Update README.md
Browse files
README.md
CHANGED
|
@@ -7,7 +7,6 @@ sdk: static
|
|
| 7 |
pinned: false
|
| 8 |
license: mit
|
| 9 |
---
|
| 10 |
-
π TrueWrite Scan: AI-Powered Writing Assistant
|
| 11 |
TrueWrite Scan is a full-stack AI application designed to enhance writing integrity. It integrates state-of-the-art NLP models to provide real-time Grammar Checking, Plagiarism Detection, and AI Content Classification.
|
| 12 |
|
| 13 |
π Key Features
|
|
@@ -27,8 +26,8 @@ Deployment: Hugging Face Spaces (Dockerized)
|
|
| 27 |
ποΈ Architecture & Deployment Strategy
|
| 28 |
Running heavy ML models (PyTorch + JVM for Grammar) requires significant RAM. I implemented a Split Deployment Strategy to optimize performance on free-tier infrastructure:
|
| 29 |
|
| 30 |
-
Frontend
|
| 31 |
-
Backend
|
| 32 |
Why Docker? To install System-level dependencies (OpenJDK 21 for Grammar check, build-essential for Auth).
|
| 33 |
Optimization: Configured to use CPU-optimized PyTorch builds to reduce slug size and memory footprint.
|
| 34 |
|
|
|
|
| 7 |
pinned: false
|
| 8 |
license: mit
|
| 9 |
---
|
|
|
|
| 10 |
TrueWrite Scan is a full-stack AI application designed to enhance writing integrity. It integrates state-of-the-art NLP models to provide real-time Grammar Checking, Plagiarism Detection, and AI Content Classification.
|
| 11 |
|
| 12 |
π Key Features
|
|
|
|
| 26 |
ποΈ Architecture & Deployment Strategy
|
| 27 |
Running heavy ML models (PyTorch + JVM for Grammar) requires significant RAM. I implemented a Split Deployment Strategy to optimize performance on free-tier infrastructure:
|
| 28 |
|
| 29 |
+
Frontend: Handles UI/UX and client-side logic.
|
| 30 |
+
Backend: A custom Docker container running FastAPI.
|
| 31 |
Why Docker? To install System-level dependencies (OpenJDK 21 for Grammar check, build-essential for Auth).
|
| 32 |
Optimization: Configured to use CPU-optimized PyTorch builds to reduce slug size and memory footprint.
|
| 33 |
|