Hugging Face's logo

begnini
/
ner-bert-large-cased-pt-contratos_tceal

Token Classification
Transformers
TensorBoard
Safetensors
bert
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
begnini commited on
Commit
759b0a0
•
1 Parent(s): 2e301a4

contratos_tceal

Browse files
Files changed (6) hide show
  1. README.md +30 -30
  2. config.json +86 -84
  3. model.safetensors +2 -2
  4. runs/Jan02_13-10-48_5f1698292a1e/events.out.tfevents.1704202755.5f1698292a1e.206.2 +3 -0
  5. runs/Jan02_13-10-48_5f1698292a1e/events.out.tfevents.1704205367.5f1698292a1e.206.3 +3 -0
  6. training_args.bin +1 -1
README.md CHANGED
@@ -23,16 +23,16 @@ model-index:
23
  metrics:
24
  - name: Precision
25
  type: precision
26
- value: 0.9134177215189874
27
  - name: Recall
28
  type: recall
29
- value: 0.9168996188055909
30
  - name: F1
31
  type: f1
32
- value: 0.9151553582752061
33
  - name: Accuracy
34
  type: accuracy
35
- value: 0.9556322655972385
36
  ---
37
 
38
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -42,11 +42,11 @@ should probably proofread and complete it, then remove this comment. -->
42
 
43
  This model was trained from scratch on the contratos_tceal dataset.
44
  It achieves the following results on the evaluation set:
45
- - Loss: 0.3141
46
- - Precision: 0.9134
47
- - Recall: 0.9169
48
- - F1: 0.9152
49
- - Accuracy: 0.9556
50
 
51
  ## Model description
52
 
@@ -77,31 +77,31 @@ The following hyperparameters were used during training:
77
 
78
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
79
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
80
- | No log | 1.0 | 252 | 0.2193 | 0.9026 | 0.8948 | 0.8987 | 0.9488 |
81
- | 0.2496 | 2.0 | 504 | 0.2110 | 0.8957 | 0.9098 | 0.9027 | 0.9494 |
82
- | 0.2496 | 3.0 | 756 | 0.2098 | 0.9166 | 0.9105 | 0.9136 | 0.9531 |
83
- | 0.1666 | 4.0 | 1008 | 0.2063 | 0.9221 | 0.9146 | 0.9183 | 0.9559 |
84
- | 0.1666 | 5.0 | 1260 | 0.2165 | 0.9219 | 0.9146 | 0.9182 | 0.9562 |
85
- | 0.1255 | 6.0 | 1512 | 0.2143 | 0.9175 | 0.9133 | 0.9154 | 0.9555 |
86
- | 0.1255 | 7.0 | 1764 | 0.2278 | 0.9181 | 0.9146 | 0.9164 | 0.9559 |
87
- | 0.092 | 8.0 | 2016 | 0.2404 | 0.9188 | 0.9174 | 0.9181 | 0.9561 |
88
- | 0.092 | 9.0 | 2268 | 0.2538 | 0.9133 | 0.9100 | 0.9117 | 0.9533 |
89
- | 0.069 | 10.0 | 2520 | 0.2654 | 0.9132 | 0.9118 | 0.9125 | 0.9543 |
90
- | 0.069 | 11.0 | 2772 | 0.2796 | 0.9085 | 0.9133 | 0.9109 | 0.9527 |
91
- | 0.0498 | 12.0 | 3024 | 0.2827 | 0.9130 | 0.9149 | 0.9139 | 0.9552 |
92
- | 0.0498 | 13.0 | 3276 | 0.2869 | 0.9127 | 0.9144 | 0.9135 | 0.9557 |
93
- | 0.0397 | 14.0 | 3528 | 0.2993 | 0.9123 | 0.9093 | 0.9108 | 0.9546 |
94
- | 0.0397 | 15.0 | 3780 | 0.2951 | 0.9056 | 0.9144 | 0.9100 | 0.9547 |
95
- | 0.0312 | 16.0 | 4032 | 0.2989 | 0.9092 | 0.9136 | 0.9114 | 0.9566 |
96
- | 0.0312 | 17.0 | 4284 | 0.3104 | 0.9115 | 0.9113 | 0.9114 | 0.9554 |
97
- | 0.0257 | 18.0 | 4536 | 0.3098 | 0.9143 | 0.9161 | 0.9152 | 0.9564 |
98
- | 0.0257 | 19.0 | 4788 | 0.3129 | 0.9141 | 0.9166 | 0.9154 | 0.9556 |
99
- | 0.0207 | 20.0 | 5040 | 0.3141 | 0.9134 | 0.9169 | 0.9152 | 0.9556 |
100
 
