Issue Logging SetFitModel with MLflow

Models
1

I’m encountering an issue when trying to log a SetFitModel with MLflow. My SetFitModel is loaded from a .pth file using PyTorch’s torch.load() function. However, when I try to log the model with mlflow.pytorch.log_model() , I’m receiving a TypeError stating that the ‘pytorch_model’ argument should be a torch.nn.Module .

import torch
import mlflow

model_path = "/path/to/my/model.pth"
model_V5 = torch.load(model_path, map_location=torch.device('cpu'))
model_V5.to('cpu')

mlflow.start_run()
mlflow.pytorch.log_model(model_V5, "my_SetFit_model_V5")

Any guidance on how to log a SetFitModel with MLflow would be greatly appreciated.

Thank you in advance!

2 Likes
2

Did you find a way to log a SetFitModelto MLflow?