Add GitHub link and usage instructions to model card

README.md

@@ -4,16 +4,32 @@ license: apache-2.0
 tags:
   - text-classification
   - sentiment-analysis
+base_model: distilbert-base-uncased
+datasets:
+  - imdb
 ---
 
 # hf-workshop-sentiment
 
-Fine-tuned DistilBERT for binary sentiment classification (positive/negative).
+Binary sentiment classifier (positive/negative) fine-tuned from `distilbert-base-uncased`.
 
 ## Usage
 
 ```python
 from transformers import pipeline
+
 classifier = pipeline("sentiment-analysis", model="marksoulier/hf-workshop-sentiment")
-classifier("I loved this!")
+classifier("I loved this film!")
+# [{'label': 'POSITIVE', 'score': 0.998}]
+```
+
+## Training
+
+Training code, dependencies, and full reproduction steps are on GitHub:
+**https://github.com/marksoulier/hf-workshop**
+
+```bash
+git clone https://github.com/marksoulier/hf-workshop
+cd hf-workshop
+uv run train.py
 ```