101
 
102
  ### Framework versions
103
 
104
  - Transformers 4.36.2
105
  - Pytorch 2.1.0+cu121
106
- - Datasets 2.16.0
107
  - Tokenizers 0.15.0
 
23
  metrics:
24
  - name: Precision
25
  type: precision
26
+ value: 0.8839137645107794
27
  - name: Recall
28
  type: recall
29
+ value: 0.9055001061796559
30
  - name: F1
31
  type: f1
32
+ value: 0.8945767334522186
33
  - name: Accuracy
34
  type: accuracy
35
+ value: 0.9379326903837688
36
  ---
37
 
38
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
42
 
43
  This model was trained from scratch on the contratos_tceal dataset.
44
  It achieves the following results on the evaluation set:
45
+ - Loss: 0.4354
46
+ - Precision: 0.8839
47
+ - Recall: 0.9055
48
+ - F1: 0.8946
49
+ - Accuracy: 0.9379
50
 
51
  ## Model description
52
 
 
77
 
78
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
79
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
80
+ | No log | 1.0 | 283 | 0.2557 | 0.8770 | 0.8824 | 0.8796 | 0.9320 |
81
+ | 0.2435 | 2.0 | 566 | 0.2473 | 0.8861 | 0.8919 | 0.8890 | 0.9381 |
82
+ | 0.2435 | 3.0 | 849 | 0.2722 | 0.8818 | 0.8951 | 0.8884 | 0.9386 |
83
+ | 0.172 | 4.0 | 1132 | 0.2934 | 0.8834 | 0.9008 | 0.8920 | 0.9364 |
84
+ | 0.172 | 5.0 | 1415 | 0.2876 | 0.8903 | 0.9010 | 0.8956 | 0.9414 |
85
+ | 0.13 | 6.0 | 1698 | 0.3027 | 0.8785 | 0.8996 | 0.8889 | 0.9360 |
86
+ | 0.13 | 7.0 | 1981 | 0.3205 | 0.8755 | 0.9004 | 0.8878 | 0.9356 |
87
+ | 0.0996 | 8.0 | 2264 | 0.3290 | 0.8818 | 0.9042 | 0.8928 | 0.9384 |
88
+ | 0.075 | 9.0 | 2547 | 0.3765 | 0.8798 | 0.9049 | 0.8922 | 0.9361 |
89
+ | 0.075 | 10.0 | 2830 | 0.3684 | 0.8779 | 0.9051 | 0.8913 | 0.9399 |
90
+ | 0.054 | 11.0 | 3113 | 0.3853 | 0.8799 | 0.9051 | 0.8923 | 0.9380 |
91
+ | 0.054 | 12.0 | 3396 | 0.3975 | 0.8774 | 0.9055 | 0.8912 | 0.9352 |
92
+ | 0.0457 | 13.0 | 3679 | 0.3974 | 0.8844 | 0.9066 | 0.8953 | 0.9387 |
93
+ | 0.0457 | 14.0 | 3962 | 0.4117 | 0.8758 | 0.9049 | 0.8901 | 0.9375 |
94
+ | 0.0326 | 15.0 | 4245 | 0.4128 | 0.8757 | 0.9036 | 0.8894 | 0.9380 |
95
+ | 0.0265 | 16.0 | 4528 | 0.4304 | 0.8838 | 0.9042 | 0.8939 | 0.9377 |
96
+ | 0.0265 | 17.0 | 4811 | 0.4258 | 0.8815 | 0.9066 | 0.8938 | 0.9384 |
97
+ | 0.0214 | 18.0 | 5094 | 0.4304 | 0.8783 | 0.9070 | 0.8924 | 0.9375 |
98
+ | 0.0214 | 19.0 | 5377 | 0.4351 | 0.8828 | 0.9055 | 0.8940 | 0.9383 |
99
+ | 0.018 | 20.0 | 5660 | 0.4354 | 0.8839 | 0.9055 | 0.8946 | 0.9379 |
100
 
