├── .dockerignore ├── Models ├── TensorBoard │ ├── .dockerignore │ ├── app │ │ ├── services │ │ │ ├── __init__.py │ │ │ ├── generate_predictions.py │ │ │ ├── generate_logs.py │ │ │ └── generate_images.py │ │ ├── utils │ │ │ ├── __init__.py │ │ │ └── image_utils.py │ │ ├── __init__.py │ │ └── routes │ │ │ ├── __init__.py │ │ │ └── logs.py │ ├── models │ │ └── MobileNet │ │ │ ├── __init__.py │ │ │ ├── model.py │ │ │ └── conv_layers │ │ │ ├── convscript.py │ │ │ ├── conv_layer_names.txt │ │ │ └── all_layers.txt │ ├── requirements.txt │ ├── server.py │ ├── start │ │ ├── get_weights.py │ │ └── start.sh │ └── Dockerfile └── TensorFlow │ ├── app │ ├── services │ │ ├── __init__.py │ │ ├── generate_predictions.py │ │ └── generate_images.py │ ├── utils │ │ ├── __init__.py │ │ ├── image_utils.py │ │ └── s3_utils.py │ ├── __init__.py │ └── routes │ │ ├── __init__.py │ │ ├── predictions.py │ │ ├── featuremaps.py │ │ ├── heatmaps.py │ │ └── preprocess.py │ ├── models │ └── MobileNet │ │ ├── __init__.py │ │ ├── model.py │ │ └── conv_layers │ │ ├── convscript.py │ │ ├── conv_layer_names.txt │ │ └── all_layers.txt │ ├── server.py │ ├── start │ ├── get_weights.py │ └── start.sh │ ├── Dockerfile │ └── requirements.txt ├── README.md ├── public ├── logo.jpg ├── logo.png ├── title.png ├── background.jpg ├── loadspinner.gif ├── logo copy.jpeg ├── logoBlack.png ├── backgroundFlare.jpg ├── backgroundFlareBW.avif ├── modal-background-1.jpg ├── modal-background-2.jpg ├── modal-background-3.jpg ├── modal-background-4.jpg ├── xflair_logo_clear.png ├── input_images_comparison.png ├── vercel.svg └── next.svg ├── src └── app │ ├── favicon.ico │ ├── globals.css │ ├── ui │ ├── resizedImage.tsx │ ├── gridMapGif.tsx │ ├── navComponents │ │ ├── AccordionLink.tsx │ │ ├── Icon.tsx │ │ ├── Accordion.tsx │ │ └── Sidenav.tsx │ ├── heatMapGif.tsx │ ├── modal.tsx │ ├── top5Classes.tsx │ ├── dotButton.tsx │ ├── heatmap.tsx │ ├── carousel.tsx │ └── newModal.tsx │ ├── lib │ ├── definitions.ts │ └── accordionData.ts │ ├── (routes) │ ├── models │ │ ├── actions │ │ │ ├── logs │ │ │ │ └── route.ts │ │ │ └── image │ │ │ │ ├── heatmaps │ │ │ │ └── route.ts │ │ │ │ ├── featmaps │ │ │ │ └── route.ts │ │ │ │ ├── predict │ │ │ │ └── route.ts │ │ │ │ ├── upload │ │ │ │ └── route.ts │ │ │ │ ├── preprocess │ │ │ │ └── route.ts │ │ │ │ └── gifs │ │ │ │ └── route.ts │ │ ├── agnostic │ │ │ ├── text │ │ │ │ ├── gpt │ │ │ │ │ └── page.tsx │ │ │ │ └── texttospeech │ │ │ │ │ └── page.tsx │ │ │ └── images │ │ │ │ ├── detection │ │ │ │ └── page.tsx │ │ │ │ ├── generation │ │ │ │ └── page.tsx │ │ │ │ └── classification │ │ │ │ └── page.tsx │ │ ├── pytorch │ │ │ ├── text │ │ │ │ ├── gpt │ │ │ │ │ └── page.tsx │ │ │ │ └── texttospeech │ │ │ │ │ └── page.tsx │ │ │ └── images │ │ │ │ ├── detection │ │ │ │ └── page.tsx │ │ │ │ ├── classification │ │ │ │ └── page.tsx │ │ │ │ └── generation │ │ │ │ └── page.tsx │ │ └── tensorflow │ │ │ ├── text │ │ │ ├── gpt │ │ │ │ └── page.tsx │ │ │ └── texttospeech │ │ │ │ └── page.tsx │ │ │ └── images │ │ │ ├── detection │ │ │ └── page.tsx │ │ │ ├── generation │ │ │ └── page.tsx │ │ │ └── classification │ │ │ └── page.tsx │ ├── about-us │ │ └── page.tsx │ ├── libraries │ │ ├── python │ │ │ └── page.tsx │ │ └── react │ │ │ ├── javascript │ │ │ └── page.tsx │ │ │ └── typescript │ │ │ └── page.tsx │ └── documentation │ │ ├── python │ │ └── page.tsx │ │ └── javascript-typescript │ │ └── page.tsx │ ├── Sidebar.module.css │ ├── page.tsx │ ├── layout.tsx │ └── page.module.css ├── next.config.mjs ├── postcss.config.js ├── .eslintrc.json ├── Dockerfile-dev ├── Dockerfile ├── docker-compose-tensorboard.yml ├── docker-compose-prod.yml ├── docker-compose-dev.yml ├── tsconfig.json ├── tailwind.config.js ├── scripts ├── newHeatmap.ts └── newRenderTest.ts ├── .gitignore └── package.json /.dockerignore: -------------------------------------------------------------------------------- 1 | node_modules 2 | Models -------------------------------------------------------------------------------- /Models/TensorBoard/.dockerignore: -------------------------------------------------------------------------------- 1 | logs/* -------------------------------------------------------------------------------- /Models/TensorBoard/app/services/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Models/TensorFlow/app/services/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Models/TensorBoard/models/MobileNet/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Models/TensorFlow/models/MobileNet/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | xFlair is still under development. More info coming soon! -------------------------------------------------------------------------------- /public/logo.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/logo.jpg -------------------------------------------------------------------------------- /public/logo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/logo.png -------------------------------------------------------------------------------- /public/title.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/title.png -------------------------------------------------------------------------------- /src/app/favicon.ico: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/src/app/favicon.ico -------------------------------------------------------------------------------- /src/app/globals.css: -------------------------------------------------------------------------------- 1 | @tailwind base; 2 | @tailwind components; 3 | @tailwind utilities; 4 | -------------------------------------------------------------------------------- /public/background.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/background.jpg -------------------------------------------------------------------------------- /public/loadspinner.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/loadspinner.gif -------------------------------------------------------------------------------- /public/logo copy.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/logo copy.jpeg -------------------------------------------------------------------------------- /public/logoBlack.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/logoBlack.png -------------------------------------------------------------------------------- /public/backgroundFlare.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/backgroundFlare.jpg -------------------------------------------------------------------------------- /public/backgroundFlareBW.avif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/backgroundFlareBW.avif -------------------------------------------------------------------------------- /public/modal-background-1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/modal-background-1.jpg -------------------------------------------------------------------------------- /public/modal-background-2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/modal-background-2.jpg -------------------------------------------------------------------------------- /public/modal-background-3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/modal-background-3.jpg -------------------------------------------------------------------------------- /public/modal-background-4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/modal-background-4.jpg -------------------------------------------------------------------------------- /public/xflair_logo_clear.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/xflair_logo_clear.png -------------------------------------------------------------------------------- /next.config.mjs: -------------------------------------------------------------------------------- 1 | /** @type {import('next').NextConfig} */ 2 | const nextConfig = {}; 3 | 4 | export default nextConfig; 5 | -------------------------------------------------------------------------------- /postcss.config.js: -------------------------------------------------------------------------------- 1 | module.exports = { 2 | plugins: { 3 | tailwindcss: {}, 4 | autoprefixer: {}, 5 | }, 6 | } 7 | -------------------------------------------------------------------------------- /public/input_images_comparison.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/oslabs-beta/xflair/HEAD/public/input_images_comparison.png -------------------------------------------------------------------------------- /.eslintrc.json: -------------------------------------------------------------------------------- 1 | { 2 | "extends": "next/core-web-vitals", 3 | "rules": { 4 | "@next/next/no-img-element": "off" 5 | } 6 | } 7 | -------------------------------------------------------------------------------- /Models/TensorBoard/requirements.txt: -------------------------------------------------------------------------------- 1 | Flask 2 | Flask-Cors 3 | tensorflow 4 | numpy 5 | Pillow 6 | opencv-python 7 | boto3 8 | gunicorn 9 | matplotlib -------------------------------------------------------------------------------- /Models/TensorBoard/server.py: -------------------------------------------------------------------------------- 1 | from app import create_app 2 | 3 | app = create_app() 4 | 5 | if __name__ == '__main__': 6 | app.run(debug=True) -------------------------------------------------------------------------------- /Models/TensorFlow/server.py: -------------------------------------------------------------------------------- 1 | from app import create_app 2 | 3 | app = create_app() 4 | 5 | if __name__ == '__main__': 6 | app.run(debug=True) -------------------------------------------------------------------------------- /Models/TensorBoard/start/get_weights.py: -------------------------------------------------------------------------------- 1 | import tensorflow as tf 2 | 3 | def get_model(): 4 | return tf.keras.applications.MobileNetV2(weights='imagenet') 5 | 6 | model = get_model() -------------------------------------------------------------------------------- /Models/TensorFlow/start/get_weights.py: -------------------------------------------------------------------------------- 1 | import tensorflow as tf 2 | 3 | def get_model(): 4 | return tf.keras.applications.MobileNetV2(weights='imagenet') 5 | 6 | model = get_model() -------------------------------------------------------------------------------- /Dockerfile-dev: -------------------------------------------------------------------------------- 1 | FROM node:latest 2 | 3 | WORKDIR /usr/src/xflair 4 | 5 | COPY . /usr/src/xflair 6 | 7 | RUN npm install 8 | 9 | EXPOSE 3000 10 | 11 | CMD ["npm", "run", "dev"] 12 | 13 | -------------------------------------------------------------------------------- /Dockerfile: -------------------------------------------------------------------------------- 1 | FROM node:latest 2 | 3 | WORKDIR /usr/src/xflair 4 | 5 | COPY . /usr/src/xflair 6 | 7 | RUN npm install 8 | 9 | RUN npm run build 10 | 11 | EXPOSE 3000 12 | 13 | CMD ["npm", "start"] 14 | 15 | 16 | 17 | -------------------------------------------------------------------------------- /Models/TensorBoard/app/utils/__init__.py: -------------------------------------------------------------------------------- 1 | 2 | # Import any modules or subpackages you want to make available 3 | # when importing the `utils` package 4 | 5 | # For example: 6 | # from .module1 import * 7 | # from .module2 import * 8 | # from .subpackage import * 9 | -------------------------------------------------------------------------------- /Models/TensorFlow/app/utils/__init__.py: -------------------------------------------------------------------------------- 1 | 2 | # Import any modules or subpackages you want to make available 3 | # when importing the `utils` package 4 | 5 | # For example: 6 | # from .module1 import * 7 | # from .module2 import * 8 | # from .subpackage import * 9 | -------------------------------------------------------------------------------- /Models/TensorBoard/app/__init__.py: -------------------------------------------------------------------------------- 1 | from flask import Flask 2 | from app.routes import routes_bp 3 | from flask_cors import CORS 4 | 5 | def create_app(): 6 | app = Flask(__name__) 7 | CORS(app) 8 | app.register_blueprint(routes_bp) 9 | return app 10 | 11 | # Path: app/routes.py 12 | -------------------------------------------------------------------------------- /Models/TensorFlow/app/__init__.py: -------------------------------------------------------------------------------- 1 | from flask import Flask 2 | from app.routes import routes_bp 3 | from flask_cors import CORS 4 | 5 | def create_app(): 6 | app = Flask(__name__) 7 | CORS(app) 8 | app.register_blueprint(routes_bp) 9 | return app 10 | 11 | # Path: app/routes.py 12 | -------------------------------------------------------------------------------- /Models/TensorBoard/app/routes/__init__.py: -------------------------------------------------------------------------------- 1 | 2 | from flask import Blueprint 3 | 4 | # Create a blueprint for the routes 5 | routes_bp = Blueprint('routes', __name__) 6 | 7 | 8 | # Import the routes 9 | from app.routes import logs 10 | 11 | # Register the routes 12 | 13 | routes_bp.register_blueprint(logs.logs) 14 | 15 | 16 | -------------------------------------------------------------------------------- /Models/TensorFlow/start/start.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | 3 | # Navigate to the directory containing the model container files 4 | cd /TensorFlow 5 | 6 | # Download the weights by running the get_weights.py script 7 | echo "Downloading MobileNetV2 weights..." 8 | python start/get_weights.py 9 | 10 | # Start Gunicorn with your Flask application 11 | echo "Starting Gunicorn..." 12 | exec gunicorn --workers=9 --timeout 120 --bind 0.0.0.0:5000 server:app -------------------------------------------------------------------------------- /Models/TensorFlow/app/routes/__init__.py: -------------------------------------------------------------------------------- 1 | 2 | from flask import Blueprint 3 | 4 | # Create a blueprint for the routes 5 | routes_bp = Blueprint('routes', __name__) 6 | 7 | 8 | # Import the routes 9 | from app.routes import featuremaps, heatmaps, preprocess, predictions 10 | 11 | # Register the routes 12 | 13 | routes_bp.register_blueprint(featuremaps.featuremaps) 14 | routes_bp.register_blueprint(heatmaps.heatmaps) 15 | routes_bp.register_blueprint(preprocess.preprocess) 16 | routes_bp.register_blueprint(predictions.predictions) 17 | 18 | 19 | -------------------------------------------------------------------------------- /Models/TensorBoard/start/start.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | 3 | # Navigate to the directory containing the model container files 4 | cd /TensorBoard 5 | 6 | # Download the weights by running the get_weights.py script 7 | echo "Downloading MobileNetV2 weights..." 8 | python start/get_weights.py 9 | 10 | # Start TensorBoard in the background 11 | echo "Starting TensorBoard..." 12 | tensorboard --logdir=./logs --bind_all --port=6006 & 13 | 14 | # Start Gunicorn with your Flask application 15 | echo "Starting Gunicorn..." 16 | exec gunicorn --workers=5 --timeout 120 --bind 0.0.0.0:5000 server:app -------------------------------------------------------------------------------- /public/vercel.svg: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /src/app/ui/resizedImage.tsx: -------------------------------------------------------------------------------- 1 | import { useEffect, useRef, useState } from 'react'; 2 | import Image from 'next/image'; 3 | import styles from '../page.module.css'; 4 | 5 | interface Props { 6 | preprocessedFilePath: string; 7 | } 8 | 9 | let resizedImage: File | undefined; 10 | 11 | export default function ResizedImage(props: Props) { 12 | return ( 13 |
14 | 21 |
22 | ); 23 | } 24 | -------------------------------------------------------------------------------- /docker-compose-tensorboard.yml: -------------------------------------------------------------------------------- 1 | version: '3' 2 | 3 | services: 4 | app: 5 | build: 6 | context: . 7 | dockerfile: Dockerfile-dev 8 | image: xflair/dev:latest 9 | environment: 10 | - mode=logs 11 | ports: 12 | - '3000:3000' 13 | depends_on: 14 | - model 15 | 16 | model: 17 | build: 18 | context: ./Models/TensorBoard 19 | dockerfile: Dockerfile 20 | image: xflair/tensorboard:latest 21 | ports: 22 | - '5000:5000' 23 | - '6006:6006' 24 | volumes: 25 | - ./Models/TensorBoard/logs:./TensorBoard/logs 26 | entrypoint: /TensorBoard/start/start.sh 27 | -------------------------------------------------------------------------------- /docker-compose-prod.yml: -------------------------------------------------------------------------------- 1 | version: '3' 2 | 3 | services: 4 | app: 5 | build: 6 | context: . 7 | dockerfile: Dockerfile 8 | image: xflair/prod 9 | ports: 10 | - "3000:3000" 11 | environment: 12 | - NODE_ENV=production 13 | - SERVICE_HOST=host.docker.internal 14 | env_file: 15 | - .env 16 | depends_on: 17 | - model 18 | 19 | model: 20 | build: 21 | context: ./Models/TensorFlow 22 | dockerfile: Dockerfile 23 | image: xflair/tensorflow:latest 24 | ports: 25 | - "5000:5000" 26 | env_file: 27 | - ./Models/TensorFlow/.env 28 | entrypoint: /TensorFlow/start/start.sh -------------------------------------------------------------------------------- /docker-compose-dev.yml: -------------------------------------------------------------------------------- 1 | version: '3' 2 | 3 | services: 4 | app: 5 | build: 6 | context: . 7 | dockerfile: Dockerfile-dev 8 | image: xflair/dev:latest 9 | ports: 10 | - '3000:3000' 11 | environment: 12 | - mode=development 13 | - SERVICE_HOST=host.docker.internal 14 | env_file: 15 | - .env 16 | depends_on: 17 | - model 18 | 19 | model: 20 | build: 21 | context: ./Models/Tensorflow 22 | dockerfile: Dockerfile 23 | image: xflair/tensorflow:latest 24 | ports: 25 | - '5000:5000' 26 | env_file: 27 | - ./Models/TensorFlow/.env 28 | entrypoint: /TensorFlow/start/start.sh 29 | -------------------------------------------------------------------------------- /src/app/lib/definitions.ts: -------------------------------------------------------------------------------- 1 | export type Upload = { 2 | id: string; 3 | name: string; 4 | size: number; 5 | image: Buffer; 6 | }; 7 | 8 | export type Heatmaps = { 9 | heatmaps: string[] | null 10 | progressbars: string[] | null 11 | } 12 | 13 | export type Featuremaps = { 14 | featuremaps: string[] | null 15 | progressbars: string[] | null 16 | } 17 | 18 | export interface Top5Obj { 19 | [key: string]: number; 20 | } 21 | 22 | export interface AccordionContent { 23 | type: 'Link' | 'Accordion'; 24 | label: string; 25 | path?: string; 26 | id?: string[]; 27 | icon?: 'gear' | 'book' | 'pages'; 28 | contents?: AccordionContent[]; 29 | } 30 | -------------------------------------------------------------------------------- /Models/TensorBoard/models/MobileNet/model.py: -------------------------------------------------------------------------------- 1 | import tensorflow as tf 2 | import os 3 | 4 | dir = os.path.dirname(__file__) 5 | 6 | def get_model(): 7 | return tf.keras.applications.MobileNetV2(weights='imagenet') 8 | 9 | def get_layer_names(): 10 | layer_names_file = os.path.join(dir, 'conv_layers', 'conv_layer_names.txt') 11 | with open(layer_names_file, "r") as f: 12 | layer_names = f.read().splitlines() 13 | return layer_names 14 | 15 | def get_all_layer_names(): 16 | layer_names_file = os.path.join(dir, 'conv_layers', 'all_layers.txt') 17 | with open(layer_names_file, "r") as f: 18 | layer_names = f.read().splitlines() 19 | return layer_names 20 | 21 | -------------------------------------------------------------------------------- /Models/TensorFlow/models/MobileNet/model.py: -------------------------------------------------------------------------------- 1 | import tensorflow as tf 2 | import os 3 | 4 | dir = os.path.dirname(__file__) 5 | 6 | def get_model(): 7 | return tf.keras.applications.MobileNetV2(weights='imagenet') 8 | 9 | def get_layer_names(): 10 | layer_names_file = os.path.join(dir, 'conv_layers', 'conv_layer_names.txt') 11 | with open(layer_names_file, "r") as f: 12 | layer_names = f.read().splitlines() 13 | return layer_names 14 | 15 | def get_all_layer_names(): 16 | layer_names_file = os.path.join(dir, 'conv_layers', 'all_layers.txt') 17 | with open(layer_names_file, "r") as f: 18 | layer_names = f.read().splitlines() 19 | return layer_names 20 | 21 | -------------------------------------------------------------------------------- /tsconfig.json: -------------------------------------------------------------------------------- 1 | { 2 | "compilerOptions": { 3 | "lib": ["dom", "dom.iterable", "esnext"], 4 | "allowJs": true, 5 | "skipLibCheck": true, 6 | "strict": true, 7 | "noEmit": true, 8 | "esModuleInterop": true, 9 | "module": "esnext", 10 | "moduleResolution": "bundler", 11 | "resolveJsonModule": true, 12 | "isolatedModules": true, 13 | "jsx": "preserve", 14 | "incremental": true, 15 | "plugins": [ 16 | { 17 | "name": "next" 18 | } 19 | ], 20 | "paths": { 21 | "@/*": ["./src/*"] 22 | } 23 | }, 24 | "include": ["next-env.d.ts", "**/*.ts", "**/*.tsx", ".next/types/**/*.ts"], 25 | "exclude": ["node_modules"] 26 | } 27 | -------------------------------------------------------------------------------- /tailwind.config.js: -------------------------------------------------------------------------------- 1 | /** @type {import('tailwindcss').Config} */ 2 | module.exports = { 3 | content: [ 4 | "./src/pages/**/*.{js,ts,jsx,tsx,mdx}", 5 | "./src/components/**/*.{js,ts,jsx,tsx,mdx}", 6 | "./src/app/**/*.{js,ts,jsx,tsx,mdx}", 7 | 'node_modules/preline/dist/*.js', 8 | ], 9 | theme: { 10 | extend: { 11 | colors: { 12 | palette: { 13 | light: '#F5F3FF', 14 | primary: '#7C3AED', 15 | dark: '#6D28D9', 16 | }, 17 | }, 18 | fontFamily: { 19 | primary: ['Poppins'], 20 | secondary: ['Open Sans'], 21 | saira: ['Saira'], 22 | }, 23 | }, 24 | }, 25 | variants: { 26 | extend: {}, 27 | }, 28 | plugins: [ 29 | require('preline/plugin'), 30 | ], 31 | } 32 | 33 | -------------------------------------------------------------------------------- /Models/TensorFlow/models/MobileNet/conv_layers/convscript.py: -------------------------------------------------------------------------------- 1 | import tensorflow as tf 2 | 3 | # Load MobileNetV2 model 4 | model = tf.keras.applications.MobileNetV2(input_shape=(224, 224, 3), 5 | include_top=True, 6 | weights='imagenet') 7 | 8 | # Specify the file path to save the layer names 9 | file_path = '/Users/charlesschubach/Development/codesmith/OSP/xflairRP/python/conv_layer_names.txt' 10 | 11 | # Open a file in write mode 12 | with open(file_path, 'w') as file: 13 | # Iterate through the model's layers 14 | for layer in model.layers: 15 | # Write the layer name followed by a newline character to the file 16 | file.write(layer.name + '\n') 17 | 18 | print(f"Layer names saved to {file_path}") -------------------------------------------------------------------------------- /Models/TensorBoard/models/MobileNet/conv_layers/convscript.py: -------------------------------------------------------------------------------- 1 | import tensorflow as tf 2 | 3 | # Load MobileNetV2 model 4 | model = tf.keras.applications.MobileNetV2(input_shape=(224, 224, 3), 5 | include_top=True, 6 | weights='imagenet') 7 | 8 | # Specify the file path to save the layer names 9 | file_path = '/Users/charlesschubach/Development/codesmith/OSP/xflairRP/python/conv_layer_names.txt' 10 | 11 | # Open a file in write mode 12 | with open(file_path, 'w') as file: 13 | # Iterate through the model's layers 14 | for layer in model.layers: 15 | # Write the layer name followed by a newline character to the file 16 | file.write(layer.name + '\n') 17 | 18 | print(f"Layer names saved to {file_path}") -------------------------------------------------------------------------------- /scripts/newHeatmap.ts: -------------------------------------------------------------------------------- 1 | const fs = require('fs'); 2 | 3 | const numRows = 224; 4 | const numCols = 224; 5 | 6 | // Create an array to hold the CSV rows 7 | const csvRows = []; 8 | 9 | // Add the CSV header 10 | csvRows.push('row,column,value'); 11 | 12 | // Generate random values and populate the CSV rows 13 | for (let row = 0; row < numRows; row++) { 14 | for (let col = 0; col < numCols; col++) { 15 | const value = Math.floor(Math.random() * 100) + 1; // Random value between 1 and 100 16 | csvRows.push(`r${row},c${col},${value}`); 17 | } 18 | } 19 | 20 | // Join rows with newline characters 21 | const csvContent = csvRows.join('\n'); 22 | 23 | // Write CSV content to a file 24 | fs.writeFileSync('heatmap_data.csv', csvContent); 25 | 26 | console.log('Heatmap data saved to heatmap_data.csv'); -------------------------------------------------------------------------------- /src/app/(routes)/models/actions/logs/route.ts: -------------------------------------------------------------------------------- 1 | import { type NextRequest } from 'next/server'; 2 | 3 | const serviceHost = process.env.SERVICE_HOST || 'localhost'; // Default to localhost if not set 4 | const serviceUrl = `http://${serviceHost}:5000`; 5 | 6 | export async function POST(req: NextRequest) { 7 | try { 8 | const reqParsed = await req.json(); 9 | const { data, modelName } = reqParsed; 10 | const response = await fetch(`${serviceUrl}/logs/${modelName}`, { 11 | method: 'POST', 12 | headers: { 13 | 'Content-Type': 'application/json', 14 | }, 15 | body: JSON.stringify({ data }), 16 | }); 17 | return response; 18 | } catch (err) { 19 | console.error('Error: ', err); 20 | return new Response('Internal Server Error', { status: 500 }); 21 | } 22 | } 23 | -------------------------------------------------------------------------------- /src/app/(routes)/models/actions/image/heatmaps/route.ts: -------------------------------------------------------------------------------- 1 | import { type NextRequest } from 'next/server'; 2 | 3 | const serviceHost = process.env.SERVICE_HOST || 'localhost'; // Default to localhost if not set 4 | const serviceUrl = `http://${serviceHost}:5000`; 5 | 6 | export async function POST(req: NextRequest) { 7 | try { 8 | const reqParsed = await req.json(); 9 | const { data, modelName } = reqParsed; 10 | const response = await fetch(`${serviceUrl}/heatmaps/${modelName}`, { 11 | method: 'POST', 12 | headers: { 13 | 'Content-Type': 'application/json', 14 | }, 15 | body: JSON.stringify({ data }), 16 | }); 17 | return response; 18 | } catch (err) { 19 | console.error('Error: ', err); 20 | return new Response('Internal Server Error', { status: 500 }); 21 | } 22 | } 23 | -------------------------------------------------------------------------------- /src/app/(routes)/models/actions/image/featmaps/route.ts: -------------------------------------------------------------------------------- 1 | import { type NextRequest } from 'next/server'; 2 | 3 | const serviceHost = process.env.SERVICE_HOST || 'localhost'; // Default to localhost if not set 4 | const serviceUrl = `http://${serviceHost}:5000`; 5 | 6 | export async function POST(req: NextRequest) { 7 | try { 8 | const reqParsed = await req.json(); 9 | const { data, modelName } = reqParsed; 10 | const response = await fetch(`${serviceUrl}/featuremaps/${modelName}`, { 11 | method: 'POST', 12 | headers: { 13 | 'Content-Type': 'application/json', 14 | }, 15 | body: JSON.stringify({ data }), 16 | }); 17 | return response; 18 | } catch (err) { 19 | console.error('Error: ', err); 20 | return new Response('Internal Server Error', { status: 500 }); 21 | } 22 | } 23 | -------------------------------------------------------------------------------- /src/app/(routes)/models/actions/image/predict/route.ts: -------------------------------------------------------------------------------- 1 | import { type NextRequest } from 'next/server'; 2 | 3 | const serviceHost = process.env.SERVICE_HOST || 'localhost'; // Default to localhost if not set 4 | const serviceUrl = `http://${serviceHost}:5000`; 5 | 6 | export async function POST(req: NextRequest) { 7 | try { 8 | const reqParsed = await req.json(); 9 | const { data, modelName } = reqParsed; 10 | 11 | const response = await fetch(`${serviceUrl}/predictions/${modelName}`, { 12 | method: 'POST', 13 | headers: { 14 | 'Content-Type': 'application/json', 15 | }, 16 | body: JSON.stringify({ data }), 17 | }); 18 | return response; 19 | } catch (err) { 20 | console.error('Error: ', err); 21 | return new Response('Internal Server Error', { status: 500 }); 22 | } 23 | } 24 | -------------------------------------------------------------------------------- /Models/TensorFlow/Dockerfile: -------------------------------------------------------------------------------- 1 | FROM python:3.12.2 2 | 3 | WORKDIR /TensorFlow 4 | 5 | COPY . /TensorFlow/ 6 | 7 | RUN apt-get update && apt-get install -y \ 8 | python3-pip \ 9 | python3-dev \ 10 | build-essential \ 11 | libssl-dev \ 12 | libffi-dev \ 13 | python3-setuptools \ 14 | libgl1-mesa-dev \ 15 | libhdf5-dev \ 16 | && apt-get clean 17 | 18 | RUN pip install --upgrade pip 19 | 20 | # Install any needed packages specified in requirements.txt 21 | RUN pip install --no-cache-dir -r requirements.txt 22 | 23 | RUN chmod +x /TensorFlow/start/start.sh 24 | 25 | RUN chmod +x /TensorFlow/start/get_weights.py 26 | 27 | # Make port 5000 available to the world outside this container 28 | EXPOSE 5000 29 | 30 | # Run server.py when the container launches 31 | ENTRYPOINT ["/TensorFlow/start/start.sh"] 32 | 33 | 34 | -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | # See https://help.github.com/articles/ignoring-files/ for more about ignoring files. 2 | 3 | # dependencies 4 | /node_modules 5 | /.pnp 6 | .pnp.js 7 | .yarn/install-state.gz 8 | 9 | # testing 10 | /coverage 11 | 12 | # next.js 13 | /.next/ 14 | /out/ 15 | 16 | # production 17 | /build 18 | 19 | # misc 20 | .DS_Store 21 | *.pem 22 | 23 | # debug 24 | npm-debug.log* 25 | yarn-debug.log* 26 | yarn-error.log* 27 | 28 | # local env files 29 | .env*.local 30 | 31 | # vercel 32 | .vercel 33 | 34 | # typescript 35 | *.tsbuildinfo 36 | next-env.d.ts 37 | 38 | # venv 39 | venv/ 40 | 41 | # env 42 | /Models/TensorFlow/.env 43 | /Models/TensorBoard/logs/* 44 | /Models/TensorFlow/app/assets/* 45 | /Models/TensorBoard/logs/* 46 | /Models/TensorFLow/logs/* 47 | .env 48 | .vscode 49 | /public/renderTest 50 | /public/RP Data 51 | /public/heatmap_data.csv 52 | 53 | -------------------------------------------------------------------------------- /Models/TensorBoard/Dockerfile: -------------------------------------------------------------------------------- 1 | FROM python:latest 2 | 3 | WORKDIR /TensorBoard 4 | 5 | COPY . /TensorBoard/ 6 | 7 | RUN apt-get update && apt-get install -y \ 8 | python3-pip \ 9 | python3-dev \ 10 | build-essential \ 11 | libssl-dev \ 12 | libffi-dev \ 13 | python3-setuptools \ 14 | libgl1-mesa-dev \ 15 | libhdf5-dev \ 16 | && apt-get clean 17 | 18 | RUN pip install --upgrade pip 19 | 20 | # Install any needed packages specified in requirements.txt 21 | RUN pip install --no-cache-dir -r requirements.txt 22 | 23 | RUN chmod +x /TensorBoard/start/start.sh 24 | 25 | RUN chmod +x /TensorBoard/start/get_weights.py 26 | 27 | # Make port 5000 available to the world outside this container 28 | EXPOSE 5000 29 | 30 | EXPOSE 6006 31 | 32 | # Run server.py when the container launches 33 | ENTRYPOINT ["/TensorBoard/start/start.sh"] 34 | 35 | 36 | -------------------------------------------------------------------------------- /Models/TensorBoard/app/services/generate_predictions.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import tensorflow as tf 3 | from app.utils.image_utils import preprocess_image 4 | 5 | 6 | def generate_predictions(model, base64_image): 7 | preprocessed_image = preprocess_image(base64_image) 8 | 9 | # Make predictions using the provided model 10 | predictions = model.predict(preprocessed_image) 11 | predicted_class = np.argmax(predictions[0]).item() # Convert to Python int for JSON serialization 12 | 13 | # Get the top 5 predictions and their probabilities 14 | top_5 = tf.keras.applications.mobilenet_v2.decode_predictions(predictions, top=5)[0] 15 | 16 | # Return a dictionary that's directly serializable to JSON 17 | return { 18 | "predicted_class": predicted_class, 19 | "top_5": top_5, 20 | "preprocessed_image": preprocessed_image 21 | } -------------------------------------------------------------------------------- /src/app/ui/gridMapGif.tsx: -------------------------------------------------------------------------------- 1 | import styles from "../page.module.css"; 2 | import Image from "next/image"; 3 | import { useState, useEffect } from "react"; 4 | 5 | interface Props { 6 | fGifURL: string; 7 | } 8 | 9 | export default function GridMapGif(props: Props) { 10 | const [fGifListener, setFGifListener] = useState(false); 11 | 12 | useEffect(() => { 13 | if (props.fGifURL) setFGifListener(true); 14 | }, [props.fGifURL]); 15 | 16 | return ( 17 | <> 18 | {!fGifListener && loading} 19 | {fGifListener && ( 20 | 21 | 27 | 28 | )} 29 | 30 | ); 31 | } 32 | -------------------------------------------------------------------------------- /src/app/ui/navComponents/AccordionLink.tsx: -------------------------------------------------------------------------------- 1 | import Link from 'next/link'; 2 | 3 | interface Props { 4 | path: string; 5 | label: string; 6 | } 7 | 8 | export default function AccordionLink(props: Props) { 9 | return ( 10 |
  • 11 | 19 | {props.label} 20 | 21 |
    • 22 | ); 23 | } 24 | -------------------------------------------------------------------------------- /src/app/ui/heatMapGif.tsx: -------------------------------------------------------------------------------- 1 | import { image } from 'd3'; 2 | import styles from '../page.module.css'; 3 | import Image from 'next/image'; 4 | import { useState, useEffect } from 'react'; 5 | 6 | interface Props { 7 | hGifURL: string; 8 | } 9 | 10 | export default function HeatMapGif(props: Props) { 11 | const [hGifListener, setHGifListener] = useState(false); 12 | 13 | 14 | useEffect(() => { 15 | if(props.hGifURL)setHGifListener(true) 16 | },[props.hGifURL]) 17 | 18 | return ( 19 | <> 20 | {(!hGifListener && 21 | loading)} 22 | {hGifListener && 23 | ( 24 | 25 | 31 | )} 32 | 33 | ); 34 | } -------------------------------------------------------------------------------- /src/app/(routes)/about-us/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } -------------------------------------------------------------------------------- /src/app/(routes)/libraries/python/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } -------------------------------------------------------------------------------- /src/app/(routes)/documentation/python/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } -------------------------------------------------------------------------------- /src/app/(routes)/libraries/react/javascript/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } -------------------------------------------------------------------------------- /src/app/(routes)/libraries/react/typescript/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } -------------------------------------------------------------------------------- /src/app/(routes)/models/agnostic/text/gpt/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/pytorch/text/gpt/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/tensorflow/text/gpt/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/documentation/javascript-typescript/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } -------------------------------------------------------------------------------- /src/app/(routes)/models/pytorch/images/detection/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/agnostic/images/detection/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/agnostic/images/generation/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/agnostic/text/texttospeech/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/pytorch/images/classification/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/pytorch/images/generation/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/pytorch/text/texttospeech/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/tensorflow/images/detection/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/tensorflow/images/generation/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/tensorflow/text/texttospeech/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /src/app/(routes)/models/agnostic/images/classification/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import Link from "next/link"; 4 | 5 | export default async function Home() { 6 | return ( 7 |
    8 | titleText 13 |

