使用LaTeX-OCR提取公式后，latex2sympy转换报错求助

问题背景 #

用latex-ocr从图片提取数学公式后，尝试用latex2sympy在Python中处理该公式，出现解析错误。

运行代码 #

from PIL import Image
from pix2tex.cli import LatexOCR
from latex2sympy2 import latex2sympy, latex2latex

i = Image.open(myfile)
model = LatexOCR()
print(model(i))
# 输出结果为  \sum_{j}^{G}\sum_{l=1}^{M_{j,s p,i}}f_{j}\bigl(X,Y_{\dots}\bigr)^{\star\kappa}\,C F_{j}

t = r"\sum_{j}^{G}\sum_{l=1}^{M_{j,s p,i}}f_{j}\bigl(X,Y_{\dots}\bigr)^{\star\kappa}\,C F_{j}"
s = latex2sympy(t)

print(s)

报错信息 #

NoViableAltException: None
During handling of the above exception, another exception occurred:

LexerNoViableAltException                 Traceback (most recent call last)
LexerNoViableAltException: LexerNoViableAltException('\\')

During handling of the above exception, another exception occurred:

Exception                                 Traceback (most recent call last)
/usr/local/lib/python3.10/dist-packages/latex2sympy2.py in syntaxError(self, recog, symbol, line, col, msg, e)
    139         else:
    140             err = fmt % ("I don't understand this", self.src, marker)
--> 141         raise Exception(err)
    142 
    143 

Exception: I don't understand this
\sum_{j}^{G}\sum_{l=1}^{M_{j,s p,i}}f_{j}\bigl(X,Y_{\dots}\bigr)^{\star\kappa} C F_{j}

解决方案 #

latex2sympy对LaTeX语法的支持有限，需要调整提取后的LaTeX表达式，适配其解析规则：

• 替换扩展括号：\bigl( → (，\bigr) → )，latex2sympy不识别这类排版用的括号
• 替换省略号：\dots → \cdots，SymPy更兼容标准数学省略号
• 替换特殊符号：\star → *，latex2sympy无法解析\star作为乘法符号
• 合并变量空格：C F → CF，空格会被解析为变量分隔，导致识别为两个独立变量
• 修正下标空格：M_{j,s p,i} → M_{j,sp,i}，下标内的空格会引发解析错误

修改后的代码示例：

from latex2sympy2 import latex2sympy

# 适配latex2sympy的LaTeX字符串
t = r"\sum_{j}^{G}\sum_{l=1}^{M_{j,sp,i}}f_{j}(X,Y_{\cdots})^{* \kappa} CF_{j}"
s = latex2sympy(t)

print(s)
# 输出可用于Python处理的SymPy表达式

内容的提问来源于stack exchange，提问作者User202010