def viz_outs_grads(model, layer_idx=1):
x, y = make_data(K.int_shape(model.input), model.layers[2].units)
grads = get_layer_gradients(model, x, y, layer_idx=layer_idx)
kws = dict(n_rows=8, title_mode='grads')
show_features_1D(grads[0], show_borders=False, **kws)
show_features_2D(grads, norm=(-1e-4, 1e-4), **kws)
def test_misc(): # test miscellaneous functionalities
units = 6
batch_shape = (8, 100, 2*units)
reset_seeds(reset_graph_with_backend=K)
model = make_model(GRU, batch_shape, activation='relu', recurrent_dropout=0.3)
x, y = make_data(batch_shape, units)
model.train_on_batch(x, y)
grads = get_layer_gradients(model, x, y, layer_idx=1)
show_features_1D(grads, subplot_samples=True, tight=True, borderwidth=2)
show_features_1D(grads[0], subplot_samples=True)
show_features_2D(grads.T, n_rows=1.5, tight=True, borderwidth=2)
show_features_2D(grads.T[:, :, 0])
rnn_histogram(model, layer_idx=1, show_xy_ticks=[0, 0], equate_axes=2)
rnn_heatmap(model, layer_idx=1, cmap=None, normalize=True, show_borders=False)
rnn_heatmap(model, layer_idx=1, cmap=None, absolute_value=True)
rnn_heatmap(model, layer_idx=1, norm=None)
rnn_heatmap(model, layer_idx=1, norm=(-.004, .004))
from see_rnn.inspect_gen import get_layer, _detect_nans
get_layer(model, layer_name='gru')
get_rnn_weights(model, layer_idx=1, concat_gates=False, as_tensors=True)
rnn_heatmap(model, layer_idx=1, input_data=x, labels=y, mode='weights')
_test_prefetched_data(model)
# test NaN detection
nan_txt = _detect_nans(np.array([1]*9999 + [np.nan])).replace('\n', ' ')
print(nan_txt) # case: print as quantity
K.set_value(model.optimizer.lr, 1e12)
train_model(model, iterations=10)
rnn_histogram(model, layer_idx=1)
rnn_heatmap(model, layer_idx=1)
del model
reset_seeds(reset_graph_with_backend=K)
# test SimpleRNN & other
_model = make_model(SimpleRNN, batch_shape, units=128, use_bias=False)
train_model(_model, iterations=1) # TF2-Keras-Graph bug workaround
rnn_histogram(_model, layer_idx=1) # test _pretty_hist
K.set_value(_model.optimizer.lr, 1e50) # SimpleRNNs seem ridiculously robust
train_model(_model, iterations=20)
rnn_heatmap(_model, layer_idx=1)
data = get_rnn_weights(_model, layer_idx=1)
rnn_heatmap(_model, layer_idx=1, input_data=x, labels=y, data=data)
os.environ["TF_KERAS"] = '0'
get_rnn_weights(_model, layer_idx=1, concat_gates=False)
del _model
assert True
cprint("\n<< MISC TESTS PASSED >>\n", 'green')
def _test_outputs_gradients(model):
x, y = make_data(K.int_shape(model.input), model.layers[2].units)
name = model.layers[1].name
grads_all = get_layer_gradients(model, x, y, layer_name=name, mode='outputs')
grads_last = get_layer_gradients(model, x, y, layer_idx=2, mode='outputs')
kwargs1 = dict(n_rows=None, show_xy_ticks=[0, 0], show_borders=True,
max_timesteps=50, title_mode='grads')
kwargs2 = dict(n_rows=2, show_xy_ticks=[1, 1], show_borders=False,
max_timesteps=None)
show_features_1D(grads_all[0], **kwargs1)
show_features_1D(grads_all[:1], **kwargs1)
show_features_1D(grads_all, **kwargs2)
show_features_2D(grads_all[0], norm=(-.01, .01), show_colorbar=True, **kwargs1)
show_features_2D(grads_all, norm=None, reflect_half=True, **kwargs2)
show_features_0D(grads_last, marker='o', color=None, title_mode='grads')
show_features_0D(grads_last, marker='x', color='blue', ylims=(-.1, .1))
print('\n') # improve separation
def view_features(model, data, layer_num):
outs = get_layer_outputs(model, data, layer_idx=layer_num)
for i in range(len(data)):
plt.plot(data[i])
plt.show()
show_features_1D(outs[i:i + 1], n_rows=8, show_borders=False)
def viz_outs(model, layer_idx=1):
x, y = make_data(K.int_shape(model.input), model.layers[2].units)
outs = get_layer_outputs(model, x, layer_idx=layer_idx)
show_features_1D(outs[:1], n_rows=8, show_borders=False)
show_features_2D(outs, n_rows=8, norm=(-1, 1))
def _test_outputs(model):
x, _ = make_data(K.int_shape(model.input), model.layers[2].units)
outs = get_layer_outputs(model, x, layer_idx=1)
show_features_1D(outs[:1], show_y_zero=True)
show_features_1D(outs[0])
show_features_2D(outs)