    14 | Coming Soon 15 |

    16 |

    17 | Feature still in progress. We apologize for the inconvenience. 18 |

    19 | 23 | Return to Homepage 24 | 25 |
    26 | ); 27 | } 28 | -------------------------------------------------------------------------------- /Models/TensorFlow/app/routes/predictions.py: -------------------------------------------------------------------------------- 1 | import os 2 | from flask import Blueprint, request, jsonify 3 | import time 4 | 5 | from app.services.generate_predictions import generate_predictions 6 | 7 | 8 | dir = os.path.dirname(__file__) 9 | 10 | predictions = Blueprint('predictions', __name__) 11 | 12 | @predictions.route('/predictions/', methods=['POST']) 13 | def upload_predictions(model_name): 14 | 15 | data = request.get_json() 16 | base64_image = data['data'] 17 | 18 | model_module = __import__(f"models.{model_name}.model", fromlist=["get_model"]) 19 | model = model_module.get_model() 20 | 21 | start_time = time.time() 22 | 23 | predictions = generate_predictions(model, base64_image) 24 | 25 | end_time = time.time() 26 | 27 | total_time = end_time - start_time 28 | 29 | print(f"Predictions: {predictions}") 30 | print(f"Total time taken: {total_time}") 31 | 32 | return jsonify({ 33 | 'predictions': predictions, 34 | 'time': total_time 35 | }) 36 | 37 | -------------------------------------------------------------------------------- /Models/TensorBoard/app/routes/logs.py: -------------------------------------------------------------------------------- 1 | from flask import Blueprint, request, jsonify 2 | import os 3 | 4 | from app.services.generate_logs import handle_tensorboard_logging 5 | 6 | dir = os.path.dirname(__file__) 7 | logs = Blueprint('logs', __name__) 8 | 9 | @logs.route('/logs/', methods=['POST']) 10 | def tensorboard_logs(model_name): 11 | data = request.get_json() 12 | base64_image = data['data'] 13 | 14 | # Invoke the get_model function from /model 15 | model_module = __import__(f"models.{model_name}.model", fromlist=["get_model"]) 16 | model = model_module.get_model() 17 | layer_names_module = __import__(f"models.{model_name}.model", fromlist=["get_all_layer_names"]) 18 | layer_names = layer_names_module.get_all_layer_names() 19 | 20 | # Generate TensorBoard logs and obtain the log directory 21 | log_dir = handle_tensorboard_logging(model, base64_image, layer_names, model_name) 22 | # model_summary(model, log_dir) 23 | 24 | return jsonify({"message": "TensorBoard logs generated successfully.", "log_dir": log_dir}) 25 | 26 | -------------------------------------------------------------------------------- /src/app/ui/modal.tsx: -------------------------------------------------------------------------------- 1 | import { Top5Obj } from '../lib/definitions'; 2 | import styles from '../page.module.css'; 3 | import { EmblaCarousel } from './carousel'; 4 | 5 | interface Props { 6 | closeViz: () => void; 7 | hGifURL: string; 8 | fGifURL: string; 9 | top5: Top5Obj; 10 | preprocessFilePath: string; 11 | } 12 | 13 | export default function Modal(props: Props) { 14 | return ( 15 |
    16 |
    17 |
    18 |

