如何更改分组条的默认分组颜色
How to change the default group color for grouped bars
我发现 的代码很有帮助，但在添加颜色信息以避免分组子栏中的栏自动颜色时出现错误。即获得重复的颜色，如果我在 plt.bar(....,color). color=['black', 'red', 'green', 'blue', 'cyan','brown','grey','goldenrod','lime','violet','indigo','coral','olive'].
中添加特定的“颜色”
在何处添加颜色细节，以便分组子栏的默认颜色及其图例根据需要是唯一的。
def grouped_barplot(df, cat,subcat, val , err):
u = df[cat].unique()
x = np.arange(len(u))
subx = df[subcat].unique()
offsets = (np.arange(len(subx))-np.arange(len(subx)).mean())/(len(subx)+1.)
width= np.diff(offsets).mean()
for i,gr in enumerate(subx):
dfg = df[df[subcat] == gr]
plt.bar(x+offsets[i], dfg[val].values, width=width,
label="{} {}".format(subcat, gr), yerr=dfg[err].values, capsize=5)
plt.xlabel("Test", fontsize=14)
plt.ylabel("value ",fontsize=14)
plt.xticks(x, u, fontsize=14)
plt.yticks((0,10,20,30,40,50,60,70,80,90,100), fontsize=14)
plt.legend(title = "ML",loc="upper center", bbox_to_anchor=(.5, 1.25), ncol=6, fontsize=12)
plt.show()
plt.title('comparision')
- 代码来自
requires 20 lines of code to do what can be done with 3 lines of code, as explained in .
- 测试于
python 3.8.12、pandas 1.3.4、matplotlib 3.4.3
导入和 DataFrame
import pandas as pd
import matplotlib as mpl
from matplotlib.patches import Patch
import matplotlib.pyplot as plt
import numpy as np
data = {'Candidate': ['X', 'Y', 'Z', 'X', 'Y', 'Z', 'X', 'Y', 'Z'],
'Sample_Set': [1, 1, 1, 2, 2, 2, 3, 3, 3],
'Values': [20, 10, 10, 200, 101, 99, 1999, 998, 1003],
'Error': [5, 2, 3, 30, 30, 30, 10, 10, 10]}
colors = ['black', 'red', 'green', 'blue', 'cyan','brown','grey','goldenrod','lime','violet','indigo','coral','olive']
df = pd.DataFrame(data)
df = df.sort_values(['Candidate', 'Sample_Set'])
# display(df)
Candidate Sample_Set Values Error
0 X 1 20 5
3 X 2 200 30
6 X 3 1999 10
1 Y 1 10 2
4 Y 2 101 30
7 Y 3 998 10
2 Z 1 10 3
5 Z 2 99 30
8 Z 3 1003 10
# reshape the dataframe into a wide format for Values
vals = df.pivot(index='Candidate', columns='Sample_Set', values='Values')
# display(vals)
Sample_Set 1 2 3
Candidate
X 20 200 1999
Y 10 101 998
Z 10 99 1003
# reshape the dataframe into a wide format for Errors
yerr = df.pivot(index='Candidate', columns='Sample_Set', values='Error')
# display(yerr)
Sample_Set 1 2 3
Candidate
X 5 30 10
Y 2 30 10
Z 3 30 10
更新答案
- 根据 OP 的评论，目标似乎是替换默认颜色，但各组的颜色应该相同。
- 使用新答案中的代码，只需将
colors 列表传递给 color 参数即可轻松完成此操作。
# plot vals with yerr
ax = vals.plot(kind='bar', yerr=yerr, logy=True, rot=0, figsize=(6, 5), ylabel='Value', title='Comparison', color=colors)
ax.legend(title='Sample Set', bbox_to_anchor=(1, 1.02), loc='upper left')
plt.show()
使用 OP
中引用的答案
for i, (gr, color) in enumerate(zip(subx, colors)): 允许将列表中的颜色分配给每个组。
def grouped_barplot(df, cat,subcat, val , err):
u = df[cat].unique()
x = np.arange(len(u))
subx = df[subcat].unique()
offsets = (np.arange(len(subx))-np.arange(len(subx)).mean())/(len(subx)+1.)
