Upload 17 files

.gitattributes

@@ -33,3 +33,8 @@ 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
+confusion_matrix_attacked_test.png filter=lfs diff=lfs merge=lfs -text
+confusion_matrix_test.png filter=lfs diff=lfs merge=lfs -text
+confusion_matrix_validation.png filter=lfs diff=lfs merge=lfs -text
+model.onnx.data filter=lfs diff=lfs merge=lfs -text
+training_history.png filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,49 @@
+{
+  "architectures": [
+    "ElectraForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "embedding_size": 768,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "O",
+    "1": "COMMA",
+    "2": "DARI",
+    "3": "QUESTION",
+    "4": "EXCLAMATION",
+    "5": "SEMICOLON",
+    "6": "COLON",
+    "7": "HYPHEN"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "COLON": 6,
+    "COMMA": 1,
+    "DARI": 2,
+    "EXCLAMATION": 4,
+    "HYPHEN": 7,
+    "O": 0,
+    "QUESTION": 3,
+    "SEMICOLON": 5
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "electra",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "summary_activation": "gelu",
+  "summary_last_dropout": 0.1,
+  "summary_type": "first",
+  "summary_use_proj": true,
+  "transformers_version": "4.57.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 32000
+}

detailed_results_attacked_test.json

@@ -0,0 +1,166 @@
+{
+  "split": "attacked_test",
+  "classification_report": "              precision    recall  f1-score   support\n\n           O     0.9832    0.9261    0.9538     16661\n       COMMA     0.2602    0.5802    0.3593       374\n        DARI     0.1744    0.6433    0.2745       157\n    QUESTION     0.2908    0.5190    0.3727        79\n EXCLAMATION     0.1957    0.5000    0.2812        18\n   SEMICOLON     0.0000    0.0000    0.0000         4\n       COLON     0.1026    0.2000    0.1356        20\n      HYPHEN     0.0000    0.0000    0.0000        31\n\n    accuracy                         0.9110     17344\n   macro avg     0.2509    0.4211    0.2971     17344\nweighted avg     0.9533    0.9110    0.9286     17344\n",
+  "confusion_matrix": [
+    [
+      15429,
+      601,
+      456,
+      96,
+      34,
+      7,
+      34,
+      4
+    ],
+    [
+      149,
+      217,
+      6,
+      1,
+      0,
+      0,
+      1,
+      0
+    ],
+    [
+      41,
+      9,
+      101,
+      3,
+      3,
+      0,
+      0,
+      0
+    ],
+    [
+      28,
+      3,
+      7,
+      41,
+      0,
+      0,
+      0,
+      0
+    ],
+    [
+      4,
+      1,
+      4,
+      0,
+      9,
+      0,
+      0,
+      0
+    ],
+    [
+      2,
+      0,
+      2,
+      0,
+      0,
+      0,
+      0,
+      0
+    ],
+    [
+      10,
+      2,
+      3,
+      0,
+      0,
+      0,
+      4,
+      1
+    ],
+    [
+      30,
+      1,
+      0,
+      0,
+      0,
+      0,
+      0,
+      0
+    ]
+  ],
+  "per_class_metrics": [
+    {
+      "class": "O",
+      "support": 16661,
+      "accuracy": 0.9260548586519417,
+      "correct": 15429
+    },
+    {
+      "class": "COMMA",
+      "support": 374,
+      "accuracy": 0.5802139037433155,
+      "correct": 217
+    },
+    {
+      "class": "DARI",
+      "support": 157,
+      "accuracy": 0.643312101910828,
+      "correct": 101
+    },
+    {
+      "class": "QUESTION",
+      "support": 79,
+      "accuracy": 0.5189873417721519,
+      "correct": 41
+    },
+    {
+      "class": "EXCLAMATION",
+      "support": 18,
+      "accuracy": 0.5,
+      "correct": 9
+    },
+    {
+      "class": "SEMICOLON",
+      "support": 4,
+      "accuracy": 0.0,
+      "correct": 0
+    },
+    {
+      "class": "COLON",
+      "support": 20,
+      "accuracy": 0.2,
+      "correct": 4
+    },
+    {
+      "class": "HYPHEN",
+      "support": 31,
+      "accuracy": 0.0,
+      "correct": 0
+    }
+  ],
+  "bleu_rouge_scores": {
+    "bleu": 82.59219598233263,
+    "rouge1_precision": 0.0,
+    "rouge1_recall": 0.0,
+    "rouge1_f": 0.0,
+    "rouge2_precision": 0.0,
+    "rouge2_recall": 0.0,
+    "rouge2_f": 0.0,
+    "rougeL_precision": 0.0,
+    "rougeL_recall": 0.0,
+    "rougeL_f": 0.0
+  },
+  "overall_metrics": {
+    "test_loss": 0.05920533090829849,
+    "test_precision": 0.9532955961063683,
+    "test_recall": 0.911035516605166,
+    "test_f1": 0.9285803514155778,
+    "test_weighted_f1": 0.9285803514155778,
+    "test_punctuation_f1": 0.6590236317324747,
+    "test_f1_comma": 0.35927152317880795,
+    "test_f1_dari": 0.27445652173913043,
+    "test_f1_question": 0.37272727272727274,
+    "test_f1_exclamation": 0.28125,
+    "test_f1_semicolon": 0.0,
+    "test_f1_colon": 0.13559322033898305,
+    "test_f1_hyphen": 0.0,
+    "test_runtime": 4.8696,
+    "test_samples_per_second": 119.724,
+    "test_steps_per_second": 2.054
+  }
+}