    Analysis

    19 |

    20 | X 21 |

    22 |
    23 | 29 |
    30 | 33 |
    34 |
    35 | ); 36 | } 37 | -------------------------------------------------------------------------------- /src/app/Sidebar.module.css: -------------------------------------------------------------------------------- 1 | /* ./ui/Sidebar.module.css */ 2 | 3 | .sidebar { 4 | background-color: black; 5 | /* border: 1px solid white; */ 6 | color: white; 7 | min-height: 100vh; 8 | width: 200px; 9 | display: flex; 10 | flex-direction: column; 11 | padding: 5px; 12 | padding-top: 30px; 13 | } 14 | 15 | .sideBarLogo { 16 | margin-bottom: 30px; 17 | text-align: center; 18 | font-size: 24px; 19 | font-weight: bold; 20 | } 21 | 22 | .menuItem { 23 | border: 1px solid white; 24 | margin: 10px 0px; 25 | padding: 10px; 26 | padding-left: 30px; 27 | cursor: pointer; 28 | border-radius: 1px; 29 | transition: background-color 0.3s ease; 30 | } 31 | 32 | .menuItem:hover { 33 | background-color: #34495e; 34 | } 35 | 36 | .footer { 37 | margin-top: auto; 38 | text-align: center; 39 | font-size: 14px; 40 | } 41 | 42 | .sideBarTitle{ 43 | height: 50px; 44 | width: 75px; 45 | } 46 | 47 | .sideBarLogoContainer { 48 | display: flex; 49 | flex-direction: row; 50 | align-items: center; 51 | justify-content: center; 52 | margin-bottom: 30px; 53 | } 54 | -------------------------------------------------------------------------------- /package.json: -------------------------------------------------------------------------------- 1 | { 2 | "name": "explai", 3 | "version": "0.1.0", 4 | "private": true, 5 | "scripts": { 6 | "dev": "next dev", 7 | "build": "next build", 8 | "start": "next start", 9 | "lint": "next lint" 10 | }, 11 | "dependencies": { 12 | "@aws-sdk/client-s3": "^3.540.0", 13 | "@aws-sdk/s3-request-presigner": "^3.540.0", 14 | "@preline/overlay": "^2.0.2", 15 | "@vercel/postgres": "^0.7.2", 16 | "aws-sdk": "^2.1591.0", 17 | "axios": "^1.6.8", 18 | "cloudinary": "^2.1.0", 19 | "cors": "^2.8.5", 20 | "d3": "^7.9.0", 21 | "dotenv": "^16.4.5", 22 | "embla-carousel-react": "^8.0.0", 23 | "next": "14.1.3", 24 | "preline": "^2.0.3", 25 | "react": "^18", 26 | "react-dom": "^18", 27 | "vercel": "^33.5.5", 28 | "zod": "^3.22.4" 29 | }, 30 | "devDependencies": { 31 | "@types/d3": "^7.4.3", 32 | "@types/node": "^20", 33 | "@types/react": "^18", 34 | "@types/react-dom": "^18", 35 | "autoprefixer": "^10.4.19", 36 | "eslint": "^8", 37 | "eslint-config-next": "14.1.3", 38 | "postcss": "^8.4.38", 39 | "tailwindcss": "^3.4.3", 40 | "typescript": "^5" 41 | } 42 | } 43 | -------------------------------------------------------------------------------- /src/app/ui/top5Classes.tsx: -------------------------------------------------------------------------------- 1 | import { useEffect, useRef, useState } from 'react'; 2 | import styles from '../page.module.css'; 3 | import { Top5Obj } from '../lib/definitions'; 4 | 5 | interface Props { 6 | top5: Top5Obj; 7 | } 8 | 9 | export default function Top5(props: Props) { 10 | const top5Table = []; 11 | 12 | // Convert object into an array of [key, value] pairs 13 | const sortedEntries = Object.entries(props.top5) 14 | // Sort the array by value in descending order 15 | .sort((a, b) => b[1] - a[1]); 16 | 17 | // Iterate over sorted entries to generate table rows 18 | for (let [key, value] of sortedEntries) { 19 | top5Table.push( 20 | 21 | {key} 22 | {(value * 100).toFixed(2)}% 23 | 24 | ); 25 | } 26 | 27 | return ( 28 |
    29 | {sortedEntries.length > 0 && ( 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | {top5Table} 38 |
    ClassCertainty
    39 | )} 40 |
    41 | ); 42 | } 43 | -------------------------------------------------------------------------------- /Models/TensorFlow/requirements.txt: -------------------------------------------------------------------------------- 1 | absl-py==2.1.0 2 | astunparse==1.6.3 3 | blinker==1.7.0 4 | boto3==1.34.78 5 | botocore==1.34.78 6 | certifi==2024.2.2 7 | charset-normalizer==3.3.2 8 | click==8.1.7 9 | contourpy==1.2.1 10 | cycler==0.12.1 11 | Flask==3.0.2 12 | Flask-Cors==4.0.0 13 | flatbuffers==24.3.25 14 | fonttools==4.50.0 15 | gast==0.5.4 16 | google-pasta==0.2.0 17 | grpcio==1.62.1 18 | gunicorn==21.2.0 19 | h5py==3.10.0 20 | idna==3.6 21 | itsdangerous==2.1.2 22 | Jinja2==3.1.3 23 | jmespath==1.0.1 24 | keras==3.1.1 25 | kiwisolver==1.4.5 26 | libclang==18.1.1 27 | Markdown==3.6 28 | markdown-it-py==3.0.0 29 | MarkupSafe==2.1.5 30 | matplotlib==3.8.4 31 | mdurl==0.1.2 32 | ml-dtypes==0.3.2 33 | namex==0.0.7 34 | numpy==1.26.4 35 | opencv-python==4.9.0.80 36 | opt-einsum==3.3.0 37 | optree==0.11.0 38 | packaging==24.0 39 | pillow==10.3.0 40 | protobuf==4.25.3 41 | Pygments==2.17.2 42 | pyparsing==3.1.2 43 | python-dateutil==2.9.0.post0 44 | requests==2.31.0 45 | rich==13.7.1 46 | s3transfer==0.10.1 47 | setuptools==69.1.1 48 | six==1.16.0 49 | tensorboard==2.16.2 50 | tensorboard-data-server==0.7.2 51 | tensorflow==2.16.1 52 | termcolor==2.4.0 53 | typing_extensions==4.10.0 54 | urllib3==2.2.1 55 | Werkzeug==3.0.2 56 | wheel==0.43.0 57 | wrapt==1.16.0 -------------------------------------------------------------------------------- /public/next.svg: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Models/TensorFlow/app/utils/image_utils.py: -------------------------------------------------------------------------------- 1 | import base64 2 | import tensorflow as tf 3 | import numpy as np 4 | 5 | def preprocess_image(base64_image_str): 6 | image_bytes = base64.b64decode(base64_image_str) 7 | image_tensor = tf.image.decode_image(image_bytes, channels=3) 8 | image_tensor = tf.image.resize(image_tensor, (224, 224)) 9 | image_tensor = tf.keras.applications.mobilenet_v2.preprocess_input(image_tensor) 10 | image_tensor = tf.expand_dims(image_tensor, 0) 11 | return image_tensor 12 | 13 | def add_progress_bar(image, current_index, total_layers, width, border_color=(255,255,255), progress_color=(50,205,50), remaining_color=(220,220,220)): 14 | 15 | img_height, img_width = image.shape[:2] 16 | progress_width = int(((current_index + 1) / total_layers) * img_width) 17 | 18 | # Create a progress bar image 19 | bar_img = np.zeros((width, img_width, 3), dtype=np.uint8) 20 | 21 | # Fill the progress part 22 | bar_img[:, :progress_width] = progress_color 23 | 24 | # Fill the remaining part 25 | bar_img[:, progress_width:] = remaining_color 26 | 27 | # Add a border 28 | bar_img[0, :] = border_color 29 | bar_img[-1, :] = border_color 30 | bar_img[:, 0] = border_color 31 | bar_img[:, -1] = border_color 32 | 33 | # Combine the progress bar with the original image 34 | combined_img = np.vstack((bar_img, image)) 35 | 36 | return combined_img -------------------------------------------------------------------------------- /Models/TensorFlow/app/services/generate_predictions.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import tensorflow as tf 3 | from app.utils.image_utils import preprocess_image 4 | 5 | 6 | def generate_predictions(model, base64_image): 7 | preprocessed_image = preprocess_image(base64_image) 8 | 9 | # Make predictions using the provided model 10 | predictions = model.predict(preprocessed_image) 11 | predicted_class = np.argmax(predictions[0]).item() # Convert to Python int for JSON serialization 12 | 13 | # Get the name of the predicted class 14 | predicted_class_name = tf.keras.applications.mobilenet_v2.decode_predictions(predictions, top=1)[0][0][1] 15 | 16 | # Get the top 5 predictions and their probabilities 17 | top_5 = tf.keras.applications.mobilenet_v2.decode_predictions(predictions, top=5)[0] 18 | class_name_probabilities = {} 19 | 20 | for _, class_name, probability in top_5: 21 | # Convert numpy float to Python float for JSON serialization 22 | class_name_probabilities[class_name] = float(probability) 23 | 24 | # Return a dictionary that's directly serializable to JSON 25 | return { 26 | "predicted_class_name": predicted_class_name, 27 | "predicted_class_probability": predictions[0][predicted_class].item(), # Ensure this is a Python float 28 | "class_name_probabilities": class_name_probabilities, 29 | "predicted_class": predicted_class 30 | } -------------------------------------------------------------------------------- /Models/TensorBoard/app/utils/image_utils.py: -------------------------------------------------------------------------------- 1 | import base64 2 | import tensorflow as tf 3 | import numpy as np 4 | 5 | def preprocess_image(base64_image_str): 6 | image_bytes = base64.b64decode(base64_image_str) 7 | image_tensor = tf.image.decode_image(image_bytes, channels=3) 8 | image_tensor = tf.image.resize(image_tensor, (224, 224)) 9 | image_tensor = tf.keras.applications.mobilenet_v2.preprocess_input(image_tensor) 10 | image_tensor = tf.expand_dims(image_tensor, 0) 11 | return image_tensor 12 | 13 | def add_progress_bar(image, current_index, total_layers, bar_height=5, border_color=(255,255,255), progress_color=(50,205,50), remaining_color=(220,220,220)): 14 | 15 | img_height, img_width = image.shape[:2] 16 | progress_width = int(((current_index + 1) / total_layers) * img_width) 17 | 18 | # Create a progress bar image 19 | bar_img = np.zeros((bar_height, img_width, 3), dtype=np.uint8) 20 | 21 | # Fill the progress part 22 | bar_img[:, :progress_width] = progress_color 23 | 24 | # Fill the remaining part 25 | bar_img[:, progress_width:] = remaining_color 26 | 27 | # Add a border 28 | bar_img[0, :] = border_color 29 | bar_img[-1, :] = border_color 30 | bar_img[:, 0] = border_color 31 | bar_img[:, -1] = border_color 32 | 33 | # Combine the progress bar with the original image 34 | combined_img = np.vstack((bar_img, image)) 35 | 36 | return combined_img -------------------------------------------------------------------------------- /Models/TensorBoard/models/MobileNet/conv_layers/conv_layer_names.txt: -------------------------------------------------------------------------------- 1 | Conv1 2 | bn_Conv1 3 | Conv1_relu 4 | expanded_conv_depthwise 5 | expanded_conv_depthwise_relu 6 | expanded_conv_project 7 | block_1_expand 8 | block_1_expand_BN 9 | block_1_expand_relu 10 | block_1_pad 11 | block_1_depthwise 12 | block_1_depthwise_BN 13 | block_1_depthwise_relu 14 | block_1_project 15 | block_1_project_BN 16 | block_2_expand 17 | block_2_expand_BN 18 | block_2_expand_relu 19 | block_2_depthwise 20 | block_2_depthwise_BN 21 | block_2_depthwise_relu 22 | block_2_project 23 | block_2_project_BN 24 | block_3_expand 25 | block_3_depthwise_relu 26 | block_3_project 27 | block_4_expand 28 | block_4_depthwise_relu 29 | block_4_project 30 | block_5_expand 31 | block_5_depthwise_relu 32 | block_5_project 33 | block_6_expand 34 | block_6_depthwise_relu 35 | block_6_project 36 | block_7_expand 37 | block_7_depthwise_relu 38 | block_7_project 39 | block_8_expand 40 | block_8_depthwise_relu 41 | block_8_project 42 | block_9_expand 43 | block_9_depthwise_relu 44 | block_9_project 45 | block_10_expand 46 | block_10_depthwise_relu 47 | block_10_project 48 | block_11_expand 49 | block_11_depthwise_relu 50 | block_11_project 51 | block_12_expand 52 | block_12_depthwise_relu 53 | block_12_project 54 | block_13_expand 55 | block_13_depthwise_relu 56 | block_13_project 57 | block_16_expand 58 | block_16_depthwise_relu 59 | block_16_project 60 | Conv_1 61 | Conv_1_bn 62 | out_relu 63 | -------------------------------------------------------------------------------- /Models/TensorFlow/models/MobileNet/conv_layers/conv_layer_names.txt: -------------------------------------------------------------------------------- 1 | Conv1 2 | bn_Conv1 3 | Conv1_relu 4 | expanded_conv_depthwise 5 | expanded_conv_depthwise_relu 6 | expanded_conv_project 7 | block_1_expand 8 | block_1_expand_BN 9 | block_1_expand_relu 10 | block_1_pad 11 | block_1_depthwise 12 | block_1_depthwise_BN 13 | block_1_depthwise_relu 14 | block_1_project 15 | block_1_project_BN 16 | block_2_expand 17 | block_2_expand_BN 18 | block_2_expand_relu 19 | block_2_depthwise 20 | block_2_depthwise_BN 21 | block_2_depthwise_relu 22 | block_2_project 23 | block_2_project_BN 24 | block_3_expand 25 | block_3_depthwise_relu 26 | block_3_project 27 | block_4_expand 28 | block_4_depthwise_relu 29 | block_4_project 30 | block_5_expand 31 | block_5_depthwise_relu 32 | block_5_project 33 | block_6_expand 34 | block_6_depthwise_relu 35 | block_6_project 36 | block_7_expand 37 | block_7_depthwise_relu 38 | block_7_project 39 | block_8_expand 40 | block_8_depthwise_relu 41 | block_8_project 42 | block_9_expand 43 | block_9_depthwise_relu 44 | block_9_project 45 | block_10_expand 46 | block_10_depthwise_relu 47 | block_10_project 48 | block_11_expand 49 | block_11_depthwise_relu 50 | block_11_project 51 | block_12_expand 52 | block_12_depthwise_relu 53 | block_12_project 54 | block_13_expand 55 | block_13_depthwise_relu 56 | block_13_project 57 | block_16_expand 58 | block_16_depthwise_relu 59 | block_16_project 60 | Conv_1 61 | Conv_1_bn 62 | out_relu 63 | -------------------------------------------------------------------------------- /src/app/page.tsx: -------------------------------------------------------------------------------- 1 | 'use server'; 2 | 3 | import './globals.css'; 4 | import Link from 'next/link'; 5 | 6 | async function StartPage() { 7 | return ( 8 |
    9 | main background image 14 |
    15 |
    16 |
    17 |

    18 | Enhance the user experience with AI integrations using{' '} 19 | xFlair. 20 |

    21 |

    22 | Provide meaningful, digestible, and real-time visualizations for AI models, from input to output. 23 |

