Problem With Model Training

[ZenMaxe]

Hello i want fine tune a pre trained model but i get many
[!] train_step() retuned None outputs. Skipping training step. errors

my code:
https://github.com/karim23657/Persian-tts-coqui/blob/main/recepies/glowtts/train_glowtts.py

my dataset:
https://www.kaggle.com/datasets/magnoliasis/persian-tts-dataset-male

full output:

 Character '1' not found in the vocabulary. Discarding it.
100%|████████████████████████████████████████████████████████████████████████████| 15840/15840 [16:21<00:00, 16.14it/s]

 > TRAINING (2024-12-13 20:54:57)

   --> TIME: 2024-12-13 20:56:40 -- STEP: 0/495 -- GLOBAL_STEP: 0
     | > current_lr: 2.5e-07
     | > step_time: 58.4767  (58.47669196128845)
     | > loader_time: 44.8332  (44.83320760726929)

 [!] `train_step()` retuned `None` outputs. Skipping training step.
 [!] `train_step()` retuned `None` outputs. Skipping training step.
 [!] `train_step()` retuned `None` outputs. Skipping training step.
 [!] `train_step()` retuned `None` outputs. Skipping training step.
 [!] `train_step()` retuned `None` outputs. Skipping training step.
 [!] `train_step()` retuned `None` outputs. Skipping training step.
 [!] `train_step()` retuned `None` outputs. Skipping training step.
 [!] `train_step()` retuned `None` outputs. Skipping training step.
 [!] `train_step()` retuned `None` outputs. Skipping training step.
 [!] `train_step()` retuned `None` outputs. Skipping training step.
 > Keyboard interrupt detected.
 > Saving model before exiting...

[ZenMaxe]

new update

i used example recipe too and i got those warning or errors too
https://github.com/idiap/coqui-ai-TTS/blob/dev/recipes/ljspeech/glow_tts/train_glowtts.py