rospy.get_param gives strings instead of negative values in a list

edit

Hello,

I am trying to pass a matrix through a param.yaml file to be read by a python file. I am following the advice here with the exception that I am using rospy rather than roscpp. So I have formatted my .yaml file to include my 6x6 matrix as a flat list, however when I go to read it in and reshape it back into a matrix using rospy.get_param, I end up with a list full of floats for the positive values, and strings for the negative ones. I want to avoid having to use a for loop to go through each entry of the matrix and typecast these strings back into floats, but I'm at a loss for how else to get my matrix through the param server. How else can I get it to read my negative values as floats instead of strings? I've put a minimal example below.

the param.yaml file:

matrix: [1.0,0.0,0.0,0.0,0.0,0.0,  0.0,-1.9,0.0,0.0,0.0,0.0,  0.0,0.0,1.0,0.0,0.0,0.0,  0.0,0.0,1.0,0.0,0.0,0.0,  0.0,0.0,1.0,0.0,0.0,1.0,  0.0,0.0,1.0,0.0,1.0,1.0]

loaded through a launch file within the appropriate node:

<rosparam file='param.yaml' command='load'/>

called within the node as:

 if rospy.has_param('~matrix'):
    listmatrix=rospy.get_param('~matrix')
    toarray=np.array(listmatrix)
    self.matrix=toarray.reshape(6,6)
    rospy.loginfo(str(type(listmatrix[0])) + str(type(listmatrix[7])))
    rospy.loginfo(str(type(toarray[0])) + str(type(toarray[7])))
    rospy.loginfo(str(type(self.matrix[0,0])) + str(type(listmatrix[1,1])))

which returns

[INFO] [########.####, ########]: <class 'float'><class 'str'>
[INFO] [########.####, ########]: <class 'numpy.str_'><class 'numpy.str_'>
[INFO] [########.####, ########]: <class 'numpy.str_'><class 'numpy.str_'>

update: I've just checked in an attempt to debug, and it actually does this with any negative value I try to pass as a parameter, not just ones in lists.

second update: I checked on values I pass through either the launch file or the terminal, and they come back as floats, it is only ones loaded through .yaml files (and I fixed a typo, as per Mike's comment)