101
 
102
  ### Framework versions
103
 
104
  - Transformers 4.36.2
105
  - Pytorch 2.1.0+cu121
106
+ - Datasets 2.16.1
107
  - Tokenizers 0.15.0
config.json CHANGED
@@ -63,52 +63,53 @@
63
  "54": "L-NUMERO_CONTRATO",
64
  "55": "U-NUMERO_CONTRATO",
65
  "56": "B-NUMERO_EDITAL",
66
- "57": "L-NUMERO_EDITAL",
67
- "58": "U-NUMERO_EDITAL",
68
- "59": "B-NUMERO_LICITACAO",
69
  "6": "L-CNPJ_CONTRATANTE",
70
- "60": "I-NUMERO_LICITACAO",
71
- "61": "L-NUMERO_LICITACAO",
72
- "62": "U-NUMERO_LICITACAO",
73
- "63": "B-NUMERO_PROCESSO",
74
- "64": "I-NUMERO_PROCESSO",
75
- "65": "L-NUMERO_PROCESSO",
76
- "66": "U-NUMERO_PROCESSO",
77
- "67": "O",
78
- "68": "B-OBJETO",
79
- "69": "I-OBJETO",
80
  "7": "U-CNPJ_CONTRATANTE",
81
- "70": "L-OBJETO",
82
- "71": "U-OBJETO",
83
- "72": "B-ORGAO",
84
- "73": "I-ORGAO",
85
- "74": "L-ORGAO",
86
- "75": "B-ORGAO_CONTRATANTE",
87
- "76": "I-ORGAO_CONTRATANTE",
88
- "77": "L-ORGAO_CONTRATANTE",
89
- "78": "U-ORGAO_CONTRATANTE",
90
- "79": "B-PRAZO",
91
  "8": "B-CRITERIO_JULGAMENTO",
92
- "80": "I-PRAZO",
93
- "81": "L-PRAZO",
94
- "82": "U-PRAZO",
95
- "83": "B-TIPO",
96
- "84": "I-TIPO",
97
- "85": "L-TIPO",
98
- "86": "U-TIPO",
99
- "87": "B-TIPO_PUBLICACAO",
100
- "88": "I-TIPO_PUBLICACAO",
101
- "89": "L-TIPO_PUBLICACAO",
102
  "9": "I-CRITERIO_JULGAMENTO",
103
- "90": "U-TIPO_PUBLICACAO",
104
- "91": "B-VALOR",
105
- "92": "I-VALOR",
106
- "93": "L-VALOR",
107
- "94": "U-VALOR",
108
- "95": "B-VIGENCIA",
109
- "96": "I-VIGENCIA",
110
- "97": "L-VIGENCIA",
111
- "98": "U-VIGENCIA"
 
112
  },
113
  "initializer_range": 0.02,
114
  "intermediate_size": 4096,
@@ -128,16 +129,16 @@
128
  "B-NUMERO_ATO": 48,
129
  "B-NUMERO_CONTRATO": 52,
130
  "B-NUMERO_EDITAL": 56,
131
- "B-NUMERO_LICITACAO": 59,
132
- "B-NUMERO_PROCESSO": 63,
133
- "B-OBJETO": 68,
134
- "B-ORGAO": 72,
135
- "B-ORGAO_CONTRATANTE": 75,
136
- "B-PRAZO": 79,
137
- "B-TIPO": 83,
138
- "B-TIPO_PUBLICACAO": 87,
139
- "B-VALOR": 91,
140
- "B-VIGENCIA": 95,
141
  "I-CNPJ_CONTRATADA": 1,
142
  "I-CNPJ_CONTRATANTE": 5,
143
  "I-CRITERIO_JULGAMENTO": 9,
@@ -152,16 +153,17 @@
152
  "I-MODALIDADE": 45,
153
  "I-NUMERO_ATO": 49,
154
  "I-NUMERO_CONTRATO": 53,
155
- "I-NUMERO_LICITACAO": 60,
156
- "I-NUMERO_PROCESSO": 64,
157
- "I-OBJETO": 69,
158
- "I-ORGAO": 73,
159
- "I-ORGAO_CONTRATANTE": 76,
160
- "I-PRAZO": 80,
161
- "I-TIPO": 84,
162
- "I-TIPO_PUBLICACAO": 88,
163
- "I-VALOR": 92,
164
- "I-VIGENCIA": 96,
 
