Value Error : target loss not found[Bug]

[Hafiz-ATHAR]

Describe the bug

Hi! I am trying to train a MultibandMelGAN model. I am using thorsten_de_v03 dataset. When epoch is finished, i get a warning "UserWarning: Detected call of lr_scheduler.step() before optimizer.step()." Then evaluation happens and After that : " ValueError: [!] Target loss not found in the keep_avg_target. You might be exiting the training loop before it is computed or set the target_loss in the model config incorrectly. "

I used thorsten's multiband mel gan recipe code. This issue also occurs with univnet.

To Reproduce

I ran : recipes/thorsten_DE/multiband_melgan/train_multiband_melgan.py

Expected behavior

No response

Logs

No response

Environment

{
    "CUDA": {
        "GPU": [
            "NVIDIA GeForce RTX 3090 Ti"
        ],
        "available": true,
        "version": "12.1"
    },
    "Packages": {
        "PyTorch_debug": false,
        "PyTorch_version": "2.1.2+cu121",
        "TTS": "0.22.0",
        "numpy": "1.22.0"
    },
    "System": {
        "OS": "Linux",
        "architecture": [
            "64bit",
            ""
        ],
        "processor": "x86_64",
        "python": "3.10.12",
        "version": "#101-Ubuntu SMP Tue Nov 14 13:30:08 UTC 2023"
    }
}

Additional context

I used dockerfile from dev branch but changed cuda image to cuda:12.2.2-base-ubuntu22.04.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. You might also look our discussion channels.

[Amith5970]

did you resolved this error?

[Naminwang]

The default target_loss is loss_0, loss_0 is discriminator loss, but default steps_to_start_discriminator is 200000 in UnivnetConfig and MultibandMelganConfig, so target loss not found error appear. And set target_loss='loss_1' in the UnivnetConfig or MultibandMelganConfig can avoid this error.

[Amith5970]

Are you able to run multiband melgan successfully ,because iam facing issues while saving the checkpoint