At the era of large surveys, it becomes more and more challenging to study the galaxy evolution and to search specific targets for the traditional methods. New methods including stacking large amount of data and the machine learning algorithms are becoming more and more important. In this talk, I will discuss how we study the galaxy evolution from different scales. At the circumgalactic scale (~ 200 kpc), I will introduce how we study the circumgalactic mediun (CGM) using large spectroscopic surveys. The CGM harbors more than 80% of the total baryons in a galaxy, is both the reservoir of gas for subsequent star formation and the depository of chemically processed gas, energy, and angular momentum from feedback. At the galactic scale (~ few kpc), I will introduce the all-sky search for dwarf galaxies in the local Universe using the DESI Imaging Legacy Survey and WISE data via themachine learning technique. The local dwarf galaxies are key to understand the formation and evolution of galaxies.
