Initial version

README.md

@@ -0,0 +1,70 @@
+---
+license: mit
+base_model: papluca/xlm-roberta-base-language-detection
+tags:
+- Italian
+- legal ruling
+- generated_from_trainer
+metrics:
+- f1
+- accuracy
+model-index:
+- name: ribesstefano/RuleBert-v0.2-k0
+  results: []
+---
+
+<!-- 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. -->
+
+# ribesstefano/RuleBert-v0.2-k0
+
+This model is a fine-tuned version of [papluca/xlm-roberta-base-language-detection](https://huggingface.co/papluca/xlm-roberta-base-language-detection) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3669
+- F1: 0.4972
+- Roc Auc: 0.6720
+- Accuracy: 0.0
+
+## 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: 1e-05
+- train_batch_size: 2
+- eval_batch_size: 64
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- training_steps: 8000
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | F1     | Roc Auc | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:|
+| 0.3463        | 0.06  | 250  | 0.3661          | 0.4972 | 0.6720  | 0.0      |
+| 0.3384        | 0.12  | 500  | 0.3619          | 0.4972 | 0.6720  | 0.0      |
+| 0.3266        | 0.19  | 750  | 0.3583          | 0.5162 | 0.6763  | 0.0533   |
+| 0.3276        | 0.25  | 1000 | 0.3594          | 0.5152 | 0.6751  | 0.06     |
+| 0.3412        | 0.31  | 1250 | 0.3669          | 0.4972 | 0.6720  | 0.0      |
+
+
+### Framework versions
+
+- Transformers 4.36.2
+- Pytorch 2.1.0+cu121
+- Datasets 2.16.1
+- Tokenizers 0.15.0

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb44c1f6bbe2d46c0916eae809f2a1095f5a3a2ca8280f7ebad51cd38963bf8c
+oid sha256:c5da60646d0f4fc1c7485fee1f827e1390d75d0b7ddc45a29ae2470bb4a9b574
 size 1112334200

runs/Jan07_19-02-52_1c5a8e80ae39/events.out.tfevents.1704654173.1c5a8e80ae39.517.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b7a3b67c3886f1c56a5b59029eaac42330e001c1f7bf5fa687a1aae2deccd30
-size 15670
+oid sha256:2518fbeb0133322922abd9b4b3fc58b029093efed5343e68b801146403246d3f
+size 18080

runs/Jan07_19-12-30_1c5a8e80ae39/events.out.tfevents.1704654751.1c5a8e80ae39.517.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11b46eb6f585e8e807a3868bb157b48fe0e65124f795b07d9af6ebc0838d7e7c
+size 14819