detailed_results_test.json

@@ -0,0 +1,166 @@
+{
+  "split": "test",
+  "classification_report": "              precision    recall  f1-score   support\n\n           O     0.9742    0.9802    0.9772     17078\n       COMMA     0.6969    0.6213    0.6569      1014\n        DARI     0.8005    0.8583    0.8284       360\n    QUESTION     0.8315    0.8407    0.8361       182\n EXCLAMATION     0.7179    0.7467    0.7320        75\n   SEMICOLON     0.0000    0.0000    0.0000        23\n       COLON     0.3529    0.4444    0.3934        54\n      HYPHEN     0.0000    0.0000    0.0000        23\n\n    accuracy                         0.9523     18809\n   macro avg     0.5467    0.5615    0.5530     18809\nweighted avg     0.9493    0.9523    0.9507     18809\n",
+  "confusion_matrix": [
+    [
+      16740,
+      241,
+      37,
+      19,
+      4,
+      1,
+      34,
+      2
+    ],
+    [
+      340,
+      630,
+      26,
+      5,
+      4,
+      1,
+      7,
+      1
+    ],
+    [
+      24,
+      13,
+      309,
+      4,
+      10,
+      0,
+      0,
+      0
+    ],
+    [
+      16,
+      3,
+      6,
+      153,
+      4,
+      0,
+      0,
+      0
+    ],
+    [
+      4,
+      4,
+      8,
+      3,
+      56,
+      0,
+      0,
+      0
+    ],
+    [
+      18,
+      5,
+      0,
+      0,
+      0,
+      0,
+      0,
+      0
+    ],
+    [
+      26,
+      4,
+      0,
+      0,
+      0,
+      0,
+      24,
+      0
+    ],
+    [
+      16,
+      4,
+      0,
+      0,
+      0,
+      0,
+      3,
+      0
+    ]
+  ],
+  "per_class_metrics": [
+    {
+      "class": "O",
+      "support": 17078,
+      "accuracy": 0.9802084553226373,
+      "correct": 16740
+    },
+    {
+      "class": "COMMA",
+      "support": 1014,
+      "accuracy": 0.621301775147929,
+      "correct": 630
+    },
+    {
+      "class": "DARI",
+      "support": 360,
+      "accuracy": 0.8583333333333333,
+      "correct": 309
+    },
+    {
+      "class": "QUESTION",
+      "support": 182,
+      "accuracy": 0.8406593406593407,
+      "correct": 153
+    },
+    {
+      "class": "EXCLAMATION",
+      "support": 75,
+      "accuracy": 0.7466666666666667,
+      "correct": 56
+    },
+    {
+      "class": "SEMICOLON",
+      "support": 23,
+      "accuracy": 0.0,
+      "correct": 0
+    },
+    {
+      "class": "COLON",
+      "support": 54,
+      "accuracy": 0.4444444444444444,
+      "correct": 24
+    },
+    {
+      "class": "HYPHEN",
+      "support": 23,
+      "accuracy": 0.0,
+      "correct": 0
+    }
+  ],
+  "bleu_rouge_scores": {
+    "bleu": 90.0686061783956,
+    "rouge1_precision": 0.0,
+    "rouge1_recall": 0.0,
+    "rouge1_f": 0.0,
+    "rouge2_precision": 0.0,
+    "rouge2_recall": 0.0,
+    "rouge2_f": 0.0,
+    "rougeL_precision": 0.0,
+    "rougeL_recall": 0.0,
+    "rougeL_f": 0.0
+  },
+  "overall_metrics": {
+    "test_loss": 0.02253320813179016,
+    "test_precision": 0.9493235462409354,
+    "test_recall": 0.9523100643309054,
+    "test_f1": 0.9506557193404622,
+    "test_weighted_f1": 0.9506557193404622,
+    "test_punctuation_f1": 0.7637091871138297,
+    "test_f1_comma": 0.656934306569343,
+    "test_f1_dari": 0.8284182305630027,
+    "test_f1_question": 0.8360655737704918,
+    "test_f1_exclamation": 0.7320261437908496,
+    "test_f1_semicolon": 0.0,
+    "test_f1_colon": 0.39344262295081966,
+    "test_f1_hyphen": 0.0,
+    "test_runtime": 4.8333,
+    "test_samples_per_second": 120.622,
+    "test_steps_per_second": 2.069
+  }
+}

