+ "sp_model_kwargs": {},
11
+ "special_tokens_map_file": null,
12
+ "split_by_punct": false,
13
+ "tokenizer_class": "DebertaV2Tokenizer",
14
+ "unk_token": "[UNK]",
15
+ "vocab_type": "spm"
16
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "train_loss": 0.217802461039144,
4
+ "train_runtime": 2284.8138,
5
+ "train_samples": 10007,
6
+ "train_samples_per_second": 43.798,
7
+ "train_steps_per_second": 2.74
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,517 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 10.0,
5
+ "global_step": 6260,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.16,
12
+ "learning_rate": 4.920127795527157e-05,
13
+ "loss": 0.7778,
14
+ "step": 100
15
+ },
16
+ {
17
+ "epoch": 0.32,
18
+ "learning_rate": 4.840255591054313e-05,
19
+ "loss": 0.5964,
20
+ "step": 200
21
+ },
22
+ {
23
+ "epoch": 0.48,
24
+ "learning_rate": 4.76038338658147e-05,
25
+ "loss": 0.5901,
26
+ "step": 300
27
+ },
28
+ {
29
+ "epoch": 0.64,
30
+ "learning_rate": 4.680511182108626e-05,
31
+ "loss": 0.4494,
32
+ "step": 400
33
+ },
34
+ {
35
+ "epoch": 0.8,
36
+ "learning_rate": 4.600638977635783e-05,
37
+ "loss": 0.4166,
38
+ "step": 500
39
+ },
40
+ {
41
+ "epoch": 0.96,
42
+ "learning_rate": 4.520766773162939e-05,
43
+ "loss": 0.4516,
44
+ "step": 600
45
+ },
46
+ {
47
+ "epoch": 1.0,
48
+ "eval_accuracy": 0.8979708054137165,
49
+ "eval_f1": 0.44450345193839613,
50
+ "eval_loss": 0.40465056896209717,
51
+ "eval_precision": 0.4332298136645963,
52
+ "eval_recall": 0.4563794983642312,
53
+ "eval_runtime": 68.5329,
54
+ "eval_samples_per_second": 146.017,
55
+ "eval_steps_per_second": 36.508,
56
+ "step": 626
57
+ },
58
+ {
59
+ "epoch": 1.12,
60
+ "learning_rate": 4.440894568690096e-05,
61
+ "loss": 0.4155,
62
+ "step": 700
63
+ },
64
+ {
65
+ "epoch": 1.28,
66
+ "learning_rate": 4.361022364217253e-05,
67
+ "loss": 0.3645,
68
+ "step": 800
69
+ },
70
+ {
71
+ "epoch": 1.44,
72
+ "learning_rate": 4.2811501597444096e-05,
73
+ "loss": 0.3712,
74
+ "step": 900
75
+ },
76
+ {
77
+ "epoch": 1.6,
78
+ "learning_rate": 4.201277955271566e-05,
79
+ "loss": 0.3669,
80
+ "step": 1000
81
+ },
82
+ {
83
+ "epoch": 1.76,
84
+ "learning_rate": 4.1214057507987225e-05,
85
+ "loss": 0.3424,
86
+ "step": 1100
87
+ },
88
+ {
89
+ "epoch": 1.92,
90
+ "learning_rate": 4.041533546325879e-05,
91
+ "loss": 0.3677,
92
+ "step": 1200
93
+ },
94
+ {
95
+ "epoch": 2.0,
96
+ "eval_accuracy": 0.9192688065906114,
97
+ "eval_f1": 0.5293376983127676,
98
+ "eval_loss": 0.2773844301700592,
99
+ "eval_precision": 0.4918109499298081,
100
+ "eval_recall": 0.5730643402399127,
101
+ "eval_runtime": 67.6554,
102
+ "eval_samples_per_second": 147.911,
103
+ "eval_steps_per_second": 36.982,
104
+ "step": 1252
105
+ },
106
+ {
107
+ "epoch": 2.08,
108
+ "learning_rate": 3.9616613418530355e-05,
109
+ "loss": 0.3039,
110
+ "step": 1300
111
+ },
112
+ {
113
+ "epoch": 2.24,
114
+ "learning_rate": 3.8817891373801916e-05,
115
+ "loss": 0.2599,
116
+ "step": 1400
117
+ },
118
+ {
119
+ "epoch": 2.4,
120
+ "learning_rate": 3.8019169329073485e-05,
121
+ "loss": 0.3243,
122
+ "step": 1500
123
+ },
124
+ {
125
