Victor Barros

FOR ACADEMIC USE ONLY!!!

This project aims to create a value forecasting model based on historical data using time series techniques.

Here are the technologies used in this project:

• Python with the following libraries:
  • os for system commands
  • pathlib for file system path manipulation
  • textwrap (useful for text formatting in command-line interfaces)
  • numpy for mathematical operations (e.g., sqrt, abs, ...)
  • pandas for CSV data manipulation
  • google.generativeai for utilizing Gemini
  • hashlib for secure hash calculation and message digests
  • user_data from Google Colab to retrieve the GOOGLE_API_KEY
  • display and Markdown from IPython for displaying formatted content
  • LinearRegression from sklearn for linear regression
  • data_table from Google Colab for table formatting

• Google Colab Notebook

Open Google Colab Notebooks: https://colab.research.google.com/drive/1ZqX9SHqcRmn2RPC2bVPPQwBufXyWZuiQ?authuser=0#scrollTo=Io4_kLs-URlW

Despite the small size of the time series dataset, both predictions were approximately correct. Our conclusion is that the provided data demonstrates the use of Google Gemini, but the low sample quantity may impact the classifier's accuracy.

• Utilize the following libraries:

  • tensorflow for neural networks
  • keras for neural networks

• Stream data acquisition

• Article: https://github.com/vicssb/Artificial-intelligence-in-INPE---2022/blob/master/Analise%20de%20dados%20estatisticos%20dos%20satelites%20Sentinel%201%20.pdf

• Repository: https://github.com/vicssb/lab_asset_forecast

  • In case of sensitive bugs like security vulnerabilities, please contact YOUR EMAIL directly instead of using issue tracker. We value your effort to improve the security and privacy of this project!

1.0.0.0

• Victor Barros:

• @vicssb (https://github.com/vicssb)
• [email protected]

Please follow github and join us! Thanks for visiting and happy coding!