Update README.md
Browse files
README.md
CHANGED
@@ -12,28 +12,41 @@ widget:
|
|
12 |
- text: "This product is great!"
|
13 |
---
|
14 |
|
15 |
-
# Fine-tuned RoBERTa for Sentiment Analysis on Amazon Reviews
|
16 |
|
17 |
This is a fine-tuned version of [cardiffnlp/twitter-roberta-base-sentiment-latest](https://huggingface.co/cardiffnlp/twitter-roberta-base-sentiment-latest) on the [Amazon Reviews dataset](https://www.kaggle.com/datasets/bittlingmayer/amazonreviews) for sentiment analysis.
|
18 |
|
19 |
-
## Model Details
|
20 |
|
21 |
-
-
|
22 |
-
-
|
23 |
-
-
|
24 |
-
-
|
25 |
|
26 |
-
## Training
|
27 |
|
28 |
The model was trained using the following parameters:
|
29 |
|
30 |
-
-
|
31 |
-
-
|
32 |
-
-
|
33 |
-
-
|
34 |
-
-
|
35 |
|
36 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
37 |
|
38 |
You can use this model directly with the Hugging Face `transformers` library:
|
39 |
|
@@ -47,7 +60,12 @@ tokenizer = RobertaTokenizer.from_pretrained(model_name)
|
|
47 |
# Example usage
|
48 |
inputs = tokenizer("This product is great!", return_tensors="pt")
|
49 |
outputs = model(**inputs)
|
|
|
|
|
|
|
|
|
50 |
```
|
51 |
-
## License
|
52 |
|
53 |
-
|
|
|
|
|
|
12 |
- text: "This product is great!"
|
13 |
---
|
14 |
|
15 |
+
# π Fine-tuned RoBERTa for Sentiment Analysis on Amazon Reviews π
|
16 |
|
17 |
This is a fine-tuned version of [cardiffnlp/twitter-roberta-base-sentiment-latest](https://huggingface.co/cardiffnlp/twitter-roberta-base-sentiment-latest) on the [Amazon Reviews dataset](https://www.kaggle.com/datasets/bittlingmayer/amazonreviews) for sentiment analysis.
|
18 |
|
19 |
+
## π Model Details
|
20 |
|
21 |
+
- **π Model Name:** `AnkitAI/reviews-roberta-base-sentiment-analysis`
|
22 |
+
- **π Base Model:** `cardiffnlp/twitter-roberta-base-sentiment-latest`
|
23 |
+
- **π Dataset:** [Amazon Reviews](https://www.kaggle.com/datasets/bittlingmayer/amazonreviews)
|
24 |
+
- **βοΈ Fine-tuning:** This model was fine-tuned for sentiment analysis with a classification head for binary sentiment classification (positive and negative).
|
25 |
|
26 |
+
## ποΈ Training
|
27 |
|
28 |
The model was trained using the following parameters:
|
29 |
|
30 |
+
- **π§ Learning Rate:** 2e-5
|
31 |
+
- **π¦ Batch Size:** 16
|
32 |
+
- **β³ Epochs:** 3
|
33 |
+
- **βοΈ Weight Decay:** 0.01
|
34 |
+
- **π
Evaluation Strategy:** Epoch
|
35 |
|
36 |
+
|
37 |
+
### ποΈ Training Details
|
38 |
+
|
39 |
+
- **π Eval Loss:** 0.1049
|
40 |
+
- **β±οΈ Eval Runtime:** 3177.538 seconds
|
41 |
+
- **π Eval Samples/Second:** 226.591
|
42 |
+
- **π Eval Steps/Second:** 7.081
|
43 |
+
- **π Epoch:** 3.0
|
44 |
+
- **π Train Runtime:** 110070.6349 seconds
|
45 |
+
- **π Train Samples/Second:** 78.495
|
46 |
+
- **π Train Steps/Second:** 2.453
|
47 |
+
- **π Train Loss:** 0.0858
|
48 |
+
|
49 |
+
## π Usage
|
50 |
|
51 |
You can use this model directly with the Hugging Face `transformers` library:
|
52 |
|
|
|
60 |
# Example usage
|
61 |
inputs = tokenizer("This product is great!", return_tensors="pt")
|
62 |
outputs = model(**inputs)
|
63 |
+
|
64 |
+
# Get sentiment
|
65 |
+
logits = outputs.logits
|
66 |
+
print(logits)
|
67 |
```
|
|
|
68 |
|
69 |
+
## π License
|
70 |
+
|
71 |
+
This model is licensed under the [MIT License](LICENSE).
|