ArabicNewsAnalyzer/Marbertv2-Hate-and-Offensive-Speech

README.md

@@ -0,0 +1,83 @@
+---
+library_name: transformers
+base_model: UBC-NLP/MARBERTv2
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: checkpoints
+  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. -->
+
+# checkpoints
+
+This model is a fine-tuned version of [UBC-NLP/MARBERTv2](https://huggingface.co/UBC-NLP/MARBERTv2) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.1854
+- Accuracy: 0.7595
+- Balanced Accuracy: 0.7595
+- Mcc: 0.6402
+- Macro F1: 0.7607
+- Macro Precision: 0.7646
+- Macro Recall: 0.7595
+- Weighted F1: 0.7607
+- Weighted Precision: 0.7646
+- Weighted Recall: 0.7595
+- Class 0 F1: 0.8113
+- Class 0 Precision: 0.86
+- Class 0 Recall: 0.7679
+- Class 1 F1: 0.7235
+- Class 1 Precision: 0.7177
+- Class 1 Recall: 0.7293
+- Class 2 F1: 0.7473
+- Class 2 Precision: 0.7162
+- Class 2 Recall: 0.7812
+
+## 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: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 32
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 0.1
+- num_epochs: 5
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | Balanced Accuracy | Mcc    | Macro F1 | Macro Precision | Macro Recall | Weighted F1 | Weighted Precision | Weighted Recall | Class 0 F1 | Class 0 Precision | Class 0 Recall | Class 1 F1 | Class 1 Precision | Class 1 Recall | Class 2 F1 | Class 2 Precision | Class 2 Recall |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------------:|:------:|:--------:|:---------------:|:------------:|:-----------:|:------------------:|:---------------:|:----------:|:-----------------:|:--------------:|:----------:|:-----------------:|:--------------:|:----------:|:-----------------:|:--------------:|
+| 0.5611        | 1.0   | 925  | 0.5775          | 0.7714   | 0.7714            | 0.6584 | 0.7699   | 0.7716          | 0.7714       | 0.7699      | 0.7716             | 0.7714          | 0.8281     | 0.8052            | 0.8523         | 0.7215     | 0.7737            | 0.6759         | 0.7602     | 0.7360            | 0.7861         |
+| 0.4367        | 2.0   | 1850 | 0.6233          | 0.7757   | 0.7757            | 0.6641 | 0.7759   | 0.7774          | 0.7757       | 0.7759      | 0.7774             | 0.7757          | 0.8223     | 0.8418            | 0.8036         | 0.7355     | 0.7504            | 0.7212         | 0.7698     | 0.7399            | 0.8023         |
+| 0.2949        | 3.0   | 2775 | 0.7892          | 0.7703   | 0.7703            | 0.6573 | 0.7717   | 0.7778          | 0.7703       | 0.7717      | 0.7777             | 0.7703          | 0.8132     | 0.8748            | 0.7597         | 0.7473     | 0.7044            | 0.7958         | 0.7547     | 0.7540            | 0.7553         |
+| 0.2141        | 4.0   | 3700 | 1.0209          | 0.7584   | 0.7584            | 0.6389 | 0.7599   | 0.7646          | 0.7584       | 0.7599      | 0.7645             | 0.7584          | 0.8093     | 0.8595            | 0.7646         | 0.7281     | 0.6920            | 0.7682         | 0.7423     | 0.7423            | 0.7423         |
+| 0.1405        | 5.0   | 4625 | 1.1854          | 0.7595   | 0.7595            | 0.6402 | 0.7607   | 0.7646          | 0.7595       | 0.7607      | 0.7646             | 0.7595          | 0.8113     | 0.86              | 0.7679         | 0.7235     | 0.7177            | 0.7293         | 0.7473     | 0.7162            | 0.7812         |
+
+
+### Framework versions
+
+- Transformers 5.13.1
+- Pytorch 2.11.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2

config.json

@@ -0,0 +1,47 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": null,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "dtype": "float32",
+  "eos_token_id": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "Normal",
+    "1": "Offensive",
+    "2": "Hate Speech"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "is_decoder": false,
+  "label2id": {
+    "Hate Speech": 2,
+    "Normal": 0,
+    "Offensive": 1
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "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",
+  "problem_type": "single_label_classification",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.13.1",
+  "type_vocab_size": 2,
+  "use_cache": false,
+  "vocab_size": 100000
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a48fca801a6f250123de33ca8fb681f640a9fe4b8000fc7b61f7e99d56cf8a3
+size 651398148

tokenizer_config.json

@@ -0,0 +1,17 @@
+{
+  "backend": "tokenizers",
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff9f61d25f86f77c1ae942b6812618cad3aab3f5d4b34880b9fd2a2b6e72661e
+size 5265