    24 |
    25 | 29 | Start 30 | 31 |
    32 |
    33 |
    34 | ); 35 | } 36 | 37 | export default StartPage; 38 | -------------------------------------------------------------------------------- /scripts/newRenderTest.ts: -------------------------------------------------------------------------------- 1 | const file = require('fs'); 2 | const path = require('path'); 3 | 4 | const x = 50; 5 | const y = 50; 6 | const rate = 4; 7 | let dataset = 1; 8 | 9 | interface GridValues { 10 | [index: string]: number; 11 | } 12 | 13 | const final: GridValues = {}; 14 | 15 | for (let row = 1; row <= x; row++) { 16 | for (let col = 1; col <= y; col++) { 17 | const distance = Math.max(row, col) - 1; 18 | const value = 100 - distance * 2; 19 | final[`r${row},c${col}`] = value; 20 | } 21 | } 22 | 23 | // file.writeFileSync( 24 | // path.join(__dirname, `../public/renderTest/dataFINAL.csv`), 25 | // final.join('\n') 26 | // ); 27 | 28 | let contents: Array = []; 29 | 30 | for (let row = 1; row <= x; row++) { 31 | for (let col = 1; col <= y; col++) { 32 | const value = Math.floor(Math.random() * 100) + 1; 33 | contents.push(`r${row},c${col},${value}`); 34 | } 35 | } 36 | 37 | file.writeFileSync( 38 | path.join(__dirname, `../public/renderTest/data${dataset}.csv`), 39 | 'row,column,value\n' + contents.join('\n') 40 | ); 41 | dataset++; 42 | 43 | for (let i = 2; i <= 30; i++) { 44 | let temp: Array = []; 45 | 46 | contents.forEach((el) => { 47 | const data: Array = el.split(','); 48 | let value: number = Number(data.pop() as string); 49 | const key: string = data.join(','); 50 | 51 | if (Math.abs(final[key] - value) < 5) value = final[key]; 52 | else if (final[key] > value) value += rate; 53 | else if (final[key] < value) value -= rate; 54 | 55 | temp.push(`${key},${value}`); 56 | }); 57 | 58 | contents = temp; 59 | 60 | file.writeFileSync( 61 | path.join(__dirname, `../public/renderTest/data${dataset}.csv`), 62 | 'row,column,value\n' + contents.join('\n') 63 | ); 64 | dataset++; 65 | } 66 | 67 | console.log('done'); 68 | -------------------------------------------------------------------------------- /Models/TensorFlow/app/utils/s3_utils.py: -------------------------------------------------------------------------------- 1 | import boto3 2 | from botocore.exceptions import ClientError, NoCredentialsError, BotoCoreError 3 | import logging 4 | import time 5 | 6 | def create_s3_client(): 7 | """Create an S3 client.""" 8 | return boto3.client('s3') 9 | 10 | def upload_file(file_path, bucket_name, object_name): 11 | s3_client = create_s3_client() 12 | try: 13 | s3_client.upload_file(file_path, bucket_name, object_name) 14 | except ClientError as e: 15 | print(f"Error uploading file to S3: {e}") 16 | raise 17 | 18 | def download_file(bucket_name, object_name, file_path): 19 | """Download a file from an S3 bucket.""" 20 | s3_client = create_s3_client() 21 | try: 22 | s3_client.download_file(bucket_name, object_name, file_path) 23 | print(f"File {object_name} downloaded from {bucket_name} to {file_path}") 24 | except ClientError as e: 25 | print(f"Error downloading file from S3: {e}") 26 | raise 27 | 28 | def list_objects_in_bucket(bucket_name, prefix=""): 29 | """List objects in an S3 bucket.""" 30 | s3_client = create_s3_client() 31 | try: 32 | response = s3_client.list_objects_v2(Bucket=bucket_name, Prefix=prefix) 33 | objects = response.get("Contents", []) 34 | return [obj["Key"] for obj in objects] 35 | except ClientError as e: 36 | print(f"Error listing objects in bucket {bucket_name}: {e}") 37 | raise 38 | 39 | def delete_object_from_bucket(bucket_name, object_name): 40 | """Delete an object from an S3 bucket.""" 41 | s3_client = create_s3_client() 42 | try: 43 | s3_client.delete_object(Bucket=bucket_name, Key=object_name) 44 | print(f"Object {object_name} deleted from {bucket_name}") 45 | except ClientError as e: 46 | print(f"Error deleting object from bucket {bucket_name}: {e}") 47 | raise -------------------------------------------------------------------------------- /src/app/ui/dotButton.tsx: -------------------------------------------------------------------------------- 1 | import React, { 2 | PropsWithChildren, 3 | useCallback, 4 | useEffect, 5 | useState 6 | } from 'react' 7 | import { EmblaCarouselType } from 'embla-carousel' 8 | import styles from '../page.module.css'; 9 | 10 | type UseDotButtonType = { 11 | selectedIndex: number 12 | scrollSnaps: number[] 13 | onDotButtonClick: (index: number) => void 14 | } 15 | 16 | export const useDotButton = ( 17 | emblaApi: EmblaCarouselType | undefined 18 | ): UseDotButtonType => { 19 | const [selectedIndex, setSelectedIndex] = useState(0) 20 | const [scrollSnaps, setScrollSnaps] = useState([]) 21 | 22 | const onDotButtonClick = useCallback( 23 | (index: number) => { 24 | if (!emblaApi) return 25 | emblaApi.scrollTo(index) 26 | }, 27 | [emblaApi] 28 | ) 29 | 30 | const onInit = useCallback((emblaApi: EmblaCarouselType) => { 31 | setScrollSnaps(emblaApi.scrollSnapList()) 32 | }, []) 33 | 34 | const onSelect = useCallback((emblaApi: EmblaCarouselType) => { 35 | setSelectedIndex(emblaApi.selectedScrollSnap()) 36 | }, []) 37 | 38 | useEffect(() => { 39 | if (!emblaApi) return 40 | 41 | onInit(emblaApi) 42 | onSelect(emblaApi) 43 | emblaApi.on('reInit', onInit) 44 | emblaApi.on('reInit', onSelect) 45 | emblaApi.on('select', onSelect) 46 | }, [emblaApi, onInit, onSelect]) 47 | 48 | return { 49 | selectedIndex, 50 | scrollSnaps, 51 | onDotButtonClick 52 | } 53 | } 54 | 55 | type PropType = PropsWithChildren< 56 | React.DetailedHTMLProps< 57 | React.ButtonHTMLAttributes, 58 | HTMLButtonElement 59 | > 60 | > 61 | 62 | export const DotButton: React.FC = (props) => { 63 | const { children, ...restProps } = props 64 | 65 | return ( 66 | 69 | ) 70 | } 71 | -------------------------------------------------------------------------------- /src/app/ui/heatmap.tsx: -------------------------------------------------------------------------------- 1 | import { useEffect, useRef, useState } from 'react'; 2 | import * as d3 from 'd3'; 3 | 4 | interface DataItem { 5 | group: string; 6 | variable: string; 7 | value: string; 8 | } 9 | 10 | export default function Heatmap() { 11 | const svgRef = useRef(null); 12 | 13 | const [containerSize, setContainerSize] = useState({ 14 | width: 650, 15 | height: 650, 16 | }); 17 | 18 | useEffect(() => { 19 | const numCells = { x: 224, y: 224 }; 20 | const margin = { top: 5, right: 5, bottom: 5, left: 5 }; 21 | const width = containerSize.width - margin.left - margin.right; 22 | const height = containerSize.height - margin.top - margin.bottom; 23 | 24 | const cellSize = { 25 | width: Math.min(width / numCells.x, height / numCells.y), 26 | height: Math.min(width / numCells.x, height / numCells.y), 27 | }; 28 | 29 | const svg = d3 30 | .select(svgRef.current) 31 | .attr('width', width + margin.left + margin.right) 32 | .attr('height', height + margin.top + margin.bottom) 33 | .append('g') 34 | .attr('transform', `translate(${margin.left}, ${margin.top})`); 35 | 36 | const myColor = d3 37 | .scaleLinear() 38 | .range(['#f5f0f0', '#f20505']) 39 | .domain([1, 100]); 40 | 41 | // d3.csv(`/renderTest/data${nextArray}.csv`).then((data) => { 42 | d3.csv(`/RP Data/gradients.csv`).then((data) => { 43 | svg 44 | .selectAll('rect') 45 | .data(data) 46 | .enter() 47 | .append('rect') 48 | .attr('x', (_, i) => (i % numCells.x) * cellSize.width) 49 | .attr('y', (_, i) => Math.floor(i / numCells.x) * cellSize.height) 50 | .attr('width', cellSize.width) 51 | .attr('height', cellSize.height) 52 | .style('fill', (d) => myColor(+d.value)); 53 | }); 54 | }); 55 | 56 | return ; 57 | } 58 | -------------------------------------------------------------------------------- /src/app/(routes)/models/actions/image/upload/route.ts: -------------------------------------------------------------------------------- 1 | import { NextRequest, NextResponse } from 'next/server'; 2 | import { S3Client, PutObjectCommand } from '@aws-sdk/client-s3'; 3 | 4 | export async function POST(req: NextRequest, res: NextResponse) { 5 | try { 6 | //Incoming request is a multipart/form-data with 'file' and 'modelName' fields 7 | //file is a File type and modelName is a string 8 | //help 9 | 10 | const formData = await req.formData(); 11 | 12 | const file = formData.get('file'); 13 | const modelName = formData.get('modelName'); 14 | 15 | if (!file || !modelName) { 16 | return new Response('Missing file or modelName', { status: 400 }); 17 | } 18 | 19 | const buffer = Buffer.from(await (file as File).arrayBuffer()); 20 | 21 | const fileName = `inputimages/${modelName}/${Date.now()}.${( 22 | file as File 23 | ).name 24 | .split('.') 25 | .pop()}`; 26 | 27 | const s3 = new S3Client({ 28 | region: process.env.AWS_DEFAULT_REGION as string, 29 | credentials: { 30 | accessKeyId: process.env.AWS_ACCESS_KEY_ID as string, 31 | secretAccessKey: process.env.AWS_SECRET_ACCESS_KEY as string, 32 | }, 33 | }); 34 | 35 | const uploadParams = { 36 | Bucket: process.env.AWS_S3_BUCKET_NAME as string, 37 | Key: fileName, 38 | Body: buffer, // Stream the file directly to S3 39 | ContentType: (file as File).type, // Set the correct content type for the file// Optional: Set ACL to control access permissions 40 | }; 41 | 42 | const command = new PutObjectCommand(uploadParams); 43 | 44 | await s3.send(command); 45 | 46 | const filePath = `https://${process.env.AWS_S3_BUCKET_NAME}.s3.amazonaws.com/${fileName}`; 47 | 48 | return new Response(JSON.stringify({ filePath }), { 49 | status: 200, 50 | }); 51 | } catch (err) { 52 | console.error('Error uploading image:', err); 53 | return new Response('Error uploading image', { status: 500 }); 54 | } 55 | } 56 | -------------------------------------------------------------------------------- /src/app/ui/navComponents/Icon.tsx: -------------------------------------------------------------------------------- 1 | interface Props { 2 | name: string; 3 | } 4 | 5 | export default function Icon({ name }: Props) { 6 | if (name === 'gear') { 7 | return ( 8 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | ); 33 | } 34 | if (name === 'book') { 35 | return ( 36 | 48 | 49 | 50 | 51 | ); 52 | } 53 | if (name === 'pages') { 54 | return ( 55 | 67 | 68 | 69 | 70 | 71 | ); 72 | } 73 | } 74 | -------------------------------------------------------------------------------- /src/app/(routes)/models/actions/image/preprocess/route.ts: -------------------------------------------------------------------------------- 1 | import { type NextRequest } from 'next/server'; 2 | import { S3Client, GetObjectCommand } from '@aws-sdk/client-s3'; 3 | import { getSignedUrl } from '@aws-sdk/s3-request-presigner'; 4 | 5 | const serviceHost = process.env.SERVICE_HOST || 'localhost'; // Default to localhost if not set 6 | const serviceUrl = `http://${serviceHost}:5000`; 7 | 8 | const s3 = new S3Client({ 9 | region: process.env.AWS_DEFAULT_REGION as string, 10 | credentials: { 11 | accessKeyId: process.env.AWS_ACCESS_KEY_ID as string, 12 | secretAccessKey: process.env.AWS_SECRET_ACCESS_KEY as string, 13 | }, 14 | }); 15 | 16 | const generateSignedUrl = async ( 17 | bucketName: string, 18 | objectKey: string, 19 | expiresInSeconds: number 20 | ) => { 21 | const command = new GetObjectCommand({ 22 | Bucket: bucketName, 23 | Key: objectKey, 24 | }); 25 | 26 | try { 27 | const signedUrl = await getSignedUrl(s3, command, { 28 | expiresIn: expiresInSeconds, 29 | }); 30 | return signedUrl; 31 | } catch (error) { 32 | console.error('Error generating signed URL:', error); 33 | throw error; // Rethrow the error for further handling 34 | } 35 | }; 36 | export async function POST(req: NextRequest) { 37 | try { 38 | const reqParsed = await req.json(); 39 | const { data, filePath, fileType } = reqParsed; 40 | 41 | // Call the preprocess endpoint 42 | const preprocessResponse = await fetch(`${serviceUrl}/preprocess`, { 43 | method: 'POST', 44 | headers: { 45 | 'Content-Type': 'application/json', 46 | }, 47 | body: JSON.stringify({ data, filePath, fileType }), 48 | }); 49 | const preprocessResult = await preprocessResponse.json(); 50 | 51 | const { preprocessed_images } = preprocessResult; 52 | const bucketName = process.env.AWS_S3_BUCKET_NAME as string; 53 | const expiresInSeconds = 60 * 5; // 5 minutes 54 | 55 | // Generate signed URLs for the preprocessed images 56 | const signedUrls = await Promise.all( 57 | preprocessed_images.map(async (image: string) => { 58 | return generateSignedUrl(bucketName, image, expiresInSeconds); 59 | }) 60 | ); 61 | 62 | // Return the signed URLs 63 | return new Response(JSON.stringify({ urls: signedUrls }), { status: 200 }); 64 | } catch (err) { 65 | console.error('Error:', err); 66 | return new Response('Internal Server Error', { status: 500 }); 67 | } 68 | } 69 | -------------------------------------------------------------------------------- /src/app/ui/carousel.tsx: -------------------------------------------------------------------------------- 1 | import React, { useEffect, useCallback, Suspense, lazy } from 'react'; 2 | import useEmblaCarousel, { 3 | type UseEmblaCarouselType, 4 | } from 'embla-carousel-react'; 5 | import styles from '../page.module.css'; 6 | import { DotButton, useDotButton } from './dotButton'; 7 | import ResizedImage from './resizedImage'; 8 | import HeatMapGif from './heatMapGif'; 9 | import GridMapGif from './gridMapGif'; 10 | import Top5Classes from './top5Classes'; 11 | import { Top5Obj } from '../lib/definitions'; 12 | 13 | interface Props { 14 | hGifURL: string; 15 | fGifURL: string; 16 | top5: Top5Obj; 17 | preprocessFilePath: string; 18 | } 19 | 20 | export function EmblaCarousel(props: Props) { 21 | const [emblaRef, emblaApi] = useEmblaCarousel({ loop: false }); 22 | 23 | useEffect(() => { 24 | if (emblaApi) { 25 | console.log(emblaApi.slideNodes()); // Access API 26 | } 27 | }, [emblaApi]); 28 | 29 | const { selectedIndex, scrollSnaps, onDotButtonClick } = 30 | useDotButton(emblaApi); 31 | 32 | const scrollPrev = useCallback(() => { 33 | if (emblaApi) emblaApi.scrollPrev(); 34 | }, [emblaApi]); 35 | const scrollNext = useCallback(() => { 36 | if (emblaApi) emblaApi.scrollNext(); 37 | }, [emblaApi]); 38 | 39 | return ( 40 |
    41 |
    42 |
    43 | 44 | 45 | 46 | 47 |
    48 |
    49 |
    50 | 53 |
    54 | {scrollSnaps.map((_, index) => { 55 | const style = 56 | index === selectedIndex ? 'embla__dot--selected' : 'embla__dot'; 57 | return ( 58 | onDotButtonClick(index)} 61 | className={styles[style]} 62 | /> 63 | ); 64 | })} 65 |
    66 | 69 |
    70 |
    71 | ); 72 | } 73 | -------------------------------------------------------------------------------- /Models/TensorBoard/app/services/generate_logs.py: -------------------------------------------------------------------------------- 1 | import tensorflow as tf 2 | import numpy as np 3 | import os 4 | from datetime import datetime\ 5 | 6 | from app.services.generate_images import make_gradcam_heatmap 7 | from app.services.generate_predictions import generate_predictions 8 | 9 | dir = os.path.dirname(__file__) 10 | 11 | def log_model_architecture(model, writer): 12 | dummy_data = tf.random.normal(shape=(1, 224, 224, 3)) 13 | 14 | # Wrap the model call with tf.function 15 | @tf.function 16 | def model_call(inputs): 17 | return model(inputs) 18 | 19 | # Get the concrete function from the wrapped model call 20 | concrete_func = model_call.get_concrete_function(dummy_data) 21 | 22 | # Create a file writer for TensorBoard summaries 23 | # Write the model architecture to TensorBoard 24 | with writer.as_default(): 25 | tf.summary.graph(concrete_func.graph) 26 | 27 | def log_comprehensive_information(model, predictions, layer_names, writer, log_dir): 28 | # Generate predictions 29 | top_5 = predictions.top_5 30 | preprocessed_image = predictions.preprocessed_image 31 | 32 | with writer.as_default(): 33 | # Log input image 34 | tf.summary.image("Input Image", preprocessed_image, step=0) 35 | # Log top 5 predictions 36 | for _, label, prob in top_5: # Replaced imagenet_id with _ 37 | tf.summary.scalar(f"Top 5 Predictions/{label}", prob, step=0) 38 | # Generate and log Grad-CAM heatmaps 39 | for layer_name in layer_names: 40 | heatmap = make_gradcam_heatmap(preprocessed_image, model, layer_name) 41 | if heatmap is not None: 42 | heatmap = np.uint8(255 * heatmap) 43 | heatmap_image = np.expand_dims(np.repeat(heatmap[:, :, np.newaxis], 3, axis=2), axis=0) 44 | tf.summary.image(f"Grad-CAM/{layer_name}", heatmap_image, step=0) 45 | 46 | def handle_tensorboard_logging(model, base64_image, layer_names, model_name): 47 | predictions = generate_predictions(model, base64_image) 48 | 49 | current_time = datetime.now().strftime("%Y%m%d-%H%M%S") 50 | log_dir = os.path.join(dir, f"../../logs/{model_name}/{current_time}") 51 | os.makedirs(log_dir, exist_ok=True) 52 | 53 | writer = tf.summary.create_file_writer(log_dir) 54 | 55 | log_comprehensive_information(model, predictions, layer_names, writer, log_dir) 56 | log_model_architecture(model, writer, log_dir) 57 | 58 | writer.flush() 59 | writer.close() 60 | 61 | return log_dir -------------------------------------------------------------------------------- /src/app/layout.tsx: -------------------------------------------------------------------------------- 1 | "use client"; 2 | 3 | import type { Metadata } from "next"; 4 | import { Inter, Saira, Roboto, Roboto_Slab } from "next/font/google"; 5 | import "./globals.css"; 6 | import Head from "next/head"; 7 | import tailwind from "tailwindcss"; 8 | 9 | import { ReactNode, useState } from "react"; 10 | import Sidenav from "./ui/navComponents/Sidenav"; 11 | 12 | interface Props { 13 | children: ReactNode; 14 | } 15 | 16 | const inter = Inter({ subsets: ["latin"] }); 17 | const saira = Saira({ subsets: ["latin"] }); 18 | const roboto = Roboto({ 19 | subsets: ["latin"], 20 | weight: "100", 21 | }); 22 | const robotoSlab = Roboto_Slab({ 23 | subsets: ["latin"], 24 | weight: "900", 25 | }); 26 | 27 | export default function RootLayout({ children }: Props) { 28 | const [sidebarOpen, setSidebarOpen] = useState(false); 29 | 30 | return ( 31 | 32 | 33 | xFlair Demo 34 | 35 | {/* */} 36 | 37 | 38 |
    39 |
    40 | 44 |
    49 | 68 | {children} 69 |
    70 |
    71 |
    72 | 73 | 74 | ); 75 | } 76 | -------------------------------------------------------------------------------- /Models/TensorFlow/app/routes/featuremaps.py: -------------------------------------------------------------------------------- 1 | import os 2 | import numpy as np 3 | from PIL import Image 4 | from flask import Blueprint, request, jsonify 5 | 6 | from app.utils.image_utils import preprocess_image 7 | from app.services.generate_images import make_feature_maps 8 | from app.utils.image_utils import add_progress_bar 9 | from app.utils.s3_utils import upload_file 10 | 11 | dir = os.path.dirname(__file__) 12 | featuremaps = Blueprint('featuremaps', __name__) 13 | 14 | @featuremaps.route('/featuremaps/', methods=['POST']) 15 | def upload_featuremaps(model_name): 16 | data = request.get_json() 17 | base64_image = data['data'] 18 | 19 | model_module = __import__(f"models.{model_name}.model", fromlist=["get_model"]) 20 | model = model_module.get_model() 21 | layer_names_module = __import__(f"models.{model_name}.model", fromlist=["get_all_layer_names"]) 22 | layer_names = layer_names_module.get_layer_names() 23 | 24 | preprocessed_image = preprocess_image(base64_image) 25 | 26 | visuals_dir = os.path.join(dir, "../visuals/") 27 | featuremaps_dir = os.path.join(visuals_dir, "featuremaps/") 28 | featuremapspb_dir = os.path.join(featuremaps_dir, "progressbars/") 29 | os.makedirs(visuals_dir, exist_ok=True) 30 | os.makedirs(featuremaps_dir, exist_ok=True) 31 | os.makedirs(featuremapspb_dir, exist_ok=True) 32 | 33 | links = [] 34 | progressbar_links = [] 35 | 36 | bucket_name = os.getenv('AWS_S3_BUCKET_NAME') 37 | if not bucket_name: 38 | return jsonify({"error": "Server configuration error. AWS S3 Bucket name is missing."}), 500 39 | 40 | for layer_name in layer_names: 41 | output_path = make_feature_maps(preprocessed_image, model, layer_name, featuremaps_dir, layer_names) 42 | if output_path is None: 43 | continue 44 | 45 | file_name = os.path.splitext(os.path.basename(output_path))[0] 46 | upload_file(output_path, bucket_name, f"featuremaps/{os.path.basename(output_path)}") 47 | links.append(f"https://{bucket_name}.s3.amazonaws.com/featuremaps/{file_name}.jpg") 48 | 49 | img_with_progress = Image.open(output_path) 50 | img_array = np.array(img_with_progress) 51 | current_index = layer_names.index(layer_name) 52 | total_layers = len(layer_names) 53 | img_with_progress_bar = add_progress_bar(img_array, current_index, total_layers, width = 20) 54 | 55 | # Resize and save the final image with the progress bar 56 | final_img = Image.fromarray(img_with_progress_bar) 57 | final_img_resized = final_img.resize((500, 500), Image.Resampling.LANCZOS) # Resizing the final image 58 | output_path_pb = os.path.join(featuremapspb_dir, file_name) 59 | final_img_resized.save(output_path_pb, format='JPEG', quality=85) # Saving the resized final image 60 | 61 | upload_file(output_path_pb, bucket_name, f"featuremaps/progressbars/{os.path.basename(output_path_pb)}.jpg") 62 | progressbar_links.append(f"https://{bucket_name}.s3.amazonaws.com/featuremaps/progressbars/{file_name}.jpg") 63 | 64 | return jsonify({'featuremaps': links, 'progressbars': progressbar_links}) 65 | 66 | 67 | 68 | -------------------------------------------------------------------------------- /Models/TensorFlow/app/routes/heatmaps.py: -------------------------------------------------------------------------------- 1 | from flask import Blueprint, request, jsonify 2 | import tensorflow as tf 3 | import cv2 4 | import os 5 | import numpy as np 6 | 7 | from app.services.generate_images import make_gradcam_heatmap 8 | from app.utils.image_utils import preprocess_image 9 | from app.services.generate_predictions import generate_predictions 10 | from app.utils.s3_utils import upload_file 11 | from app.utils.image_utils import add_progress_bar 12 | 13 | 14 | dir = os.path.dirname(__file__) 15 | heatmaps = Blueprint('heatmaps', __name__) 16 | 17 | @heatmaps.route('/heatmaps/', methods=['POST']) 18 | def upload_heatmaps(model_name): 19 | data = request.get_json() 20 | base64_image = data['data'] 21 | 22 | model_module = __import__(f"models.{model_name}.model", fromlist=["get_model"]) 23 | model = model_module.get_model() 24 | layer_names_module = __import__(f"models.{model_name}.model", fromlist=["get_all_layer_names"]) 25 | layer_names = layer_names_module.get_layer_names() 26 | 27 | preprocessed_image = preprocess_image(base64_image) 28 | pred_index = generate_predictions(model, base64_image)['predicted_class'] 29 | 30 | visuals_dir = os.path.join(dir, "../visuals/") 31 | heatmaps_dir = os.path.join(visuals_dir, "heatmaps/") 32 | heatmapspb_dir = os.path.join(heatmaps_dir, "progressbars/") 33 | os.makedirs(visuals_dir, exist_ok=True) 34 | os.makedirs(heatmaps_dir, exist_ok=True) 35 | os.makedirs(heatmapspb_dir, exist_ok=True) 36 | 37 | links = [] 38 | progessbar_links = [] 39 | 40 | for layer_name in layer_names: 41 | heatmap = make_gradcam_heatmap(preprocessed_image, model, layer_name, pred_index) 42 | if heatmap is None: 43 | continue 44 | 45 | img = tf.squeeze(preprocessed_image) 46 | img = (img * 0.5) + 0.5 47 | heatmap_resized = cv2.resize(heatmap, (img.shape[1], img.shape[0])) 48 | heatmap_resized = np.uint8(255 * heatmap_resized) 49 | heatmap_colored = cv2.applyColorMap(heatmap_resized, cv2.COLORMAP_JET) 50 | superimposed_img = heatmap_colored * 0.4 + np.uint8(img * 255) 51 | 52 | file_name = str(layer_names.index(layer_name) + 1).zfill(3) + layer_name 53 | output_path = os.path.join(heatmaps_dir, f"{file_name}.jpg") 54 | cv2.imwrite(output_path, superimposed_img) 55 | 56 | bucket_name = os.getenv('AWS_S3_BUCKET_NAME') 57 | upload_file(output_path, bucket_name, f"heatmaps/{file_name}.jpg") 58 | links.append(f"https://{bucket_name}.s3.amazonaws.com/heatmaps/{file_name}.jpg") 59 | 60 | output_path_pb = os.path.join(heatmapspb_dir, f"{file_name}.jpg") 61 | current_index = layer_names.index(layer_name) 62 | total_layers = len(layer_names) 63 | superimposed_img_with_progress = add_progress_bar(superimposed_img, current_index, total_layers, width=4) 64 | cv2.imwrite(output_path_pb, superimposed_img_with_progress) # Correctly save the image with progress bar 65 | 66 | upload_file(output_path_pb, bucket_name, f"heatmaps/progressbars/{file_name}.jpg") 67 | progessbar_links.append(f"https://{bucket_name}.s3.amazonaws.com/heatmaps/progressbars/{file_name}.jpg") 68 | 69 | return jsonify({'heatmaps': links, 'progressbars': progessbar_links}) -------------------------------------------------------------------------------- /src/app/ui/navComponents/Accordion.tsx: -------------------------------------------------------------------------------- 1 | import AccordionLink from './AccordionLink'; 2 | import Icon from './Icon'; 3 | import { AccordionContent } from '@/app/lib/definitions'; 4 | 5 | interface Props { 6 | activeAccordion: string[]; 7 | toggleAccordion: (path: string[]) => void; 8 | label: string; 9 | contents: AccordionContent[]; 10 | depth: number; 11 | id: string[]; 12 | icon?: 'gear' | 'book' | 'pages'; 13 | } 14 | 15 | export default function Accordion(props: Props) { 16 | const contents: JSX.Element[] = []; 17 | props.contents.forEach((el) => { 18 | if (el.type === 'Link') { 19 | contents.push( 20 | 21 | ); 22 | } else { 23 | contents.push( 24 | 32 | ); 33 | } 34 | }); 35 | 36 | return ( 37 |
