Hugging Face's logo

datalawyer
/
pedidos-v5.3-distill

Token Classification
Transformers
Safetensors
roberta
Generated from Trainer
Inference Endpoints
Model card Files Community
adalbertojunior commited on
Commit
23ac9f2
•
1 Parent(s): f452269

Upload 16 files

Browse files
Files changed (7) hide show
  1. predictions.txt +0 -0
  2. predictions_classification_report.txt +23 -0
  3. tokenizer.json +1 -6
  4. tokenizer_config.json +0 -1
  5. train_results.json +6 -6
  6. trainer_state.json +119 -96
  7. training_args.bin +2 -2
predictions.txt ADDED
The diff for this file is too large to render. See raw diff
 
predictions_classification_report.txt ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ precision recall f1-score support
2
+
3
+ ATRIBUICAO 0.82 0.82 0.82 221
4
+ DECISAO 0.81 0.82 0.82 544
5
+ FUNCAO 0.94 0.89 0.91 486
6
+ FUNDAMENTO 0.89 0.83 0.86 1501
7
+ LOCAL 0.85 0.84 0.85 245
8
+ ORGANIZACAO 0.90 0.86 0.88 626
9
+ PEDIDO 0.86 0.81 0.83 4341
10
+ PESSOA 0.95 0.94 0.95 654
11
+ REFLEXO 0.85 0.84 0.85 358
12
+ TIPO_ACAO 0.93 0.89 0.91 341
13
+ TRIBUNAL 0.96 0.92 0.94 190
14
+ VALOR_ACORDO 0.91 0.71 0.79 41
15
+ VALOR_CAUSA 0.89 0.92 0.90 62
16
+ VALOR_CONDENACAO 0.89 0.76 0.82 72
17
+ VALOR_CUSTAS 0.95 0.93 0.94 134
18
+ VALOR_PEDIDO 0.94 0.81 0.87 308
19
+ VARA 0.95 0.96 0.96 81
20
+
21
+ micro avg 0.88 0.84 0.86 10205
22
+ macro avg 0.90 0.86 0.88 10205
23
+ weighted avg 0.88 0.84 0.86 10205
tokenizer.json CHANGED
@@ -1,11 +1,6 @@
1
  {
2
  "version": "1.0",
3
- "truncation": {
4
- "direction": "Right",
5
- "max_length": 512,
6
- "strategy": "LongestFirst",
7
- "stride": 0
8
- },
9
  "padding": null,
10
  "added_tokens": [
11
  {
 
1
  {
2
  "version": "1.0",
3
+ "truncation": null,
 
 
 
 
 
4
  "padding": null,
5
  "added_tokens": [
6
  {
tokenizer_config.json CHANGED
@@ -47,7 +47,6 @@
47
  "cls_token": "<s>",
48
  "eos_token": "</s>",
49
  "errors": "replace",
50
- "extra_special_tokens": {},
51
  "mask_token": "<mask>",
52
  "max_length": 512,
53
  "model_max_length": 512,
 
47
  "cls_token": "<s>",
48
  "eos_token": "</s>",
49
  "errors": "replace",
 
50
  "mask_token": "<mask>",
51
  "max_length": 512,
52
  "model_max_length": 512,
train_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 19.997926744989634,
3
- "total_flos": 6132708568035504.0,
4
- "train_loss": 0.8169754918592458,
5
- "train_runtime": 915.3849,
6
  "train_samples": 23148,
7
- "train_samples_per_second": 505.754,
8
- "train_steps_per_second": 7.887
9
  }
 
