Python for Scientific Research

Learn to run Python from command line, use the interactive shell, and run scripts saved as .py. Create programs with print, save with notepad, and execute to see hello world.

Explore control flow in Python by using conditional statements, loops, and function calls to drive program behavior, with examples of if-else, while loops, and comparisons.

Learn how to define and name a function in Python, pass arguments, and reuse the same function across calls.

Learn to work with modules in Python by creating packages from related modules, saving and importing models, and using built-in and installer package imports.

Learn Python file handling with the open function, using modes for reading, appending, writing, and creating files, including opening and closing a sample data file.

Learn how to manage directories in Python by using the os module to change the current working directory and list files in the home directory.

Import a crop health csv in pandas, inspect for missing values and outliers, fill gaps with means, clip ndVi to -1 and 1, and convert health to binary.

Apply NumPy for scientific computing by creating arrays, computing max, min, mean, and median, and analyzing correlations with random data.

Advance scientific computing with SciPy by describing data and performing interpolation and integration. Handle data formats MATLAB nets and IDF, import and generate X and Y data, and evaluate p-values.

Explore descriptive statistics in Python by computing mean, median, variance, standard deviation, and percentiles; analyze correlations, covariance, and linear regression using pandas and SciPy.

Learn to compute zonal statistics in Python by summarizing raster data within shapefile boundaries, producing metrics like mean, max, min, and standard deviation per zone.

Learn plotting in python with matplotlib and seaborn to create time series and two-variable plots, read data from github, and customize labels, titles, colors, and figure sizes in jupyter.

Explore creating heatmaps, scatterplots, 3D plots, and joint plots with seaborn in Python, including data cleaning, subplots, color mapping, and gender and happiness analyses.

Explore specialized plotting techniques in Python, including pie charts with palettes, animated scatter plots of life expectancy vs GDP per capita over time, and categorical plots using Titanic data.

Set up a gpu-enabled TensorFlow environment for machine learning and deep learning with conda, CUDA 11.3, cuDNN 8.2, verify GPU availability, and train a simple MNIST neural network.

Master geospatial data processing to detect and count plants with Python, using Geopandas, Rasterio, and OpenCV to read shapefiles, mask rasters, and extract plant coordinates.

Develop a python-based workflow to predict crop health from remote sensing data using ndvi, zonal statistics, and a three-layer neural network with TensorFlow and rasterio, geopandas, and numpy.

Remove outliers using the IQR method and group data by state to compute mean pollutant levels (SO2, NO2, SPM) with pandas. Impute missing values and derive AQI for state-level results.

Explore exploratory data analysis (EDA) of India's air quality data using visualizations, correlations, and regression models to forecast AQI with feature selection and cross-validation.

Explore how multicollinearity affects regression models for India's air quality. Apply variance inflation factor, cross-validation, and ridge regression within machine learning to detect, evaluate, and mitigate overfitting.

Explore geospatial and climate data analysis with Python using xray to work with multi-dimensional arrays, covering daily gridded observation data, precipitation, temperatures, and cmip6 climate projections.

Learn to use Xarray to open multiple data files, concatenate datasets, and save netcdf, then work with monthly and annual precipitation analyses for Kansas using masking, plotting, and region averages.

Explore interpolation between CPC club and grade met minimum air temperature datasets at different resolutions, convert longitude ranges, scale kelvin to celsius, and compute moving averages for Texas regions.

Calculate ndvi, ndbi, and nbr from Landsat 8 imagery in Kurdistan, Iraq, using Python; read red, near-infrared, and swir bands, export nbr as geotiff, and visualize with matplotlib.

Recap the key python for scientific research concepts: miniconda environments, jupyter notebook, numpy and scipy, file handling, zonal statistics, fourcastnet AI, and climate data case studies.