Step to run MultiLanguage Invoice Extractor in local/windows

• Clone google-gemini in your local machine
• Download and install Anaconda from https://www.anaconda.com/download
• Type anaconda on windows search and open anaconda command prompt
• Navigate to mcqgen progect (in step 1) from conda prompt and/by follow below commands
  • cd /google-gemini
  • conda create -n google-gemini python=3.11 -y
  • conda activate google-gemini
  • pip install -r requirement.txt
• Create a file with name '.env' in google-gemini folder
• Add below line in .env file
  • GOOGLE_API_KEY=""
• Run MultiLanguage Invoice Extractor with below command
  • streamlit run app.py --server.port 8080
  • Access the application on http://localhost:8080 from any of your favorite browser
  • Upload any png, jpg or jpeg invoice file of any language and in prompt ask any question e.g. what is the total bill?
  • Click on 'Tell me about the invoice' to extract the information.

Step to run MultiLanguage Invoice Extractor in AWS EC2

• Login to the AWS: https://aws.amazon.com/console/
• Search for EC2 service
• Choose UBUNTU Machine
• Launch EC2 instance
• Run below command to update teh OS
  • sudo apt update
  • sudo apt-get update
  • sudo apt upgrade -y
  • sudo apt install git curl unzip tar make sudo vim wget -y
  • sudo apt install python3-pip
  • git clone git@github.com:ThirdEyeInfo/google-gemini.git
  • cd google-gemini
  • pip3 install -r requirement.txt
• Add GOOGLE_API_KEY with value in .env file
  • vi .env
  • Click 'insert' button and add below line
  • GOOGLE_API_KEY=""
  • Click on 'escape' button and type ':wq!' to save the file and exit
• To run the application execute below command
  • python3 -m streamlit run app.py --server.port 8080
• To make 8080 port accessible from public domain navigate to EC2->Network & Security->Security Groups->Security group ID (click on your security group id)
• This will open a new page, from here click on 'Edit inbound rules' button
• This again will open a new page where you can click on 'Add rule' button and follow below instructions
  • Type == Custom TCP
  • Port range == 8080
  • Source == don't select anything from dropdown instead click on 'search bar' and select '0.0.0.0/0'
  • click on 'Save rules' button and the bottom of the page and access it using 'Public IPv4 address' which is located under EC2->Instances->(instance id)