tmp-checkpoint-2000/config.json

@@ -0,0 +1,121 @@
+{
+  "_name_or_path": "papluca/xlm-roberta-base-language-detection",
+  "architectures": [
+    "XLMRobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "andamento retromarcia imputato",
+    "1": "attraversamento corretto della vittima",
+    "2": "attraversamento vittima con semaforo rosso",
+    "3": "attraversamento vittima senza strisce o sbucata improvvisamente",
+    "4": "centro abitato",
+    "5": "colpo di sonno",
+    "6": "eccesso verso imputato",
+    "7": "eccesso verso vittima",
+    "8": "fuga imputato",
+    "9": "giorno",
+    "10": "giovane eta imputato",
+    "11": "guida brusca o manovra brusca della vittima",
+    "12": "guida brusca o manovra brusca imputato",
+    "13": "guida senza patente valida imputato",
+    "14": "guida senza patente valida vittima",
+    "15": "imputato perde controllo",
+    "16": "imputato ubriaco o drogato",
+    "17": "invasione corsia opposta imputato",
+    "18": "lavori stradali di manutenzione",
+    "19": "localita commerciale o luogo di lavoro",
+    "20": "maltempo",
+    "21": "mancanza cintura di sicurezza o casco vittima",
+    "22": "neopat imputato",
+    "23": "notte",
+    "24": "ostacolo evitabile dall imputato",
+    "25": "ostacolo prevedibile dall imputato",
+    "26": "ostacolo prevedibile dalla vittima",
+    "27": "piu imputati",
+    "28": "poca illuminazione o scarsa visibilita",
+    "29": "presenza rimorchio imputato",
+    "30": "prossimita di un incrocio o curva",
+    "31": "segnaletica non visibile o assente",
+    "32": "strada a doppio senso di marcia",
+    "33": "strada rovinata o bagnata",
+    "34": "uso telefono alla guida",
+    "35": "veicoli non abilitato a circolare o senza assicurazione imputato",
+    "36": "veicoli non abilitato a circolare o senza assicurazione vittima",
+    "37": "violazione stop o semaforo o codice della strada da parte della vittima",
+    "38": "violazione stop o semaforo o codice della strada da parte imputato",
+    "39": "vittima eta avanzata",
+    "40": "vittima eta giovane",
+    "41": "vittima perde controllo",
+    "42": "vittima priva di giubbino e bretelle riflettenti",
+    "43": "vittima ubriaca o drogata"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "andamento retromarcia imputato": 0,
+    "attraversamento corretto della vittima": 1,
+    "attraversamento vittima con semaforo rosso": 2,
+    "attraversamento vittima senza strisce o sbucata improvvisamente": 3,
+    "centro abitato": 4,
+    "colpo di sonno": 5,
+    "eccesso verso imputato": 6,
+    "eccesso verso vittima": 7,
+    "fuga imputato": 8,
+    "giorno": 9,
+    "giovane eta imputato": 10,
+    "guida brusca o manovra brusca della vittima": 11,
+    "guida brusca o manovra brusca imputato": 12,
+    "guida senza patente valida imputato": 13,
+    "guida senza patente valida vittima": 14,
+    "imputato perde controllo": 15,
+    "imputato ubriaco o drogato": 16,
+    "invasione corsia opposta imputato": 17,
+    "lavori stradali di manutenzione": 18,
+    "localita commerciale o luogo di lavoro": 19,
+    "maltempo": 20,
+    "mancanza cintura di sicurezza o casco vittima": 21,
+    "neopat imputato": 22,
+    "notte": 23,
+    "ostacolo evitabile dall imputato": 24,
+    "ostacolo prevedibile dall imputato": 25,
+    "ostacolo prevedibile dalla vittima": 26,
+    "piu imputati": 27,
+    "poca illuminazione o scarsa visibilita": 28,
+    "presenza rimorchio imputato": 29,
+    "prossimita di un incrocio o curva": 30,
+    "segnaletica non visibile o assente": 31,
+    "strada a doppio senso di marcia": 32,
+    "strada rovinata o bagnata": 33,
+    "uso telefono alla guida": 34,
+    "veicoli non abilitato a circolare o senza assicurazione imputato": 35,
+    "veicoli non abilitato a circolare o senza assicurazione vittima": 36,
+    "violazione stop o semaforo o codice della strada da parte della vittima": 37,
+    "violazione stop o semaforo o codice della strada da parte imputato": 38,
+    "vittima eta avanzata": 39,
+    "vittima eta giovane": 40,
+    "vittima perde controllo": 41,
+    "vittima priva di giubbino e bretelle riflettenti": 42,
+    "vittima ubriaca o drogata": 43
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "multi_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

tmp-checkpoint-2000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc97342d439840d7034d7b7d1bbcbef587a7136355915c0fc023732bf7c7adf8
+size 1112334200

tmp-checkpoint-2000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a476bdcabe4a45ce58b1040b783da6913af5dc50b0765b009bef66594b7e9f49
+size 2224787962

tmp-checkpoint-2000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05217f29dcb6fb2cd6c57db3cfec0e9aa935d8ff06c19993652ff71a0f2dbc30
+size 14244

