Big Moments

Project for the Big Moments: Developer Edition GCP Challenge | #INSIDEGoogleCloud

Big Moments is a Google Photos Clone for the INSIDE Google Cloud ‘21 Challenge.

Broad improvements:

• Better UI
  
• Better object detection
  
• Better CI/CD in GCP
  
• Custom domain
  
  

🚀 Features

Search by the number of people in the picture. ⭐

Not enough results? What about these suggestions?

🔗 link broken?

Identify objects, places, logos, animal breeds, etc.

🔗 link broken?

The puppy is using an iPad 🐶!

The typos will not stop you!

Get results based on memes.

Search by the language you desire.

Search the image by the text they contain.

🏗 Infrastructure

🏁 Roadmap

• Create a bucket in Cloud Storage.
  
• Create a trigger in Cloud Build to build the Docker image on ./app changes.
  
  • Save the build image on Container Registry.
    
• Instance a VM with Compute Engine for the container imagen.
  
• Create a repo for the photo_analysis_service function in Cloud Source Repositories.
  
• Create a trigger to run Vision API tagging.
  
• Create a DB in Firestore to keep the pictures url their related tags.
  
  

🎉 Extra

• Detect more objects with Vision AI.
  
• Get the number of people in a picture with Vision AI.
  
• Analyze the user query with Cloud Natural Language.
  
  • Suggest results based on the analysis.
    
• Add multilingual support with Cloud Translation.
  

• [x] Add a custom domain.

  • Add an HTTPS Load Balancing.
    
    

• [x] Update function from Cloud Functions.

  gcloud functions deploy photo_analysis_service \
     --source=https://source.developers.google.com/projects/project-id/repos/repo-name/moveable-aliases/main/paths/function/ \
     --runtime=python39 --trigger-resource=big-moments-bucket \
     --trigger-event=google.storage.object.finalize \
     --set-env-vars BUCKET=big-moments-bucket

• [x] Update container images from Compute Engine VM instance.

  gcloud compute instances update-container instance-name --zone us-central1-a \
     --container-image=gcr.io/project-id/repo@sha256:hash

Dijkstra’s shortest path algorithm

Simple (really simple) web application that plots the shortest path obtained from the Dijkstra’s algorithm.

Table of Contents

• Dijkstra’s shortest path algorithm
  
• Table of Contents
  
  • Install🚀
    
  • About🤔
    
  • To mess around you need🛠
    
  • To do📝
    
  • Contributing🤝🏼
    
  • The Team🙌🏼
    
  • License📄
    
    

Install🚀

1. Fork and clone the Dijkstra’s shortest path repo

   git clone https://github.com/TanZng/dijkstras-shortest-path.git

2. Create your virtualenv on the proyect root

   virtualenv flask_env

3. Activate virtualenv

   # Windows
   cd flask_env/Scripts
   activate

   # MacOS, Ubuntu, Debian
   cd flask_env/bin
   source activate

To leave virtualenv we use deactivate

1. Run Flask App

   cd flaskr/
   python app.py

2. Open the direction http://127.0.0.1:5000/ on your browser! 🥳

About🤔

WIP (Work in Progress)

To mess around you need🛠

Git

Python 3

I’m using 3.8.3rc1, but i think > 3.4 its okay

pip

To download other dependencies. I’m using pip 20.1.1

python -m pip install --upgrade pip

Pandas

I’m using 1.0.5

pip install pandas

Flask

I’m at 1.1.2

pip install Flask

NetworkX

pip install networkx

Plotly

pip install plotly==4.9.0

virtualenv

pip install virtualenv

To do📝

• Load and Read graphs on txt files.
  
• Render different graphs from a file.
  
• Set a default example.
  
• Form to choose u (star) and v (end).
  
• Create a simple README.md
  
• Render shortest path - edge trace - with a diferent color.
  
• Upload Dijkstra’s shortest path algorithm.
  
• Upload the web app.
  
• Add “How to use” on the README.
  
• Add simple CONTRIBUTING.md
  
• Make it responsive.
  
• Add CSS Style.
  
• Write a Post about this.
  
  

Contributing🤝🏼

Feel free to contribute. I am open to suggestion. Feel free to create an issue or open a PR

WIP (Work in Progress)

Please read CONTRIBUTING.md for details of our code of conduct, and the process to send us pull requests.

The Team🙌🏼

Tania R. Zuñiga - Initial Work - TanZng

Becris Icon designer

License📄

• MIT license
  
• Copyright 2020 © Tania R Zuniga.