Update README.md
Browse files
README.md
CHANGED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language: en
|
3 |
+
license: mit
|
4 |
+
datasets: imdb
|
5 |
+
tags:
|
6 |
+
- sentiment-analysis
|
7 |
+
- transformers
|
8 |
+
- huggingface
|
9 |
+
---
|
10 |
+
|
11 |
+
# π¬ IMDb Sentiment Classifier
|
12 |
+
This is a fine-tuned **DistilBERT model** for analyzing sentiment in IMDb movie reviews.
|
13 |
+
|
14 |
+
## π Dataset
|
15 |
+
- **Source:** IMDb dataset from Hugging Face Datasets
|
16 |
+
- **Task:** Binary classification (Positive / Negative)
|
17 |
+
|
18 |
+
## π Training Details
|
19 |
+
- **Model:** `distilbert-base-uncased`
|
20 |
+
- **Learning rate:** `2e-5`
|
21 |
+
- **Batch size:** `4`
|
22 |
+
- **Epochs:** `1`
|
23 |
+
- **Loss function:** CrossEntropyLoss
|
24 |
+
|
25 |
+
## π Evaluation Results
|
26 |
+
| Metric | Score |
|
27 |
+
|------------|--------|
|
28 |
+
| Accuracy | 92.5% |
|
29 |
+
| F1-score | 92.6% |
|
30 |
+
| Precision | 92.9% |
|
31 |
+
| Recall | 92.3% |
|
32 |
+
|
33 |
+
## π How to Use
|
34 |
+
```python
|
35 |
+
from transformers import pipeline
|
36 |
+
|
37 |
+
classifier = pipeline("text-classification", model="Camilla9000/imdb-sentiment-classifier")
|
38 |
+
print(classifier("This movie was amazing!"))
|