Upload 13 files

README.md

@@ -0,0 +1,93 @@
+---
+library_name: transformers
+base_model: models/distill-robertalex-3L-trained
+tags:
+- generated_from_trainer
+datasets:
+- adalbertojunior/entities
+metrics:
+- precision
+- recall
+- f1
+- accuracy
+model-index:
+- name: test_v6
+  results:
+  - task:
+      name: Token Classification
+      type: token-classification
+    dataset:
+      name: adalbertojunior/entities
+      type: adalbertojunior/entities
+      config: segmentacao
+      split: validation
+      args: segmentacao
+    metrics:
+    - name: Precision
+      type: precision
+      value: 0.7678083439606486
+    - name: Recall
+      type: recall
+      value: 0.8550415905863258
+    - name: F1
+      type: f1
+      value: 0.8090804377039739
+    - name: Accuracy
+      type: accuracy
+      value: 0.9699217442249749
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# test_v6
+
+This model is a fine-tuned version of [models/distill-robertalex-3L-trained](https://huggingface.co/models/distill-robertalex-3L-trained) on the adalbertojunior/entities dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1536
+- Precision: 0.7678
+- Recall: 0.8550
+- F1: 0.8091
+- Accuracy: 0.9699
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 8
+- eval_batch_size: 1
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 64
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 500
+- num_epochs: 20.0
+
+### Training results
+
+| Training Loss | Epoch   | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.0925        | 19.3898 | 7000 | 0.1536          | 0.7678    | 0.8550 | 0.8091 | 0.9699   |
+
+
+### Framework versions
+
+- Transformers 4.47.1
+- Pytorch 2.5.1+cu121
+- Datasets 3.0.1
+- Tokenizers 0.21.0

all_results.json

@@ -0,0 +1,18 @@
+{
+    "epoch": 19.997926744989634,
+    "eval_accuracy": 0.9699217442249749,
+    "eval_f1": 0.8090804377039739,
+    "eval_loss": 0.15359356999397278,
+    "eval_precision": 0.7678083439606486,
+    "eval_recall": 0.8550415905863258,
+    "eval_runtime": 65.5854,
+    "eval_samples": 4960,
+    "eval_samples_per_second": 75.627,
+    "eval_steps_per_second": 75.627,
+    "total_flos": 6132708568035504.0,
+    "train_loss": 0.8169754918592458,
+    "train_runtime": 915.3849,
+    "train_samples": 23148,
+    "train_samples_per_second": 505.754,
+    "train_steps_per_second": 7.887
+}