+ "epoch": 2.56,
126
+ "learning_rate": 3.722044728434505e-05,
127
+ "loss": 0.2701,
128
+ "step": 1600
129
+ },
130
+ {
131
+ "epoch": 2.72,
132
+ "learning_rate": 3.6421725239616614e-05,
133
+ "loss": 0.2634,
134
+ "step": 1700
135
+ },
136
+ {
137
+ "epoch": 2.88,
138
+ "learning_rate": 3.562300319488818e-05,
139
+ "loss": 0.2892,
140
+ "step": 1800
141
+ },
142
+ {
143
+ "epoch": 3.0,
144
+ "eval_accuracy": 0.9383648753820163,
145
+ "eval_f1": 0.6352631578947368,
146
+ "eval_loss": 0.21329015493392944,
147
+ "eval_precision": 0.6139369277721262,
148
+ "eval_recall": 0.6581243184296619,
149
+ "eval_runtime": 67.955,
150
+ "eval_samples_per_second": 147.259,
151
+ "eval_steps_per_second": 36.818,
152
+ "step": 1878
153
+ },
154
+ {
155
+ "epoch": 3.04,
156
+ "learning_rate": 3.482428115015975e-05,
157
+ "loss": 0.315,
158
+ "step": 1900
159
+ },
160
+ {
161
+ "epoch": 3.19,
162
+ "learning_rate": 3.402555910543131e-05,
163
+ "loss": 0.242,
164
+ "step": 2000
165
+ },
166
+ {
167
+ "epoch": 3.35,
168
+ "learning_rate": 3.322683706070287e-05,
169
+ "loss": 0.2366,
170
+ "step": 2100
171
+ },
172
+ {
173
+ "epoch": 3.51,
174
+ "learning_rate": 3.242811501597444e-05,
175
+ "loss": 0.2233,
176
+ "step": 2200
177
+ },
178
+ {
179
+ "epoch": 3.67,
180
+ "learning_rate": 3.162939297124601e-05,
181
+ "loss": 0.233,
182
+ "step": 2300
183
+ },
184
+ {
185
+ "epoch": 3.83,
186
+ "learning_rate": 3.083067092651757e-05,
187
+ "loss": 0.2469,
188
+ "step": 2400
189
+ },
190
+ {
191
+ "epoch": 3.99,
192
+ "learning_rate": 3.003194888178914e-05,
193
+ "loss": 0.2736,
194
+ "step": 2500
195
+ },
196
+ {
197
+ "epoch": 4.0,
198
+ "eval_accuracy": 0.9488050720373569,
199
+ "eval_f1": 0.6536661466458659,
200
+ "eval_loss": 0.17724330723285675,
201
+ "eval_precision": 0.6247514910536779,
202
+ "eval_recall": 0.6853871319520175,
203
+ "eval_runtime": 68.1206,
204
+ "eval_samples_per_second": 146.901,
205
+ "eval_steps_per_second": 36.729,
206
+ "step": 2504
207
+ },
208
+ {
209
+ "epoch": 4.15,
210
+ "learning_rate": 2.9233226837060707e-05,
211
+ "loss": 0.1691,
212
+ "step": 2600
213
+ },
214
+ {
215
+ "epoch": 4.31,
216
+ "learning_rate": 2.843450479233227e-05,
217
+ "loss": 0.1941,
218
+ "step": 2700
219
+ },
220
+ {
221
+ "epoch": 4.47,
222
+ "learning_rate": 2.7635782747603834e-05,
223
+ "loss": 0.1891,
224
+ "step": 2800
225
+ },
226
+ {
227
+ "epoch": 4.63,
228
+ "learning_rate": 2.68370607028754e-05,
229
+ "loss": 0.2037,
230
+ "step": 2900
231
+ },
232
+ {
233
+ "epoch": 4.79,
234
+ "learning_rate": 2.6038338658146967e-05,
235
+ "loss": 0.2222,
236
+ "step": 3000
237
+ },
238
+ {
239
+ "epoch": 4.95,
240
+ "learning_rate": 2.523961661341853e-05,
241
+ "loss": 0.221,
242
+ "step": 3100
243
+ },
244
+ {
245
+ "epoch": 5.0,
246
+ "eval_accuracy": 0.9560372809931474,
247
+ "eval_f1": 0.6772486772486773,
248
+ "eval_loss": 0.15026314556598663,
249
+ "eval_precision": 0.6295081967213115,
250
+ "eval_recall": 0.732824427480916,
251
+ "eval_runtime": 68.8869,
252
+ "eval_samples_per_second": 145.267,
253
+ "eval_steps_per_second": 36.32,
254
+ "step": 3130
255
+ },
256
+ {
257
+ "epoch": 5.11,
