mvaldenegro / keras-uncertainty Goto Github PK

Hi Matitas,

How are you ?

I have a few codes as follows:

def build_model():
model = merge_audio_video()

model.compile(optimizer="adam",
              loss='categorical_crossentropy',
              #loss=triplet_loss_adapted_from_tf,
              metrics=['accuracy', precision, recall, f1]
              )
from keras_uncertainty.models import DeepEnsembleClassifier
num_ensembles = 5
model = DeepEnsembleClassifier(model, num_ensembles)
return model

When using your code, there is one TypeError: call() missing 1 required positional argument: 'inputs'.
Can you help me? Thank you so much.

Dear Matitas,

I am trying to use the keras-uncertainty library but I am getting the following error.

"Exception has occurred: TypeError (note: full exception trace is shown but execution is paused at: tf__train_function)

Cannot convert a symbolic Keras input/output to a numpy array. This error may indicate that you're trying to pass a symbolic value to a NumPy call, which is not supported. Or, you may be trying to pass Keras symbolic inputs/outputs to a TF API that does not register dispatching, preventing Keras from automatically converting the API call to a lambda layer in the Functional Model."

I found this answer in stack overflow which solves the problem in the model structure you have in the regression_deep_ensemble example:
(https://stackoverflow.com/questions/65366442/cannot-convert-a-symbolic-keras-input-output-to-a-numpy-array-typeerror-when-usi)

In the first comment of the solution though it is mentioned that:
"That's not a solution, in my case I'm running on GPU a model with LSTM layer, once I disable the eager execution, another error come's to be LSTM cannot use GPU it's not respecting the criteria"

In my case I also use a structure with LSTM layers, and when I call the DeepEnsembleRegressor constructor, like DeepEnsembleRegressor(lambda: define_stacked_model(members), 5) (since my define_stacked_model also needs an argument).
My define_stacked_model is constructing a model with concatenate and when I am trying to fit the DeepEnsembleRegressor I get the following error:
"Calling Model.fit in graph mode is not supported when the Model instance was constructed with eager mode enabled. Please construct your Model instance in graph mode or call Model.fit with eager mode enabled."

Since the eager execution cancelation is not an option, do you have any idea what could solve the first error?

Thanks in advance!

Hi @mvaldenegro -
I returned to this repo to prepare some examples for a lecture series I am giving. When I went through your code, I stumbled to this line:
https://github.com/mvaldenegro/keras-uncertainty/blob/42f50a36c70003b16b7f343002766708ad2a289a/keras_uncertainty/models/DeepEnsembleClassifier.py#L219C1-L220C73

predictions = np.concatenate(predictions)
mean_pred = np.mean(predictions, axis=0)
mean_pred = mean_pred / np.sum(mean_pred, axis=1, keepdims=True)

according to the DeepEnsemble paper, one should average the predicted probabilities in the classification case. I guess, that is the line which instantiates the variable mean_pred. But I am a bit lost, why you "normalize" this again in the last line of the excerpt above. Can you comment please?

Hi @mvaldenegro , I had already installed your great library keras-uncertainty.
just I need a documentation for this library, because I do not know how can I get more information about my model , I used DeepEnsembleClassifie() and I want to know each class probability and std() like this :

Thank you.