Failed to import model trained on external device

additionally, the joblib version is 1.3.2

Sure, I can guide you on how to retrieve a stored model using joblib in QuantConnect. Here are the steps:

1. First, you need to check if the model key exists in the Object Store. You can do this by calling the ContainsKey method with the model key. This method returns a boolean that represents if the model_key is in the Object Store.

qb.ObjectStore.ContainsKey(model_key)

1. If the Object Store contains the model_key, you can proceed to get the file path of the stored model. Call the GetFilePath method with the key.

file_name = qb.ObjectStore.GetFilePath(model_key)

1. Now, you can load the saved model by calling the load method with the file path.

loaded_model = joblib.load(file_name)

Please replace model_key with the actual key you used when saving the model. If the Object Store does not contain the model_key, you need to save the model using the model_key before you proceed.

Here are the sources for your reference:

• Scikit-Learn
• XGBoost
• Hmmlearn
• Aesera

The joblib version shouldn't affect the process of loading the model from the Object Store in QuantConnect. The steps to load the model remain the same regardless of the joblib version. Here they are again for your reference:

1. Check if the model key exists in the Object Store:

qb.ObjectStore.ContainsKey(model_key)

1. If the Object Store contains the model_key, get the file path of the stored model:

file_name = qb.ObjectStore.GetFilePath(model_key)

1. Load the saved model:

loaded_model = joblib.load(file_name)

Please replace model_key with the actual key you used when saving the model. If the Object Store does not contain the model_key, you need to save the model using the model_key before you proceed.

If you're still having trouble, it might be due to compatibility issues between the joblib version used to save the model and the one used to load it. In that case, you might need to retrain and save your model using the same joblib version.

just for reference, joblib load works on other python objects

It's good to know that joblib load is working for other Python objects. This suggests that your joblib installation and basic functionality are working correctly.

If you're having trouble specifically with loading sklearn models, it could be due to a compatibility issue between the sklearn version used to train and save the model and the one used to load it.

As a troubleshooting step, you could try retraining and saving your model with the same sklearn and joblib versions that you're using to load the model.

If the issue persists, it might be helpful to check the specific error message you're getting when trying to load the sklearn model. This could provide more insight into what might be going wrong.