def log_validation(self, reduced_loss, model, y, y_pred, iteration):
self.add_scalar("validation.loss", reduced_loss, iteration)
_, mel_outputs, gate_outputs, alignments = y_pred
mel_targets, gate_targets = y
# plot distribution of parameters
for tag, value in model.named_parameters():
tag = tag.replace('.', '/')
self.add_histogram(tag, value.data.cpu().numpy(), iteration)
# plot alignment, mel target and predicted, gate target and predicted
idx = random.randint(0, alignments.size(0) - 1)
self.add_image("alignment",
plot_alignment_to_numpy(
alignments[idx].data.cpu().numpy().T),
iteration,
dataformats='HWC')
self.add_image("mel_target",
plot_spectrogram_to_numpy(
mel_targets[idx].data.cpu().numpy()),
iteration,
dataformats='HWC')
self.add_image("mel_predicted",
plot_spectrogram_to_numpy(
mel_outputs[idx].data.cpu().numpy()),
iteration,
dataformats='HWC')
self.add_image("gate",
plot_gate_outputs_to_numpy(
gate_targets[idx].data.cpu().numpy(),
torch.sigmoid(
gate_outputs[idx]).data.cpu().numpy()),
iteration,
dataformats='HWC')
def log(self, y, y_pred, idx, iteration):
_, mel_outputs, gate_outputs, alignments = y_pred
mel_targets, gate_targets = y
self.add_image("alignment",
plot_alignment_to_numpy(
alignments[idx].data.cpu().numpy().T),
iteration,
dataformats='HWC')
self.add_image("mel_target",
plot_spectrogram_to_numpy(
mel_targets[idx].data.cpu().numpy()),
iteration,
dataformats='HWC')
self.add_image("mel_predicted",
plot_spectrogram_to_numpy(
mel_outputs[idx].data.cpu().numpy()),
iteration,
dataformats='HWC')
self.add_image("gate",
plot_gate_outputs_to_numpy(
gate_targets[idx].data.cpu().numpy(),
torch.sigmoid(
gate_outputs[idx]).data.cpu().numpy()),
iteration,
dataformats='HWC')
def sample_training(self, output, iteration):
mel_outputs = to_arr(output[0][0])
mel_outputs_postnet = to_arr(output[1][0])
alignments = to_arr(output[2][0]).T
# plot alignment, mel and postnet output
self.add_image("alignment", plot_alignment_to_numpy(alignments),
iteration)
self.add_image("mel_outputs", plot_spectrogram_to_numpy(mel_outputs),
iteration)
self.add_image("mel_outputs_postnet",
plot_spectrogram_to_numpy(mel_outputs_postnet),
iteration)
# save audio
try: # sometimes error
wav = inv_melspectrogram(mel_outputs)
wav /= max(0.01, np.max(np.abs(wav)))
wav_postnet = inv_melspectrogram(mel_outputs_postnet)
wav_postnet /= max(0.01, np.max(np.abs(wav_postnet)))
self.add_audio('pred', wav, iteration, hps.sample_rate)
self.add_audio('pred_postnet', wav_postnet, iteration,
hps.sample_rate)
except:
pass
def sample_training(self, output, target, iteration):
mel_outputs = to_arr(output[0][0])
mel_target = to_arr(target[0][0])
mel_outputs_postnet = to_arr(output[1][0])
alignments = to_arr(output[2][0]).T
# plot alignment, mel and postnet output
self.add_image("alignment_test", plot_alignment_to_numpy(alignments),
iteration)
self.add_image("mel_outputs_test",
plot_spectrogram_to_numpy(mel_outputs), iteration)
self.add_image("mel_outputs_postnet_test",
plot_spectrogram_to_numpy(mel_outputs_postnet),
iteration)
self.add_image("mel_target_test",
plot_spectrogram_to_numpy(mel_target), iteration)
# save audio
# try: # sometimes error
wav = inv_mel_spectrogram(mel_outputs, hps)
# wav *= 32767 / max(0.01, np.max(np.abs(wav)))
# wav /= max(0.01, np.max(np.abs(wav)))
wav_postnet = inv_mel_spectrogram(mel_outputs_postnet, hps)
# wav_postnet *= 32767 / max(0.01, np.max(np.abs(wav_postnet)))
# wav_postnet /= max(0.01, np.max(np.abs(wav_postnet)))
wav_target = inv_mel_spectrogram(mel_target, hps)
# wav_target *= 32767 / max(0.01, np.max(np.abs(wav_target)))
# wav_target /= max(0.01, np.max(np.abs(wav_target)))
self.add_audio('pred_test', wav, iteration, hps.sample_rate)
