1

I'm using Pandas and am very new to programming. I'm plotting Energy Deposited (eDep) as a function of its x,y and z positions. So far, was successful in getting it to plot, but it won't let me plot the colormap beside my scatter plot! Any help is much appreciated

%matplotlib inline
import pandas as pd
import numpy as np

IncubatorBelow = "./Analysis.Test.csv"

df = pd.read_csv(IncubatorBelow, sep = ',', names['Name','TrackID','ParentID','xPos','yPos','zPos','eDep','DeltaE','Einit','EventID'],low_memory=False,error_bad_lines=False)
df["xPos"] = df["xPos"].str.replace("(","")
df["zPos"] = df["zPos"].str.replace(")","")

df.sort_values(by='Name', ascending=[False])
df.dropna(how='any',axis=0,subset=['Name','TrackID','ParentID','xPos','yPos','zPos','eDep','DeltaE','Einit','EventID'], inplace=True)

import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
df['xPos'] = df['xPos'].astype(float)
df['yPos'] = df['yPos'].astype(float)
df['zPos'] = df['zPos'].astype(float)
#df10[df10['Name'].str.contains("e-")]

threedee = plt.figure().gca(projection='3d')
threedee.scatter(df["xPos"], df["yPos"], df["zPos"], c=df["eDep"], cmap=plt.cm.coolwarm)
threedee.set_xlabel("x(mm)")
threedee.set_ylabel("y(mm)")
threedee.set_zlabel("z(mm)")
plt.show()

Heres what the plot looks like!

4D plot

Its from a particle physics simulation using GEANT4. The actual files are extremely large (3.7GB's that I've chunked into 40ish MB's) and this plot only represents a small fraction of the data.

Kaj
  • 89
  • 10
Konnor John
  • 11
  • 3
  • If anyone has any suggestions on how to better plot this stuff, would be much appreciated! I'm a noob, and started with Pandas and haven't really learnt a ton of Python yet... – Konnor John Mar 08 '20 at 02:58
  • This is difficult to interpret as a scatter plot. I think you should look into 3d kernel density estimation (KDE) methods. – sjc Mar 08 '20 at 03:09
  • @sjc ok, thanks so much! I'll definitely look into it. I was kinda going for a heat map esque plot but I guess I failed... – Konnor John Mar 08 '20 at 03:13
  • I think I'm going to drop one of my directional positions (probably Z) and then make this a heatmap in X and Y – Konnor John Mar 08 '20 at 04:27
  • It's not uncommon to see several 2d projections for these kinds of plots – sjc Mar 09 '20 at 03:21

0 Answers0