165
  "L-CNPJ_CONTRATADA": 2,
166
  "L-CNPJ_CONTRATANTE": 6,
167
  "L-CRITERIO_JULGAMENTO": 10,
@@ -176,18 +178,18 @@
176
  "L-MODALIDADE": 46,
177
  "L-NUMERO_ATO": 50,
178
  "L-NUMERO_CONTRATO": 54,
179
- "L-NUMERO_EDITAL": 57,
180
- "L-NUMERO_LICITACAO": 61,
181
- "L-NUMERO_PROCESSO": 65,
182
- "L-OBJETO": 70,
183
- "L-ORGAO": 74,
184
- "L-ORGAO_CONTRATANTE": 77,
185
- "L-PRAZO": 81,
186
- "L-TIPO": 85,
187
- "L-TIPO_PUBLICACAO": 89,
188
- "L-VALOR": 93,
189
- "L-VIGENCIA": 97,
190
- "O": 67,
191
  "U-CNPJ_CONTRATADA": 3,
192
  "U-CNPJ_CONTRATANTE": 7,
193
  "U-CRITERIO_JULGAMENTO": 11,
@@ -202,16 +204,16 @@
202
  "U-MODALIDADE": 47,
203
  "U-NUMERO_ATO": 51,
204
  "U-NUMERO_CONTRATO": 55,
205
- "U-NUMERO_EDITAL": 58,
206
- "U-NUMERO_LICITACAO": 62,
207
- "U-NUMERO_PROCESSO": 66,
208
- "U-OBJETO": 71,
209
- "U-ORGAO_CONTRATANTE": 78,
210
- "U-PRAZO": 82,
211
- "U-TIPO": 86,
212
- "U-TIPO_PUBLICACAO": 90,
213
- "U-VALOR": 94,
214
- "U-VIGENCIA": 98
215
  },
216
  "layer_norm_eps": 1e-12,
217
  "max_position_embeddings": 512,
 
63
  "54": "L-NUMERO_CONTRATO",
64
  "55": "U-NUMERO_CONTRATO",
65
  "56": "B-NUMERO_EDITAL",
66
+ "57": "I-NUMERO_EDITAL",
67
+ "58": "L-NUMERO_EDITAL",
68
+ "59": "U-NUMERO_EDITAL",
69
  "6": "L-CNPJ_CONTRATANTE",
70
+ "60": "B-NUMERO_LICITACAO",
71
+ "61": "I-NUMERO_LICITACAO",
72
+ "62": "L-NUMERO_LICITACAO",
73
+ "63": "U-NUMERO_LICITACAO",
74
+ "64": "B-NUMERO_PROCESSO",
75
+ "65": "I-NUMERO_PROCESSO",
76
+ "66": "L-NUMERO_PROCESSO",
77
+ "67": "U-NUMERO_PROCESSO",
78
+ "68": "O",
79
+ "69": "B-OBJETO",
80
  "7": "U-CNPJ_CONTRATANTE",
81
+ "70": "I-OBJETO",
82
+ "71": "L-OBJETO",
83
+ "72": "U-OBJETO",
84
+ "73": "B-ORGAO",
85
+ "74": "I-ORGAO",
86
+ "75": "L-ORGAO",
87
+ "76": "B-ORGAO_CONTRATANTE",
88
+ "77": "I-ORGAO_CONTRATANTE",
89
+ "78": "L-ORGAO_CONTRATANTE",
90
+ "79": "U-ORGAO_CONTRATANTE",
91
  "8": "B-CRITERIO_JULGAMENTO",
92
+ "80": "B-PRAZO",
93
+ "81": "I-PRAZO",
94
+ "82": "L-PRAZO",
95
+ "83": "U-PRAZO",
96
+ "84": "B-TIPO",
97
+ "85": "I-TIPO",
98
+ "86": "L-TIPO",
99
+ "87": "U-TIPO",
100
+ "88": "B-TIPO_PUBLICACAO",
101
+ "89": "I-TIPO_PUBLICACAO",
102
  "9": "I-CRITERIO_JULGAMENTO",
103
+ "90": "L-TIPO_PUBLICACAO",
104
+ "91": "U-TIPO_PUBLICACAO",
105
+ "92": "B-VALOR",
106
+ "93": "I-VALOR",
107
+ "94": "L-VALOR",
108
+ "95": "U-VALOR",
109
+ "96": "B-VIGENCIA",
110
+ "97": "I-VIGENCIA",
111
+ "98": "L-VIGENCIA",
112
+ "99": "U-VIGENCIA"
113
  },
114
  "initializer_range": 0.02,
115
  "intermediate_size": 4096,
 