detailed_results_validation.json

@@ -0,0 +1,166 @@
+{
+  "split": "validation",
+  "classification_report": "              precision    recall  f1-score   support\n\n           O     0.9736    0.9813    0.9774     17717\n       COMMA     0.6741    0.6105    0.6408       986\n        DARI     0.7895    0.7971    0.7933       414\n    QUESTION     0.8565    0.8685    0.8625       213\n EXCLAMATION     0.5909    0.5417    0.5652        48\n   SEMICOLON     0.7333    0.4400    0.5500        25\n       COLON     0.5614    0.4211    0.4812        76\n      HYPHEN     0.4286    0.1111    0.1765        27\n\n    accuracy                         0.9522     19506\n   macro avg     0.7010    0.5964    0.6309     19506\nweighted avg     0.9497    0.9522    0.9506     19506\n",
+  "confusion_matrix": [
+    [
+      17385,
+      246,
+      47,
+      17,
+      3,
+      1,
+      17,
+      1
+    ],
+    [
+      342,
+      602,
+      28,
+      3,
+      2,
+      2,
+      4,
+      3
+    ],
+    [
+      40,
+      28,
+      330,
+      4,
+      10,
+      1,
+      1,
+      0
+    ],
+    [
+      18,
+      1,
+      6,
+      185,
+      3,
+      0,
+      0,
+      0
+    ],
+    [
+      5,
+      3,
+      7,
+      7,
+      26,
+      0,
+      0,
+      0
+    ],
+    [
+      12,
+      2,
+      0,
+      0,
+      0,
+      11,
+      0,
+      0
+    ],
+    [
+      34,
+      10,
+      0,
+      0,
+      0,
+      0,
+      32,
+      0
+    ],
+    [
+      20,
+      1,
+      0,
+      0,
+      0,
+      0,
+      3,
+      3
+    ]
+  ],
+  "per_class_metrics": [
+    {
+      "class": "O",
+      "support": 17717,
+      "accuracy": 0.9812609358243495,
+      "correct": 17385
+    },
+    {
+      "class": "COMMA",
+      "support": 986,
+      "accuracy": 0.6105476673427992,
+      "correct": 602
+    },
+    {
+      "class": "DARI",
+      "support": 414,
+      "accuracy": 0.7971014492753623,
+      "correct": 330
+    },
+    {
+      "class": "QUESTION",
+      "support": 213,
+      "accuracy": 0.8685446009389671,
+      "correct": 185
+    },
+    {
+      "class": "EXCLAMATION",
+      "support": 48,
+      "accuracy": 0.5416666666666666,
+      "correct": 26
+    },
+    {
+      "class": "SEMICOLON",
+      "support": 25,
+      "accuracy": 0.44,
+      "correct": 11
+    },
+    {
+      "class": "COLON",
+      "support": 76,
+      "accuracy": 0.42105263157894735,
+      "correct": 32
+    },
+    {
+      "class": "HYPHEN",
+      "support": 27,
+      "accuracy": 0.1111111111111111,
+      "correct": 3
+    }
+  ],
+  "bleu_rouge_scores": {
+    "bleu": 90.20234005231055,
+    "rouge1_precision": 0.0,
+    "rouge1_recall": 0.0,
+    "rouge1_f": 0.0,
+    "rouge2_precision": 0.0,
+    "rouge2_recall": 0.0,
+    "rouge2_f": 0.0,
+    "rougeL_precision": 0.0,
+    "rougeL_recall": 0.0,
+    "rougeL_f": 0.0
+  },
+  "overall_metrics": {
+    "test_loss": 0.023327844217419624,
+    "test_precision": 0.9496856615053557,
+    "test_recall": 0.9522198297959602,
+    "test_f1": 0.950640869722443,
+    "test_weighted_f1": 0.950640869722443,
+    "test_punctuation_f1": 0.7575861893899899,
+    "test_f1_comma": 0.6407663650878127,
+    "test_f1_dari": 0.7932692307692307,
+    "test_f1_question": 0.8624708624708625,
+    "test_f1_exclamation": 0.5652173913043478,
+    "test_f1_semicolon": 0.55,
+    "test_f1_colon": 0.48120300751879697,
+    "test_f1_hyphen": 0.17647058823529413,
+    "test_runtime": 4.8056,
+    "test_samples_per_second": 121.108,
+    "test_steps_per_second": 2.081
+  }
+}

