ashandilgith commited on
Commit
8fad80c
·
1 Parent(s): e3dac0c

fixed readme for hf

Browse files
Files changed (1) hide show
  1. README.md +20 -25
README.md CHANGED
@@ -1,4 +1,14 @@
1
- # 🌊 Piranaware: Acoustic Digital Twin for Marine Engines
 
 
 
 
 
 
 
 
 
 
2
 
3
  **Predictive Maintenance / Anomaly Detection System v1.0**
4
  *Listen to what your engine is telling you—before it breaks.*
@@ -6,7 +16,7 @@
6
  ---
7
 
8
  ## 📖 Overview
9
- Piranaware is an AI-powered diagnostic tool designed for marine vessels. It creates a "Digital Twin" of an engine's acoustic signature using Deep Learning (Autoencoders). By comparing current engine sounds against a healthy baseline, it can detect mechanical degradation (bearing wear, valve drift, misfires) weeks before catastrophic failure.
10
 
11
  **Deployed on:** Hugging Face Spaces (Docker)
12
  **Backend:** Google Cloud Platform (Storage)
@@ -14,7 +24,7 @@ Piranaware is an AI-powered diagnostic tool designed for marine vessels. It crea
14
 
15
  ---
16
 
17
- ## 🚀 Key Features
18
  * **Tri-Sonic Calibration:** Trains distinct models for **IDLE**, **SLOW**, and **FAST** RPM states.
19
  * **Cloud Sync:** Automatically uploads trained models to secure Google Cloud Storage buckets.
20
  * **Statistical Thresholding:** Uses dynamic safety limits (`Mean + 2σ`) to filter out false positives.
@@ -22,7 +32,7 @@ Piranaware is an AI-powered diagnostic tool designed for marine vessels. It crea
22
 
23
  ---
24
 
25
- ## 🛠️ Tech Stack
26
  * **Frontend:** Streamlit
27
  * **ML Core:** TensorFlow, Librosa (Audio Processing), NumPy
28
  * **Infrastructure:** Docker, Google Cloud Storage (GCS)
@@ -30,25 +40,10 @@ Piranaware is an AI-powered diagnostic tool designed for marine vessels. It crea
30
 
31
  ---
32
 
33
- ## ⚙️ Setup & Installation
34
-
35
- ### Prerequisites
36
- * Python 3.9+
37
- * FFmpeg (for audio processing)
38
- * Google Cloud Service Account Key
39
-
40
- ### 1. Local Development
41
- ```bash
42
- # Clone the repo
43
- git clone [https://github.com/your-username/piranaware.git](https://github.com/your-username/piranaware.git)
44
- cd piranaware
45
-
46
- # Create virtual environment
47
- python -m venv venv
48
- source venv/bin/activate # On Windows: venv\Scripts\activate
49
-
50
- # Install dependencies
51
- pip install -r requirements.txt
52
 
53
- # Run the app
54
- streamlit run app.py
 
 
 
1
+ ---
2
+ title: Piranaware V3
3
+ emoji: 🐠
4
+ colorFrom: blue
5
+ colorTo: yellow
6
+ sdk: docker
7
+ pinned: false
8
+ app_port: 7860
9
+ ---
10
+
11
+ # PiranaWare: Preventing marine engine breakdown with vibration analysis
12
 
13
  **Predictive Maintenance / Anomaly Detection System v1.0**
14
  *Listen to what your engine is telling you—before it breaks.*
 
16
  ---
17
 
18
  ## 📖 Overview
19
+ PiranaWare is an AI-powered diagnostic tool designed for marine vessels. It creates a "Digital Twin" of an engine's acoustic signature using Deep Learning (Autoencoders). By comparing current engine sounds against a healthy baseline, it can detect mechanical degradation (bearing wear, valve drift, misfires) weeks before catastrophic failure.
20
 
21
  **Deployed on:** Hugging Face Spaces (Docker)
22
  **Backend:** Google Cloud Platform (Storage)
 
24
 
25
  ---
26
 
27
+ ## Key Features
28
  * **Tri-Sonic Calibration:** Trains distinct models for **IDLE**, **SLOW**, and **FAST** RPM states.
29
  * **Cloud Sync:** Automatically uploads trained models to secure Google Cloud Storage buckets.
30
  * **Statistical Thresholding:** Uses dynamic safety limits (`Mean + 2σ`) to filter out false positives.
 
32
 
33
  ---
34
 
35
+ ## Tech Stack
36
  * **Frontend:** Streamlit
37
  * **ML Core:** TensorFlow, Librosa (Audio Processing), NumPy
38
  * **Infrastructure:** Docker, Google Cloud Storage (GCS)
 
40
 
41
  ---
42
 
43
+ ## License
44
+ MIT License. Free for educational and testing use.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
45
 
46
+ ## Contact
47
+ **Piranaware Tech**
48
+ * **Email:** founder@piranaware.tech
49
+ * **Status:** MVP (Active Development)