CODEWITHSUNDEEP
tensorflowkerasgoogle-colaboratorytransfer-learning
SashimiDélicieux
SashimiDélicieux

Reputation: 475

Keras EfficientNet transfer learning code example not working

My code were working perfectly for months but today I realized it's not working anymore. In my code, I just copy-pasted this keras code example : https://keras.io/examples/vision/image_classification_efficientnet_fine_tuning/#example-efficientnetb0-for-stanford-dogs

So "my" code looks like this :

import tensorflow as tf
import keras
from keras.layers import *
from keras import Sequential
from keras.layers.experimental import preprocessing
from keras import layers
from tensorflow.keras.applications import EfficientNetB0

img_augmentation = Sequential(
    [
        preprocessing.RandomRotation(factor=0.15),
        preprocessing.RandomTranslation(height_factor=0.1, width_factor=0.1),
        preprocessing.RandomFlip(),
        preprocessing.RandomContrast(factor=0.1),
    ],
    name="img_augmentation",
)

inputs = layers.Input(shape=(224, 224, 3))
x = img_augmentation(inputs)
outputs = EfficientNetB0(include_top=True, weights=None, classes=5)(x)

model = tf.keras.Model(inputs, outputs)
model.compile(
    optimizer="adam", loss="categorical_crossentropy", metrics=["accuracy"]
)

However, today when I run this cell in my colab, I get a lot of warnings like this :

WARNING:tensorflow:
The following Variables were used a Lambda layer's call (tf.compat.v1.nn.fused_batch_norm_422), but
are not present in its tracked objects:
  <tf.Variable 'top_bn/gamma:0' shape=(1280,) dtype=float32>
  <tf.Variable 'top_bn/beta:0' shape=(1280,) dtype=float32>
It is possible that this is intended behavior, but it is more likely
an omission. This is a strong indication that this layer should be
formulated as a subclassed Layer rather than a Lambda layer

And this error :

TypeError: 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 think google colab updated keras and tensorflow, now they are both version 2.5.0

How can I make my code works again ?

Upvotes: 0

Views: 707

Answers (2)

Innat
Innat

Reputation: 17219

You should not mix tf 2.x and standalone keras. You should import your libraries as follows, thus you won't get any issue.

import tensorflow as tf
from tensorflow.keras.layers import *
from tensorflow.keras import Sequential
from tensorflow.keras.layers.experimental import preprocessing
from tensorflow.keras import layers
from tensorflow.keras.applications import EfficientNetB0

img_augmentation = Sequential(
    [
        preprocessing.RandomRotation(factor=0.15),
        preprocessing.RandomTranslation(height_factor=0.1, width_factor=0.1),
        preprocessing.RandomFlip(),
        preprocessing.RandomContrast(factor=0.1),
    ],
    name="img_augmentation",
)

inputs = layers.Input(shape=(224, 224, 3))
x = img_augmentation(inputs)
outputs = EfficientNetB0(include_top=True, weights=None, classes=5)(x)

model = tf.keras.Model(inputs, outputs)

Upvotes: 1

SashimiD&#233;licieux
SashimiD&#233;licieux

Reputation: 475

So after some research, I realized it comes from the imports :

from tensorflow.keras.applications import EfficientNetB0

I don't know why but it does not throw any error, but breaks the entire code. Instead, I have to import :

from keras.applications.efficientnet import EfficientNetB0

And it works perfectly.

Upvotes: 0

Related Questions