Basic Examples (2) 

Compute the determinant of a matrix in NumPy:

In[1]:=

In[2]:=

Out[2]=

Or using the built-in Wolfram Language function:

In[3]:=

Out[3]=

Scope (6) 

Compute the determinant of a real-valued matrix:

In[4]:=

Out[4]=

In[5]:=

Out[5]=

Sparse array:

In[6]:=

Out[6]=

NumericArray object:

In[7]:=

Out[7]=

Complex-valued array:

In[8]:=

Out[8]=

A tensor representing a list of matrices:

In[9]:=

Out[9]=

In[10]:=

Out[10]=

In[11]:=

Out[11]=

Compare with the built-in Det:

In[12]:=

Out[12]=

Make several calls to NumPyDet in the same external session:

In[13]:=

Out[13]=

In[14]:=

Out[14]=

In[15]:=

Out[15]=

End the session:

In[16]:=

Properties and Relations (3) 

For relatively large matrices, the determinant cannot be computed using machine arithmetic:

In[17]:=

In[18]:=

Out[18]=

Use the resource function NumPySignLogDet instead:

In[19]:=

Out[20]=

In[21]:=

Out[21]=

Or the built-in Det:

In[22]:=

Out[22]=

Possible Issues (3) 

Automatic conversion of the input array to a NumericArray object can fail:

In[23]:=

Out[23]=

In[24]:=

Out[24]=

Convert the array to a NumericArray before passing it to NumPyDet:

In[25]:=

Out[25]=

A call to NumPyDet on an arbitrary precision array fails:

In[26]:=

Out[26]=

In[27]:=

Out[27]=

Convert the array to a NumericArray before passing it to NumPyDet:

In[28]:=

Out[28]=

Or use the machine-precision array:

In[29]:=

Out[29]=

NumPyDet does not issue a warning when precision is lost:

In[30]:=

In[31]:=

Out[31]=

Compare with the built-in Det computed with machine precision:

In[32]:=

Out[32]=

Obtain a more accurate result by computing the determinant with a higher precision using the built-in Det:

In[33]:=

Out[33]=

Verify the result by using still higher precision:

In[34]:=

Out[34]=