1
  {
2
+ "epoch": 9.979267449896337,
3
+ "total_flos": 6112468258365600.0,
4
+ "train_loss": 0.10573644003214268,
5
+ "train_runtime": 1226.1602,
6
  "train_samples": 23148,
7
+ "train_samples_per_second": 188.784,
8
+ "train_steps_per_second": 2.944
9
  }
trainer_state.json CHANGED
@@ -1,138 +1,161 @@
1
  {
2
- "best_metric": 0.8090804377039739,
3
- "best_model_checkpoint": "./models/entities/test_v6/checkpoint-7000",
4
- "epoch": 19.997926744989634,
5
- "eval_steps": 7000,
6
- "global_step": 7220,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 1.3842432619212164,
13
- "grad_norm": 7.1875,
14
  "learning_rate": 0.0002,
15
- "loss": 8.369,
16
  "step": 500
17
  },
18
  {
19
- "epoch": 2.7684865238424328,
20
- "grad_norm": 4.84375,
21
- "learning_rate": 0.00018511904761904765,
22
- "loss": 1.059,
23
- "step": 1000
24
- },
25
- {
26
- "epoch": 4.1548030407740155,
27
- "grad_norm": 5.28125,
28
- "learning_rate": 0.00017023809523809523,
29
- "loss": 0.6034,
30
- "step": 1500
31
  },
32
  {
33
- "epoch": 5.5390463026952315,
34
- "grad_norm": 3.90625,
35
- "learning_rate": 0.00015535714285714287,
36
- "loss": 0.3745,
37
- "step": 2000
38
  },
39
  {
40
- "epoch": 6.923289564616448,
41
- "grad_norm": 3.515625,
42
- "learning_rate": 0.00014047619047619049,
43
- "loss": 0.2708,
44
- "step": 2500
 
 
 
 
 
45
  },
46
  {
47
- "epoch": 8.309606081548031,
48
- "grad_norm": 3.203125,
49
- "learning_rate": 0.0001255952380952381,
50
- "loss": 0.1967,
51
- "step": 3000
52
  },
53
  {
54
- "epoch": 9.693849343469246,
55
- "grad_norm": 2.3125,
56
- "learning_rate": 0.00011071428571428572,
57
- "loss": 0.1566,
58
- "step": 3500
 
 
 
 
 
59
  },
60
  {
61
- "epoch": 11.080165860400829,
62
- "grad_norm": 3.03125,
63
- "learning_rate": 9.583333333333334e-05,
64
- "loss": 0.1318,
65
- "step": 4000
66
  },
67
  {
68
- "epoch": 12.464409122322046,
69
- "grad_norm": 4.84375,
70
- "learning_rate": 8.095238095238096e-05,
71
- "loss": 0.1131,
72
- "step": 4500
 
 
 
 
 
73
  },
74
  {
75
- "epoch": 13.848652384243263,
76
- "grad_norm": 2.484375,
77
- "learning_rate": 6.607142857142857e-05,
78
- "loss": 0.1041,
79
- "step": 5000
80
  },
81
  {
82
- "epoch": 15.234968901174845,
83
- "grad_norm": 3.078125,
84
- "learning_rate": 5.119047619047619e-05,
85
- "loss": 0.0986,
86
- "step": 5500
 
 
 
 
 
87
  },
88
  {
89
- "epoch": 16.619212163096062,
90
- "grad_norm": 2.53125,
91
- "learning_rate": 3.630952380952381e-05,
92
- "loss": 0.0951,
93
- "step": 6000
94
  },
95
  {
96
- "epoch": 18.005528680027645,
97
- "grad_norm": 3.53125,
98
- "learning_rate": 2.1428571428571428e-05,
99
- "loss": 0.0921,
100
- "step": 6500
 
 
 
 
 
