Medix-AI/coqui-tts
Medix-AI
/
coqui-tts
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

optional proirity_freq

This commit is contained in:
Eren Golge 2018-04-26 05:35:25 -07:00
parent 7bfdc32b7b
commit c894919120
2 changed files with 7 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
config.json
View File

@ -17,6 +17,7 @@
"eval_batch_size":32, "eval_batch_size":32,
"r": 5, "r": 5,
"mk": 1.0, "mk": 1.0,
"priority_freq": false,
"griffin_lim_iters": 60, "griffin_lim_iters": 60,
"power": 1.2, "power": 1.2,

8
train.py
View File

@ -54,6 +54,12 @@ pickle.dump(c, open(tmp_path, "wb"))
LOG_DIR = OUT_PATH LOG_DIR = OUT_PATH
tb = SummaryWriter(LOG_DIR) tb = SummaryWriter(LOG_DIR)
if c.priority_freq:
n_priority_freq = int(3000 / (c.sample_rate * 0.5) * c.num_freq)
print(" > Using num priority freq. : {}".format(n_priority_freq))
else:
print(" > Priority freq. is disabled.")
def signal_handler(signal, frame): def signal_handler(signal, frame):
"""Ctrl+C handler to remove empty experiment folder""" """Ctrl+C handler to remove empty experiment folder"""
@ -71,7 +77,6 @@ def train(model, criterion, data_loader, optimizer, epoch):
print(" | > Epoch {}/{}".format(epoch, c.epochs)) print(" | > Epoch {}/{}".format(epoch, c.epochs))
progbar = Progbar(len(data_loader.dataset) / c.batch_size) progbar = Progbar(len(data_loader.dataset) / c.batch_size)
n_priority_freq = int(3000 / (c.sample_rate * 0.5) * c.num_freq)
progbar_display = {} progbar_display = {}
for num_iter, data in enumerate(data_loader): for num_iter, data in enumerate(data_loader):
start_time = time.time() start_time = time.time()
@ -214,7 +219,6 @@ def evaluate(model, criterion, data_loader, current_step):
print("\n | > Validation") print("\n | > Validation")
progbar = Progbar(len(data_loader.dataset) / c.batch_size) progbar = Progbar(len(data_loader.dataset) / c.batch_size)
n_priority_freq = int(3000 / (c.sample_rate * 0.5) * c.num_freq)
with torch.no_grad(): with torch.no_grad():
for num_iter, data in enumerate(data_loader): for num_iter, data in enumerate(data_loader):
start_time = time.time() start_time = time.time()