End of training

README.md

@@ -0,0 +1,77 @@
+---
+base_model: asafaya/bert-base-arabic
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: Improved-Arabic-bert-base
+  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. -->
+
+# Improved-Arabic-bert-base
+
+This model is a fine-tuned version of [asafaya/bert-base-arabic](https://huggingface.co/asafaya/bert-base-arabic) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.7595
+- Accuracy: 0.86
+
+## 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: 32
+- eval_batch_size: 32
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 15
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 0.5197        | 0.55  | 50   | 0.3977          | 0.8      |
+| 0.3323        | 1.1   | 100  | 0.3298          | 0.86     |
+| 0.2844        | 1.65  | 150  | 0.3401          | 0.84     |
+| 0.2128        | 2.2   | 200  | 0.4569          | 0.8      |
+| 0.1539        | 2.75  | 250  | 0.4315          | 0.83     |
+| 0.1346        | 3.3   | 300  | 0.5178          | 0.81     |
+| 0.0933        | 3.85  | 350  | 0.5167          | 0.84     |
+| 0.0641        | 4.4   | 400  | 0.6903          | 0.82     |
+| 0.0698        | 4.95  | 450  | 0.5628          | 0.85     |
+| 0.028         | 5.49  | 500  | 0.6472          | 0.86     |
+| 0.0449        | 6.04  | 550  | 0.6739          | 0.85     |
+| 0.0133        | 6.59  | 600  | 0.6925          | 0.84     |
+| 0.0177        | 7.14  | 650  | 0.6716          | 0.87     |
+| 0.0209        | 7.69  | 700  | 0.6644          | 0.89     |
+| 0.0226        | 8.24  | 750  | 0.7650          | 0.84     |
+| 0.0137        | 8.79  | 800  | 0.8186          | 0.86     |
+| 0.0164        | 9.34  | 850  | 0.7771          | 0.86     |
+| 0.006         | 9.89  | 900  | 0.7805          | 0.85     |
+| 0.0069        | 10.44 | 950  | 0.7595          | 0.86     |
+
+
+### Framework versions
+
+- Transformers 4.34.1
+- Pytorch 2.1.0+cu118
+- Datasets 2.14.7
+- Tokenizers 0.14.1

config.json

@@ -1,7 +1,7 @@
 {
   "_name_or_path": "asafaya/bert-base-arabic",
   "architectures": [
-    "BertForMaskedLM"
+    "BertForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
   "classifier_dropout": null,
@@ -9,16 +9,8 @@
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
-  "id2label": {
-    "0": "Negative",
-    "1": "Positive"
-  },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
-  "label2id": {
-    "Negative": 0,
-    "Positive": 1
-  },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
   "model_type": "bert",
@@ -27,6 +19,8 @@
   "output_past": true,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
   "transformers_version": "4.34.1",
   "type_vocab_size": 2,
   "use_cache": true,