101
  },
102
  {
103
- "epoch": 19.389771941948858,
104
- "grad_norm": 1.671875,
105
- "learning_rate": 6.547619047619048e-06,
106
- "loss": 0.0925,
107
- "step": 7000
108
  },
109
  {
110
- "epoch": 19.389771941948858,
111
- "eval_accuracy": 0.9699217442249749,
112
- "eval_f1": 0.8090804377039739,
113
- "eval_loss": 0.15359356999397278,
114
- "eval_precision": 0.7678083439606486,
115
- "eval_recall": 0.8550415905863258,
116
- "eval_runtime": 169.5033,
117
- "eval_samples_per_second": 29.262,
118
- "eval_steps_per_second": 29.262,
119
- "step": 7000
120
  },
121
  {
122
- "epoch": 19.997926744989634,
123
- "step": 7220,
124
- "total_flos": 6132708568035504.0,
125
- "train_loss": 0.8169754918592458,
126
- "train_runtime": 915.3849,
127
- "train_samples_per_second": 505.754,
128
- "train_steps_per_second": 7.887
129
  }
130
  ],
131
  "logging_steps": 500,
132
- "max_steps": 7220,
133
  "num_input_tokens_seen": 0,
134
- "num_train_epochs": 20,
135
- "save_steps": 7000,
136
  "stateful_callbacks": {
137
  "TrainerControl": {
138
  "args": {
@@ -145,7 +168,7 @@
145
  "attributes": {}
146
  }
147
  },
148
- "total_flos": 6132708568035504.0,
149
  "train_batch_size": 8,
150
  "trial_name": null,
151
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.8648728562980484,
3
+ "best_model_checkpoint": "./models/entities/test_v4/checkpoint-3500",
4
+ "epoch": 9.979267449896337,
5
+ "eval_steps": 500,
6
+ "global_step": 3610,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 1.38217000691085,
13
+ "grad_norm": 0.9774827361106873,
14
  "learning_rate": 0.0002,
15
+ "loss": 0.5764,
16
  "step": 500
17
  },
18
  {
19
+ "epoch": 1.38217000691085,
20
+ "eval_accuracy": 0.9593812772012053,
21
+ "eval_f1": 0.7594292979234355,
22
+ "eval_loss": 0.1411387324333191,
23
+ "eval_precision": 0.708673872923807,
24
+ "eval_recall": 0.8180158247108947,
25
+ "eval_runtime": 199.3049,
26
+ "eval_samples_per_second": 24.886,
27
+ "eval_steps_per_second": 24.886,
28
+ "step": 500
 
