eraldoluis commited on
Commit
1e5d80c
1 Parent(s): 61c9406

First version

Browse files
.gitattributes CHANGED
@@ -30,3 +30,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
30
  *.zip filter=lfs diff=lfs merge=lfs -text
31
  *.zst filter=lfs diff=lfs merge=lfs -text
32
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
30
  *.zip filter=lfs diff=lfs merge=lfs -text
31
  *.zst filter=lfs diff=lfs merge=lfs -text
32
  *tfevents* filter=lfs diff=lfs merge=lfs -text
33
+ pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -1,3 +1,72 @@
1
  ---
2
  license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: apache-2.0
3
+ language: br
4
+ library_name: transformers
5
+ tags:
6
+ - question-answering
7
+ datasets:
8
+ - eraldoluis/faquad
9
+ metrics:
10
+ - squad
11
+
12
+ model-index:
13
+ - name: faquad-bert-base-portuguese-cased
14
+ results:
15
+ - task:
16
+ type: extractive-qa
17
+ name: Extractive Question-Answering
18
+ dataset:
19
+ type: eraldoluis/faquad
20
+ name: FaQuAD
21
+ split: eval
22
+ metrics:
23
+ - type: squad
24
+ value:
25
+ - f1: 83.0912959832023
26
+ exact_match: 74.53169347209082
27
+ name: Eval F1 and ExactMatch
28
+ verified: false
29
  ---
30
+
31
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
32
+ should probably proofread and complete it, then remove this comment. -->
33
+
34
+ # tmp_exs_faquad
35
+
36
+ This model is a fine-tuned version of [neuralmind/bert-base-portuguese-cased](https://huggingface.co/neuralmind/bert-base-portuguese-cased) on the [FaQuAD dataset](https://huggingface.co/datasets/eraldoluis/faquad).
37
+
38
+ ## Model description
39
+
40
+ More information needed
41
+
42
+ ## Intended uses & limitations
43
+
44
+ More information needed
45
+
46
+ ## Training and evaluation data
47
+
48
+ The model was trained on the `train` split and evaluated on the `eval` split.
49
+
50
+ ## Training procedure
51
+
52
+ ### Training hyperparameters
53
+
54
+ The following hyperparameters were used during training:
55
+ - learning_rate: 3e-05
56
+ - train_batch_size: 64
57
+ - eval_batch_size: 32
58
+ - seed: 42
59
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
+ - lr_scheduler_type: linear
61
+ - num_epochs: 3.0
62
+
63
+ ### Training results
64
+
65
+
66
+
67
+ ### Framework versions
68
+
69
+ - Transformers 4.21.3
70
+ - Pytorch 1.12.1+cu113
71
+ - Datasets 2.4.0
72
+ - Tokenizers 0.12.1
all_results.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_exact_match": 74.53169347209082,
4
+ "eval_f1": 83.0912959832023,
5
+ "eval_samples": 12198,
6
+ "train_loss": 1.1721088012826149,
7
+ "train_runtime": 6793.9155,
8
+ "train_samples": 99470,
9
+ "train_samples_per_second": 43.923,
10
+ "train_steps_per_second": 0.687
11
+ }
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "neuralmind/bert-base-portuguese-cased",
3
+ "architectures": [
4
+ "BertForQuestionAnswering"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "directionality": "bidi",
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_norm_eps": 1e-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "output_past": true,
20
+ "pad_token_id": 0,
21
+ "pooler_fc_size": 768,
22
+ "pooler_num_attention_heads": 12,
23
+ "pooler_num_fc_layers": 3,
24
+ "pooler_size_per_head": 128,
25
+ "pooler_type": "first_token_transform",
26
+ "position_embedding_type": "absolute",
27
+ "torch_dtype": "float32",
28
+ "transformers_version": "4.21.3",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 29794
32
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7caff164835d55e8d5cab822d83e6832ad27cc30ccb2bda330ee09da011bda4
3
+ size 433406769
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "do_basic_tokenize": true,