self.add_audio('pred_postnet_test', wav_postnet, iteration,
hps.sample_rate)
self.add_audio('target_test', wav_target, iteration, hps.sample_rate)
def log_spec(self, specs, olens, iteration, name=None, num=None):
for spec in specs:
for k in spec.keys():
s = spec[k].transpose(1, 2)[0][:, :olens[0]].cpu().data
if name is not None:
self.add_figure(name + '_' + num + '_' + k,
plot_spectrogram_to_numpy(s), iteration)
else:
self.add_figure(k, plot_spectrogram_to_numpy(s), iteration)
def log_training_vid(self, output, target, reduced_loss, grad_norm,
learning_rate, iteration):
mel_loss, mel_loss_post, l1_loss, gate_loss = reduced_loss
self.add_scalar("training.mel_loss", mel_loss, iteration)
self.add_scalar("training.mel_loss_post", mel_loss_post, iteration)
self.add_scalar("training.l1_loss", l1_loss, iteration)
self.add_scalar("training.gate_loss", gate_loss, iteration)
self.add_scalar("grad.norm", grad_norm, iteration)
self.add_scalar("learning.rate", learning_rate, iteration)
mel_outputs = to_arr(output[0][0])
mel_target = to_arr(target[0][0])
mel_outputs_postnet = to_arr(output[1][0])
alignments = to_arr(output[3][0]).T
# plot alignment, mel and postnet output
self.add_image("alignment", plot_alignment_to_numpy(alignments),
iteration)
self.add_image("mel_outputs", plot_spectrogram_to_numpy(mel_outputs),
iteration)
self.add_image("mel_outputs_postnet",
plot_spectrogram_to_numpy(mel_outputs_postnet),
iteration)
self.add_image("mel_target", plot_spectrogram_to_numpy(mel_target),
iteration)
# save audio
# try: # sometimes error
wav = inv_mel_spectrogram(mel_outputs, hps)
wav *= 32767 / max(0.01, np.max(np.abs(wav)))
# wav /= max(0.01, np.max(np.abs(wav)))
wav_postnet = inv_mel_spectrogram(mel_outputs_postnet, hps)
wav_postnet *= 32767 / max(0.01, np.max(np.abs(wav_postnet)))
# wav_postnet /= max(0.01, np.max(np.abs(wav_postnet)))
wav_target = inv_mel_spectrogram(mel_target, hps)
wav_target *= 32767 / max(0.01, np.max(np.abs(wav_target)))
# wav_target /= max(0.01, np.max(np.abs(wav_target)))
self.add_audio('pred', wav, iteration, hps.sample_rate)
self.add_audio('pred_postnet', wav_postnet, iteration, hps.sample_rate)
self.add_audio('target', wav_target, iteration, hps.sample_rate)
def train(args, hp, hp_str, logger, vocoder):
os.makedirs(os.path.join(hp.train.chkpt_dir, args.name), exist_ok=True)
os.makedirs(os.path.join(args.outdir, args.name), exist_ok=True)
os.makedirs(os.path.join(args.outdir, args.name, "assets"), exist_ok=True)
device = torch.device("cuda" if hp.train.ngpu > 0 else "cpu")
dataloader = loader.get_tts_dataset(hp.data.data_dir, hp.train.batch_size,
hp)
validloader = loader.get_tts_dataset(hp.data.data_dir, 1, hp, True)
idim = len(valid_symbols)
odim = hp.audio.num_mels
model = fastspeech.FeedForwardTransformer(idim, odim, hp)
# set torch device
model = model.to(device)
print("Model is loaded ...")
githash = get_commit_hash()
if args.checkpoint_path is not None:
if os.path.exists(args.checkpoint_path):
logger.info("Resuming from checkpoint: %s" % args.checkpoint_path)
checkpoint = torch.load(args.checkpoint_path)
model.load_state_dict(checkpoint["model"])
optimizer = get_std_opt(
model,
hp.model.adim,
hp.model.transformer_warmup_steps,
hp.model.transformer_lr,
)
optimizer.load_state_dict(checkpoint["optim"])
global_step = checkpoint["step"]
if hp_str != checkpoint["hp_str"]:
logger.warning(
"New hparams is different from checkpoint. Will use new.")
if githash != checkpoint["githash"]:
logger.warning(
"Code might be different: git hash is different.")
