#Day9 of #DataAnalytics Finished a last section of learning #Pandas, and did extracting data with: - str.contain(' ', case=False) to make it case-insensitive - .head(n) to get the first n rows, usually paired with .value_counts - len(df[’col2’].unique()) / df[’col2’].nunique()

#Day10 of #DataAnalytics Started #Matplotlib visualization tool for Python! View: matplotlib.org/2.0.2/gallery.… to see the whole list of figures that can be done + source code (eg statistical plots & scientific figures) import matplotlib.pyplot as plt %matplotlib inline plt.plot()

#Day11 of #DataAnalytics I'm struggling with #Matplotlib because my kernel keeps restarting/dying whenever I try to import matplotlib... this was the same problem I faced the previous time when I was learning this too...

#Day14 of #DataAnalytics #Seaborns kdeplot - kernel density estimation. Idea is to replace each data point (represented by dashmark in rugplot) with a small Gaussian (Normal) distribution centered around that value, then summing the Gaussians for smooth estimate of the distributi

#Day15 of #DataAnalytics #Seaborns Place data in matrix form by .pivot_table() .heatmap to plot data in color-encoded matrices. annot=True for annotation of the values to be presented on the grid. cmap to change color variation VS .clustermap data grouped based on similarity