  • 38 | 79 | 80 |
    88 |
      {contents}
    89 |
    90 |
    • 91 | ); 92 | } 93 | -------------------------------------------------------------------------------- /Models/TensorFlow/app/routes/preprocess.py: -------------------------------------------------------------------------------- 1 | import os 2 | from flask import Blueprint, request, jsonify 3 | from botocore.exceptions import ClientError 4 | import base64 5 | import glob 6 | 7 | from app.services.generate_images import make_preprocess_image, make_preprocess_images 8 | from app.utils.image_utils import preprocess_image 9 | from app.utils.s3_utils import download_file, upload_file 10 | 11 | 12 | dir = os.path.dirname(__file__) 13 | preprocess = Blueprint('preprocess', __name__) 14 | 15 | @preprocess.route('/preprocess', methods=['POST']) 16 | def upload_preprocess(): 17 | data = request.get_json() 18 | print(f"Received data: {data}") 19 | base64_image = data['data'] 20 | s3_url = str(data['filePath']) # Convert s3_url to string 21 | image_type = data['fileType'] 22 | 23 | print(f"Received image from {s3_url}") 24 | print(f"Image type is {image_type}") 25 | 26 | # Process the base64-encoded image 27 | preprocessed = preprocess_image(base64_image) 28 | 29 | # Prepare directories and filenames 30 | assets_dir = os.path.join(dir, '../assets') 31 | if not os.path.exists(assets_dir): 32 | os.makedirs(assets_dir) 33 | 34 | print(f"Assets directory is {assets_dir}") 35 | 36 | object_name = s3_url.split('.com/')[-1] 37 | print(f"Object name is {object_name}") 38 | 39 | download_path = os.path.join(assets_dir, f'upload.{image_type}') # Fixed string formatting 40 | 41 | s3_bucket = os.environ.get('AWS_S3_BUCKET_NAME') 42 | 43 | try: 44 | # Attempt to download the file from S3 45 | download_file(s3_bucket, object_name, download_path) 46 | except Exception as e: 47 | # If download fails, fallback to saving the file from base64 data 48 | print(f"Failed to download {object_name} from S3: {e}. Fallback to base64.") 49 | try: 50 | image_data = base64.b64decode(base64_image) 51 | with open(download_path, 'wb') as file: 52 | file.write(image_data) 53 | except Exception as e: 54 | print(f"Failed to save file from base64 data: {e}") 55 | return jsonify({'error': 'Failed to process image.'}), 500 56 | 57 | # Error handling for S3 operations 58 | try: 59 | print(f"Downloaded file from s3://{s3_bucket}/{object_name}") 60 | file_names = make_preprocess_images(base64_image, assets_dir) 61 | # upload_file(upload_path, s3_bucket, 'preprocess/image.jpg') 62 | if os.path.exists(download_path): 63 | os.remove(download_path) 64 | print(f"Deleted upload.jpg from {assets_dir}") 65 | 66 | file_list = glob.glob(os.path.join(assets_dir, '*')) 67 | # Loop through each file and upload it to S3 68 | for file_path in file_list: 69 | try: 70 | # Get the filename from the file path 71 | filename = os.path.basename(file_path) 72 | 73 | # Upload the file to S3 74 | upload_file(file_path, s3_bucket, f'preprocess/{filename}') 75 | 76 | print(f"Uploaded {filename} to s3://{s3_bucket}/preprocess/{filename}") 77 | except ClientError as e: 78 | print(f"Failed to upload {filename} to S3: {e}") 79 | 80 | print(f"Uploaded preprocessed image to s3://{s3_bucket}/preprocess/image.jpg") 81 | object_names = [f"preprocess/{file_name}" for file_name in file_names] 82 | return jsonify({'preprocessed_images': object_names}) 83 | except ClientError as e: 84 | return jsonify({'error': str(e)}), 500 -------------------------------------------------------------------------------- /Models/TensorBoard/models/MobileNet/conv_layers/all_layers.txt: -------------------------------------------------------------------------------- 1 | Conv1 2 | bn_Conv1 3 | Conv1_relu 4 | expanded_conv_depthwise 5 | expanded_conv_depthwise_BN 6 | expanded_conv_depthwise_relu 7 | expanded_conv_project 8 | expanded_conv_project_BN 9 | block_1_expand 10 | block_1_expand_BN 11 | block_1_expand_relu 12 | block_1_pad 13 | block_1_depthwise 14 | block_1_depthwise_BN 15 | block_1_depthwise_relu 16 | block_1_project 17 | block_1_project_BN 18 | block_2_expand 19 | block_2_expand_BN 20 | block_2_expand_relu 21 | block_2_depthwise 22 | block_2_depthwise_BN 23 | block_2_depthwise_relu 24 | block_2_project 25 | block_2_project_BN 26 | block_2_add 27 | block_3_expand 28 | block_3_expand_BN 29 | block_3_expand_relu 30 | block_3_pad 31 | block_3_depthwise 32 | block_3_depthwise_BN 33 | block_3_depthwise_relu 34 | block_3_project 35 | block_3_project_BN 36 | block_4_expand 37 | block_4_expand_BN 38 | block_4_expand_relu 39 | block_4_depthwise 40 | block_4_depthwise_BN 41 | block_4_depthwise_relu 42 | block_4_project 43 | block_4_project_BN 44 | block_4_add 45 | block_5_expand 46 | block_5_expand_BN 47 | block_5_expand_relu 48 | block_5_depthwise 49 | block_5_depthwise_BN 50 | block_5_depthwise_relu 51 | block_5_project 52 | block_5_project_BN 53 | block_5_add 54 | block_6_expand 55 | block_6_expand_BN 56 | block_6_expand_relu 57 | block_6_pad 58 | block_6_depthwise 59 | block_6_depthwise_BN 60 | block_6_depthwise_relu 61 | block_6_project 62 | block_6_project_BN 63 | block_7_expand 64 | block_7_expand_BN 65 | block_7_expand_relu 66 | block_7_depthwise 67 | block_7_depthwise_BN 68 | block_7_depthwise_relu 69 | block_7_project 70 | block_7_project_BN 71 | block_7_add 72 | block_8_expand 73 | block_8_expand_BN 74 | block_8_expand_relu 75 | block_8_depthwise 76 | block_8_depthwise_BN 77 | block_8_depthwise_relu 78 | block_8_project 79 | block_8_project_BN 80 | block_8_add 81 | block_9_expand 82 | block_9_expand_BN 83 | block_9_expand_relu 84 | block_9_depthwise 85 | block_9_depthwise_BN 86 | block_9_depthwise_relu 87 | block_9_project 88 | block_9_project_BN 89 | block_9_add 90 | block_10_expand 91 | block_10_expand_BN 92 | block_10_expand_relu 93 | block_10_depthwise 94 | block_10_depthwise_BN 95 | block_10_depthwise_relu 96 | block_10_project 97 | block_10_project_BN 98 | block_11_expand 99 | block_11_expand_BN 100 | block_11_expand_relu 101 | block_11_depthwise 102 | block_11_depthwise_BN 103 | block_11_depthwise_relu 104 | block_11_project 105 | block_11_project_BN 106 | block_11_add 107 | block_12_expand 108 | block_12_expand_BN 109 | block_12_expand_relu 110 | block_12_depthwise 111 | block_12_depthwise_BN 112 | block_12_depthwise_relu 113 | block_12_project 114 | block_12_project_BN 115 | block_12_add 116 | block_13_expand 117 | block_13_expand_BN 118 | block_13_expand_relu 119 | block_13_pad 120 | block_13_depthwise 121 | block_13_depthwise_BN 122 | block_13_depthwise_relu 123 | block_13_project 124 | block_13_project_BN 125 | block_14_expand 126 | block_14_expand_BN 127 | block_14_expand_relu 128 | block_14_depthwise 129 | block_14_depthwise_BN 130 | block_14_depthwise_relu 131 | block_14_project 132 | block_14_project_BN 133 | block_14_add 134 | block_15_expand 135 | block_15_expand_BN 136 | block_15_expand_relu 137 | block_15_depthwise 138 | block_15_depthwise_BN 139 | block_15_depthwise_relu 140 | block_15_project 141 | block_15_project_BN 142 | block_15_add 143 | block_16_expand 144 | block_16_expand_BN 145 | block_16_expand_relu 146 | block_16_depthwise 147 | block_16_depthwise_BN 148 | block_16_depthwise_relu 149 | block_16_project 150 | block_16_project_BN 151 | Conv_1 152 | Conv_1_bn 153 | out_relu 154 | -------------------------------------------------------------------------------- /Models/TensorFlow/models/MobileNet/conv_layers/all_layers.txt: -------------------------------------------------------------------------------- 1 | Conv1 2 | bn_Conv1 3 | Conv1_relu 4 | expanded_conv_depthwise 5 | expanded_conv_depthwise_BN 6 | expanded_conv_depthwise_relu 7 | expanded_conv_project 8 | expanded_conv_project_BN 9 | block_1_expand 10 | block_1_expand_BN 11 | block_1_expand_relu 12 | block_1_pad 13 | block_1_depthwise 14 | block_1_depthwise_BN 15 | block_1_depthwise_relu 16 | block_1_project 17 | block_1_project_BN 18 | block_2_expand 19 | block_2_expand_BN 20 | block_2_expand_relu 21 | block_2_depthwise 22 | block_2_depthwise_BN 23 | block_2_depthwise_relu 24 | block_2_project 25 | block_2_project_BN 26 | block_2_add 27 | block_3_expand 28 | block_3_expand_BN 29 | block_3_expand_relu 30 | block_3_pad 31 | block_3_depthwise 32 | block_3_depthwise_BN 33 | block_3_depthwise_relu 34 | block_3_project 35 | block_3_project_BN 36 | block_4_expand 37 | block_4_expand_BN 38 | block_4_expand_relu 39 | block_4_depthwise 40 | block_4_depthwise_BN 41 | block_4_depthwise_relu 42 | block_4_project 43 | block_4_project_BN 44 | block_4_add 45 | block_5_expand 46 | block_5_expand_BN 47 | block_5_expand_relu 48 | block_5_depthwise 49 | block_5_depthwise_BN 50 | block_5_depthwise_relu 51 | block_5_project 52 | block_5_project_BN 53 | block_5_add 54 | block_6_expand 55 | block_6_expand_BN 56 | block_6_expand_relu 57 | block_6_pad 58 | block_6_depthwise 59 | block_6_depthwise_BN 60 | block_6_depthwise_relu 61 | block_6_project 62 | block_6_project_BN 63 | block_7_expand 64 | block_7_expand_BN 65 | block_7_expand_relu 66 | block_7_depthwise 67 | block_7_depthwise_BN 68 | block_7_depthwise_relu 69 | block_7_project 70 | block_7_project_BN 71 | block_7_add 72 | block_8_expand 73 | block_8_expand_BN 74 | block_8_expand_relu 75 | block_8_depthwise 76 | block_8_depthwise_BN 77 | block_8_depthwise_relu 78 | block_8_project 79 | block_8_project_BN 80 | block_8_add 81 | block_9_expand 82 | block_9_expand_BN 83 | block_9_expand_relu 84 | block_9_depthwise 85 | block_9_depthwise_BN 86 | block_9_depthwise_relu 87 | block_9_project 88 | block_9_project_BN 89 | block_9_add 90 | block_10_expand 91 | block_10_expand_BN 92 | block_10_expand_relu 93 | block_10_depthwise 94 | block_10_depthwise_BN 95 | block_10_depthwise_relu 96 | block_10_project 97 | block_10_project_BN 98 | block_11_expand 99 | block_11_expand_BN 100 | block_11_expand_relu 101 | block_11_depthwise 102 | block_11_depthwise_BN 103 | block_11_depthwise_relu 104 | block_11_project 105 | block_11_project_BN 106 | block_11_add 107 | block_12_expand 108 | block_12_expand_BN 109 | block_12_expand_relu 110 | block_12_depthwise 111 | block_12_depthwise_BN 112 | block_12_depthwise_relu 113 | block_12_project 114 | block_12_project_BN 115 | block_12_add 116 | block_13_expand 117 | block_13_expand_BN 118 | block_13_expand_relu 119 | block_13_pad 120 | block_13_depthwise 121 | block_13_depthwise_BN 122 | block_13_depthwise_relu 123 | block_13_project 124 | block_13_project_BN 125 | block_14_expand 126 | block_14_expand_BN 127 | block_14_expand_relu 128 | block_14_depthwise 129 | block_14_depthwise_BN 130 | block_14_depthwise_relu 131 | block_14_project 132 | block_14_project_BN 133 | block_14_add 134 | block_15_expand 135 | block_15_expand_BN 136 | block_15_expand_relu 137 | block_15_depthwise 138 | block_15_depthwise_BN 139 | block_15_depthwise_relu 140 | block_15_project 141 | block_15_project_BN 142 | block_15_add 143 | block_16_expand 144 | block_16_expand_BN 145 | block_16_expand_relu 146 | block_16_depthwise 147 | block_16_depthwise_BN 148 | block_16_depthwise_relu 149 | block_16_project 150 | block_16_project_BN 151 | Conv_1 152 | Conv_1_bn 153 | out_relu 154 | -------------------------------------------------------------------------------- /src/app/(routes)/models/actions/image/gifs/route.ts: -------------------------------------------------------------------------------- 1 | import { type NextRequest } from 'next/server'; 2 | import { v2 as cloudinary } from 'cloudinary'; 3 | import { S3Client, GetObjectCommand } from '@aws-sdk/client-s3'; 4 | import { getSignedUrl } from '@aws-sdk/s3-request-presigner'; 5 | import { ImageAndVideoFormatOptions } from 'cloudinary'; 6 | 7 | cloudinary.config({ 8 | cloud_name: process.env.CLOUDINARY_CLOUD_NAME, 9 | api_key: process.env.CLOUDINARY_API_KEY, 10 | api_secret: process.env.CLOUDINARY_API_SECRET, 11 | }); 12 | 13 | const s3 = new S3Client({ 14 | region: process.env.AWS_DEFAULT_REGION as string, 15 | credentials: { 16 | accessKeyId: process.env.AWS_ACCESS_KEY_ID as string, 17 | secretAccessKey: process.env.AWS_SECRET_ACCESS_KEY as string, 18 | }, 19 | }); 20 | 21 | const generateSignedUrl = async ( 22 | bucketName: string, 23 | objectKey: string, 24 | expiresInSeconds: number 25 | ) => { 26 | const command = new GetObjectCommand({ 27 | Bucket: bucketName, 28 | Key: objectKey, 29 | }); 30 | 31 | // Generate the signed URL 32 | try { 33 | const signedUrl = await getSignedUrl(s3, command, { 34 | expiresIn: expiresInSeconds, 35 | }); 36 | return signedUrl; 37 | } catch (error) { 38 | console.error('Error generating signed URL:', error); 39 | throw error; // Rethrow the error for further handling 40 | } 41 | }; 42 | 43 | const uploadImage = async ( 44 | imagePath: string, 45 | publicId: string, 46 | tag: string 47 | ) => { 48 | return cloudinary.uploader.upload(imagePath, { 49 | folder: `${tag}_images`, 50 | tags: tag, 51 | public_id: publicId, 52 | }); 53 | }; 54 | 55 | const delay = (ms: number) => { 56 | return new Promise(resolve => setTimeout(resolve, ms)); 57 | } 58 | 59 | export async function POST(req: NextRequest) { 60 | try { 61 | const { urls, tag } = await req.json(); 62 | 63 | const objectKeys = urls.map((url: string) => { 64 | const startIndex = url.indexOf('.com') + 5; 65 | const endIndex = url.length; 66 | const substring = url.substring(startIndex, endIndex); 67 | return substring; 68 | }); 69 | 70 | if (!objectKeys.length || !tag) { 71 | return new Response('Missing URLs', { status: 400 }); 72 | } 73 | 74 | const signedUrls = await Promise.all( 75 | objectKeys.map((objectKey: string) => 76 | generateSignedUrl( 77 | process.env.AWS_S3_BUCKET_NAME as string, 78 | objectKey, 79 | 180 80 | ) 81 | ) 82 | ); 83 | 84 | // Creating upload promises for all URLs 85 | const uploadPromises = signedUrls.map((url: string, index: number) => { 86 | const formattedIndex = String(index + 1).padStart(3, '0'); 87 | uploadImage(url, `${tag}_${formattedIndex}`, tag); 88 | }); 89 | 90 | // Await all upload promises 91 | await Promise.all(uploadPromises); 92 | console.log('All images uploaded.'); 93 | 94 | await delay(1000); 95 | 96 | // Create GIF from uploaded images and cleanup 97 | const gifCreationResult = await cloudinary.uploader.multi(tag as string, { 98 | format: 'gif' as ImageAndVideoFormatOptions, 99 | }); 100 | 101 | // Delete uploaded images after creating the GIF 102 | cloudinary.api.delete_resources_by_tag(tag); 103 | console.log('Uploaded images cleaned up successfully'); 104 | 105 | // Return the URL of the created GIF 106 | return new Response( 107 | JSON.stringify({ 108 | url: gifCreationResult.secure_url as string, 109 | tag: tag as string, 110 | }), 111 | { status: 200 } 112 | ); 113 | } catch (error) { 114 | console.error('Error:', error); 115 | return new Response('Internal Server Error', { status: 500 }); 116 | } 117 | } 118 | -------------------------------------------------------------------------------- /Models/TensorBoard/app/services/generate_images.py: -------------------------------------------------------------------------------- 1 | import tensorflow as tf 2 | import os 3 | import matplotlib.pyplot as plt 4 | import cv2 5 | 6 | def make_gradcam_heatmap(img_array, model, conv_layer_name, pred_index=None): 7 | # Create a model that maps the input image to the activations of the conv layer 8 | # as well as the output predictions 9 | grad_model = tf.keras.models.Model(inputs=model.inputs, outputs=[model.get_layer(conv_layer_name).output, model.output]) 10 | 11 | # Record operations for automatic differentiation to compute gradients later 12 | with tf.GradientTape() as tape: 13 | # Cast the input image to float32 in case it's not 14 | img_array = tf.cast(img_array, tf.float32) 15 | tape.watch(img_array) 16 | # Compute the model's output (including the output of the conv layer) 17 | conv_layer_output, preds = grad_model(img_array) 18 | if pred_index is None: 19 | pred_index = tf.argmax(preds[0]) 20 | # Use the gradient of the output neuron (for the predicted class or specified class) 21 | # relative to the conv layer output to compute the gradients 22 | class_channel = preds[:, pred_index] 23 | 24 | # Compute gradients towards the class channel 25 | grads = tape.gradient(class_channel, conv_layer_output) 26 | 27 | # Pooling and normalization of gradients 28 | pooled_grads = tf.reduce_mean(grads, axis=(0, 1, 2)) 29 | conv_layer_output = conv_layer_output[0] 30 | 31 | # Weight the channels by the gradients and then sum them 32 | heatmap = tf.matmul(conv_layer_output, pooled_grads[..., tf.newaxis]) 33 | heatmap = tf.squeeze(heatmap) 34 | 35 | # For visualization, make sure the heatmap is in a [0, 1] range 36 | heatmap = tf.maximum(heatmap, 0) / tf.math.reduce_max(heatmap) 37 | return heatmap.numpy() 38 | 39 | def make_feature_maps(preprocessed_image, model, layer_name, output_dir, layer_names, max_feature_maps=16): 40 | intermediate_layer_model = tf.keras.Model(inputs=model.input, outputs=model.get_layer(layer_name).output) 41 | intermediate_output = intermediate_layer_model(preprocessed_image) 42 | num_feature_maps = intermediate_output.shape[-1] 43 | num_feature_maps_to_visualize = min(num_feature_maps, max_feature_maps) 44 | 45 | if num_feature_maps_to_visualize >= 16: 46 | nrows, ncols = 4, 4 47 | fig, axes = plt.subplots(nrows=nrows, ncols=ncols, figsize=(20, 20)) 48 | axes = axes.flatten() 49 | 50 | for i in range(num_feature_maps_to_visualize): 51 | ax = axes[i] 52 | feature_map = intermediate_output[0, :, :, i] 53 | ax.imshow(feature_map, cmap='viridis') 54 | ax.set_title(f'Feature Map {i+1}') 55 | ax.axis('off') 56 | 57 | file_name = f"{str(layer_names.index(layer_name) + 1).zfill(3) + layer_name}.jpg" 58 | output_path = os.path.join(output_dir, file_name) 59 | plt.savefig(output_path, format='jpeg') 60 | plt.close(fig) 61 | else: 62 | return 63 | 64 | return output_path 65 | 66 | 67 | def make_preprocess_image(original_image_path, preprocessed_image, output_dir): 68 | original_image = cv2.imread(original_image_path) 69 | preprocessed_image_np = preprocessed_image.numpy().squeeze() 70 | 71 | fig, ax = plt.subplots(1, 2, figsize=(10, 5)) 72 | ax[0].imshow(cv2.cvtColor(original_image, cv2.COLOR_BGR2RGB)) 73 | ax[0].set_title('Original Image') 74 | ax[0].axis('off') 75 | 76 | ax[1].imshow((preprocessed_image_np * 0.5) + 0.5) # Assuming preprocessing includes normalization 77 | ax[1].set_title('Preprocessed Image') 78 | ax[1].axis('off') 79 | file_name = 'input_images_comparison.png' 80 | file_path = os.path.join(output_dir, file_name) 81 | print(file_name) 82 | # Save the figure 83 | plt.savefig(file_path) 84 | plt.close(fig) # Close the figure to free memory -------------------------------------------------------------------------------- /src/app/ui/navComponents/Sidenav.tsx: -------------------------------------------------------------------------------- 1 | // Import necessary libraries and components 2 | import React, { useState, useEffect, useRef } from 'react'; 3 | import Link from 'next/link'; 4 | import Accordion from './Accordion'; 5 | import { AccordionContent } from '@/app/lib/definitions'; 6 | import { 7 | modAccordion, 8 | libAccordion, 9 | docAccordion, 10 | } from '@/app/lib/accordionData'; 11 | 12 | interface props { 13 | sidebarOpen: boolean; 14 | setSidebarOpen: React.Dispatch>; 15 | } 16 | 17 | // Define the Sidenav component 18 | export default function Sidenav({ sidebarOpen, setSidebarOpen }: props) { 19 | // Define state for sidebar expansion 20 | const [sidebarExpanded, setSidebarExpanded] = useState(false); 21 | const [activeAccordion, setActiveAccordion] = useState([]); 22 | 23 | // Create a reference to the sidebar element 24 | const sidebar = useRef(null); 25 | 26 | // Effect to add or remove a className to the body element based on sidebar expansion 27 | useEffect(() => { 28 | if (sidebarExpanded) { 29 | document.body.classList.add('sidebar-expanded'); 30 | } else { 31 | document.body.classList.remove('sidebar-expanded'); 32 | } 33 | }, [sidebarExpanded]); 34 | 35 | // Function to toggle accordion item 36 | const toggleAccordion = (path: string[]) => { 37 | const newPath = path.slice(); 38 | if (activeAccordion.includes(newPath[newPath.length - 1])) newPath.pop(); 39 | setActiveAccordion(newPath); 40 | }; 41 | 42 | return ( 43 |
    50 |
    51 |