logger.warning("%s -> %s" % (checkpoint["githash"], githash))
else:
print("Checkpoint does not exixts")
global_step = 0
return None
else:
print("New Training")
global_step = 0
optimizer = get_std_opt(
model,
hp.model.adim,
hp.model.transformer_warmup_steps,
hp.model.transformer_lr,
)
print("Batch Size :", hp.train.batch_size)
num_params(model)
os.makedirs(os.path.join(hp.train.log_dir, args.name), exist_ok=True)
writer = SummaryWriter(os.path.join(hp.train.log_dir, args.name))
model.train()
forward_count = 0
# print(model)
for epoch in range(hp.train.epochs):
start = time.time()
running_loss = 0
j = 0
pbar = tqdm.tqdm(dataloader, desc="Loading train data")
for data in pbar:
global_step += 1
x, input_length, y, _, out_length, _, dur, e, p = data
# x : [batch , num_char], input_length : [batch], y : [batch, T_in, num_mel]
# # stop_token : [batch, T_in], out_length : [batch]
loss, report_dict = model(
x.cuda(),
input_length.cuda(),
y.cuda(),
out_length.cuda(),
dur.cuda(),
e.cuda(),
p.cuda(),
)
loss = loss.mean() / hp.train.accum_grad
running_loss += loss.item()
loss.backward()
# update parameters
forward_count += 1
j = j + 1
if forward_count != hp.train.accum_grad:
continue
forward_count = 0
step = global_step
# compute the gradient norm to check if it is normal or not
grad_norm = torch.nn.utils.clip_grad_norm_(model.parameters(),
hp.train.grad_clip)
logging.debug("grad norm={}".format(grad_norm))
if math.isnan(grad_norm):
logging.warning("grad norm is nan. Do not update model.")
else:
optimizer.step()
optimizer.zero_grad()
if step % hp.train.summary_interval == 0:
pbar.set_description(
"Average Loss %.04f Loss %.04f | step %d" %
(running_loss / j, loss.item(), step))
for r in report_dict:
for k, v in r.items():
if k is not None and v is not None:
if "cupy" in str(type(v)):
v = v.get()
if "cupy" in str(type(k)):
k = k.get()
writer.add_scalar("main/{}".format(k), v, step)
if step % hp.train.validation_step == 0:
for valid in validloader:
x_, input_length_, y_, _, out_length_, ids_, dur_, e_, p_ = valid
model.eval()
with torch.no_grad():
loss_, report_dict_ = model(
x_.cuda(),
input_length_.cuda(),
y_.cuda(),
out_length_.cuda(),
dur_.cuda(),
e_.cuda(),
p_.cuda(),
)
mels_ = model.inference(x_[-1].cuda()) # [T, num_mel]
model.train()
for r in report_dict_:
for k, v in r.items():
if k is not None and v is not None:
if "cupy" in str(type(v)):
v = v.get()
if "cupy" in str(type(k)):
k = k.get()
writer.add_scalar("validation/{}".format(k), v,
step)
mels_ = mels_.T # Out: [num_mels, T]
writer.add_image(
"melspectrogram_target_{}".format(ids_[-1]),
plot_spectrogram_to_numpy(
y_[-1].T.data.cpu().numpy()[:, :out_length_[-1]]),
step,
dataformats="HWC",
)
writer.add_image(
"melspectrogram_prediction_{}".format(ids_[-1]),
plot_spectrogram_to_numpy(mels_.data.cpu().numpy()),
step,
dataformats="HWC",
)
# print(mels.unsqueeze(0).shape)
audio = generate_audio(
mels_.unsqueeze(0), vocoder
) # selecting the last data point to match mel generated above
audio = audio.cpu().float().numpy()
audio = audio / (audio.max() - audio.min()
) # get values between -1 and 1
writer.add_audio(
tag="generated_audio_{}".format(ids_[-1]),
snd_tensor=torch.Tensor(audio),
global_step=step,
sample_rate=hp.audio.sample_rate,
)
_, target = read_wav_np(
hp.data.wav_dir + f"{ids_[-1]}.wav",
sample_rate=hp.audio.sample_rate,
)
writer.add_audio(
tag=" target_audio_{}".format(ids_[-1]),
snd_tensor=torch.Tensor(target),
global_step=step,
sample_rate=hp.audio.sample_rate,
)
##
if step % hp.train.save_interval == 0:
avg_p, avg_e, avg_d = evaluate(hp, validloader, model)
writer.add_scalar("evaluation/Pitch Loss", avg_p, step)
writer.add_scalar("evaluation/Energy Loss", avg_e, step)
writer.add_scalar("evaluation/Dur Loss", avg_d, step)
save_path = os.path.join(
hp.train.chkpt_dir,
args.name,
"{}_fastspeech_{}_{}k_steps.pyt".format(
args.name, githash, step // 1000),
)
torch.save(
{
"model": model.state_dict(),
"optim": optimizer.state_dict(),
"step": step,
"hp_str": hp_str,
"githash": githash,
},
save_path,
)
logger.info("Saved checkpoint to: %s" % save_path)
print("Time taken for epoch {} is {} sec\n".format(
epoch + 1, int(time.time() - start)))