model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a92f5e10999e782700003fbd7a945fd8642fcdccf4c82db86587571cca41739
+size 1217718

model.onnx.data

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2259b4911ff5c543d618adaf3862ae1d88269fc636158cebefece8f06c8454a
+size 440139776

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9424e9b37281d9a3476c04ca37518ec08e6bf340923cea4c40d740d4b4fa16e
+size 440155536

special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,66 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "extra_special_tokens": {},
+  "full_tokenizer_file": null,
+  "mask_token": "[MASK]",
+  "max_length": 200,
+  "model_max_length": 1000000000000000019884624838656,
+  "never_split": null,
+  "pad_to_multiple_of": null,
+  "pad_token": "[PAD]",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "[SEP]",
+  "stride": 0,
+  "strip_accents": null,
+  "tokenize_chinese_chars": false,
+  "tokenizer_class": "ElectraTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ab1461816db7bef705da8566020ca27553f0459ecebc07928dff4aeed60c45d
+size 5841

training_config.json

@@ -0,0 +1 @@
+"{'__module__': '__main__', 'TOKENIZED_DATA_DIR': '/kaggle/working/Hishab_Dataset/custom-pr-bn-v1-tokenized', 'OUTPUT_DIR': '/kaggle/working/models/custom-pr-bn-v1', 'LOGGING_DIR': '/kaggle/working/logs/custom-pr-bn-v1', 'MODEL_NAME': 'csebuetnlp/banglabert', 'NUM_LABELS': 8, 'LEARNING_RATE': 2e-05, 'WEIGHT_DECAY': 0.01, 'NUM_EPOCHS': 50, 'WARMUP_RATIO': 0.1, 'TRAIN_BATCH_SIZE': 16, 'EVAL_BATCH_SIZE': 32, 'GRADIENT_ACCUMULATION_STEPS': 2, 'OPTIMIZER': 'adamw_torch', 'LR_SCHEDULER': 'cosine', 'MAX_GRAD_NORM': 1.0, 'DROPOUT': 0.1, 'ATTENTION_DROPOUT': 0.1, 'HIDDEN_DROPOUT': 0.1, 'EARLY_STOPPING_PATIENCE': 5, 'EARLY_STOPPING_THRESHOLD': 0.001, 'EVAL_STRATEGY': 'epoch', 'SAVE_STRATEGY': 'epoch', 'SAVE_TOTAL_LIMIT': 3, 'LOAD_BEST_MODEL': True, 'METRIC_FOR_BEST_MODEL': 'eval_weighted_f1', 'GREATER_IS_BETTER': True, 'LOGGING_STEPS': 50, 'REPORT_TO': 'tensorboard', 'USE_CLASS_WEIGHTS': True, 'FOCAL_LOSS': True, 'FOCAL_ALPHA': 0.25, 'FOCAL_GAMMA': 2.0, 'SEED': 42, '__dict__': <attribute '__dict__' of 'TrainingConfig' objects>, '__weakref__': <attribute '__weakref__' of 'TrainingConfig' objects>, '__doc__': None}"