    55 | xFlair 56 |

    57 |
    58 | 141 |
    142 | titleText 147 |
    148 |
    149 | ); 150 | } 151 | -------------------------------------------------------------------------------- /Models/TensorFlow/app/services/generate_images.py: -------------------------------------------------------------------------------- 1 | import tensorflow as tf 2 | import os 3 | import matplotlib.pyplot as plt 4 | import cv2 5 | import base64 6 | import numpy as np 7 | 8 | def make_gradcam_heatmap(img_array, model, conv_layer_name, pred_index=None): 9 | # Create a model that maps the input image to the activations of the conv layer 10 | # as well as the output predictions 11 | grad_model = tf.keras.models.Model(inputs=model.inputs, outputs=[model.get_layer(conv_layer_name).output, model.output]) 12 | 13 | # Record operations for automatic differentiation to compute gradients later 14 | with tf.GradientTape() as tape: 15 | # Cast the input image to float32 in case it's not 16 | img_array = tf.cast(img_array, tf.float32) 17 | tape.watch(img_array) 18 | # Compute the model's output (including the output of the conv layer) 19 | conv_layer_output, preds = grad_model(img_array) 20 | if pred_index is None: 21 | pred_index = tf.argmax(preds[0]) 22 | # Use the gradient of the output neuron (for the predicted class or specified class) 23 | # relative to the conv layer output to compute the gradients 24 | class_channel = preds[:, pred_index] 25 | 26 | # Compute gradients towards the class channel 27 | grads = tape.gradient(class_channel, conv_layer_output) 28 | 29 | # Pooling and normalization of gradients 30 | pooled_grads = tf.reduce_mean(grads, axis=(0, 1, 2)) 31 | conv_layer_output = conv_layer_output[0] 32 | 33 | # Weight the channels by the gradients and then sum them 34 | heatmap = tf.matmul(conv_layer_output, pooled_grads[..., tf.newaxis]) 35 | heatmap = tf.squeeze(heatmap) 36 | 37 | # For visualization, make sure the heatmap is in a [0, 1] range 38 | heatmap = tf.maximum(heatmap, 0) / tf.math.reduce_max(heatmap) 39 | return heatmap.numpy() 40 | 41 | def make_feature_maps(preprocessed_image, model, layer_name, output_dir, layer_names, max_feature_maps=16): 42 | intermediate_layer_model = tf.keras.Model(inputs=model.input, outputs=model.get_layer(layer_name).output) 43 | intermediate_output = intermediate_layer_model(preprocessed_image) 44 | num_feature_maps = intermediate_output.shape[-1] 45 | num_feature_maps_to_visualize = min(num_feature_maps, max_feature_maps) 46 | 47 | if num_feature_maps_to_visualize >= 16: 48 | nrows, ncols = 4, 4 49 | fig, axes = plt.subplots(nrows=nrows, ncols=ncols, figsize=(20, 20)) 50 | axes = axes.flatten() 51 | 52 | for i in range(num_feature_maps_to_visualize): 53 | ax = axes[i] 54 | feature_map = intermediate_output[0, :, :, i] 55 | ax.imshow(feature_map, cmap='viridis') 56 | ax.set_title(f'Feature Map {i+1}') 57 | ax.axis('off') 58 | 59 | file_name = f"{str(layer_names.index(layer_name) + 1).zfill(3) + layer_name}.jpg" 60 | output_path = os.path.join(output_dir, file_name) 61 | plt.savefig(output_path, format='jpeg') 62 | plt.close(fig) 63 | else: 64 | return 65 | 66 | return output_path 67 | 68 | 69 | def make_preprocess_image(original_image_path, preprocessed_image, output_dir): 70 | original_image = cv2.imread(original_image_path) 71 | preprocessed_image_np = preprocessed_image.numpy().squeeze() 72 | 73 | fig, ax = plt.subplots(1, 2, figsize=(10, 5)) 74 | ax[0].imshow(cv2.cvtColor(original_image, cv2.COLOR_BGR2RGB)) 75 | ax[0].set_title('Original Image') 76 | ax[0].axis('off') 77 | 78 | ax[1].imshow((preprocessed_image_np * 0.5) + 0.5) # Assuming preprocessing includes normalization 79 | ax[1].axis('off') 80 | # Save the figure 81 | print (output_dir) 82 | print ('saving') 83 | plt.savefig(output_dir, format='jpeg') 84 | print ('saved') 85 | plt.close(fig) # Close the figure to free memory 86 | 87 | def preprocess_image_steps(base64_image_str): 88 | # Decode the image from base64 89 | image_bytes = base64.b64decode(base64_image_str) 90 | # Decode image to tensor and resize 91 | image_tensor = tf.image.decode_image(image_bytes, channels=3) 92 | resized_image_tensor = tf.image.resize(image_tensor, (224, 224)) 93 | # Normalize 94 | preprocessed_image_tensor = tf.keras.applications.mobilenet_v2.preprocess_input(resized_image_tensor) 95 | # Add batch dimension 96 | final_image_tensor = tf.expand_dims(preprocessed_image_tensor, 0) 97 | 98 | # Convert tensors to numpy for visualization 99 | original_np = image_tensor.numpy() 100 | resized_np = resized_image_tensor.numpy() 101 | preprocessed_np = (preprocessed_image_tensor.numpy()) # Undo normalization for visualization 102 | 103 | return original_np, resized_np, preprocessed_np 104 | 105 | def save_image(image_np, output_path): 106 | fig, ax = plt.subplots(figsize=(5, 5)) 107 | ax.imshow(image_np) 108 | ax.axis('off') # Remove axes and borders 109 | plt.savefig(output_path, format='jpeg', bbox_inches='tight', pad_inches=0) 110 | plt.close(fig) 111 | 112 | def make_preprocess_images(base64_image_str, output_dir): 113 | original_np, resized_np, preprocessed_np = preprocess_image_steps(base64_image_str) 114 | 115 | original_file_name = 'original_image.jpg' 116 | resized_file_name = 'resized_image.jpg' 117 | preprocessed_file_name = 'preprocessed_image.jpg' 118 | # Adjust these lines to change the format or naming convention 119 | original_output_path = os.path.join(output_dir, original_file_name) 120 | resized_output_path = os.path.join(output_dir, resized_file_name) 121 | preprocessed_output_path = os.path.join(output_dir, preprocessed_file_name) 122 | 123 | # Save the images 124 | save_image(cv2.cvtColor(original_np.astype(np.uint8), cv2.COLOR_BGR2RGB), original_output_path) 125 | save_image(cv2.cvtColor(resized_np.astype(np.uint8), cv2.COLOR_BGR2RGB), resized_output_path) 126 | save_image(preprocessed_np, preprocessed_output_path) 127 | 128 | return [original_file_name, resized_file_name, preprocessed_file_name] -------------------------------------------------------------------------------- /src/app/lib/accordionData.ts: -------------------------------------------------------------------------------- 1 | import { AccordionContent } from "./definitions"; 2 | 3 | 4 | export const modAccordion: AccordionContent = { 5 | type: 'Accordion', 6 | label: 'Models', 7 | id: ['mod-accordion'], 8 | icon: 'gear', 9 | contents: [ 10 | { 11 | type: 'Accordion', 12 | label: 'TensorFlow', 13 | id: ['mod-accordion', 'mod-tf-accordion'], 14 | contents: [ 15 | { 16 | type: 'Accordion', 17 | label: 'Images', 18 | id: ['mod-accordion', 'mod-tf-accordion', 'mod-tf-img-accordion'], 19 | contents: [ 20 | { 21 | type: 'Link', 22 | label: 'Classification', 23 | path: '/models/tensorflow/images/classification', 24 | }, 25 | { 26 | type: 'Link', 27 | label: 'Detection', 28 | path: '/models/tensorflow/images/detection', 29 | }, 30 | { 31 | type: 'Link', 32 | label: 'Generation', 33 | path: '/models/tensorflow/images/generation', 34 | }, 35 | ], 36 | }, 37 | { 38 | type: 'Accordion', 39 | label: 'Text', 40 | id: ['mod-accordion', 'mod-tf-accordion', 'mod-tf-txt-accordion'], 41 | contents: [ 42 | { 43 | type: 'Link', 44 | label: 'GPTs', 45 | path: '/models/tensorflow/text/gpt', 46 | }, 47 | { 48 | type: 'Link', 49 | label: 'Detection', 50 | path: '/models/tensorflow/text/texttospeech', 51 | }, 52 | ], 53 | }, 54 | ], 55 | }, 56 | { 57 | type: 'Accordion', 58 | label: 'PyTorch', 59 | id: ['mod-accordion', 'mod-py-accordion'], 60 | contents: [ 61 | { 62 | type: 'Accordion', 63 | label: 'Images', 64 | id: ['mod-accordion', 'mod-py-accordion', 'mod-py-img-accordion'], 65 | contents: [ 66 | { 67 | type: 'Link', 68 | label: 'Classification', 69 | path: '/models/pytorch/images/classification', 70 | }, 71 | { 72 | type: 'Link', 73 | label: 'Detection', 74 | path: '/models/pytorch/images/detection', 75 | }, 76 | { 77 | type: 'Link', 78 | label: 'Generation', 79 | path: '/models/pytorch/images/generation', 80 | }, 81 | ], 82 | }, 83 | { 84 | type: 'Accordion', 85 | label: 'Text', 86 | id: ['mod-accordion', 'mod-py-accordion', 'mod-py-txt-accordion'], 87 | contents: [ 88 | { 89 | type: 'Link', 90 | label: 'GPTs', 91 | path: '/models/pytorch/text/gpt', 92 | }, 93 | { 94 | type: 'Link', 95 | label: 'Detection', 96 | path: '/models/pytorch/text/texttospeech', 97 | }, 98 | ], 99 | }, 100 | ], 101 | }, 102 | { 103 | type: 'Accordion', 104 | label: 'Model Agnostic', 105 | id: ['mod-accordion', 'mod-ag-accordion'], 106 | contents: [ 107 | { 108 | type: 'Accordion', 109 | label: 'Images', 110 | id: ['mod-accordion', 'mod-ag-accordion', 'mod-ag-img-accordion'], 111 | contents: [ 112 | { 113 | type: 'Link', 114 | label: 'Classification', 115 | path: '/models/agnostic/images/classification', 116 | }, 117 | { 118 | type: 'Link', 119 | label: 'Detection', 120 | path: '/models/agnostic/images/detection', 121 | }, 122 | { 123 | type: 'Link', 124 | label: 'Generation', 125 | path: '/models/agnostic/images/generation', 126 | }, 127 | ], 128 | }, 129 | { 130 | type: 'Accordion', 131 | label: 'Text', 132 | id: ['mod-accordion', 'mod-ag-accordion', 'mod-ag-txt-accordion'], 133 | contents: [ 134 | { 135 | type: 'Link', 136 | label: 'GPTs', 137 | path: '/models/agnostic/text/gpt', 138 | }, 139 | { 140 | type: 'Link', 141 | label: 'Detection', 142 | path: '/models/agnostic/text/texttospeech', 143 | }, 144 | ], 145 | }, 146 | ], 147 | }, 148 | ], 149 | }; 150 | 151 | export const libAccordion: AccordionContent = { 152 | type: 'Accordion', 153 | label: 'Libraries', 154 | id: ['lib-accordion'], 155 | icon: 'book', 156 | contents: [ 157 | { 158 | type: 'Link', 159 | label: 'Python', 160 | path: '/libraries/python', 161 | }, 162 | { 163 | type: 'Accordion', 164 | label: 'React', 165 | id: ['lib-accordion', 'lib-react-accordion'], 166 | contents: [ 167 | { 168 | type: 'Link', 169 | label: 'Javascript', 170 | path: '/libraries/react/javascript', 171 | }, 172 | { 173 | type: 'Link', 174 | label: 'Typescript', 175 | path: '/libraries/react/typescript', 176 | }, 177 | ], 178 | }, 179 | ], 180 | }; 181 | 182 | export const docAccordion: AccordionContent = { 183 | type: 'Accordion', 184 | label: 'Documentation', 185 | id: ['doc-accordion'], 186 | icon: 'pages', 187 | contents: [ 188 | { 189 | type: 'Link', 190 | label: 'Python', 191 | path: '/documentation/python', 192 | }, 193 | { 194 | type: 'Link', 195 | label: 'Javascript / Typescript', 196 | path: '/documentation/javascript-typescript', 197 | }, 198 | ], 199 | }; -------------------------------------------------------------------------------- /src/app/ui/newModal.tsx: -------------------------------------------------------------------------------- 1 | import { Top5Obj } from '../lib/definitions'; 2 | import styles from '../page.module.css'; 3 | import HeatMapGif from './heatMapGif'; 4 | import GridMapGif from './gridMapGif'; 5 | import Top5 from './top5Classes'; 6 | 7 | interface Props { 8 | closeViz: () => void; 9 | hGifURL: string; 10 | fGifURL: string; 11 | top5: Top5Obj; 12 | preprocessFilePath: string[]; 13 | } 14 | 15 | export default function NewModal(props: Props) { 16 | return ( 17 | // modal main container 18 |
    19 | {/* exit button */} 20 | 29 | {/* intro slide */} 30 |
    31 | {/* title */} 32 |
    33 |