4
+ "do_lower_case": false,
5
+ "mask_token": "[MASK]",
6
+ "name_or_path": "neuralmind/bert-base-portuguese-cased",
7
+ "never_split": null,
8
+ "pad_token": "[PAD]",
9
+ "sep_token": "[SEP]",
10
+ "special_tokens_map_file": "/home/eraldo/.cache/huggingface/transformers/eecc45187d085a1169eed91017d358cc0e9cbdd5dc236bcd710059dbf0a2f816.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d",
11
+ "strip_accents": null,
12
+ "tokenize_chinese_chars": true,
13
+ "tokenizer_class": "BertTokenizer",
14
+ "unk_token": "[UNK]"
15
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.1721088012826149,
4
+ "train_runtime": 6793.9155,
5
+ "train_samples": 99470,
6
+ "train_samples_per_second": 43.923,
7
+ "train_steps_per_second": 0.687
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "global_step": 4665,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.32,
12
+ "learning_rate": 2.6784565916398713e-05,
13
+ "loss": 2.1905,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.32,
18
+ "eval_exact_match": 61.28666035950804,
19
+ "eval_f1": 71.363458978244,
20
+ "step": 500
21
+ },
22
+ {
23
+ "epoch": 0.64,
24
+ "learning_rate": 2.356913183279743e-05,
25
+ "loss": 1.4634,
26
+ "step": 1000
27
+ },
28
+ {
29
+ "epoch": 0.64,
30
+ "eval_exact_match": 67.86187322611164,
31
+ "eval_f1": 77.44564129629615,
32
+ "step": 1000
33
+ },
34
+ {
35
+ "epoch": 0.96,
36
+ "learning_rate": 2.0353697749196142e-05,
37
+ "loss": 1.2925,
38
+ "step": 1500
39
+ },
40
+ {
41
+ "epoch": 0.96,
42
+ "eval_exact_match": 70.5771050141911,
43
+ "eval_f1": 79.53930047182527,
44
+ "step": 1500
45
+ },
46
+ {
47
+ "epoch": 1.29,
48
+ "learning_rate": 1.7138263665594855e-05,
49
+ "loss": 1.0802,
50
+ "step": 2000
51
+ },
52
+ {
53
+ "epoch": 1.29,
54
+ "eval_exact_match": 72.1286660359508,
55
+ "eval_f1": 80.97163819236762,
56
+ "step": 2000
57
+ },
58
+ {
59
+ "epoch": 1.61,
60
+ "learning_rate": 1.392282958199357e-05,
61
+ "loss": 1.0438,
62
+ "step": 2500
63
+ },
64
+ {
65
+ "epoch": 1.61,
66
+ "eval_exact_match": 73.28287606433302,
67
+ "eval_f1": 81.68746817914159,
68
+ "step": 2500
69
+ },
70
+ {
71
+ "epoch": 1.93,
72
+ "learning_rate": 1.0707395498392284e-05,
73
+ "loss": 1.0276,
74
+ "step": 3000
75
+ },
76
+ {
77
+ "epoch": 1.93,
78
+ "eval_exact_match": 72.99905392620624,
79
+ "eval_f1": 81.97780926342594,
80
+ "step": 3000
81
+ },
82
+ {
83
+ "epoch": 2.25,
84
+ "learning_rate": 7.491961414790997e-06,
85
+ "loss": 0.8789,
86
+ "step": 3500
87
+ },
88
+ {
89
+ "epoch": 2.25,
90
+ "eval_exact_match": 74.03973509933775,
91
+ "eval_f1": 82.60781680674411,
92
+ "step": 3500
93
+ },
94
+ {
95
+ "epoch": 2.57,
96
+ "learning_rate": 4.276527331189711e-06,
97
+ "loss": 0.8502,
98
+ "step": 4000
99
+ },
100
+ {
101
+ "epoch": 2.57,
102
+ "eval_exact_match": 74.33301797540209,
103
+ "eval_f1": 82.95704498318004,
104
+ "step": 4000
105
+ },
106
+ {
107
+ "epoch": 2.89,
108
+ "learning_rate": 1.0610932475884245e-06,
109
+ "loss": 0.8325,
110
+ "step": 4500
111
+ },
112
+ {
113
+ "epoch": 2.89,
114
+ "eval_exact_match": 74.5600756859035,
115
+ "eval_f1": 83.17536432572425,
116
+ "step": 4500
117
+ },
118
+ {
119
+ "epoch": 3.0,
120
+ "step": 4665,
121
+ "total_flos": 5.848017388319232e+16,
122
+ "train_loss": 1.1721088012826149,
123
+ "train_runtime": 6793.9155,
124
+ "train_samples_per_second": 43.923,
125
+ "train_steps_per_second": 0.687
126
+ }
127
+ ],
128
+ "max_steps": 4665,
129
+ "num_train_epochs": 3,
130
+ "total_flos": 5.848017388319232e+16,
131
+ "trial_name": null,
132
+ "trial_params": null
133
+ }
training_args.bin ADDED
Binary file (3.31 kB). View file
 
vocab.txt ADDED
The diff for this file is too large to render. See raw diff