hrushi11 / real-time-skin-type-detection Goto Github PK

Tensorflow versiom 2.16.1
keras version 3.3.3

error loading trying to load the model.

tried:
model = load_model()

def load_model:
model = tf.keras.models.load_model("Skin-Type-Recognition-Degraded")
return model

error:
File format not supported: filepath=Skin-Type-Recognition-Degraded. Keras 3 only supports V3 .keras files and legacy H5 format files (.h5 extension). Note that the legacy SavedModel format is not supported by load_model() in Keras 3. In order to reload a TensorFlow SavedModel as an inference-only layer in Keras 3, use keras.layers.TFSMLayer(Skin-Type-Recognition-Degraded, call_endpoint='serving_default') (note that your call_endpoint might have a different name).

tried:
model = load_model()

def load_model:
model = keras.layers.TFSMLayer("Skin-Type-Recognition-Degraded")
return model

error:
AttributeError: '_UserObject' object has no attribute 'add_slot'

More info:
Model Name is save.model.pb

Any help would be greatly appreciated.

Thanks