def train(hlayers, logpath, batch, epchs):
model = K.Sequential()
if len(hlayers) == 0:
model.add(K.layers.Dense(tgt.shape[1], activation=act, input_shape=(ipt.shape[1],)))
else:
model.add(K.layers.Dense(hlayers[0], activation=act, input_shape=(ipt.shape[1],)))
# model.add(K.layers.Dropout(0.01))
for hn in hlayers[1:]:
model.add(K.layers.Dense(hn, activation=act))
# model.add(K.layers.Dropout(0.01))
model.add(K.layers.Dense(tgt.shape[1], activation=act))
# model.add(K.layers.Dropout(0.01))
model.compile(loss='mean_squared_error', optimizer=opt)
tb = K.callbacks.TensorBoard(logpath)
K.utils.print_summary(model)
history = model.fit(ipt, tgt, batch_size=batch, epochs=epchs, verbose=0, callbacks=[tb], shuffle=True)
save_model(model, logpath, 'model')
return model
names = names[shuffled_idx]
for i in range(samples): # for every sample
batch_ipt = []
batch_tgt = []
for step in range(timesteps - channels - 1):
batch_ipt.append(signals[i, step:step + channels])
batch_tgt.append([signals[i, step + channels + 1]])
batch_ipt = np.array(batch_ipt)
batch_tgt = np.array(batch_tgt)
batch_ipt = np.expand_dims(batch_ipt, 0)
batch_tgt = np.expand_dims(batch_tgt, 0)
batch_loss = model.train_on_batch(batch_ipt, batch_tgt) ### TRAINNING
loss.append([time.time() - initial_time, batch_loss])
print('LOSS:', loss[-1][1], 'Time:', loss[-1][0])
save_model(model, path, 'model')
np.savetxt(path + 'loss.csv', np.array(loss), fmt='%.18e', delimiter=',')
P_model.set_weights(model.get_weights())
save_model(P_model, path, 'P_model')
for i in range(samples):
seed = np.array(signals[i][0:channels])
save_wav(max_amp * seed, name=path + 'seed_' + names[i] + '.wav')
seed = np.expand_dims(seed, 0)
seed = np.expand_dims(seed, 0)
pred_wave = np.zeros(timesteps)
for j in range(channels, timesteps):
predicted = P_model.predict(seed)
pred_wave[j] = predicted[0, 0, 0]
seed = np.roll(seed, -1)
seed[0, 0, channels - 1] = predicted[0, 0, 0]
activation=act,
kernel_initializer=init,
bias_initializer=init)(layer)
model = K.models.Model(input, output)
K.utils.print_summary(model)
model.compile(loss='mean_squared_error', optimizer=opt)
tb = K.callbacks.TensorBoard(path + '/amplitudes')
history = model.fit(ipt,
tgt,
batch_size=ipt.shape[0],
epochs=epcs,
verbose=1,
callbacks=[tb],
shuffle=True)
save_model(model, path + '/amplitudes/', 'model')
pred = np.reshape(model.predict(complete_ipt), (-1, partials)) * (M - m) + m
print(m, M)
X = np.reshape(complete_ipt[:, 1], (-1, partials))
plt.plot(X, amplitudes, 'ko')
plt.plot(X, pred, 'r.')
plt.savefig(path + 'fractions_of_max_amps.png')
plt.close()
np.savetxt(path + 'fractions_of_max_amps.csv', pred, delimiter=',')
activation=act,
kernel_initializer=init,
bias_initializer=init)(layer)
model = K.models.Model(input, output)
K.utils.print_summary(model)
model.compile(loss='mean_squared_error', optimizer=opt)
tb = K.callbacks.TensorBoard(path + '/decays')
history = model.fit(ipt,
tgt,
batch_size=ipt.shape[0],
epochs=epcs,
verbose=1,
callbacks=[tb],
shuffle=True)
save_model(model, path + '/decays/', 'model')
pred = np.reshape(model.predict(complete_ipt), (-1, partials)) * (M - m) + m
print(m, M)
np.savetxt(path + 'fractions_of_max_decays.csv', pred, delimiter=',')
X = np.reshape(complete_ipt[:, 1], (-1, partials))
plt.plot(X, decays, 'ko')
plt.plot(X, pred, 'r.')
plt.savefig(path + 'fractions_of_max_decays.png')
plt.close()
activation=act,
kernel_initializer=init,
bias_initializer=init)(layer)
model = K.models.Model(input, output)
K.utils.print_summary(model)
model.compile(loss='mean_squared_error', optimizer=opt)
tb = K.callbacks.TensorBoard(path + '/frequencies')
history = model.fit(ipt,
tgt,
batch_size=ipt.shape[0],
epochs=epcs,
verbose=1,
callbacks=[tb],
shuffle=True)
save_model(model, path + '/frequencies/', 'model')
# tgt = (tgt - m) / (M - m)
pred = np.reshape(model.predict(complete_ipt), (-1, partials)) * (M - m) + m
print(m, M)
plt.plot(frequencies, 'ko')
plt.plot(pred, 'r.')
plt.savefig(path + 'Mfreqs_over_Tfreqs.png')
plt.close()
np.savetxt(path + 'Mfreqs_over_Tfreqs.csv', pred, delimiter=',')