llSourcell / Image_classifier_template

Credits

Credits go to the FastAI team, Naveen Chanakya, and the Flutter team. I integrated a few different tutorials together to form a SaaS pipeline template. This is the code for this video on Youtube by Siraj Raval on building an image classification startup. There are 3 components here; A web API, a model training script, and a mobile app. The code in this repository is for the starter flutter app. Let's go through the 5 step process below. Pull Requests are always welcome!

Step 1: Find an Image Dataset

What is the image classification service you'd like to build? Once you decide, find a related dataset using these tools

• Dataset Search
• Awesome Datasets

Step 2: Transfer Learning

• Run this notebook on your local machine or upload and run it to colab. Replace the bear dataset with your own image dataset. It's retraining a 'resnet34' image classification model. This is transfer learning.
• Save the resulting model pkl file to google drive, save the download link.

Step 3: Signup for Firebase + Stripe

• Firebase
• Stripe

Step 4: Deploy the Web API

• Fork this repository.
• Follow the instructions in its README to deploy it to render
• Once deployed, check that it works.
• Then replace line 12 in 'server.py' of the web example with a link to your own classifier pkl file and re-deploy
• Make any cosmetic changes to the front-end inteface that you'd like

Step 5: Build the Mobile App

• Install Flutter here
• Download this code
• Open it in android studio as a new flutter project
• it will ask you to 'get' all dependencies, say yes and it'll will all be installed automatically
• Replace the default render link in 'main.dart' to the link to your deployed render app
• Notice the 2 functions for signup and login. This is where your stripe and firebase authentication code will be placed
• See this and this