258
+ "learning_rate": 2.44408945686901e-05,
259
+ "loss": 0.1524,
260
+ "step": 3200
261
+ },
262
+ {
263
+ "epoch": 5.27,
264
+ "learning_rate": 2.364217252396166e-05,
265
+ "loss": 0.1575,
266
+ "step": 3300
267
+ },
268
+ {
269
+ "epoch": 5.43,
270
+ "learning_rate": 2.284345047923323e-05,
271
+ "loss": 0.1606,
272
+ "step": 3400
273
+ },
274
+ {
275
+ "epoch": 5.59,
276
+ "learning_rate": 2.2044728434504794e-05,
277
+ "loss": 0.1314,
278
+ "step": 3500
279
+ },
280
+ {
281
+ "epoch": 5.75,
282
+ "learning_rate": 2.124600638977636e-05,
283
+ "loss": 0.1845,
284
+ "step": 3600
285
+ },
286
+ {
287
+ "epoch": 5.91,
288
+ "learning_rate": 2.0447284345047924e-05,
289
+ "loss": 0.1569,
290
+ "step": 3700
291
+ },
292
+ {
293
+ "epoch": 6.0,
294
+ "eval_accuracy": 0.9622824168106149,
295
+ "eval_f1": 0.7409068261086198,
296
+ "eval_loss": 0.1283087134361267,
297
+ "eval_precision": 0.6821100917431193,
298
+ "eval_recall": 0.8107960741548528,
299
+ "eval_runtime": 69.0068,
300
+ "eval_samples_per_second": 145.015,
301
+ "eval_steps_per_second": 36.257,
302
+ "step": 3756
303
+ },
304
+ {
305
+ "epoch": 6.07,
306
+ "learning_rate": 1.964856230031949e-05,
307
+ "loss": 0.1495,
308
+ "step": 3800
309
+ },
310
+ {
311
+ "epoch": 6.23,
312
+ "learning_rate": 1.8849840255591057e-05,
313
+ "loss": 0.1309,
314
+ "step": 3900
315
+ },
316
+ {
317
+ "epoch": 6.39,
318
+ "learning_rate": 1.805111821086262e-05,
319
+ "loss": 0.131,
320
+ "step": 4000
321
+ },
322
+ {
323
+ "epoch": 6.55,
324
+ "learning_rate": 1.7252396166134186e-05,
325
+ "loss": 0.1177,
326
+ "step": 4100
327
+ },
328
+ {
329
+ "epoch": 6.71,
330
+ "learning_rate": 1.645367412140575e-05,
331
+ "loss": 0.1046,
332
+ "step": 4200
333
+ },
334
+ {
335
+ "epoch": 6.87,
336
+ "learning_rate": 1.565495207667732e-05,
337
+ "loss": 0.1534,
338
+ "step": 4300
339
+ },
340
+ {
341
+ "epoch": 7.0,
342
+ "eval_accuracy": 0.9707674493650462,
343
+ "eval_f1": 0.7749154306531356,
344
+ "eval_loss": 0.09951327741146088,
345
+ "eval_precision": 0.7411647585863613,
346
+ "eval_recall": 0.811886586695747,
347
+ "eval_runtime": 67.5487,
348
+ "eval_samples_per_second": 148.145,
349
+ "eval_steps_per_second": 37.04,
350
+ "step": 4382
351
+ },
352
+ {
353
+ "epoch": 7.03,
354
+ "learning_rate": 1.485623003194888e-05,
355
+ "loss": 0.1147,
356
+ "step": 4400
357
+ },
358
+ {
359
+ "epoch": 7.19,
360
+ "learning_rate": 1.4057507987220447e-05,
361
+ "loss": 0.1158,
362
+ "step": 4500
363
+ },
364
+ {
365
+ "epoch": 7.35,
366
+ "learning_rate": 1.3258785942492014e-05,
367
+ "loss": 0.0993,
368
+ "step": 4600
369
+ },
370
+ {
371
+ "epoch": 7.51,
372
+ "learning_rate": 1.2460063897763578e-05,
373
+ "loss": 0.1288,
374
+ "step": 4700
375
+ },
376
+ {
377
+ "epoch": 7.67,
378
+ "learning_rate": 1.1661341853035145e-05,
379
+ "loss": 0.0874,
380
+ "step": 4800
381
+ },
382
+ {
383
+ "epoch": 7.83,
384
+ "learning_rate": 1.086261980830671e-05,
385
+ "loss": 0.105,
386
+ "step": 4900
387
+ },
388
+ {
389
+ "epoch": 7.99,
390
+ "learning_rate": 1.0063897763578276e-05,
391
+ "loss": 0.089,
392
+ "step": 5000
393
+ },
394
+ {
395
+ "epoch": 8.0,
