Upload README.md with huggingface_hub

README.md

@@ -2,38 +2,74 @@
 library_name: transformers
 license: apache-2.0
 base_model: distilbert-base-uncased
+language:
+- id
+datasets:
+- Bangkah/atha-text-dataset
 tags:
 - generated_from_trainer
+- sentiment-analysis
+- text-classification
+- indonesian
 metrics:
 - accuracy
 - f1
 model-index:
 - name: atha-text-classifier
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: Atha Text Dataset
+      type: Bangkah/atha-text-dataset
+      split: validation
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8222
+    - name: Macro F1
+      type: f1
+      value: 0.8237
 ---
 
-<!-- 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. -->
-
 # atha-text-classifier
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
-It achieves the following results on the evaluation set:
+Model ini adalah fine-tuned [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) untuk klasifikasi sentimen Bahasa Indonesia 3 kelas:
+
+- `negative`
+- `neutral`
+- `positive`
+
+Training data: https://huggingface.co/datasets/Bangkah/atha-text-dataset
+
+Evaluasi pada split validation:
+
 - Loss: 0.7220
 - Accuracy: 0.8222
-- F1: 0.8237
+- Macro F1: 0.8237
 
 ## Model description
 
-More information needed
+Model ini dibuat sebagai baseline untuk pipeline NLP end-to-end: dataset, training, publishing model, dan demo app.
 
 ## Intended uses & limitations
 
-More information needed
+- Cocok untuk demo, prototyping, dan portfolio project.
+- Tidak disarankan untuk use-case produksi tanpa retraining di data nyata.
+- Karena data sintetis, model bisa bias terhadap pola template.
 
 ## Training and evaluation data
 
-More information needed
+Dataset publik:
+
+- https://huggingface.co/datasets/Bangkah/atha-text-dataset
+
+Ringkasan:
+
+- Train: 360 contoh
+- Validation: 90 contoh
+- Kelas seimbang (negative, neutral, positive)
 
 ## Training procedure
 
@@ -50,9 +86,9 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
-| 0.8508        | 1.0   | 45   | 0.7220          | 0.8222   | 0.8237 |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | Macro F1 |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|
+| 0.8508        | 1.0   | 45   | 0.7220          | 0.8222   | 0.8237   |
 
 
 ### Framework versions