    The Path Between Input and Output

    34 |
    35 |
    36 |
    37 |

    Unraveling AI: A Visual Journey

    38 |

    Discover how artificial intelligence

    39 |

    perceives and interprets your image.

    40 |
    41 |
    42 |

    43 | This immersive experience will guide you through the fascinating 44 | process of AI image analysis. From preprocessing to feature 45 | extraction and classification, you'll gain a deeper 46 | understanding of how AI models work behind the scenes. 47 |

    48 |
    49 |
    50 | {/*

    Scroll Down For More Information

    */} 51 | 59 | 63 | 64 |
    65 |
    66 |
    67 |
    68 |
    69 |
    70 |
    71 |
    72 | {/*preprocess image explanation*/} 73 |
    74 |
    75 |

    Preparing Your Image

    76 |

    77 | The essential steps that prepare your image for AI analysis. 78 |

    79 |
    80 |
    81 |

    82 | Before an AI model can analyze your image, it needs to be 83 | preprocessed. This crucial step ensures that the image is 84 | converted into an array of parameters that the nueral network expects. In 85 | this slide, notice how your image is transformed through resizing, 86 | normalization, and other preprocessing techniques. These steps are 87 | essential for the AI model to properly interpret your image. 88 |

    89 |
    90 |
    91 | {/*

    Scroll Down For More Information

    */} 92 | 100 | 104 | 105 |
    106 |
    107 |
    108 |
    109 |
    110 | {/*preprocess image*/} 111 |
    112 |
    Initial Formatting
    113 | 119 | 120 |
    121 |
    122 |
    Image Resizing
    123 | 124 | 130 | 131 |
    132 |
    133 |
    Image Tensor
    134 | 135 | 141 | 142 |
    143 |
    144 |
    145 | {/*explain gifs*/} 146 |
    147 |
    148 |

    The Brain of AI

    149 |

    Your image entering the complex world of the neural network

    150 |
    151 |
    152 |

    153 | In the slide below, we show how your 154 | preprocessed image is analyzed by the AI model's neural 155 | network. The sequences showcase activation heatmaps at different 156 | layers, representing the regions of the image that the AI focuses 157 | on. As your image flows through the layers, the network assigns 158 | weights to different regions, ultimately leading to a 159 | classification decision. 160 |

    161 |
    162 |
    163 | {/*

    Scroll Down For More Information

    */} 164 | 172 | 176 | 177 |
    178 |
    179 |
    180 |
    181 |
    182 |
    183 |
    184 |
    185 |
    Heat Map
    186 | 187 |
    188 |
    189 |
    Feature Map
    190 | 191 |
    192 |
    193 |

    194 | On the left side, you can see which pixels the AI model focuses on 195 | as it moves through the layers of the neural network. The network 196 | provides weights to important regions, and the larger the weight, 197 | the hotter the region. The right is a visualization of how the 198 | neural network interprets different features of the image. Feature 199 | determination is broad in earlier stages, and becomes more focused 200 | on smaller areas towards the end. 201 |

    202 |
    203 |
    204 | 205 |
    206 | ); 207 | } 208 | -------------------------------------------------------------------------------- /src/app/(routes)/models/tensorflow/images/classification/page.tsx: -------------------------------------------------------------------------------- 1 | 'use client'; 2 | 3 | import React, { useState, useEffect, ChangeEvent, use } from 'react'; 4 | import axios from 'axios'; 5 | import NewModal from '@/app/ui/newModal'; 6 | import { Heatmaps, Featuremaps } from '@/app/lib/definitions'; 7 | import { Top5Obj } from '@/app/lib/definitions'; 8 | 9 | let inputImage: File | undefined; 10 | let count = 0; 11 | 12 | let modelName: string = 'MobileNet'; 13 | 14 | export default function Home() { 15 | console.log(`===== RENDER ${count++} ======`); 16 | const states = [ 17 | 'original', 18 | 'preprocessing', 19 | 'preprocessed', 20 | 'loadingMesh', 21 | 'featureMap', 22 | 'resizing', 23 | 'classifying', 24 | 'output', 25 | ]; 26 | 27 | const [imgName, setImgName] = useState('Browse...'); 28 | const [imgURL, setImgURL] = useState(''); 29 | const [vizState, setVizState] = useState(false); 30 | const [predictionName, setPredictionName] = useState(''); 31 | const [viz, openViz] = useState(false); 32 | const [hGifURL, setHGifURL] = useState(''); 33 | const [fGifURL, setFGifURL] = useState(''); 34 | const [time, setTime] = useState(0); 35 | const [top5, setTop5] = useState({} as Top5Obj); 36 | const [preprocessFilePath, setPreprocessFilePath] = useState([]); 37 | const [filePath, setFilePath] = useState(''); 38 | const [buttonState, setButtonState] = useState(0); 39 | const [heatmapLinks, setHeatmapLinks] = useState({} as Heatmaps); 40 | const [featuremapLinks, setFeaturemapLinks] = useState({} as Featuremaps); 41 | const [data, setData] = useState(''); 42 | const [fileType, setFileType] = useState(''); 43 | const [top5Formatted, setTop5Formatted] = useState({} as Top5Obj); 44 | 45 | const browse = (e: ChangeEvent) => { 46 | inputImage = e.currentTarget.files?.[0]; 47 | setImgName(`File Name: ${inputImage?.name}`); 48 | if (inputImage) { 49 | setImgURL(URL.createObjectURL(inputImage)); 50 | setButtonState(1); 51 | } 52 | }; 53 | 54 | async function imgUpload(inputImage: File) { 55 | const formData = new FormData(); 56 | formData.append('file', inputImage as File); 57 | formData.append('modelName', modelName); 58 | axios 59 | .post('/models/actions/image/upload', formData, { 60 | headers: { 61 | 'Content-Type': 'multipart/form-data', 62 | }, 63 | }) 64 | .then((response) => { 65 | console.log('response:', response.data); 66 | setFilePath(response.data.filePath as string); 67 | }) 68 | .catch((error) => { 69 | console.error(error); 70 | }); 71 | } 72 | 73 | async function predict(data: string, modelName: string) { 74 | fetch('/models/actions/image/predict', { 75 | method: 'POST', 76 | headers: { 77 | 'Content-Type': 'application/json', 78 | }, 79 | body: JSON.stringify({ data, modelName }), 80 | }) 81 | .then((response) => response.json()) 82 | .then((response) => { 83 | console.log('response:', response); 84 | setPredictionName(response.predictions.predicted_class_name); 85 | setTop5(response.predictions.class_name_probabilities); 86 | setTime(response.time); 87 | }) 88 | .catch((error) => { 89 | console.error(error); 90 | }); 91 | } 92 | 93 | function preprocess(data: string, filePath: string, fileType: string) { 94 | fetch('/models/actions/image/preprocess', { 95 | method: 'POST', 96 | headers: { 97 | 'Content-Type': 'application/json', 98 | }, 99 | body: JSON.stringify({ data, filePath, fileType }), 100 | }) 101 | .then((response) => response.json()) 102 | .then((response) => { 103 | console.log('response:', response); 104 | setPreprocessFilePath(response.urls as string[]); 105 | }) 106 | .catch((error) => { 107 | console.error(error); 108 | }); 109 | } 110 | 111 | function heatmaps(data: string, modelName: string) { 112 | fetch('/models/actions/image/heatmaps', { 113 | method: 'POST', 114 | headers: { 115 | 'Content-Type': 'application/json', 116 | }, 117 | body: JSON.stringify({ data, modelName }), 118 | }) 119 | .then((response) => response.json()) 120 | .then((response) => { 121 | console.log('response:', response); 122 | setHeatmapLinks(response); 123 | }) 124 | .catch((error) => { 125 | console.error(error); 126 | }); 127 | } 128 | 129 | function gifs(urls: string[], tag: string) { 130 | fetch('/models/actions/image/gifs', { 131 | method: 'POST', 132 | headers: { 133 | 'Content-Type': 'application/json', 134 | }, 135 | body: JSON.stringify({ urls, tag }), 136 | }) 137 | .then((response) => response.json()) 138 | .then((response) => { 139 | console.log('response:', response); 140 | if (tag === 'heatmap_gif') { 141 | setHGifURL(response.url); 142 | } else { 143 | setFGifURL(response.url); 144 | } 145 | }) 146 | .catch((error) => { 147 | console.error(error); 148 | }); 149 | } 150 | 151 | function featureMaps(data: string, modelName: string) { 152 | fetch('/models/actions/image/featmaps', { 153 | method: 'POST', 154 | headers: { 155 | 'Content-Type': 'application/json', 156 | }, 157 | body: JSON.stringify({ data, modelName }), 158 | }) 159 | .then((response) => response.json()) 160 | .then((response) => { 161 | console.log('response:', response); 162 | setFeaturemapLinks(response); 163 | }) 164 | .catch((error) => { 165 | console.error(error); 166 | }); 167 | } 168 | 169 | function logs(data: string, modelName: string) { 170 | fetch('/models/actions/logs', { 171 | method: 'POST', 172 | headers: { 173 | 'Content-Type': 'application/json', 174 | }, 175 | body: JSON.stringify({ data, modelName }), 176 | }) 177 | .then((response) => response.json()) 178 | .then((response) => { 179 | console.log('response:', response); 180 | }) 181 | .then(() => { 182 | setVizState(true); 183 | setButtonState(2); 184 | }) 185 | .catch((error) => { 186 | console.error(error); 187 | }); 188 | } 189 | 190 | function formatString(inputString: string): string { 191 | return inputString 192 | .split('_') 193 | .map((word) => word.charAt(0).toUpperCase() + word.slice(1).toLowerCase()) 194 | .join(' '); 195 | } 196 | 197 | const uploadClick = async () => { 198 | if (inputImage) { 199 | const reader = new FileReader(); 200 | reader.readAsDataURL(inputImage as File); 201 | 202 | reader.onloadend = () => { 203 | const data = (reader.result as string).split(',')[1]; 204 | setData(data as string); 205 | const fileType = inputImage?.type.split('/')[1]; 206 | setFileType(fileType as string); 207 | if (process.env.mode === 'logs') { 208 | logs(data, modelName); 209 | } else { 210 | heatmaps(data, modelName); 211 | featureMaps(data, modelName); 212 | imgUpload(inputImage as File); 213 | predict(data, modelName) 214 | .then(() => { 215 | setVizState(true); 216 | setButtonState(2); 217 | }) 218 | .catch((error) => { 219 | console.error(error); 220 | }); 221 | } 222 | }; 223 | } 224 | }; 225 | 226 | useEffect(() => { 227 | if (filePath) { 228 | preprocess(data, filePath, fileType); 229 | } 230 | }, [filePath, data, fileType]); 231 | 232 | useEffect(() => { 233 | if (heatmapLinks.progressbars && !hGifURL) { 234 | gifs(heatmapLinks.progressbars as string[], 'heatmap_gif'); 235 | } 236 | }, [heatmapLinks, hGifURL]); 237 | 238 | useEffect(() => { 239 | if (featuremapLinks.progressbars && hGifURL && !fGifURL) { 240 | gifs(featuremapLinks.progressbars as string[], 'featuremap_gif'); 241 | } 242 | }, [hGifURL, featuremapLinks, fGifURL]); 243 | 244 | useEffect(() => { 245 | if (top5) { 246 | const formattedTop5: Top5Obj = {}; 247 | for (const key in top5) { 248 | const formattedKey = formatString(key); 249 | formattedTop5[formattedKey] = top5[key]; 250 | } 251 | setTop5Formatted(formattedTop5); 252 | } 253 | }, [top5]); 254 | 255 | useEffect(() => { 256 | if (fGifURL) { 257 | 258 | let hGif = hGifURL; 259 | setHGifURL(''); 260 | setHGifURL(hGif) 261 | } 262 | } 263 | , [fGifURL, hGifURL]); 264 | 265 | const clearClick = () => { 266 | inputImage = undefined; 267 | setImgName('Browse...'); 268 | setImgURL(''); 269 | setFilePath(''); 270 | setHGifURL(''); 271 | setFGifURL(''); 272 | setVizState(false); 273 | setButtonState(0); 274 | setHeatmapLinks({} as Heatmaps); 275 | setFeaturemapLinks({} as Featuremaps); 276 | setPredictionName(''); 277 | setTime(0); 278 | }; 279 | 280 | const vizClick = () => { 281 | openViz(true); 282 | }; 283 | 284 | const closeViz = () => { 285 | openViz(false); 286 | }; 287 | 288 | return ( 289 |
    290 |
    291 |
    292 |
    293 |
    294 | {/* Your image and logo components here */} 295 |
    296 | {/*
    */} 297 | titleText 302 | {/*
    */} 303 |
    304 |
    305 |
    306 | {viz && ( 307 | 314 | )}{' '} 315 |
    316 | {imgURL && ( 317 | UploadedImage 322 | )} 323 |
    324 | {vizState && ( 325 | <> 326 |

