Skip to content

Commit cb21c1d

Browse files
deepjyotipaulheredeepjyotipaulhere
committed
Create project Greenify in Showcase for Cookbook
1 parent 9da3edd commit cb21c1d

File tree

1 file changed

+204
-0
lines changed

1 file changed

+204
-0
lines changed

‎docs/showcase/greenify.mdx

Lines changed: 204 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,204 @@
1+
---
2+
title: Greenify | Localized community-driven greenification/plantation solution with AI
3+
description: Greenify is a mobile application designed to encourage and facilitate sustainable practices by analyzing live image and building community via Perplexity Sonar API.
4+
sidebar_position: 6
5+
keywords: [image processing, community, maps, expo, react native, flask, Perplexity, sonar]
6+
---
7+
8+
# Greenify
9+
10+
![Greenify](https://raw.githubusercontent.com/deepjyotipaulhere/greenify/master/assets/images/gallery.jpg)
11+
12+
The inspiration for Greenify stems from the growing need to address environmental challenges and promote sustainable living. With the rise of urbanization and technology, we wanted to create a platform that merges innovation with eco-consciousness. Greenify aims to empower individuals and communities to take actionable steps toward a greener future by leveraging technology to make sustainability accessible and engaging.
13+
14+
## Features
15+
16+
Greenify is a mobile application designed to encourage and facilitate sustainable practices. It provides users with tools and resources to:
17+
18+
- Participate in community-driven eco-friendly challenges and initiatives.
19+
- Access a curated library of tips, guides, and resources for sustainable living.
20+
- Connect with like-minded individuals through a community platform to share ideas and inspire action.
21+
22+
The app is designed to be user-friendly, visually appealing, and impactful, making it easier for users to integrate sustainability into their daily lives.
23+
24+
## Prerequisites
25+
26+
- NodeJS 20.19.4 or later
27+
- Python 3.10.0 or later
28+
- Perplexity API key for Sonar integration
29+
- Expo (SDK version 51 or later) ([Setup guide](https://docs.expo.dev/))
30+
- Android SDK/studio set up for local build or simulator run ([Setup guide](https://developer.android.com/about/versions/14/setup-sdk))
31+
- Xcode installed if using Mac and for simulator run ([Setup guide](https://developer.apple.com/documentation/safari-developer-tools/installing-xcode-and-simulators))
32+
- An Android/iPhone device for image capture with camera
33+
34+
## Installation
35+
36+
This is an [Expo](https://expo.dev) project created with [`create-expo-app`](https://www.npmjs.com/package/create-expo-app). Also there is a ```/service``` folder in the root directory of the project which contains the Flask API for communicating between frontend and Perplexity API.
37+
38+
1. Install dependencies
39+
40+
```bash
41+
npm install
42+
```
43+
44+
2. Start the app
45+
46+
```bash
47+
npx expo start
48+
```
49+
50+
In the output, you'll find options to open the app in a
51+
52+
- [development build](https://docs.expo.dev/develop/development-builds/introduction/)
53+
- [Android emulator](https://docs.expo.dev/workflow/android-studio-emulator/)
54+
- [iOS simulator](https://docs.expo.dev/workflow/ios-simulator/)
55+
- [Expo Go](https://expo.dev/go), a limited sandbox for trying out app development with Expo
56+
57+
3. In another terminal navigate to ```/service``` folder and install dependencies
58+
```bash
59+
pip install -r requirements.txt
60+
```
61+
4. Set ```PPLX_API_KEY``` in ```.env``` file inside the ```/service``` folder (create ```.env``` file if doesn't exist)
62+
4. Run Flask app
63+
```bash
64+
python app.py
65+
```
66+
5. To open app in mobile
67+
##### Option 1
68+
* Install Expo Go app from Play Store or App Store
69+
* Scan the QR code shown in the terminal
70+
71+
##### Option 2
72+
Open web browser in your smartphone and navigate to the URL shown in the console.
73+
74+
## Abstract Data Flow Diagram
75+
![Abstract Data Flow Diagram](https://d112y698adiu2z.cloudfront.net/photos/production/software_photos/003/418/290/datas/gallery.jpg)
76+
77+
78+
## Usage
79+
80+
After running the app in your own setup or through the hosted URL, the following steps can be followed:
81+
82+
### First Step: Greenification
83+
84+
| | |
85+
|------|-------------|
86+
| ![First Step](https://raw.githubusercontent.com/deepjyotipaulhere/greenify/refs/heads/master/assets/images/ezgif-4647c5467e6dac.gif) | <ul><li>The user opens the app, which requests permission to capture a photo and access the user's location.</li><li>The user takes a photo of their space (e.g., a balcony), and the app automatically captures their coordinates (latitude, longitude, and altitude).</li><li>The image is sent to the Perplexity <code>sonar-pro</code> model to analyze the environment (e.g., balcony, roadside), sunlight exposure, and available space for plants.</li><li>The resulting analysis and user coordinates are sent to the Perplexity <code>sonar-deep-research</code> model to get real-time weather information, assess plant growth suitability, and receive up to five plant suggestions.</li><li>The API returns an analysis of which plants are suitable based on the location type, sunlight, and whether it's an indoor or outdoor space.</li><li>The API also provides a list of suitable plants based on sunlight and average weather conditions for the location.</li><li>Finally, the user can share these plant suggestions with other users in the same locality.</li></ul> |