396
+ "eval_accuracy": 0.9760065298684535,
397
+ "eval_f1": 0.8010457516339871,
398
+ "eval_loss": 0.08459383249282837,
399
+ "eval_precision": 0.7694625816172778,
400
+ "eval_recall": 0.8353326063249727,
401
+ "eval_runtime": 67.5952,
402
+ "eval_samples_per_second": 148.043,
403
+ "eval_steps_per_second": 37.014,
404
+ "step": 5008
405
+ },
406
+ {
407
+ "epoch": 8.15,
408
+ "learning_rate": 9.265175718849841e-06,
409
+ "loss": 0.0766,
410
+ "step": 5100
411
+ },
412
+ {
413
+ "epoch": 8.31,
414
+ "learning_rate": 8.466453674121406e-06,
415
+ "loss": 0.0929,
416
+ "step": 5200
417
+ },
418
+ {
419
+ "epoch": 8.47,
420
+ "learning_rate": 7.66773162939297e-06,
421
+ "loss": 0.089,
422
+ "step": 5300
423
+ },
424
+ {
425
+ "epoch": 8.63,
426
+ "learning_rate": 6.869009584664538e-06,
427
+ "loss": 0.0946,
428
+ "step": 5400
429
+ },
430
+ {
431
+ "epoch": 8.79,
432
+ "learning_rate": 6.070287539936103e-06,
433
+ "loss": 0.0757,
434
+ "step": 5500
435
+ },
436
+ {
437
+ "epoch": 8.95,
438
+ "learning_rate": 5.2715654952076674e-06,
439
+ "loss": 0.0923,
440
+ "step": 5600
441
+ },
442
+ {
443
+ "epoch": 9.0,
444
+ "eval_accuracy": 0.9789108027562119,
445
+ "eval_f1": 0.828852119958635,
446
+ "eval_loss": 0.07430661469697952,
447
+ "eval_precision": 0.788102261553589,
448
+ "eval_recall": 0.8740458015267175,
449
+ "eval_runtime": 67.8851,
450
+ "eval_samples_per_second": 147.411,
451
+ "eval_steps_per_second": 36.856,
452
+ "step": 5634
453
+ },
454
+ {
455
+ "epoch": 9.11,
456
+ "learning_rate": 4.472843450479233e-06,
457
+ "loss": 0.0578,
458
+ "step": 5700
459
+ },
460
+ {
461
+ "epoch": 9.27,
462
+ "learning_rate": 3.6741214057507987e-06,
463
+ "loss": 0.0664,
464
+ "step": 5800
465
+ },
466
+ {
467
+ "epoch": 9.42,
468
+ "learning_rate": 2.8753993610223644e-06,
469
+ "loss": 0.0616,
470
+ "step": 5900
471
+ },
472
+ {
473
+ "epoch": 9.58,
474
+ "learning_rate": 2.0766773162939296e-06,
475
+ "loss": 0.0759,
476
+ "step": 6000
477
+ },
478
+ {
479
+ "epoch": 9.74,
480
+ "learning_rate": 1.2779552715654952e-06,
481
+ "loss": 0.0769,
482
+ "step": 6100
483
+ },
484
+ {
485
+ "epoch": 9.9,
486
+ "learning_rate": 4.792332268370607e-07,
487
+ "loss": 0.0711,
488
+ "step": 6200
489
+ },
490
+ {
491
+ "epoch": 10.0,
492
+ "eval_accuracy": 0.9819289686983922,
493
+ "eval_f1": 0.8444211629125196,
494
+ "eval_loss": 0.06683139503002167,
495
+ "eval_precision": 0.8125,
496
+ "eval_recall": 0.8789531079607416,
497
+ "eval_runtime": 67.6072,
498
+ "eval_samples_per_second": 148.017,
499
+ "eval_steps_per_second": 37.008,
500
+ "step": 6260
501
+ },
502
+ {
503
+ "epoch": 10.0,
504
+ "step": 6260,
505
+ "total_flos": 1.307859275810304e+16,
506
+ "train_loss": 0.217802461039144,
507
+ "train_runtime": 2284.8138,
508
+ "train_samples_per_second": 43.798,
509
+ "train_steps_per_second": 2.74
510
+ }
511
+ ],
512
+ "max_steps": 6260,
513
+ "num_train_epochs": 10,
514
+ "total_flos": 1.307859275810304e+16,
515
+ "trial_name": null,
516
+ "trial_params": null
517
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:809dfcb1e2f8c5c4d1e1187ff6c6c59fb502aab130e01409079ca67d7c9e76af
3
+ size 3375