feature: Adiciona arquivos do modelo

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+predict_wiki_section_disease_max_seq512_ts_score_lt.txt filter=lfs diff=lfs merge=lfs -text

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "[BOS]": 29794
+}

all_results.json

@@ -0,0 +1,66 @@
+{
+    "EOP_f1": 0.7776948501972493,
+    "EOP_number": 4774,
+    "EOP_precision": 0.7919652551574375,
+    "EOP_recall": 0.7639296187683284,
+    "accuracy": 0.9479881258263277,
+    "epoch": 5.0,
+    "eval_EOP_f1": 0.7617828961399019,
+    "eval_EOP_number": 2446,
+    "eval_EOP_precision": 0.7962550156041016,
+    "eval_EOP_recall": 0.7301717089125103,
+    "eval_da_EOP_f1": 0.7858523386161578,
+    "eval_da_EOP_number": 2715,
+    "eval_da_EOP_precision": 0.826758845058967,
+    "eval_da_EOP_recall": 0.7488029465930018,
+    "eval_da_overall_accuracy": 0.9486466444197256,
+    "eval_da_overall_f1": 0.7858523386161578,
+    "eval_da_overall_precision": 0.826758845058967,
+    "eval_da_overall_recall": 0.7488029465930018,
+    "eval_loss": NaN,
+    "eval_overall_accuracy": 0.9458476753769332,
+    "eval_overall_f1": 0.7617828961399019,
+    "eval_overall_precision": 0.7962550156041016,
+    "eval_overall_recall": 0.7301717089125103,
+    "eval_runtime": 370.561,
+    "eval_samples": 1524,
+    "eval_samples_per_second": 4.113,
+    "eval_steps_per_second": 1.028,
+    "overall_accuracy": 0.9479881258263277,
+    "overall_f1": 0.7776948501972493,
+    "overall_precision": 0.7919652551574375,
+    "overall_recall": 0.7639296187683284,
+    "predict_EOP_f1": 0.7776948501972493,
+    "predict_EOP_number": 4774,
+    "predict_EOP_precision": 0.7919652551574375,
+    "predict_EOP_recall": 0.7639296187683284,
+    "predict_da_EOP_f1": 0.8040997872751885,
+    "predict_da_EOP_number": 5324,
+    "predict_da_EOP_precision": 0.8286169788760462,
+    "predict_da_EOP_recall": 0.78099173553719,
+    "predict_da_overall_accuracy": 0.9520234909659239,
+    "predict_da_overall_f1": 0.8040997872751885,
+    "predict_da_overall_precision": 0.8286169788760462,
+    "predict_da_overall_recall": 0.78099173553719,
+    "predict_examples": 718,
+    "predict_loss": NaN,
+    "predict_overall_accuracy": 0.9479881258263277,
+    "predict_overall_f1": 0.7776948501972493,
+    "predict_overall_precision": 0.7919652551574375,
+    "predict_overall_recall": 0.7639296187683284,
+    "predict_runtime": 708.5614,
+    "predict_samples_per_second": 4.118,
+    "predict_steps_per_second": 1.03,
+    "threshold_0.5_example_level_1-pk": 0.8344,
+    "threshold_0.5_example_level_1-wd": 0.8123,
+    "threshold_0.5_example_level_f1": 0.7777,
+    "threshold_0.5_example_level_pk": 0.16559999999999997,
+    "threshold_0.5_example_level_precision": 0.792,
+    "threshold_0.5_example_level_recall": 0.7639,
+    "threshold_0.5_example_level_wd": 0.18769999999999998,
+    "train_loss": 0.5592155038844274,
+    "train_runtime": 39662.9553,
+    "train_samples": 10740,
+    "train_samples_per_second": 1.354,
+    "train_steps_per_second": 0.338
+}

