根据 matplotlib 设置散点的颜色

Question

Why are all points hotpink, please? Where is a mistake. The desired result is that each point is with different colour according to the first column. Thanks

file:

fer00001.txt     -30.5598   1  51183.7316 0.0    0.88884
her00034.txt     -12.9113   1  50124.7613 0.0    0.93370
occ00043.txt     -37.9350   1  50094.5721 0.0    0.94562
omc15564.txt     -9.53485   1  51576.4297 0.0    0.56777

Code:

import numpy as np  
import matplotlib.pyplot as plt
import pandas as pd

BVS, RV = np.loadtxt('file', unpack=True, usecols=[1, 5])
sp = np.loadtxt('file', unpack=True, dtype='str', usecols=[0])

kratke = [w[:3] for w in sp]

fig, ax = plt.subplots(figsize=[10,6.5])

string = [i.replace(',', ', ').replace('fer', 'FEROS').replace('her', 'HEROS').replace('occ', 'CES').replace('omc', 'RETICON') for i in kratke]

d = {'BVS': BVS, 'RV': RV, 'sp': string}
df = pd.DataFrame(data=d)

colors = {'FEROS': 'purple', 'HEROS': 'blue', 'CES': 'green', 'RETICON' : 'hotpink'}

for i in np.unique(df['sp']):
    color = colors[i]
    df1 = df[df['sp'] == i] 
    ax.scatter(df['BVS'], df['RV'], color=color, marker='o', s=6, label=i)

plt.show()

EDIT after advice

import numpy as np  
import matplotlib.pyplot as plt
import pandas as pd

BVS, RV = np.loadtxt('file', unpack=True, usecols=[1, 5])
sp = np.loadtxt('file', unpack=True, dtype='str', usecols=[0])

kratke = [w[:3] for w in sp]

fig, (ax1, ax2) = plt.subplots(nrows=1, ncols=2, figsize=[10,6.5])

string = [i.replace(',', ', ').replace('fer', 'FEROS').replace('her', 'HEROS').replace('occ', 'CES').replace('omc', 'RETICON') for i in kratke]

d = {'BVS': BVS, 'RV': RV, 'sp': string}
df = pd.DataFrame(data=d)

colors = {'FEROS': 'purple', 'HEROS': 'blue', 'CES': 'green', 'RETICON' : 'hotpink'}

for i in np.unique(df['sp']):
    color = colors[i]
    df1 = df[df['sp'] == i] 
    ax1.scatter(df1['BVS'], df1['RV'], color=color, marker='o', s=6, label=i)
    ax2.scatter(df1['BVS'][0:2], df1['RV'][0:2], color=color, marker='o', s=6, label=i)
plt.show()

Why slider does not work, please? All points are displayed again. In my original data, not points are displayed then.

Answer 1

• The main issue is ax.scatter(df['BVS'], df['RV'], color=color, marker='o', s=6, label=i) should be ax.scatter(df1['BVS'], df1['RV'], color=color, marker='o', s=6, label=i)
  • df was plotted instead of df1
• The same thing can be accomplished more easily with the following

import pandas as pd
import seaborn as sns  # high level API for matplotlib

# load the columns with pandas instead of numpy
df = pd.read_csv('file', sep='\\s+', header=None, usecols=[0, 1, 5])

# name the columns
df.columns = ['sp', 'BVS', 'RV']

# use only the first 3 values of sp
df['sp'] = df['sp'].str[:3]

# sorted list of values to map sp to
mappings = ['FEROS', 'HEROS', 'CES', 'RETICON']

# create a dict mapping the unique values in sp to mappings
mapped = dict(zip(sorted(df['sp'].unique()), mappings))

# map sp to the new values
df['sp'] = df['sp'].map(mapped)

# display(df)
        sp       BVS       RV
0    FEROS -30.55980  0.88884
1    HEROS -12.91130  0.93370
2      CES -37.93500  0.94562
3  RETICON  -9.53485  0.56777

p = sns.scatterplot(data=df, x='BVS', y='RV', hue='sp')
p.legend(bbox_to_anchor=(1.01, 1.02), loc='upper left')