How to fix incompatible shapes?

To fix incompatible shapes in NumPy, you need to ensure that the arrays you are trying to operate on have compatible dimensions. Here are some common strategies to resolve shape incompatibility:

1. Reshape Arrays: Use the reshape() method to change the shape of an array to make it compatible with another array.

   import numpy as np
   
   a = np.array([[1, 2], [3, 4]])  # Shape (2, 2)
   b = np.array([5, 6])             # Shape (2,)
   
   # Reshape b to (2, 1) to make it compatible for broadcasting
   b = b.reshape(2, 1)
   
   result = a + b  # Now compatible

2. Broadcasting: NumPy automatically expands the dimensions of smaller arrays to match the larger array's shape. Ensure that the dimensions are compatible for broadcasting.

   a = np.array([[1, 2], [3, 4]])  # Shape (2, 2)
   b = np.array([5, 6])             # Shape (2,)
   
   result = a + b  # Broadcasting works here

3. Check Dimensions: Use the shape attribute to check the dimensions of your arrays before performing operations.

   print(a.shape)  # Output: (2, 2)
   print(b.shape)  # Output: (2,)

4. Use np.newaxis: You can add a new axis to an array to make its shape compatible.

   b = np.array([5, 6])             # Shape (2,)
   b = b[:, np.newaxis]             # Now shape is (2, 1)
   
   result = a + b  # Now compatible

5. Concatenation: If you are trying to concatenate arrays, ensure they have the same number of dimensions along the concatenation axis.

   a = np.array([[1, 2], [3, 4]])  # Shape (2, 2)
   b = np.array([[5, 6]])           # Shape (1, 2)
   
   result = np.concatenate((a, b), axis=0)  # Now compatible

By following these strategies, you can resolve shape incompatibility issues in your NumPy operations.