config.json

@@ -0,0 +1,67 @@
+{
+  "_name_or_path": "neuralmind/bert-large-portuguese-cased",
+  "architectures": [
+    "BertWithDAForSentenceLabelingTopicSegmentation"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "cache_dir": null,
+  "cl_anchor_level": "eop_list",
+  "cl_loss_weight": 0.5,
+  "cl_negative_k": 3,
+  "cl_positive_k": 1,
+  "cl_temp": 0.1,
+  "classifier_dropout": null,
+  "config_name": null,
+  "directionality": "bidi",
+  "do_cssl": true,
+  "do_da_ts": true,
+  "do_tssp": true,
+  "finetuning_task": "topic_segment",
+  "focal_loss_gamma": 0.0,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "B-EOP",
+    "1": "O"
+  },
+  "ignore_mismatched_sizes": false,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "B-EOP": 0,
+    "O": 1
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_name_or_path": "neuralmind/bert-large-portuguese-cased",
+  "model_revision": "main",
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_gpu": 1,
+  "num_hidden_layers": 24,
+  "num_topic_labels": 0,
+  "num_tssp_labels": 3,
+  "output_past": true,
+  "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
+  "sentence_pooler_type": null,
+  "tokenizer_name": null,
+  "torch_dtype": "float32",
+  "transformers_version": "4.26.0",
+  "ts_loss_weight": 1.0,
+  "ts_score_predictor": "lt",
+  "ts_score_predictor_cos_temp": 1,
+  "tssp_ablation": "none",
+  "tssp_loss_weight": 1.0,
+  "type_vocab_size": 2,
+  "use_auth_token": false,
+  "use_cache": true,
+  "vocab_size": 29795,
+  "weight_label_zero": 0.5
+}

eval_results.json

@@ -0,0 +1,24 @@
+{
+    "epoch": 5.0,
+    "eval_EOP_f1": 0.7617828961399019,
+    "eval_EOP_number": 2446,
+    "eval_EOP_precision": 0.7962550156041016,
+    "eval_EOP_recall": 0.7301717089125103,
+    "eval_da_EOP_f1": 0.7858523386161578,
+    "eval_da_EOP_number": 2715,
+    "eval_da_EOP_precision": 0.826758845058967,
+    "eval_da_EOP_recall": 0.7488029465930018,
+    "eval_da_overall_accuracy": 0.9486466444197256,
+    "eval_da_overall_f1": 0.7858523386161578,
+    "eval_da_overall_precision": 0.826758845058967,
+    "eval_da_overall_recall": 0.7488029465930018,
+    "eval_loss": NaN,
+    "eval_overall_accuracy": 0.9458476753769332,
+    "eval_overall_f1": 0.7617828961399019,
+    "eval_overall_precision": 0.7962550156041016,
+    "eval_overall_recall": 0.7301717089125103,
+    "eval_runtime": 370.561,
+    "eval_samples": 1524,
+    "eval_samples_per_second": 4.113,
+    "eval_steps_per_second": 1.028
+}

example_level_predict_wiki_section_disease_max_seq512_ts_score_lt_results.json

@@ -0,0 +1,19 @@
+{
+    "EOP_f1": 0.7776948501972493,
+    "EOP_number": 4774,
+    "EOP_precision": 0.7919652551574375,
+    "EOP_recall": 0.7639296187683284,
+    "accuracy": 0.9479881258263277,
+    "overall_accuracy": 0.9479881258263277,
+    "overall_f1": 0.7776948501972493,
+    "overall_precision": 0.7919652551574375,
+    "overall_recall": 0.7639296187683284,
+    "predict_examples": 718,
+    "threshold_0.5_example_level_1-pk": 0.8344,
+    "threshold_0.5_example_level_1-wd": 0.8123,
+    "threshold_0.5_example_level_f1": 0.7777,
+    "threshold_0.5_example_level_pk": 0.16559999999999997,
+    "threshold_0.5_example_level_precision": 0.792,
+    "threshold_0.5_example_level_recall": 0.7639,
+    "threshold_0.5_example_level_wd": 0.18769999999999998
+}

example_level_predict_wiki_section_disease_max_seq512_ts_score_lt_results_str_metric.txt

@@ -0,0 +1,4 @@
+p / r / f / pk / wd
+threshold_0.5_example_level_metric
+79.20 / 76.39 / 77.77 / 16.56 / 18.77
+

predict_wiki_section_disease_max_seq512_ts_score_lt.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d621e9acb93baa87996d8e67d34f1ff8fd8f58a10f66f909dfd37e2e43e5f6c
+size 15558757

