Model save

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: Yousefmd/arabert-sentiment-classification
+base_model: aubmindlab/bert-large-arabertv2
 tags:
 - generated_from_trainer
 metrics:
@@ -14,11 +14,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # feedback-classification
 
-This model is a fine-tuned version of [Yousefmd/arabert-sentiment-classification](https://huggingface.co/Yousefmd/arabert-sentiment-classification) on the None dataset.
+This model is a fine-tuned version of [aubmindlab/bert-large-arabertv2](https://huggingface.co/aubmindlab/bert-large-arabertv2) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5129
-- Macro F1: 0.8172
-- Accuracy: 0.8340
+- Loss: 0.3881
+- Macro F1: 0.8586
+- Accuracy: 0.8586
 
 ## Model description
 
@@ -51,13 +51,13 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Macro F1 | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|
-| 0.532         | 1.0   | 1238 | 0.4517          | 0.8098   | 0.8279   |
-| 0.3349        | 2.0   | 2476 | 0.5129          | 0.8172   | 0.8340   |
+| No log        | 1.0   | 339  | 0.4434          | 0.8448   | 0.8468   |
+| 0.6508        | 2.0   | 678  | 0.3881          | 0.8586   | 0.8586   |
 
 
 ### Framework versions
 
-- Transformers 4.33.3
-- Pytorch 1.12.0
+- Transformers 4.34.0
+- Pytorch 2.0.1+cu118
 - Datasets 2.14.5
-- Tokenizers 0.13.3
+- Tokenizers 0.14.1

config.json

@@ -9,16 +9,16 @@
   "hidden_dropout_prob": 0.1,
   "hidden_size": 1024,
   "id2label": {
-    "0": "LABEL_0",
-    "1": "LABEL_1",
-    "2": "LABEL_2"
+    "0": "request",
+    "1": "complaint",
+    "2": "satisfaction"
   },
   "initializer_range": 0.02,
   "intermediate_size": 4096,
   "label2id": {
-    "LABEL_0": 0,
-    "LABEL_1": 1,
-    "LABEL_2": 2
+    "complaint": 1,
+    "request": 0,
+    "satisfaction": 2
   },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,