# discretize.CurvilinearMesh.faces¶

property CurvilinearMesh.faces

Gridded face locations (staggered grid)

This property returns a numpy array of shape (n_faces, dim) containing gridded locations for all faces in the mesh (staggered grid). This is equivalent to calling np.r_[faces_x, faces_y, faces_z].

Returns
(`n_faces`, `dim`) `numpy.ndarray` `of` `float`

Gridded face locations (staggered grid)

Examples

Here, we provide an example of a minimally staggered curvilinear mesh. In this case, the x and y-faces have normal vectors that are primarily along the x and y-directions, respectively.

```>>> from discretize import CurvilinearMesh
>>> from discretize.utils import example_curvilinear_grid, mkvc
>>> from matplotlib import pyplot as plt
```
```>>> x, y = example_curvilinear_grid([10, 10], "rotate")
>>> mesh1 = CurvilinearMesh([x, y])
>>> faces = mesh1.faces
>>> x_faces = faces[:mesh1.n_faces_x]
>>> y_faces = faces[mesh1.n_faces_x:]
```
```>>> fig1 = plt.figure(figsize=(5, 5))
>>> mesh1.plot_grid(ax=ax1)
>>> ax1.scatter(x_faces[:, 0], x_faces[:, 1], 30, 'r')
>>> ax1.scatter(y_faces[:, 0], y_faces[:, 1], 30, 'g')
>>> ax1.legend(['Mesh', 'X-faces', 'Y-faces'], fontsize=16)
>>> plt.plot()
```

Here, we provide an example of a highly irregular curvilinear mesh. In this case, the y-faces are not defined by normal vectors along a particular direction.

```>>> x, y = example_curvilinear_grid([10, 10], "sphere")
>>> mesh2 = CurvilinearMesh([x, y])
>>> faces = mesh2.faces
>>> x_faces = faces[:mesh2.n_faces_x]
>>> y_faces = faces[mesh2.n_faces_x:]
```
```>>> fig2 = plt.figure(figsize=(5, 5))  