Plot weights of convolutional layer in Keras

plot_conv_weights.py

%matplotlib inline
import pylab as pl
import matplotlib.cm as cm
import numpy.ma as ma
def make_mosaic(imgs, nrows, ncols, border=1):
    """
    Given a set of images with all the same shape, makes a
    mosaic with nrows and ncols
    """

    nimgs = imgs.shape[0]
    imshape = imgs.shape[1:]
    
    mosaic = ma.masked_all((nrows * imshape[0] + (nrows - 1) * border,
                            ncols * imshape[1] + (ncols - 1) * border),
                            dtype=np.float32)
    
    paddedh = imshape[0] + border
    paddedw = imshape[1] + border
    for i in xrange(nimgs):
        row = int(np.floor(i / ncols))
        col = i % ncols
        
        mosaic[row * paddedh:row * paddedh + imshape[0],
               col * paddedw:col * paddedw + imshape[1]] = imgs[i]
    return mosaic
# utility functions
from mpl_toolkits.axes_grid1 import make_axes_locatable

def nice_imshow(ax, data, vmin=None, vmax=None, cmap=None):
    """Wrapper around pl.imshow"""
    if cmap is None:
        cmap = cm.jet
    if vmin is None:
        vmin = data.min()
    if vmax is None:
        vmax = data.max()
    divider = make_axes_locatable(ax)
    cax = divider.append_axes("right", size="5%", pad=0.05)
    im = ax.imshow(data, vmin=vmin, vmax=vmax, interpolation='nearest', cmap=cmap)
    pl.colorbar(im, cax=cax)
#pl.imshow(make_mosaic(np.random.random((9, 10, 10)), 3, 3, border=1))

def plot_conv_weights(model, layer):
    # Visualize weights
    W = model.layers[layer].W.get_value(borrow=True)
    W = np.squeeze(W)

    if len(W.shape) == 4:
        W = W.reshape((-1,W.shape[2],W.shape[3]))
    print("W shape : ", W.shape)

    pl.figure(figsize=(15, 15))
    pl.title('conv weights')
    s = int(np.sqrt(W.shape[0])+1)
    nice_imshow(pl.gca(), make_mosaic(W, s, s), cmap=cm.binary)

# usage
plot_conv_weights(model, layer=2)

It seems that W is deprecated (Keras 2.0+). How should I change it?

i came out with this code, it takes model, and the layer name

def plot_conv_weights(model, layer):
    W = model.get_layer(name=layer).get_weights()[0]
    if len(W.shape) == 4:
        W = np.squeeze(W)
        W = W.reshape((W.shape[0], W.shape[1], W.shape[2]*W.shape[3])) 
        fig, axs = plt.subplots(5,5, figsize=(8,8))
        fig.subplots_adjust(hspace = .5, wspace=.001)
        axs = axs.ravel()
        for i in range(25):
            axs[i].imshow(W[:,:,i])
            axs[i].set_title(str(i))