129
  "B-NUMERO_ATO": 48,
130
  "B-NUMERO_CONTRATO": 52,
131
  "B-NUMERO_EDITAL": 56,
132
+ "B-NUMERO_LICITACAO": 60,
133
+ "B-NUMERO_PROCESSO": 64,
134
+ "B-OBJETO": 69,
135
+ "B-ORGAO": 73,
136
+ "B-ORGAO_CONTRATANTE": 76,
137
+ "B-PRAZO": 80,
138
+ "B-TIPO": 84,
139
+ "B-TIPO_PUBLICACAO": 88,
140
+ "B-VALOR": 92,
141
+ "B-VIGENCIA": 96,
142
  "I-CNPJ_CONTRATADA": 1,
143
  "I-CNPJ_CONTRATANTE": 5,
144
  "I-CRITERIO_JULGAMENTO": 9,
 
153
  "I-MODALIDADE": 45,
154
  "I-NUMERO_ATO": 49,
155
  "I-NUMERO_CONTRATO": 53,
156
+ "I-NUMERO_EDITAL": 57,
157
+ "I-NUMERO_LICITACAO": 61,
158
+ "I-NUMERO_PROCESSO": 65,
159
+ "I-OBJETO": 70,
160
+ "I-ORGAO": 74,
161
+ "I-ORGAO_CONTRATANTE": 77,
162
+ "I-PRAZO": 81,
163
+ "I-TIPO": 85,
164
+ "I-TIPO_PUBLICACAO": 89,
165
+ "I-VALOR": 93,
166
+ "I-VIGENCIA": 97,
167
  "L-CNPJ_CONTRATADA": 2,
168
  "L-CNPJ_CONTRATANTE": 6,
169
  "L-CRITERIO_JULGAMENTO": 10,
 
178
  "L-MODALIDADE": 46,
179
  "L-NUMERO_ATO": 50,
180
  "L-NUMERO_CONTRATO": 54,
181
+ "L-NUMERO_EDITAL": 58,
182
+ "L-NUMERO_LICITACAO": 62,
183
+ "L-NUMERO_PROCESSO": 66,
184
+ "L-OBJETO": 71,
185
+ "L-ORGAO": 75,
186
+ "L-ORGAO_CONTRATANTE": 78,
187
+ "L-PRAZO": 82,
188
+ "L-TIPO": 86,
189
+ "L-TIPO_PUBLICACAO": 90,
190
+ "L-VALOR": 94,
191
+ "L-VIGENCIA": 98,
192
+ "O": 68,
193
  "U-CNPJ_CONTRATADA": 3,
194
  "U-CNPJ_CONTRATANTE": 7,
195
  "U-CRITERIO_JULGAMENTO": 11,
 
204
  "U-MODALIDADE": 47,
205
  "U-NUMERO_ATO": 51,
206
  "U-NUMERO_CONTRATO": 55,
207
+ "U-NUMERO_EDITAL": 59,
208
+ "U-NUMERO_LICITACAO": 63,
209
+ "U-NUMERO_PROCESSO": 67,
210
+ "U-OBJETO": 72,
211
+ "U-ORGAO_CONTRATANTE": 79,
212
+ "U-PRAZO": 83,
213
+ "U-TIPO": 87,
214
+ "U-TIPO_PUBLICACAO": 91,
215
+ "U-VALOR": 95,
216
+ "U-VIGENCIA": 99
217
  },
218
  "layer_norm_eps": 1e-12,
219
  "max_position_embeddings": 512,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:40154cfcc4dca7adba2328d21c48df308d4a0a918750919f5fd63b76de17022b
3
- size 1333839980
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07f1daa10f8b3aaa622f914a0dcaacb8d45c8a52fd69f08f83bf4c40dba5bc1c
3
+ size 1333844080
runs/Jan02_13-10-48_5f1698292a1e/events.out.tfevents.1704202755.5f1698292a1e.206.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33ea39dc7f7ad43b536a154ea323504278b66e61ba2fefa17364427f928d971f
3
+ size 21548
runs/Jan02_13-10-48_5f1698292a1e/events.out.tfevents.1704205367.5f1698292a1e.206.3 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77aadaee5f1b0c8ff24611e66d9d48776b22135a8752938ca730b704b7361767
3
+ size 560
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b9b11df3867a155c276c495f9df086abef6be28c311798fdf4acab64dcd8f7b0
3
  size 4728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f21e902394efda8e7257c858ca94236105e41ba1cd3abf516c876936bd0172b9
3
  size 4728