tmp-checkpoint-2000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:897cfb9e0b9becedae8564ff10efd3085fde96969e7d73be69c93d081e0d2dbd
+size 1064

tmp-checkpoint-2000/sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

tmp-checkpoint-2000/special_tokens_map.json

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

tmp-checkpoint-2000/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6f76fe13d42f80dcee0cb86a1aeb5f14f8909bb8a8782f7a4a4ad76697ef164
+size 17083021

tmp-checkpoint-2000/tokenizer_config.json

@@ -0,0 +1,54 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}

tmp-checkpoint-2000/trainer_state.json

@@ -0,0 +1,349 @@
+{
+  "best_metric": 0.5166051660516605,
+  "best_model_checkpoint": "RuleBert-v0.2-k0/checkpoint-1500",
+  "epoch": 0.49382716049382713,
+  "eval_steps": 250,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.937500000000001e-06,
+      "loss": 0.5636,
+      "step": 50
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.875000000000001e-06,
+      "loss": 0.3677,
+      "step": 100
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.8125e-06,
+      "loss": 0.3585,
+      "step": 150
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.75e-06,
+      "loss": 0.3372,
+      "step": 200
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.6875e-06,
+      "loss": 0.3226,
+      "step": 250
+    },
+    {
+      "epoch": 0.06,
+      "eval_accuracy": 0.06666666666666667,
+      "eval_f1": 0.5180722891566265,
+      "eval_loss": 0.36232367157936096,
+      "eval_roc_auc": 0.6758306628436499,
+      "eval_runtime": 1.6645,
+      "eval_samples_per_second": 90.118,
+      "eval_steps_per_second": 1.802,
+      "step": 250
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.625e-06,
+      "loss": 0.3323,
+      "step": 300
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.562500000000002e-06,
+      "loss": 0.3451,
+      "step": 350
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.5e-06,
+      "loss": 0.3317,
+      "step": 400
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.4375e-06,
+      "loss": 0.3297,
+      "step": 450
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.375000000000001e-06,
+      "loss": 0.3287,
+      "step": 500
+    },
+    {
+      "epoch": 0.12,
+      "eval_accuracy": 0.0,
+      "eval_f1": 0.49723756906077354,
+      "eval_loss": 0.3665144741535187,
+      "eval_roc_auc": 0.6720357564513407,
+      "eval_runtime": 1.6779,
+      "eval_samples_per_second": 89.396,
+      "eval_steps_per_second": 1.788,
+      "step": 500
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.312500000000001e-06,
+      "loss": 0.3359,
+      "step": 550
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.25125e-06,
+      "loss": 0.3331,
+      "step": 600
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.18875e-06,
+      "loss": 0.3374,
+      "step": 650
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.126250000000001e-06,
+      "loss": 0.3367,
+      "step": 700
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.065e-06,
+      "loss": 0.3195,
+      "step": 750
+    },
+    {
+      "epoch": 0.19,
+      "eval_accuracy": 0.04,
+      "eval_f1": 0.5126993502658004,
+      "eval_loss": 0.3585328757762909,
+      "eval_roc_auc": 0.6747933884297521,
+      "eval_runtime": 1.6851,
+      "eval_samples_per_second": 89.013,
+      "eval_steps_per_second": 1.78,
+      "step": 750
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.0025e-06,
+      "loss": 0.3226,
+      "step": 800
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 8.94e-06,
+      "loss": 0.3435,
+      "step": 850
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 8.87875e-06,
+      "loss": 0.3213,
+      "step": 900
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 8.817500000000002e-06,
+      "loss": 0.3313,
+      "step": 950
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.755e-06,
+      "loss": 0.3241,
+      "step": 1000
+    },
+    {
+      "epoch": 0.25,
+      "eval_accuracy": 0.03333333333333333,
+      "eval_f1": 0.5134189031505251,
+      "eval_loss": 0.3658309876918793,
+      "eval_roc_auc": 0.6758812615955473,
+      "eval_runtime": 1.6782,
