I've read many posts on this site but not yet posted so I apologize for first-timer mistakes. I'm writing a class that has created a pandas dataframe to store information about an instance of the class. I call the dataframe _gamestats. Its index was a column called 'faces' and the one column is the values, or the 'weights'. I was not having any success trying to alter the value in the dataframe. I need to write a method to alter the value of the 'weight' of one of the 'faces.' To do this I though I could convert the dataframe to a dictionary, use the key to alter the value and then reconvert to a dataframe. However, I'm getting an error message that I do not understand. AttributeError: 'float' object has no attribute 'items' # method: change the weight of a single side: def change_facewt(self, face, nwt): self.nwt = nwt self.face = face x = len(self.faces) # check if nwt is int or float - OR castable | TypeError if not isinstance(nwt, (int, float)): try: nwt = float(nwt) except: raise TypeError("New weight must be numeric.") # change the weight of self.face with self.nwt stat_dict = self._gamestats.to_dict('index') stat_dict[self.face] = self.nwt self._gamestats = pd.DataFrame.from_dict(stat_dict, orient = 'index') print(self._gamestats()) Continue reading...