 
29
  },
30
  {
31
+ "epoch": 2.7643400138217,
32
+ "grad_norm": 0.7694200873374939,
33
+ "learning_rate": 0.00016784565916398716,
34
+ "loss": 0.09,
35
+ "step": 1000
36
  },
37
  {
38
+ "epoch": 2.7643400138217,
39
+ "eval_accuracy": 0.9687447689989956,
40
+ "eval_f1": 0.8259538274605103,
41
+ "eval_loss": 0.11496158689260483,
42
+ "eval_precision": 0.792852477372399,
43
+ "eval_recall": 0.8619395414891459,
44
+ "eval_runtime": 48.3856,
45
+ "eval_samples_per_second": 102.51,
46
+ "eval_steps_per_second": 102.51,
47
+ "step": 1000
48
  },
49
  {
50
+ "epoch": 4.14651002073255,
51
+ "grad_norm": 0.625320315361023,
52
+ "learning_rate": 0.0001356913183279743,
53
+ "loss": 0.0463,
54
+ "step": 1500
55
  },
56
  {
57
+ "epoch": 4.14651002073255,
58
+ "eval_accuracy": 0.9703716103113492,
59
+ "eval_f1": 0.8380395025603511,
60
+ "eval_loss": 0.13039253652095795,
61
+ "eval_precision": 0.8069878839109609,
62
+ "eval_recall": 0.8715763846622033,
63
+ "eval_runtime": 46.9044,
64
+ "eval_samples_per_second": 105.747,
65
+ "eval_steps_per_second": 105.747,
66
+ "step": 1500
67
  },
68
  {
69
+ "epoch": 5.5286800276434,
70
+ "grad_norm": 0.4629702866077423,
71
+ "learning_rate": 0.00010353697749196143,
72
+ "loss": 0.023,
73
+ "step": 2000
74
  },
75
  {
76
+ "epoch": 5.5286800276434,
77
+ "eval_accuracy": 0.9716950535654503,
78
+ "eval_f1": 0.8417248620403379,
79
+ "eval_loss": 0.1349513679742813,
80
+ "eval_precision": 0.8115641774178359,
81
+ "eval_recall": 0.8742138364779874,
82
+ "eval_runtime": 47.4889,
83
+ "eval_samples_per_second": 104.446,
84
+ "eval_steps_per_second": 104.446,
85
+ "step": 2000
86
  },
87
  {
88
+ "epoch": 6.91085003455425,
89
+ "grad_norm": 0.1776706427335739,
90
+ "learning_rate": 7.138263665594856e-05,
91
+ "loss": 0.0147,
92
+ "step": 2500
93
  },
94
  {
95
+ "epoch": 6.91085003455425,
96
+ "eval_accuracy": 0.9736305239370606,
97
+ "eval_f1": 0.8591340450771056,
98
+ "eval_loss": 0.14083334803581238,
99
+ "eval_precision": 0.8377674956622325,
100
+ "eval_recall": 0.8816189896530736,
101
+ "eval_runtime": 47.4474,
102
+ "eval_samples_per_second": 104.537,
103
+ "eval_steps_per_second": 104.537,
104
+ "step": 2500
105
  },
106
  {
107
+ "epoch": 8.2930200414651,
108
+ "grad_norm": 0.7263190746307373,
109
+ "learning_rate": 3.92282958199357e-05,
110
+ "loss": 0.0078,
111
+ "step": 3000
112
  },
113
  {
114
+ "epoch": 8.2930200414651,
115
+ "eval_accuracy": 0.9740594660194175,
116
+ "eval_f1": 0.8621834580504485,
117
+ "eval_loss": 0.14879639446735382,
118
+ "eval_precision": 0.8428446855924814,
119
+ "eval_recall": 0.8824305132886995,
120
+ "eval_runtime": 49.4794,
121
+ "eval_samples_per_second": 100.244,
122
+ "eval_steps_per_second": 100.244,
123
+ "step": 3000
124
  },
125
  {
126
+ "epoch": 9.675190048375951,
127
+ "grad_norm": 0.21800673007965088,
128
+ "learning_rate": 7.07395498392283e-06,
129
+ "loss": 0.0045,
130
+ "step": 3500
131
  },
132
  {
133
+ "epoch": 9.675190048375951,
134
+ "eval_accuracy": 0.9743680950786743,
135
+ "eval_f1": 0.8648728562980484,
136
+ "eval_loss": 0.16495844721794128,
137
+ "eval_precision": 0.8410005750431282,
138
+ "eval_recall": 0.8901399878271454,
139
+ "eval_runtime": 46.0436,
140
+ "eval_samples_per_second": 107.724,
141
+ "eval_steps_per_second": 107.724,
142
+ "step": 3500
143
  },
144
  {
145
+ "epoch": 9.979267449896337,
146
+ "step": 3610,
147
+ "total_flos": 6112468258365600.0,
148
+ "train_loss": 0.10573644003214268,
149
+ "train_runtime": 1226.1602,
150
+ "train_samples_per_second": 188.784,
151
+ "train_steps_per_second": 2.944
152
  }
153
  ],
154
  "logging_steps": 500,
155
+ "max_steps": 3610,
156
  "num_input_tokens_seen": 0,
157
+ "num_train_epochs": 10,
158
+ "save_steps": 500,
159
  "stateful_callbacks": {
160
  "TrainerControl": {
161
  "args": {
 
168
  "attributes": {}
169
  }
170
  },
171
+ "total_flos": 6112468258365600.0,
172
  "train_batch_size": 8,
173
  "trial_name": null,
174
  "trial_params": null
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:885427ad062bd54a33ec4c0969e5b2302b29c0637541f9098bc04c866f12a2cf
3
- size 5432
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7b5b580be52a23cc8fd6464c850212c94c3c5d774845e938169137e4340fb07
3
+ size 5240