87+
88+
### Second Step: Community Building
89+
90+
| | |
91+
|-------------|------|
92+
| <ul><li>The community screen displays nearby users and the plant suggestions they received from the photos they uploaded.</li><li>The user clicks the “Start Matching” button to get AI-suggested matches with other users based on similar interests, plant suggestions, and the potential for sharing resources.</li><li>The API suggests community groups for matched users based on shared interests and recommended plants.</li><li>Each community group shows the matched users and an explanation for why they were matched.</li><li>The app also explains how the community can positively impact the local ecosystem and promote economic growth.</li></ul> | ![Second Step](https://raw.githubusercontent.com/deepjyotipaulhere/greenify/refs/heads/master/assets/images/ezgif-439e19e10fb6f4.gif) |
93+
94+
## Code Explanation
95+
96+
Greenify was built using the [Expo](https://expo.dev) framework, which allowed us to create a cross-platform application for Android, iOS, and the web. Key technologies and tools used include:
97+
98+
- **Frontend**: React Native with Expo for building the user interface and ensuring a seamless user experience.
99+
- **Backend**: A Python-based service using Flask to handle data processing and API endpoints.
100+
- **Perplexity AI**: Using Perplexity AI's sonar-pro and sonar-deep-research models to classify image, plant suggestions based on image and coordinates by realtime research, creating a community by matching users of similar plant suggestions
101+
- **Design**: Using React Native UI Kitten for custom themes and assets, including fonts and icons, to create a visually cohesive and engaging interface.
102+
- **File-based routing**: Leveraging Expo's file-based routing system for intuitive navigation.
103+
- **Community features**: Implemented using React Native components and hooks for real-time interaction.
104+
105+
** Pydantic models **
106+
```python
107+
from pydantic import BaseModel, Field
108+
109+
110+
class Plant(BaseModel):
111+
name: str
112+
image: str = Field(description="Image URL of the plant")
113+
description: str = Field(description="Description of the plant")
114+
care_instructions: str = Field(description="Care instructions for the plant")
115+
care_tips: str = Field(description="Care tips for the plant")
116+
AR_model: str = Field(description="AR model URL for the plant")
117+
118+
119+
class Answer1(BaseModel):
120+
description: str
121+
122+
class Answer2(BaseModel):
123+
plants: list[Plant]
124+
125+
126+
class Benefit(BaseModel):
127+
type: str = Field(description="Type of the environmental benefit")
128+
amount: str = Field(description="How much percentage of improvement")
129+
direction: bool = Field(description="True means increasing, False means decreasing")
130+
131+
132+
class Group(BaseModel):
133+
users: list[str] = Field(
134+
description="List at least 2 or more users with similar plant suggestions and how they can combine same job in term of place, activities and plantation"
135+
)
136+
description: list[str] = Field(
137+
description="Short description of how these people match with each other"
138+
)
139+
benefits: list[Benefit] = Field(
140+
description="How this combination helps benefit the environment with parameter, percentage value"
141+
)
142+
143+
144+
class Community(BaseModel):
145+
match: list[Group]
146+
```
147+
148+
** Image Analysis and Insights about the captured image using sonar-pro and structured JSON output **
149+
```python
150+
payload = {
151+
"model": "sonar-pro",
152+
"messages": [
153+
{
154+
"role": "user",
155+
"content": [
156+
{
157+
"type": "text",
158+
"text": f"Analyze this image and return short description of the place with respect to suitability of plant growth ",
159+
},
160+
{"type": "image_url", "image_url": {"url": image}},
161+
],
162+
},
163+
],
164+
"stream": False,
165+
"response_format": {
166+
"type": "json_schema",
167+
"json_schema": {"schema": Answer1.model_json_schema()},
168+
},
169+
}
170+
```
171+
172+
** Matching people with similar plant suggestions and interests and create a matching community using sonar-deep-research **
173+
```python
174+
payload_research = {
175+
"model": "sonar-deep-research",
176+
"messages": [
177+
{
178+
"role": "system",
179+
"content": "You are a plant growth expert. You are given a description of a place where an user want to grow some plants. You are also given latitude, longitude and altitude of the user. Your task is to suggest at most 5 plant that can be grown by the user in that particular place according to average weather.",
180+
},
181+
{
182+
"role": "user",
183+
"content": f"I am standing in a place having coordinates [{lat}, {lng}] and altitude {alt}]. The place can be described as follows: {answer1}"
184+
"Suggest at most five suitable plants that can be grown here.",
185+
},
186+
],
187+
"stream": False,
188+
"response_format": {
189+
"type": "json_schema",
190+
"json_schema": {"schema": Answer2.model_json_schema()},
191+
},
192+
}
193+
```
194+
195+
196+
## Links
197+
198+
- [GitHub Repository](https://github.com/deepjyotipaulhere/greenify)
199+
- [Live Demo](https://greenify.expo.app)
200+
201+
![Live Demo URL](https://raw.githubusercontent.com/deepjyotipaulhere/greenify/refs/heads/master/assets/images/frame.png)
202+
- Youtube Demo
203+
204+
[![IMAGE ALT TEXT HERE](https://img.youtube.com/vi/IFP0EiHqd7Y/0.jpg)](https://www.youtube.com/watch?v=IFP0EiHqd7Y)

0 commit comments

Comments
 (0)