+      "eval_samples_per_second": 89.383,
+      "eval_steps_per_second": 1.788,
+      "step": 1000
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.6925e-06,
+      "loss": 0.3257,
+      "step": 1050
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.63e-06,
+      "loss": 0.3288,
+      "step": 1100
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.5675e-06,
+      "loss": 0.3331,
+      "step": 1150
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.505e-06,
+      "loss": 0.3529,
+      "step": 1200
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.442500000000001e-06,
+      "loss": 0.3321,
+      "step": 1250
+    },
+    {
+      "epoch": 0.31,
+      "eval_accuracy": 0.0,
+      "eval_f1": 0.49723756906077354,
+      "eval_loss": 0.3737262487411499,
+      "eval_roc_auc": 0.6720357564513407,
+      "eval_runtime": 1.69,
+      "eval_samples_per_second": 88.759,
+      "eval_steps_per_second": 1.775,
+      "step": 1250
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 8.380000000000001e-06,
+      "loss": 0.3339,
+      "step": 1300
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 8.317500000000001e-06,
+      "loss": 0.3153,
+      "step": 1350
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 8.255000000000001e-06,
+      "loss": 0.3406,
+      "step": 1400
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 8.192500000000002e-06,
+      "loss": 0.3109,
+      "step": 1450
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 8.13e-06,
+      "loss": 0.3315,
+      "step": 1500
+    },
+    {
+      "epoch": 0.37,
+      "eval_accuracy": 0.02666666666666667,
+      "eval_f1": 0.5166051660516605,
+      "eval_loss": 0.3748493194580078,
+      "eval_roc_auc": 0.6842047562826783,
+      "eval_runtime": 1.6814,
+      "eval_samples_per_second": 89.211,
+      "eval_steps_per_second": 1.784,
+      "step": 1500
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 8.0675e-06,
+      "loss": 0.3066,
+      "step": 1550
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 8.005e-06,
+      "loss": 0.3143,
+      "step": 1600
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 7.9425e-06,
+      "loss": 0.3286,
+      "step": 1650
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 7.88e-06,
+      "loss": 0.3125,
+      "step": 1700
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 7.8175e-06,
+      "loss": 0.3384,
+      "step": 1750
+    },
+    {
+      "epoch": 0.43,
+      "eval_accuracy": 0.02666666666666667,
+      "eval_f1": 0.5051546391752577,
+      "eval_loss": 0.367872029542923,
+      "eval_roc_auc": 0.6734187890032046,
+      "eval_runtime": 1.6925,
+      "eval_samples_per_second": 88.625,
+      "eval_steps_per_second": 1.772,
+      "step": 1750
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 7.755000000000001e-06,
+      "loss": 0.3061,
+      "step": 1800
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 7.692500000000001e-06,
+      "loss": 0.32,
+      "step": 1850
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 7.630000000000001e-06,
+      "loss": 0.3211,
+      "step": 1900
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 7.567500000000001e-06,
+      "loss": 0.3104,
+      "step": 1950
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 7.505e-06,
+      "loss": 0.3338,
+      "step": 2000
+    },
+    {
+      "epoch": 0.49,
+      "eval_accuracy": 0.02,
+      "eval_f1": 0.5050732807215332,
+      "eval_loss": 0.3726758658885956,
+      "eval_roc_auc": 0.6743632990386238,
+      "eval_runtime": 1.6832,
+      "eval_samples_per_second": 89.116,
+      "eval_steps_per_second": 1.782,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 8000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "total_flos": 1052841099264000.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

tmp-checkpoint-2000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffbef30dded80bb1fb6c1bafda8ff435451a67434a47238bca24230889797d1f
+size 4792

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ffbef30dded80bb1fb6c1bafda8ff435451a67434a47238bca24230889797d1f
+oid sha256:0264de8903a39797f4341a1200f207d511a304b358ff04d3674582665636005b
 size 4792