config.json

@@ -0,0 +1,171 @@
+{
+  "_name_or_path": "models/distill-robertalex-3L-trained",
+  "architectures": [
+    "RobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dropout_task": null,
+  "eos_token_id": 2,
+  "finetuning_task": "ner",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "B-ATRIBUICAO",
+    "1": "B-DECISAO",
+    "2": "B-FUNCAO",
+    "3": "B-FUNDAMENTO",
+    "4": "B-LOCAL",
+    "5": "B-ORGANIZACAO",
+    "6": "B-PEDIDO",
+    "7": "B-PESSOA",
+    "8": "B-REFLEXO",
+    "9": "B-TIPO_ACAO",
+    "10": "B-TRIBUNAL",
+    "11": "B-VALOR_ACORDO",
+    "12": "B-VALOR_CAUSA",
+    "13": "B-VALOR_CONDENACAO",
+    "14": "B-VALOR_CUSTAS",
+    "15": "B-VALOR_PEDIDO",
+    "16": "B-VARA",
+    "17": "E-ATRIBUICAO",
+    "18": "E-DECISAO",
+    "19": "E-FUNCAO",
+    "20": "E-FUNDAMENTO",
+    "21": "E-LOCAL",
+    "22": "E-ORGANIZACAO",
+    "23": "E-PEDIDO",
+    "24": "E-PESSOA",
+    "25": "E-REFLEXO",
+    "26": "E-TIPO_ACAO",
+    "27": "E-TRIBUNAL",
+    "28": "E-VALOR_ACORDO",
+    "29": "E-VALOR_CAUSA",
+    "30": "E-VALOR_CONDENACAO",
+    "31": "E-VALOR_CUSTAS",
+    "32": "E-VALOR_PEDIDO",
+    "33": "E-VARA",
+    "34": "I-ATRIBUICAO",
+    "35": "I-DECISAO",
+    "36": "I-FUNCAO",
+    "37": "I-FUNDAMENTO",
+    "38": "I-LOCAL",
+    "39": "I-ORGANIZACAO",
+    "40": "I-PEDIDO",
+    "41": "I-PESSOA",
+    "42": "I-REFLEXO",
+    "43": "I-TIPO_ACAO",
+    "44": "I-TRIBUNAL",
+    "45": "I-VALOR_ACORDO",
+    "46": "I-VALOR_CAUSA",
+    "47": "I-VALOR_CONDENACAO",
+    "48": "I-VALOR_CUSTAS",
+    "49": "I-VALOR_PEDIDO",
+    "50": "I-VARA",
+    "51": "O",
+    "52": "S-ATRIBUICAO",
+    "53": "S-DECISAO",
+    "54": "S-FUNCAO",
+    "55": "S-FUNDAMENTO",
+    "56": "S-LOCAL",
+    "57": "S-ORGANIZACAO",
+    "58": "S-PEDIDO",
+    "59": "S-PESSOA",
+    "60": "S-REFLEXO",
+    "61": "S-TIPO_ACAO",
+    "62": "S-TRIBUNAL",
+    "63": "S-VALOR_ACORDO",
+    "64": "S-VALOR_CAUSA",
+    "65": "S-VALOR_CONDENACAO",
+    "66": "S-VALOR_CUSTAS",
+    "67": "S-VALOR_PEDIDO",
+    "68": "S-VARA"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "B-ATRIBUICAO": 0,
+    "B-DECISAO": 1,
+    "B-FUNCAO": 2,
+    "B-FUNDAMENTO": 3,
+    "B-LOCAL": 4,
+    "B-ORGANIZACAO": 5,
+    "B-PEDIDO": 6,
+    "B-PESSOA": 7,
+    "B-REFLEXO": 8,
+    "B-TIPO_ACAO": 9,
+    "B-TRIBUNAL": 10,
+    "B-VALOR_ACORDO": 11,
+    "B-VALOR_CAUSA": 12,
+    "B-VALOR_CONDENACAO": 13,
+    "B-VALOR_CUSTAS": 14,
+    "B-VALOR_PEDIDO": 15,
+    "B-VARA": 16,
+    "E-ATRIBUICAO": 17,
+    "E-DECISAO": 18,
+    "E-FUNCAO": 19,
+    "E-FUNDAMENTO": 20,
+    "E-LOCAL": 21,
+    "E-ORGANIZACAO": 22,
+    "E-PEDIDO": 23,
+    "E-PESSOA": 24,
+    "E-REFLEXO": 25,
+    "E-TIPO_ACAO": 26,
+    "E-TRIBUNAL": 27,
+    "E-VALOR_ACORDO": 28,
+    "E-VALOR_CAUSA": 29,
+    "E-VALOR_CONDENACAO": 30,
+    "E-VALOR_CUSTAS": 31,
+    "E-VALOR_PEDIDO": 32,
+    "E-VARA": 33,
+    "I-ATRIBUICAO": 34,
+    "I-DECISAO": 35,
+    "I-FUNCAO": 36,
+    "I-FUNDAMENTO": 37,
+    "I-LOCAL": 38,
+    "I-ORGANIZACAO": 39,
+    "I-PEDIDO": 40,
+    "I-PESSOA": 41,
+    "I-REFLEXO": 42,
+    "I-TIPO_ACAO": 43,
+    "I-TRIBUNAL": 44,
+    "I-VALOR_ACORDO": 45,
+    "I-VALOR_CAUSA": 46,
+    "I-VALOR_CONDENACAO": 47,
+    "I-VALOR_CUSTAS": 48,
+    "I-VALOR_PEDIDO": 49,
+    "I-VARA": 50,
+    "O": 51,
+    "S-ATRIBUICAO": 52,
+    "S-DECISAO": 53,
+    "S-FUNCAO": 54,
+    "S-FUNDAMENTO": 55,
+    "S-LOCAL": 56,
+    "S-ORGANIZACAO": 57,
+    "S-PEDIDO": 58,
+    "S-PESSOA": 59,
+    "S-REFLEXO": 60,
+    "S-TIPO_ACAO": 61,
+    "S-TRIBUNAL": 62,
+    "S-VALOR_ACORDO": 63,
+    "S-VALOR_CAUSA": 64,
+    "S-VALOR_CONDENACAO": 65,
+    "S-VALOR_CUSTAS": 66,
+    "S-VALOR_PEDIDO": 67,
+    "S-VARA": 68
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 8194,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 3,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.47.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 19.997926744989634,
+    "eval_accuracy": 0.9699217442249749,
+    "eval_f1": 0.8090804377039739,
+    "eval_loss": 0.15359356999397278,
+    "eval_precision": 0.7678083439606486,
+    "eval_recall": 0.8550415905863258,
+    "eval_runtime": 65.5854,
+    "eval_samples": 4960,
+    "eval_samples_per_second": 75.627,
+    "eval_steps_per_second": 75.627
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1027ea711a64f6a4164bf67bfa8191750217d66373efabaf4d1eab321556837d
+size 132437538