    327 | Class: {formatString(predictionName)} 328 |

    329 | {time > 0 && ( 330 |

    331 | Time: {time.toFixed(2)} seconds 332 |

    333 | )} 334 | 335 | )} 336 |
    337 |
    338 | 345 | {!vizState && ( 346 | 349 | )} 350 | {vizState && ( 351 | 358 | )} 359 | 360 | 366 |
    367 |
    368 | {/*
    */} 369 |
    370 | {/*
    */} 371 |
    372 | {buttonState === 0 && ( 373 | 382 | )} 383 | {buttonState === 1 && ( 384 | 390 | )} 391 | {buttonState === 2 && ( 392 | 398 | )} 399 |
    400 |
    401 | ); 402 | } 403 | -------------------------------------------------------------------------------- /src/app/page.module.css: -------------------------------------------------------------------------------- 1 | .main { 2 | display: flex; 3 | flex-direction: row; 4 | justify-content: space-evenly; 5 | align-items: center; 6 | background-color: black; 7 | /* border: 1px solid white; */ 8 | /* padding: ; */ 9 | /* min-height: 100vh; */ 10 | /* background-image: url(/public/ai-generated-7914562_1920.jpg); */ 11 | } 12 | 13 | .Logo { 14 | display: flex; 15 | flex-direction: column; 16 | object-fit: contain; 17 | align-items: center; 18 | } 19 | 20 | .column { 21 | display: flex; 22 | flex-direction: column; 23 | flex: 20%; 24 | align-items: center; 25 | justify-content: center; 26 | /* border: 1px solid white; */ 27 | height: 30vh; 28 | } 29 | 30 | .majorDiv { 31 | display: flex; 32 | flex-direction: column; 33 | align-items: center; 34 | background-color: transparent; 35 | /* border: 1px solid white; */ 36 | height: 100px; 37 | width: 1vh; 38 | margin-left: 5%; 39 | } 40 | 41 | .title { 42 | font-size: xx-large; 43 | display: flex; 44 | flex-direction: row; 45 | /* align-items: row; */ 46 | background-color: #f3ec78; 47 | background-image: linear-gradient(15deg, #af4261, #f3ec78); 48 | background-size: 50%; 49 | -webkit-background-clip: text; 50 | -moz-background-clip: text; 51 | -webkit-text-fill-color: transparent; 52 | -moz-text-fill-color: transparent; 53 | /* border: 1px solid white; */ 54 | margin-bottom: 2%; 55 | } 56 | 57 | .titleImg { 58 | height: 200px; 59 | width: 300px; 60 | object-fit: contain; 61 | } 62 | 63 | .inputBox { 64 | display: flex; 65 | align-items: center; 66 | justify-content: center; 67 | height: 500px; /* This height must be enough to contain the fixed-size background */ 68 | /* width: 100%; Adjust width as needed */ 69 | /* border: 1px solid white; */ 70 | background-image: url(/backgroundFlareBW.avif); 71 | background-position: -6px; 72 | background-repeat: no-repeat; 73 | background-size: cover; /* or 'cover' if you want it to fill the entire container */ 74 | /* border-radius: 5px; */ 75 | /* background-image: linear-gradient( 76 | 130deg, 77 | hsl(28deg 52% 53%) 0%, 78 | hsl(26deg 51% 53%) 8%, 79 | hsl(24deg 48% 54%) 15%, 80 | hsl(21deg 46% 54%) 22%, 81 | hsl(19deg 44% 54%) 28%, 82 | hsl(17deg 41% 54%) 34%, 83 | hsl(15deg 38% 54%) 40%, 84 | hsl(13deg 35% 53%) 45%, 85 | hsl(11deg 32% 53%) 50%, 86 | hsl(9deg 29% 52%) 55%, 87 | hsl(7deg 26% 52%) 59%, 88 | hsl(6deg 25% 51%) 63%, 89 | hsl(5deg 24% 52%) 67%, 90 | hsl(4deg 23% 52%) 71%, 91 | hsl(4deg 22% 52%) 75%, 92 | hsl(3deg 21% 52%) 78%, 93 | hsl(3deg 20% 52%) 82%, 94 | hsl(2deg 19% 52%) 86%, 95 | hsl(2deg 18% 52%) 89%, 96 | hsl(1deg 17% 52%) 93%, 97 | hsl(1deg 16% 53%) 96%, 98 | hsl(0deg 15% 53%) 100% 99 | ); */ 100 | /* box-shadow: 0px 0px 10px white; */ 101 | /* height: 60vh; */ 102 | /* padding: 5%; */ 103 | width: 70vw; 104 | 105 | min-height: 500px; 106 | } 107 | 108 | .hide { 109 | /* background-image: url(/public/backgroundFlareBW.avif); */ 110 | display: none; 111 | } 112 | 113 | .grid-item { 114 | display: flex; 115 | flex-direction: column; /* Stack items vertically */ 116 | align-items: center; /* Center items horizontally */ 117 | justify-content: center; /* Center items vertically if there's extra space */ 118 | margin: 10px; /* Add some space around each grid item */ 119 | } 120 | 121 | .grid-item img { 122 | width: 100%; 123 | height: auto; 124 | } 125 | 126 | .gridItem div { 127 | margin-bottom: 10px; /* Space between title and image */ 128 | color: white; /* Assuming you want the text white as per modal theme */ 129 | font-size: 1em; /* Adjust based on your design */ 130 | } 131 | 132 | .placeholderText { 133 | font-size: 30px; 134 | font-weight: bolder; 135 | color: white; 136 | /* background-color: #f3ec78; 137 | background-image: linear-gradient( 138 | 180deg, 139 | hsl(0deg 100% 100%) 0%, 140 | hsl(17deg 100% 78%) 36%, 141 | hsl(26deg 83% 53%) 59%, 142 | hsl(15deg 95% 47%) 78%, 143 | hsl(2deg 80% 53%) 92%, 144 | hsl(0deg 75% 58%) 100% 145 | ); 146 | background-size: 100%; 147 | -webkit-background-clip: text; 148 | -moz-background-clip: text; 149 | -webkit-text-fill-color: transparent; 150 | -moz-text-fill-color: transparent; */ 151 | padding-top: 5%; 152 | margin-bottom: 2%; 153 | } 154 | 155 | .primaryBtn:active { 156 | color: white; 157 | background-color: rgba(100, 100, 100, 0.5); 158 | } 159 | 160 | .altBtn { 161 | display: inline-block; 162 | font-family: 'Hanken Grotesk', sans-serif; 163 | text-align: left; 164 | color: white; 165 | font-size: medium; 166 | background-color: transparent; 167 | border: none; 168 | border-radius: 50%; 169 | height: 100px; 170 | width: 100px; 171 | /* background-image: linear-gradient( 172 | 220deg, 173 | hsl(0deg 100% 100%) 0%, 174 | hsl(17deg 100% 78%) 18%, 175 | hsl(26deg 83% 53%) 59%, 176 | hsl(15deg 95% 47%) 78%, 177 | hsl(2deg 80% 53%) 92%, 178 | hsl(0deg 75% 58%) 100% 179 | ); */ 180 | padding: 10px; 181 | margin: 10px; 182 | } 183 | 184 | .altBtn:hover { 185 | color: white; 186 | /* background-color: rgba(200, 200, 200, 0.5); */ 187 | background-image: linear-gradient( 188 | 220deg, 189 | hsl(0deg 100% 100%) 0%, 190 | hsl(17deg 100% 78%) 18%, 191 | hsl(26deg 83% 53%) 59%, 192 | hsl(15deg 95% 47%) 78%, 193 | hsl(2deg 80% 53%) 92%, 194 | hsl(0deg 75% 58%) 100% 195 | ); 196 | } 197 | 198 | .altBtn:active { 199 | color: white; 200 | /* background-color: rgba(100, 100, 100, 0.5); */ 201 | background-image: linear-gradient( 202 | 220deg, 203 | hsl(0deg 100% 100%) 0%, 204 | hsl(17deg 100% 78%) 18%, 205 | hsl(26deg 83% 53%) 59%, 206 | hsl(15deg 95% 47%) 78%, 207 | hsl(2deg 80% 53%) 92%, 208 | hsl(0deg 75% 58%) 100% 209 | ); 210 | } 211 | 212 | .uploadImg { 213 | height: 150px; 214 | width: 150px; 215 | object-fit: contain; 216 | } 217 | 218 | .imgInput, 219 | .primaryBtn { 220 | display: flex; 221 | justify-content: center; 222 | align-items: center; 223 | text-align: center; 224 | border-radius: 50%; 225 | height: 11em; 226 | width: 11em; 227 | cursor: pointer; 228 | background-color: transparent; 229 | border: none; 230 | padding: 10px; 231 | font-family: 'Hanken Grotesk', sans-serif; 232 | font-size: medium; 233 | position: relative; 234 | } 235 | 236 | .imgInput:hover, 237 | .primaryBtn:hover { 238 | box-shadow: 0 0 0 5px white inset; 239 | } 240 | 241 | .browseText { 242 | color: white; 243 | } 244 | 245 | /* .imgInput { 246 | /* border: 1px solid #ccc; 247 | display: inline-block; 248 | cursor: pointer; 249 | color: white; 250 | border-radius: 50%; 251 | height: 100px; 252 | width: 100px; 253 | /* background-color: transparent; */ 254 | /* background-image: url(/backgroundFlareBW.avif); 255 | padding: 10px; 256 | margin: 10px; 257 | 258 | font-family: 'Hanken Grotesk', sans-serif; 259 | font-size: medium; 260 | text-align: center; 261 | } 262 | .imgInput:hover { 263 | color: rgb(255, 255, 255); 264 | background-color: rgba(200, 200, 200, 0.5); 265 | background-image: radial-gradient( 266 | #f5f5f5 100%, 267 | #6b6b6b 90%, 268 | rgb(42, 42, 42) 50%, 269 | rgb(0, 0%, 0%) 00%, 270 | ); 271 | background-image: linear-gradient( 272 | 220deg, 273 | hsl(0deg 100% 100%) 0%, 274 | hsl(17deg 100% 78%) 18%, 275 | hsl(26deg 83% 53%) 59%, 276 | hsl(15deg 95% 47%) 78%, 277 | hsl(2deg 80% 53%) 92%, 278 | hsl(0deg 75% 58%) 100% 279 | ); 280 | } */ 281 | 282 | /* .imgInput:active { 283 | color: white; 284 | background-color: rgba(100, 100, 100, 0.5); 285 | } */ 286 | 287 | .buttonBox { 288 | display: flex; 289 | flex-direction: column; /* Stack buttons vertically */ 290 | align-items: center; /* Center buttons horizontally */ 291 | justify-content: center; 292 | border-radius: 50%; /* Center the .buttonBox vertically within its container */ 293 | width: 100%; /* Take full width to center content properly */ 294 | } 295 | 296 | /* MODAL */ 297 | .exitmodalbutton { 298 | background-color: black; /* Green */ 299 | border: none; 300 | position: sticky; 301 | top: 5%; 302 | left: 95%; 303 | color: white; 304 | max-height: 1rem; 305 | max-width: 1rem; 306 | text-align: center; 307 | text-decoration: none; 308 | display: inline-block; 309 | font-size: 16px; 310 | /* margin: 4px 2px; */ 311 | z-index: 20; 312 | cursor: pointer; 313 | transition-duration: 0.4s; 314 | } 315 | 316 | .exitmodalbutton:hover { 317 | background-color: #e92a04; 318 | } 319 | 320 | .custom-button:hover { 321 | background-color: #45a049; 322 | } 323 | 324 | .modalclosed { 325 | display: none; 326 | } 327 | 328 | .modalcontainer { 329 | color: white; 330 | display: flex; 331 | flex-direction: column; 332 | align-items: center; 333 | justify-content: space-around; 334 | position: absolute; 335 | max-width: 80vw; 336 | max-height: 100vh; 337 | top: 55%; 338 | left: 50%; 339 | transform: translate(-50%, -50%); 340 | background-image: url(../../public/modal-background-1.jpg); 341 | background-size: cover; 342 | background-position: -15px center; 343 | z-index: 10; 344 | box-shadow: inset 0 0 16px 16px #000000; 345 | overflow: scroll; 346 | /* box-sizing: border-box; */ 347 | } 348 | 349 | .slide { 350 | border: 1px solid goldenrod; 351 | display: flex; 352 | flex-direction: column; 353 | align-items: center; 354 | justify-content: center; 355 | margin-top: 20px; 356 | position: relative; 357 | top: 0; 358 | background-color: rgba(0, 0, 0, 0.75); 359 | color: white; 360 | width: 80vw; 361 | min-height: 80vh; 362 | max-height: 80vh; 363 | /* padding: 20px; */ 364 | font-family: 'Roboto', sans-serif; 365 | } 366 | 367 | .slideRow { 368 | margin-top: 20px; 369 | border: 1px solid goldenrod; 370 | display: flex; 371 | flex-direction: row; 372 | align-items: center; 373 | justify-content: center; 374 | position: relative; 375 | top: 0; 376 | background-color: rgba(0, 0, 0, 0.5); 377 | color: white; 378 | width: 80vw; 379 | min-height: 80vh; 380 | max-height: 80vh; 381 | /* padding: 20px; */ 382 | font-family: 'Roboto', sans-serif; 383 | } 384 | 385 | .modalcontainertext { 386 | background-color: rgba(0, 0, 0, 0.5); 387 | color: white; 388 | width: 40vw; 389 | height: 45vh; 390 | display: flex; 391 | flex-direction: column; 392 | align-items: center; 393 | justify-content: center; 394 | padding: 20px; 395 | font-family: 'Roboto', sans-serif; 396 | } 397 | 398 | .modalcontainertext h1 { 399 | font-size: 1.75em; 400 | padding-bottom: 20px; 401 | padding-top: 20px; 402 | font-family: 'RobotoSlab', sans-serif; 403 | } 404 | 405 | .modalcontainertext h2 { 406 | font-size: 1.25em; 407 | } 408 | 409 | .modalcontainertext p { 410 | font-size: 1em; 411 | } 412 | 413 | /* Style for Chrome, Safari, and Opera */ 414 | .modalcontainer::-webkit-scrollbar { 415 | width: 12px; /* Adjust the width of the scrollbar */ 416 | } 417 | 418 | .modalcontainer::-webkit-scrollbar-track { 419 | background: #000; /* Track color */ 420 | } 421 | 422 | .modalcontainer::-webkit-scrollbar-thumb { 423 | background-color: #555; /* Thumb color */ 424 | border-radius: 6px; /* Roundness of the scrollbar thumb */ 425 | border: 3px solid #000; /* Creates padding around the scroll thumb */ 426 | } 427 | 428 | /* Style for Firefox */ 429 | .modalcontainer { 430 | scrollbar-width: thin; /* Can be 'auto', 'thin', or 'none' */ 431 | scrollbar-color: #555 #000; /* Thumb and track color */ 432 | } 433 | 434 | /* Hide scrollbar for IE, Edge, and Firefox (where applicable) */ 435 | .modalcontainer { 436 | -ms-overflow-style: none; /* IE and Edge */ 437 | scrollbar-width: none; /* Firefox */ 438 | } 439 | 440 | .modalcontainer::-webkit-scrollbar { 441 | display: none; /* Chrome, Safari, and Opera */ 442 | } 443 | 444 | .modalheader { 445 | display: flex; 446 | flex-direction: column; 447 | align-items: center; 448 | justify-content: space-between; 449 | text-align: center; 450 | } 451 | 452 | .modalbody { 453 | display: flex; 454 | flex-direction: column; 455 | align-items: center; 456 | text-align: center; 457 | justify-content: space-between; 458 | padding: 20px; 459 | } 460 | 461 | /* .modalcontainer { 462 | border: 1px solid blue; 463 | position: fixed !important; 464 | background: rgba(20, 20, 20, 0.98); 465 | color: white; 466 | z-index: 10; 467 | top: 50% !important; 468 | left: 50% !important; 469 | transform: translate(-50%, -50%); 470 | width: 80vw; 471 | height: 100vh; 472 | padding: 20px; 473 | overflow: none; 474 | display: flex; 475 | flex-direction: column; 476 | align-items: center; 477 | justify-content: space-between; 478 | border-radius: 5px; 479 | box-shadow: inset 0 0 30px 10px gray; 480 | } */ 481 | 482 | /* .modalcontents { 483 | border: 1px solid green; 484 | overflow: scroll; 485 | } */ 486 | 487 | .modaltitlecontainer { 488 | display: flex; 489 | flex-direction: row; 490 | align-items: center; 491 | justify-content: center; 492 | margin-bottom: 25px; 493 | width: 100%; 494 | padding: 10px; 495 | } 496 | 497 | .startJourneyButton { 498 | color: white; 499 | border: none; 500 | padding: 10px 20px; 501 | text-align: center; 502 | text-decoration: none; 503 | display: flex; 504 | align-items: center; 505 | justify-content: center; 506 | font-size: 16px; 507 | margin: 4px 2px; 508 | cursor: pointer; 509 | } 510 | 511 | .modaltitle { 512 | display: flex; 513 | align-items: center; 514 | position: absolute; 515 | font-size: 2.5em; 516 | font-weight: 1000; 517 | position: sticky; 518 | font-family: 'RobotoSlab', sans-serif; 519 | } 520 | 521 | /* CAROUSEL */ 522 | 523 | .embla { 524 | overflow: hidden; 525 | } 526 | .embla__container { 527 | display: flex; 528 | width: 50vw; 529 | height: 50vh; 530 | } 531 | .embla__slide { 532 | flex: 0 0 80%; 533 | min-width: 0; 534 | margin-right: 10px; 535 | margin-left: 10px; 536 | border: 1px solid white; 537 | } 538 | 539 | .embla__controls { 540 | display: flex; 541 | flex-direction: row; 542 | justify-content: center; 543 | align-items: center; 544 | } 545 | 546 | .embla__dots { 547 | display: flex; 548 | flex-wrap: wrap; 549 | justify-content: flex-end; 550 | align-items: center; 551 | /* margin-right: calc((2.6rem - 1.4rem) / 2 * -1); */ 552 | } 553 | 554 | .embla__dot { 555 | touch-action: manipulation; 556 | display: inline-flex; 557 | text-decoration: none; 558 | cursor: pointer; 559 | border: 0; 560 | padding: 0; 561 | margin: 0.5rem; 562 | width: 1.5rem; 563 | height: 1.5rem; 564 | display: flex; 565 | align-items: center; 566 | justify-content: center; 567 | border-radius: 50%; 568 | background-color: black; 569 | /* background-image: url(/public/ai-generated-7914562_1920.jpg); */ 570 | border: 0.2rem solid gray; 571 | } 572 | .embla__dot:hover { 573 | border-color: white; 574 | } 575 | .embla__dot--selected { 576 | touch-action: manipulation; 577 | display: inline-flex; 578 | text-decoration: none; 579 | cursor: pointer; 580 | border: 0; 581 | padding: 0; 582 | margin: 0.5rem; 583 | width: 1.5rem; 584 | height: 1.5rem; 585 | display: flex; 586 | align-items: center; 587 | justify-content: center; 588 | border-radius: 50%; 589 | background-color: white; 590 | border: 0.2rem solid white; 591 | } 592 | 593 | /* .test { 594 | border: 1px solid white; 595 | } */ 596 | 597 | .otherImg { 598 | height: 350px; 599 | width: 350x; 600 | object-fit: contain; 601 | } 602 | 603 | .table { 604 | width: 60%; /* Full-width */ 605 | border-collapse: collapse; /* Collapses the border */ 606 | margin: 20px 0; /* Adds some space around the table */ 607 | font-size: 1.1em; /* Adjusts the font size */ 608 | font-family: sans-serif; /* Uses a sans-serif font for a modern look */ 609 | min-width: 400px; /* Ensures it doesn’t get too small */ 610 | box-shadow: 0 0 20px rgba(0, 0, 0, 0.15); /* Adds a subtle shadow for depth */ 611 | } 612 | 613 | .table th, 614 | .table td { 615 | padding: 12px 15px; /* Adds space inside each cell */ 616 | text-align: left; /* Aligns text to the left */ 617 | border-bottom: 1px solid #dddddd; /* Adds a light line between rows */ 618 | } 619 | 620 | .table th { 621 | background-color: goldenrod; /* A pleasant green for the header background */ 622 | color: #ffffff; /* White text for contrast */ 623 | text-transform: uppercase; /* Makes the header text uppercase */ 624 | } 625 | 626 | .table tbody tr:last-of-type { 627 | border-bottom: 2px solid goldenrod; /* A thicker green line under the last row */ 628 | } 629 | 630 | .table th:first-child, 631 | .table td:first-child { 632 | width: 60%; /* Adjusts the width of the first column */ 633 | } 634 | 635 | .table th:nth-child(2), 636 | .table td:nth-child(2) { 637 | text-align: center; 638 | width: 40%; /* Adjusts the width of the second column, ensuring the total does not exceed 100% */ 639 | } 640 | .gifContainer { 641 | display: flex; 642 | flex-direction: column; 643 | align-items: center; 644 | justify-content: center; 645 | background-color: rgba(0, 0, 0, 0.5); 646 | } 647 | 648 | .gifDisplay { 649 | width: 100%; 650 | display: flex; 651 | flex-direction: row; 652 | align-items: center; 653 | justify-content: space-around; 654 | } 655 | 656 | .gifSlide { 657 | position: absolute; 658 | height: 100%; 659 | width: 100%; 660 | display: flex; 661 | flex-direction: column; 662 | align-items: center; 663 | justify-content: space-around; 664 | } 665 | 666 | .gifTitle { 667 | margin: 10%; 668 | font-size: x-large; 669 | font-weight: 400; 670 | } 671 | 672 | .gifDescription { 673 | background-color: rgba(0, 0, 0, 0.5); 674 | color: white; 675 | display: flex; 676 | flex-direction: column; 677 | align-items: center; 678 | justify-content: center; 679 | text-align: center; 680 | padding: 20px; 681 | margin-left: 5%; 682 | margin-right: 5%; 683 | font-family: 'Roboto', sans-serif; 684 | } 685 | --------------------------------------------------------------------------------