Spaces:
Sleeping
Sleeping
Fixed bug
Browse filesREMOVED "app_file" ENTRY
README.md
CHANGED
|
@@ -8,13 +8,36 @@ app_port: 8501
|
|
| 8 |
tags:
|
| 9 |
- streamlit
|
| 10 |
pinned: false
|
| 11 |
-
short_description: Streamlit
|
| 12 |
license: apache-2.0
|
| 13 |
---
|
| 14 |
|
| 15 |
-
#
|
| 16 |
|
| 17 |
-
|
| 18 |
|
| 19 |
-
|
| 20 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
tags:
|
| 9 |
- streamlit
|
| 10 |
pinned: false
|
| 11 |
+
short_description: Streamlit Demo App
|
| 12 |
license: apache-2.0
|
| 13 |
---
|
| 14 |
|
| 15 |
+
# 📧 Spam Classifier Agent Live Demo
|
| 16 |
|
| 17 |
+
A Streamlit web application that leverages a fine-tuned GPT-2 model to classify text messages as either "SPAM" or "NOT SPAM". This agent helps users quickly determine the nature of a message. Try it here @ [SoggyBurritos/Spam_Classifier_Agent](https://huggingface.co/spaces/SoggyBurritos/Spam_Classifier_Agent)
|
| 18 |
|
| 19 |
+
## ✨ Features
|
| 20 |
+
|
| 21 |
+
* **Text Input:** Easily enter any text message for classification.
|
| 22 |
+
* **Real-time Prediction:** Get instant results on whether the message is spam or not.
|
| 23 |
+
* **Probability Scores:** View the confidence scores for both "SPAM" and "NOT SPAM" categories.
|
| 24 |
+
* **Intuitive UI:** A clean and user-friendly interface built with Streamlit.
|
| 25 |
+
* **Fine-tuned GPT-2 Model:** Utilizes a powerful transformer model for accurate classification.
|
| 26 |
+
|
| 27 |
+
## 💻 How to Use
|
| 28 |
+
|
| 29 |
+
1. **Enter Text:** Type or paste the message you want to classify into the provided text area.
|
| 30 |
+
2. **Analyze:** Click the "Analyze Text" button.
|
| 31 |
+
3. **View Results:** The application will display the prediction (SPAM or NOT SPAM) along with the probability scores.
|
| 32 |
+
|
| 33 |
+
## ⚙️ Technical Details for Hugging Face Spaces
|
| 34 |
+
|
| 35 |
+
**App File Location:**
|
| 36 |
+
The main Streamlit application script is located at `streamlit/app/main.py`. Hugging Face Spaces will be configured to run this specific file.
|
| 37 |
+
|
| 38 |
+
**Dependencies:**
|
| 39 |
+
All required Python packages for this deployment is listed in `requirements.txt`.
|
| 40 |
+
|
| 41 |
+
**Model & Data:**
|
| 42 |
+
* The fine-tuned GPT-2 model weights (`Spam-Classifier-GPT2-Model.pt`) are expected to be in the `models/` directory at the root of the repository.
|
| 43 |
+
* The application also relies on data preparation scripts which may download or create necessary data files.
|