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,65 @@
+{
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50264": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "max_length": 512,
+  "model_max_length": 512,
+  "pad_to_multiple_of": null,
+  "pad_token": "<pad>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "</s>",
+  "stride": 0,
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 19.997926744989634,
+    "total_flos": 6132708568035504.0,
+    "train_loss": 0.8169754918592458,
+    "train_runtime": 915.3849,
+    "train_samples": 23148,
+    "train_samples_per_second": 505.754,
+    "train_steps_per_second": 7.887
+}

trainer_state.json

@@ -0,0 +1,152 @@
+{
+  "best_metric": 0.8090804377039739,
+  "best_model_checkpoint": "./models/entities/test_v6/checkpoint-7000",
+  "epoch": 19.997926744989634,
+  "eval_steps": 7000,
+  "global_step": 7220,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.3842432619212164,
+      "grad_norm": 7.1875,
+      "learning_rate": 0.0002,
+      "loss": 8.369,
+      "step": 500
+    },
+    {
+      "epoch": 2.7684865238424328,
+      "grad_norm": 4.84375,
+      "learning_rate": 0.00018511904761904765,
+      "loss": 1.059,
+      "step": 1000
+    },
+    {
+      "epoch": 4.1548030407740155,
+      "grad_norm": 5.28125,
+      "learning_rate": 0.00017023809523809523,
+      "loss": 0.6034,
+      "step": 1500
+    },
+    {
+      "epoch": 5.5390463026952315,
+      "grad_norm": 3.90625,
+      "learning_rate": 0.00015535714285714287,
+      "loss": 0.3745,
+      "step": 2000
+    },
+    {
+      "epoch": 6.923289564616448,
+      "grad_norm": 3.515625,
+      "learning_rate": 0.00014047619047619049,
+      "loss": 0.2708,
+      "step": 2500
+    },
+    {
+      "epoch": 8.309606081548031,
+      "grad_norm": 3.203125,
+      "learning_rate": 0.0001255952380952381,
+      "loss": 0.1967,
+      "step": 3000
+    },
+    {
+      "epoch": 9.693849343469246,
+      "grad_norm": 2.3125,
+      "learning_rate": 0.00011071428571428572,
+      "loss": 0.1566,
+      "step": 3500
+    },
+    {
+      "epoch": 11.080165860400829,
+      "grad_norm": 3.03125,
+      "learning_rate": 9.583333333333334e-05,
+      "loss": 0.1318,
+      "step": 4000
+    },
+    {
+      "epoch": 12.464409122322046,
+      "grad_norm": 4.84375,
+      "learning_rate": 8.095238095238096e-05,
+      "loss": 0.1131,
+      "step": 4500
+    },
+    {
+      "epoch": 13.848652384243263,
+      "grad_norm": 2.484375,
+      "learning_rate": 6.607142857142857e-05,
+      "loss": 0.1041,
+      "step": 5000
+    },
+    {
+      "epoch": 15.234968901174845,
+      "grad_norm": 3.078125,
+      "learning_rate": 5.119047619047619e-05,
+      "loss": 0.0986,
+      "step": 5500
+    },
+    {
+      "epoch": 16.619212163096062,
+      "grad_norm": 2.53125,
+      "learning_rate": 3.630952380952381e-05,
+      "loss": 0.0951,
+      "step": 6000
+    },
+    {
+      "epoch": 18.005528680027645,
+      "grad_norm": 3.53125,
+      "learning_rate": 2.1428571428571428e-05,
+      "loss": 0.0921,
+      "step": 6500
+    },
+    {
+      "epoch": 19.389771941948858,
+      "grad_norm": 1.671875,
+      "learning_rate": 6.547619047619048e-06,
+      "loss": 0.0925,
+      "step": 7000
+    },
+    {
+      "epoch": 19.389771941948858,
+      "eval_accuracy": 0.9699217442249749,
+      "eval_f1": 0.8090804377039739,
+      "eval_loss": 0.15359356999397278,
+      "eval_precision": 0.7678083439606486,
+      "eval_recall": 0.8550415905863258,
+      "eval_runtime": 169.5033,
+      "eval_samples_per_second": 29.262,
+      "eval_steps_per_second": 29.262,
+      "step": 7000
+    },
+    {
+      "epoch": 19.997926744989634,
+      "step": 7220,
+      "total_flos": 6132708568035504.0,
+      "train_loss": 0.8169754918592458,
+      "train_runtime": 915.3849,
+      "train_samples_per_second": 505.754,
+      "train_steps_per_second": 7.887
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 7000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6132708568035504.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:885427ad062bd54a33ec4c0969e5b2302b29c0637541f9098bc04c866f12a2cf
+size 5432