width= np.diff(offsets).mean()
colors = ['black', 'red', 'green', 'blue', 'cyan','brown','grey','goldenrod','lime','violet','indigo','coral','olive']
# add the colors to the loop
for i, (gr, color) in enumerate(zip(subx, colors)):
dfg = df[df[subcat].eq(gr)]
plt.bar(x+offsets[i], dfg[val], width=width, yerr=dfg[err], color=color, label=gr)
plt.legend(title='Sample_Set', bbox_to_anchor=(1, 1.02), loc='upper left')
plt.yscale('log')
plt.xlabel(cat)
plt.ylabel(val)
plt.xticks(x, u)
plt.show()
cat = "Candidate"
subcat = "Sample_Set"
val = "Values"
err = "Error"
grouped_barplot(df, cat, subcat, val, err )
原答案
- 请求可以执行，但不应该：
- 这不是一个很好的可视化实践。给定组中的条形应该都具有相同的颜色。可视化的重点是传达信息。这只会让情节变得难以阅读，从而违背初衷。
- 因为条形是按组绘制的：
'X'、'Y'、'Z' of 'Sample_Set 1'、'X'、'Y'、 'Sample_Set 2'的'Z'和'Sample_Set 3'的'X'、'Y'、'Z'，图例中只会创建3个label，即自定义需要创建具有适当句柄和标签的补丁图例。
绘图和自定义
rects的顺序和df的顺序不一样，所以df排序不同，为了zip正确的颜色到正确的[=37] =]
# add a colors column to the dataframe
df['color'] = colors[:len(df)]
# plot vals with yerr
ax = vals.plot(kind='bar', yerr=yerr, logy=True, rot=0, figsize=(6, 5), legend=False, ylabel='Value', title='Comparison')
# extract the Rectangle bar objects
rects = [c for c in ax.get_children() if isinstance(c, mpl.patches.Rectangle)]
# change the face color of the bar
for rect, color in zip(rects, df.sort_values(['Sample_Set', 'Candidate'])['color']):
rect.set_fc(color)
# create a custom handle for the legend
handles = list()
for idx, v in df.iterrows():
patch = Patch(color=v.color, label=f'{v.Candidate} {v.Sample_Set}')
handles.append(patch)
# add the legend
ax.legend(title='Candidate Sample_Set', handles=handles, bbox_to_anchor=(1, 1.02), loc='upper left')
plt.show()
使用
- 使用上面的
df。
def grouped_barplot(df, cat,subcat, val , err):
u = df[cat].unique()
x = np.arange(len(u))
subx = df[subcat].unique()
offsets = (np.arange(len(subx))-np.arange(len(subx)).mean())/(len(subx)+1.)
width= np.diff(offsets).mean()
# group the colors: range(3) because there are 3 groups
colors = [df.color[n::3] for n in range(3)]
for i, (gr, color) in enumerate(zip(subx, colors)):
dfg = df[df[subcat].eq(gr)]
plt.bar(x+offsets[i], dfg[val], width=width, yerr=dfg[err], color=color)
handles = list()
for idx, v in df.iterrows():
patch = Patch(color=v.color, label=f'{v.Candidate} {v.Sample_Set}')
handles.append(patch)
plt.legend(title='Candidate Sample_Set', handles=handles, bbox_to_anchor=(1, 1.02), loc='upper left')
plt.yscale('log')
plt.xlabel(cat)
plt.ylabel(val)
plt.xticks(x, u)
plt.show()
cat = "Candidate"
subcat = "Sample_Set"
val = "Values"
err = "Error"
# add a colors column to the dataframe
df['color'] = colors[:len(df)]
grouped_barplot(df, cat, subcat, val, err )
我发现 plt.bar(....,color). color=['black', 'red', 'green', 'blue', 'cyan','brown','grey','goldenrod','lime','violet','indigo','coral','olive'].
在何处添加颜色细节，以便分组子栏的默认颜色及其图例根据需要是唯一的。
def grouped_barplot(df, cat,subcat, val , err):
u = df[cat].unique()
x = np.arange(len(u))
subx = df[subcat].unique()
offsets = (np.arange(len(subx))-np.arange(len(subx)).mean())/(len(subx)+1.)
width= np.diff(offsets).mean()
for i,gr in enumerate(subx):
dfg = df[df[subcat] == gr]
plt.bar(x+offsets[i], dfg[val].values, width=width,
label="{} {}".format(subcat, gr), yerr=dfg[err].values, capsize=5)
plt.xlabel("Test", fontsize=14)
plt.ylabel("value ",fontsize=14)
plt.xticks(x, u, fontsize=14)
plt.yticks((0,10,20,30,40,50,60,70,80,90,100), fontsize=14)
plt.legend(title = "ML",loc="upper center", bbox_to_anchor=(.5, 1.25), ncol=6, fontsize=12)
plt.show()
plt.title('comparision')
- 代码来自
- 测试于
python 3.8.12、pandas 1.3.4、matplotlib 3.4.3
导入和 DataFrame
import pandas as pd
import matplotlib as mpl
from matplotlib.patches import Patch
import matplotlib.pyplot as plt
import numpy as np
data = {'Candidate': ['X', 'Y', 'Z', 'X', 'Y', 'Z', 'X', 'Y', 'Z'],
'Sample_Set': [1, 1, 1, 2, 2, 2, 3, 3, 3],
'Values': [20, 10, 10, 200, 101, 99, 1999, 998, 1003],
'Error': [5, 2, 3, 30, 30, 30, 10, 10, 10]}
colors = ['black', 'red', 'green', 'blue', 'cyan','brown','grey','goldenrod','lime','violet','indigo','coral','olive']
df = pd.DataFrame(data)
df = df.sort_values(['Candidate', 'Sample_Set'])
# display(df)
Candidate Sample_Set Values Error
0 X 1 20 5
3 X 2 200 30
6 X 3 1999 10
1 Y 1 10 2
4 Y 2 101 30
7 Y 3 998 10
2 Z 1 10 3
5 Z 2 99 30
8 Z 3 1003 10
# reshape the dataframe into a wide format for Values
vals = df.pivot(index='Candidate', columns='Sample_Set', values='Values')
# display(vals)
Sample_Set 1 2 3
Candidate
X 20 200 1999
Y 10 101 998
Z 10 99 1003
# reshape the dataframe into a wide format for Errors
yerr = df.pivot(index='Candidate', columns='Sample_Set', values='Error')
# display(yerr)
Sample_Set 1 2 3
Candidate
X 5 30 10
Y 2 30 10
Z 3 30 10
更新答案
- 根据 OP 的评论，目标似乎是替换默认颜色，但各组的颜色应该相同。
- 使用新答案中的代码，只需将
colors列表传递给color参数即可轻松完成此操作。
- 使用新答案中的代码，只需将
# plot vals with yerr
ax = vals.plot(kind='bar', yerr=yerr, logy=True, rot=0, figsize=(6, 5), ylabel='Value', title='Comparison', color=colors)
ax.legend(title='Sample Set', bbox_to_anchor=(1, 1.02), loc='upper left')
plt.show()
使用 OP
中引用的答案for i, (gr, color) in enumerate(zip(subx, colors)):允许将列表中的颜色分配给每个组。
def grouped_barplot(df, cat,subcat, val , err):
u = df[cat].unique()
x = np.arange(len(u))
subx = df[subcat].unique()
offsets = (np.arange(len(subx))-np.arange(len(subx)).mean())/(len(subx)+1.)
