Documentation
- Create ML model & API (for information and health facilities)
- Convert ML model to tflite
- Deploy API to google cloud using Google app engine
- Create android project using android studio
- Import some library that we need
- Download and copy tflite model to asset folder in android studio
- Using tflite library to get data prediction
- Using retrofit to get data from API
Android
Versions Used
Android Studio : 4.2.1
Kotlin : 1.5.10
Target SDK : 30
Min SDK : 21
How to run in your own pc :
- Clone this branch using this command :
git clone -b front-end https://github.com/ahmadsufyan455/MedDev.git
-
Open in Android Studio
-
Sync project with gradle files (under file tab in Android Studio)
===============================================================================================
Machine Learning
- System Requirement
- Working on : Google Colab
- Tensorflow : 2.5.0
- Dataset : https://www.kaggle.com/jr2ngb/cataractdataset
the steps to create an ML model
- Check version tensorflow
- Import all required library
- Download Dataset
- Split Dataset
- Create image augmentation
- Create Model or using transfer learning
- Accuracy, loss, and score of model
- Visualize the plot
- Convert to TFlite
===============================================================================================
Google Cloud
Before you begin
- In the Google Cloud Console, on the project selector page, select or create a Google Cloud project.
- Make sure that billing is enabled for your Cloud project. Learn how to confirm that billing is enabled for your project.
- Enable the Cloud Build API.
- Enable the API
- Install and initialize the Cloud SDK.
Deploy and run Hello World on App Engine To deploy your app to the App Engine standard environment: how to deploy:
- gcloud app deploy <name_your_file.yaml>
- choose the region where the server is made, try to be close to your location
*Now your Application is ready
API nearby location
- First you have to enable Place API in the GPC API library
- After that klik in credential and make new key for you API you can find Credential in same menu of API Library