predict_wiki_section_disease_max_seq512_ts_score_lt_results.json

@@ -0,0 +1,22 @@
+{
+    "predict_EOP_f1": 0.7776948501972493,
+    "predict_EOP_number": 4774,
+    "predict_EOP_precision": 0.7919652551574375,
+    "predict_EOP_recall": 0.7639296187683284,
+    "predict_da_EOP_f1": 0.8040997872751885,
+    "predict_da_EOP_number": 5324,
+    "predict_da_EOP_precision": 0.8286169788760462,
+    "predict_da_EOP_recall": 0.78099173553719,
+    "predict_da_overall_accuracy": 0.9520234909659239,
+    "predict_da_overall_f1": 0.8040997872751885,
+    "predict_da_overall_precision": 0.8286169788760462,
+    "predict_da_overall_recall": 0.78099173553719,
+    "predict_loss": NaN,
+    "predict_overall_accuracy": 0.9479881258263277,
+    "predict_overall_f1": 0.7776948501972493,
+    "predict_overall_precision": 0.7919652551574375,
+    "predict_overall_recall": 0.7639296187683284,
+    "predict_runtime": 708.5614,
+    "predict_samples_per_second": 4.118,
+    "predict_steps_per_second": 1.03
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a661948f9c69d9bc740c3b229127ae19b93f55a1579d611aacbfa49a979d910d
+size 1337746921

special_tokens_map.json

@@ -0,0 +1,8 @@
+{
+  "bos_token": "[BOS]",
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "name_or_path": "neuralmind/bert-large-portuguese-cased",
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": "/home/ext_luti4497_gmail_com/.cache/huggingface/hub/models--neuralmind--bert-large-portuguese-cased/snapshots/aa302f6ea73b759f7df9cad58bd272127b67ec28/special_tokens_map.json",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "train_loss": 0.5592155038844274,
+    "train_runtime": 39662.9553,
+    "train_samples": 10740,
+    "train_samples_per_second": 1.354,
+    "train_steps_per_second": 0.338
+}

