Imported model cannot be loaded

[ericstheguy]

Was exporting a model beforehand. (Using a google colab notebook.)

Type: ValueError
Text: invalid literal for int() with base 10: 'checkpoints'
Full: Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 1950, in full_dispatch_request rv = self.dispatch_request() File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 1936, in dispatch_request return self.view_functionsrule.endpoint File "/content/Voice-Cloning-App/application/views.py", line 391, in download_model model_path = get_latest_checkpoint(os.path.join(paths["models"], model_name)) File "/content/Voice-Cloning-App/training/checkpoint.py", line 27, in get_latest_checkpoint if int(checkpoint.split("")[1].split(".")[0]) > int(latest_checkpoint.split("")[1].split(".")[0]): ValueError: invalid literal for int() with base 10: 'checkpoints'

[BenAAndrew]

Hi @ericstheguy, do you mean that you have exported a model from a colab notebook and want to use it in the app?

[ericstheguy]

Not really. Generally the error above displays whenever I try to do anything with the model. Like training.

[BenAAndrew]

@ericstheguy Do you start with an existing checkpoint that you've imported or is it from scratch?

[ericstheguy]

Was using a checkpoint made from scratch in colab. Trained to 10,000 iterations (Batch Size = 78)

[BenAAndrew]

@ericstheguy Could you upload the checkpoint to google drive and reply with the link so I can verify

[ericstheguy]

@ericstheguy Could you upload the checkpoint to google drive and reply with the link so I can verify

https://drive.google.com/file/d/1MjISOaMuEcsInYbTUJzoKFCCQHtsRy2f

[BenAAndrew]

@ericstheguy I can synthesize with the model. Are you trying to train or synthesize?

2021-05-01.12-24-58.mp4

[ericstheguy]

Figured it out, really strange. Just had to import the model (upload) into the app with a new name, and import the dataset with the same name as the model. All through the colab notebook.