upload files

configs/finetune_speaker.json

@@ -1,7 +1,7 @@
 {
   "train": {
     "log_interval": 100,
-    "eval_interval": 200,
+    "eval_interval": 1000,
     "seed": 1234,
     "epochs": 10000,
     "learning_rate": 2e-4,

finetune_speaker.py

@@ -243,14 +243,14 @@ def train_and_evaluate(rank, epoch, hps, nets, optims, schedulers, scaler, loade
         utils.save_checkpoint(net_g, None, hps.train.learning_rate, epoch,
                               os.path.join(hps.model_dir, "G_latest.pth".format(global_step)))
         # utils.save_checkpoint(net_d, optim_d, hps.train.learning_rate, epoch, os.path.join(hps.model_dir, "D_{}.pth".format(global_step)))
-        old_g=os.path.join(hps.model_dir, "G_{}.pth".format(global_step-400))
+        old_g=os.path.join(hps.model_dir, "G_{}.pth".format(global_step-4000))
         # old_d=os.path.join(hps.model_dir, "D_{}.pth".format(global_step-400))
         if os.path.exists(old_g):
           os.remove(old_g)
         # if os.path.exists(old_d):
         #   os.remove(old_d)
     global_step += 1
-    if global_step == 4001:
+    if global_step == hps.n_steps + 1:
         exit()
 
   if rank == 0:

utils.py

@@ -152,6 +152,8 @@ def get_hparams(init=True):
                         help='JSON file for configuration')
     parser.add_argument('-m', '--model', type=str, default="pretrained_models",
                         help='Model name')
+    parser.add_argument('-n', '--n_steps', type=str, default="2000",
+                        help='finetune steps')
 
     args = parser.parse_args()
     model_dir = os.path.join("./", args.model)