trainer_state.json

@@ -0,0 +1,175 @@
+{
+  "best_metric": 0.7617828961399019,
+  "best_model_checkpoint": "./output/neuralmind/bert-large-portuguese-cased-finetune-wiki_section_disease/seed42-seq512-lr5e-05-epoch5-bs4-ts1.0-tssp1.0-cl0.5-2024-07-14_00:16:54/checkpoint-10740",
+  "epoch": 5.0,
+  "global_step": 13425,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4e-05,
+      "loss": 0.9837,
+      "step": 2685
+    },
+    {
+      "epoch": 1.0,
+      "eval_EOP_f1": 0.7398610799910375,
+      "eval_EOP_number": 2446,
+      "eval_EOP_precision": 0.8185423896876549,
+      "eval_EOP_recall": 0.6749795584627964,
+      "eval_da_EOP_f1": 0.7520470291832878,
+      "eval_da_EOP_number": 2715,
+      "eval_da_EOP_precision": 0.87451171875,
+      "eval_da_EOP_recall": 0.6596685082872928,
+      "eval_da_overall_accuracy": 0.9452632554690397,
+      "eval_da_overall_f1": 0.7520470291832878,
+      "eval_da_overall_precision": 0.87451171875,
+      "eval_da_overall_recall": 0.6596685082872928,
+      "eval_loss": NaN,
+      "eval_overall_accuracy": 0.9437145488922286,
+      "eval_overall_f1": 0.7398610799910375,
+      "eval_overall_precision": 0.8185423896876549,
+      "eval_overall_recall": 0.6749795584627964,
+      "eval_runtime": 369.4165,
+      "eval_samples_per_second": 4.125,
+      "eval_steps_per_second": 1.031,
+      "step": 2685
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3e-05,
+      "loss": 0.6831,
+      "step": 5370
+    },
+    {
+      "epoch": 2.0,
+      "eval_EOP_f1": 0.7486033519553074,
+      "eval_EOP_number": 2446,
+      "eval_EOP_precision": 0.7889492753623188,
+      "eval_EOP_recall": 0.7121831561733443,
+      "eval_da_EOP_f1": 0.775566502463054,
+      "eval_da_EOP_number": 2715,
+      "eval_da_EOP_precision": 0.8338983050847457,
+      "eval_da_EOP_recall": 0.7248618784530386,
+      "eval_da_overall_accuracy": 0.9472098628105302,
+      "eval_da_overall_f1": 0.775566502463054,
+      "eval_da_overall_precision": 0.8338983050847457,
+      "eval_da_overall_recall": 0.7248618784530386,
+      "eval_loss": NaN,
+      "eval_overall_accuracy": 0.9432782275658118,
+      "eval_overall_f1": 0.7486033519553074,
+      "eval_overall_precision": 0.7889492753623188,
+      "eval_overall_recall": 0.7121831561733443,
+      "eval_runtime": 368.9458,
+      "eval_samples_per_second": 4.131,
+      "eval_steps_per_second": 1.033,
+      "step": 5370
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 2e-05,
+      "loss": 0.5066,
+      "step": 8055
+    },
+    {
+      "epoch": 3.0,
+      "eval_EOP_f1": 0.7590260285474391,
+      "eval_EOP_number": 2446,
+      "eval_EOP_precision": 0.7799827437446074,
+      "eval_EOP_recall": 0.7391659852820932,
+      "eval_da_EOP_f1": 0.7927998434748582,
+      "eval_da_EOP_number": 2715,
+      "eval_da_EOP_precision": 0.845575959933222,
+      "eval_da_EOP_recall": 0.7462246777163904,
+      "eval_da_overall_accuracy": 0.9509176863181312,
+      "eval_da_overall_f1": 0.7927998434748582,
+      "eval_da_overall_precision": 0.845575959933222,
+      "eval_da_overall_recall": 0.7462246777163904,
+      "eval_loss": NaN,
+      "eval_overall_accuracy": 0.944344790808164,
+      "eval_overall_f1": 0.7590260285474391,
+      "eval_overall_precision": 0.7799827437446074,
+      "eval_overall_recall": 0.7391659852820932,
+      "eval_runtime": 369.3175,
+      "eval_samples_per_second": 4.127,
+      "eval_steps_per_second": 1.032,
+      "step": 8055
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 1e-05,
+      "loss": 0.3684,
+      "step": 10740
+    },
+    {
+      "epoch": 4.0,
+      "eval_EOP_f1": 0.7617828961399019,
+      "eval_EOP_number": 2446,
+      "eval_EOP_precision": 0.7962550156041016,
+      "eval_EOP_recall": 0.7301717089125103,
+      "eval_da_EOP_f1": 0.7858523386161578,
+      "eval_da_EOP_number": 2715,
+      "eval_da_EOP_precision": 0.826758845058967,
+      "eval_da_EOP_recall": 0.7488029465930018,
+      "eval_da_overall_accuracy": 0.9486466444197256,
+      "eval_da_overall_f1": 0.7858523386161578,
+      "eval_da_overall_precision": 0.826758845058967,
+      "eval_da_overall_recall": 0.7488029465930018,
+      "eval_loss": NaN,
+      "eval_overall_accuracy": 0.9458476753769332,
+      "eval_overall_f1": 0.7617828961399019,
+      "eval_overall_precision": 0.7962550156041016,
+      "eval_overall_recall": 0.7301717089125103,
+      "eval_runtime": 369.7052,
+      "eval_samples_per_second": 4.122,
+      "eval_steps_per_second": 1.031,
+      "step": 10740
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.0,
+      "loss": 0.2543,
+      "step": 13425
+    },
+    {
+      "epoch": 5.0,
+      "eval_EOP_f1": 0.7591393357008565,
+      "eval_EOP_number": 2446,
+      "eval_EOP_precision": 0.7761640324647586,
+      "eval_EOP_recall": 0.7428454619787408,
+      "eval_da_EOP_f1": 0.7873163186388245,
+      "eval_da_EOP_number": 2715,
+      "eval_da_EOP_precision": 0.8286528286528286,
+      "eval_da_EOP_recall": 0.7499079189686925,
+      "eval_da_overall_accuracy": 0.9490174267704857,
+      "eval_da_overall_f1": 0.7873163186388245,
+      "eval_da_overall_precision": 0.8286528286528286,
+      "eval_da_overall_recall": 0.7499079189686925,
+      "eval_loss": NaN,
+      "eval_overall_accuracy": 0.9441023900712658,
+      "eval_overall_f1": 0.7591393357008565,
+      "eval_overall_precision": 0.7761640324647586,
+      "eval_overall_recall": 0.7428454619787408,
+      "eval_runtime": 370.5788,
+      "eval_samples_per_second": 4.112,
+      "eval_steps_per_second": 1.028,
+      "step": 13425
+    },
+    {
+      "epoch": 5.0,
+      "step": 13425,
+      "total_flos": 1.000904429647872e+17,
+      "train_loss": 0.5592155038844274,
+      "train_runtime": 39662.9553,
+      "train_samples_per_second": 1.354,
+      "train_steps_per_second": 0.338
+    }
+  ],
+  "max_steps": 13425,
+  "num_train_epochs": 5,
+  "total_flos": 1.000904429647872e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ccfd407b9b35da416ca9166a2b02ece702094218a284939a6148c2d49004a1c
+size 3759