假设我绝对需要使用tex来生成我的数字文本,并且我使用以下代码:
#!/usr/bin/env python3
import matplotlib as mpl
mpl.use("PDF")
from matplotlib.backends.backend_pgf import FigureCanvasPgf
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
# https://stackoverflow.com/a/42768093/1878788
mpl.backend_bases.register_backend('pdf', FigureCanvasPgf)
TEX_PARAMS = {
"text.usetex": True, # use LaTeX to write all text
"pgf.rcfonts": False, # Ignore Matplotlibrc
"pgf.texsystem": "lualatex", # hoping to avoid memory issues
}
mpl.rcParams.update(TEX_PARAMS)
def main():
data = pd.DataFrame({
"A1": np.random.randn(4),
"B1": np.random.randn(4),
"A2": np.random.randn(4),
"B2": np.random.randn(4)})
data.plot.box()
plt.savefig("test_tex_boxplot.pdf")
if __name__ == "__main__":
exit(main())
这有效。
现在,如果我的DataFrame中的列名包含下划线(A_1
而不是A1
),则tex失败:
RuntimeError: LaTeX was not able to process the following string:
b'A_1'
我怎样才能解决这个问题?
一种解决方案是在调用绘图函数之前通过转义下划线来更改列名:
#!/usr/bin/env python3
import matplotlib as mpl
mpl.use("PDF")
from matplotlib.backends.backend_pgf import FigureCanvasPgf
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from re import sub
# https://stackoverflow.com/a/42768093/1878788
mpl.backend_bases.register_backend('pdf', FigureCanvasPgf)
TEX_PARAMS = {
"text.usetex": True, # use LaTeX to write all text
"pgf.rcfonts": False, # Ignore Matplotlibrc
"pgf.texsystem": "lualatex", # hoping to avoid memory issues
}
mpl.rcParams.update(TEX_PARAMS)
def main():
data = pd.DataFrame({
"A_1": np.random.randn(4),
"B_1": np.random.randn(4),
"A_2": np.random.randn(4),
"B_2": np.random.randn(4)})
usetex = mpl.rcParams.get("text.usetex", False)
if usetex:
data.columns = [sub("_", "\_", colname) for colname in data.columns]
data.plot.box()
plt.savefig("test_tex_boxplot.pdf")
if __name__ == "__main__":
exit(main())