Skip to content

Latest commit

 

History

History
61 lines (47 loc) · 2.92 KB

README.md

File metadata and controls

61 lines (47 loc) · 2.92 KB
Raw

LSD.a [Large Scale Data Analysis]

Data Analysis for large datasets and other adventures with kluless the computational marine scientist

| ><(((º> | >°)))彡 | ^ ˘ ω ˘ ^ | <コ:彡 | V=(° °)=V | ( ´・ω・)o |

LSD.a

📸 Image Data Aggregation and Analysis

This repository contains a collection of tools and scripts for aggregating image data and performing various analytics on it.

Overview

The project began with a series of CSV datasets containing image metadata, labels, and various environmental attributes. Our primary focus was to aggregate image data based on certain criteria, perform data cleaning, and integrate metadata for deeper insights. We also ventured into deriving additional columns from existing data and visualized relationships between different columns using linear regression models.

🌟 Features

🧹 Data Cleaning (｡◕‿◕｡)

  • Normalization: Our tools are designed to standardize datasets by removing specific prefixes. This ensures that data from different sources aligns well and can be compared or merged effectively.
  • Handling Missing Values: The tools can identify and manage missing or incomplete data, ensuring that the datasets are comprehensive and reliable.
  • Outlier Detection: Identify and manage unusual data points that might distort the analysis.

🎨 Data Aggregation (✿◠‿◠)

  • Flexible Grouping: Depending on the analysis needs, images can be grouped in various ways, such as in sets of 6 or 60. This flexibility allows for different levels of granularity in the analysis.
  • Interval-based Aggregation: Data can be aggregated based on specific intervals, providing a summarized view of larger datasets.

🔗 Data Merging (｡♥‿♥｡)

  • Common Column Merging: Different datasets can be seamlessly merged based on common columns like 'Name' and 'filename', ensuring data integrity.
  • Handling Column Variations: The tools can manage datasets with different column names, ensuring they are correctly aligned and merged.

📊 Linear Regression Plots (｡•́︿•̀｡)

  • Visual Insights: These plots provide a visual representation of the relationships between different factors, aiding in understanding correlations and trends.
  • Factor Analysis: Dive deep into specific factors, such as the relationship between %LICO and depth, to derive meaningful insights.

✨ Additional Data Derivation (｡>﹏<｡)

  • Column Operations: New columns can be derived based on operations between existing columns. For instance, calculating the seabed by summing depth and altitude provides additional data points for analysis.
  • Data Transformation: Transform data into more useful formats or units, enhancing the depth and breadth of the analysis.

Requirements

  • Python 3.x
  • Pandas
  • Matplotlib, Seaborn (for visualization)