width= np.diff(offsets).mean()
colors = ['black', 'red', 'green', 'blue', 'cyan','brown','grey','goldenrod','lime','violet','indigo','coral','olive']
# add the colors to the loop
for i, (gr, color) in enumerate(zip(subx, colors)):
dfg = df[df[subcat].eq(gr)]
plt.bar(x+offsets[i], dfg[val], width=width, yerr=dfg[err], color=color, label=gr)
plt.legend(title='Sample_Set', bbox_to_anchor=(1, 1.02), loc='upper left')
plt.yscale('log')
plt.xlabel(cat)
plt.ylabel(val)
plt.xticks(x, u)
plt.show()
cat = "Candidate"
subcat = "Sample_Set"
val = "Values"
err = "Error"
grouped_barplot(df, cat, subcat, val, err )
原答案
- 请求可以执行，但不应该：
- 这不是一个很好的可视化实践。给定组中的条形应该都具有相同的颜色。可视化的重点是传达信息。这只会让情节变得难以阅读，从而违背初衷。
- 因为条形是按组绘制的：
'X'、'Y'、'Z'of'Sample_Set 1'、'X'、'Y'、'Sample_Set 2'的'Z'和'Sample_Set 3'的'X'、'Y'、'Z'，图例中只会创建3个label，即自定义需要创建具有适当句柄和标签的补丁图例。
绘图和自定义
rects的顺序和df的顺序不一样，所以df排序不同，为了zip正确的颜色到正确的[=37] =]
# add a colors column to the dataframe
df['color'] = colors[:len(df)]
# plot vals with yerr
ax = vals.plot(kind='bar', yerr=yerr, logy=True, rot=0, figsize=(6, 5), legend=False, ylabel='Value', title='Comparison')
# extract the Rectangle bar objects
rects = [c for c in ax.get_children() if isinstance(c, mpl.patches.Rectangle)]
# change the face color of the bar
for rect, color in zip(rects, df.sort_values(['Sample_Set', 'Candidate'])['color']):
rect.set_fc(color)
# create a custom handle for the legend
handles = list()
for idx, v in df.iterrows():
patch = Patch(color=v.color, label=f'{v.Candidate} {v.Sample_Set}')
handles.append(patch)
# add the legend
ax.legend(title='Candidate Sample_Set', handles=handles, bbox_to_anchor=(1, 1.02), loc='upper left')
plt.show()
使用
- 使用上面的
df。
def grouped_barplot(df, cat,subcat, val , err):
u = df[cat].unique()
x = np.arange(len(u))
subx = df[subcat].unique()
offsets = (np.arange(len(subx))-np.arange(len(subx)).mean())/(len(subx)+1.)
width= np.diff(offsets).mean()
# group the colors: range(3) because there are 3 groups
colors = [df.color[n::3] for n in range(3)]
for i, (gr, color) in enumerate(zip(subx, colors)):
dfg = df[df[subcat].eq(gr)]
plt.bar(x+offsets[i], dfg[val], width=width, yerr=dfg[err], color=color)
handles = list()
for idx, v in df.iterrows():
patch = Patch(color=v.color, label=f'{v.Candidate} {v.Sample_Set}')
handles.append(patch)
plt.legend(title='Candidate Sample_Set', handles=handles, bbox_to_anchor=(1, 1.02), loc='upper left')
plt.yscale('log')
plt.xlabel(cat)
plt.ylabel(val)
plt.xticks(x, u)
plt.show()
cat = "Candidate"
subcat = "Sample_Set"
val = "Values"
err = "Error"
# add a colors column to the dataframe
df['color'] = colors[:len(df)]
grouped_barplot(df, cat, subcat, val, err )