Load keras model h5 unknown metrics

Question

I have trained a keras CNN monitoring the metrics as follow:

METRICS = [
  TruePositives(name=\'tp\'),
  FalsePositives(name=\'fp\'),
  TrueNegatives(name=\'tn\'         


        

Answer 1

It looks like you are playing with a tensorflow tutorial. I also used these exact metrics and had the same problem. What worked for me was to load the model with compile = False and then compile it with the custom metrics. Then you should be able to use model.predict(....) as expected.

import keras

model = keras.models.load_model('model.h5', compile = False)

METRICS = [
      keras.metrics.TruePositives(name='tp'),
      keras.metrics.FalsePositives(name='fp'),
      keras.metrics.TrueNegatives(name='tn'),
      keras.metrics.FalseNegatives(name='fn'), 
      keras.metrics.BinaryAccuracy(name='accuracy'),
      keras.metrics.Precision(name='precision'),
      keras.metrics.Recall(name='recall'),
      keras.metrics.AUC(name='auc'),
]

model.compile(optimizer = keras.optimizers.Adam(learning_rate=1e-4),
              loss = 'binary_crossentropy',
              metrics = METRICS
             )