tape.watch DenseNet121 on a layer's output throws ValueError #104521
Description
Issue type
Bug
Have you reproduced the bug with TensorFlow Nightly?
No
Source
binary
TensorFlow version
2.20.0
Custom code
Yes
OS platform and distribution
Ubuntu 25.10
Mobile device
No response
Python version
3.13.7
Bazel version
No response
GCC/compiler version
No response
CUDA/cuDNN version
No response
GPU model and memory
No response
Current behavior?
tape.watch(spatial_map_layer)
~~~~~~~~~~^^^^^^^^^^^^^^^^^^^
File "/usr/src/Python/pAIthon/.venv/lib/python3.13/site-packages/tensorflow/python/eager/backprop.py", line 873, in watch
for t in _extract_tensors_and_variables(tensor):
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^
File "/usr/src/Python/pAIthon/.venv/lib/python3.13/site-packages/tensorflow/python/eager/backprop.py", line 700, in _extract_tensors_and_variables
raise ValueError(f"Passed in object {obj} of type {type(obj).__name__!r}"
f", not tf.Tensor or tf.Variable or ExtensionType.")
ValueError: Passed in object <KerasTensor shape=(None, None, None, 1024), dtype=float32, sparse=False, ragged=False, name=keras_tensor_424> of type 'KerasTensor', not tf.Tensor or tf.Variable or ExtensionType.
Standalone code to reproduce the issue
base_model = DenseNet121(weights='models/densenet.hdf5', include_top=False)
print("Loaded DenseNet")
# add a global spatial average pooling layer
x = base_model.output
x = GlobalAveragePooling2D()(x)
# and a logistic layer
predictions = Dense(len(self._labels), activation="sigmoid")(x)
print("Added layers")
self._model = Model(inputs=base_model.input, outputs=predictions)
def get_weighted_loss(neg_weights, pos_weights, epsilon=1e-7):
def weighted_loss(y_true, y_pred):
# L(X, y) = −w * y log p(Y = 1|X) − w * (1 − y) log p(Y = 0|X)
# from https://arxiv.org/pdf/1711.05225.pdf
loss = 0
for i in range(len(neg_weights)):
loss -= (neg_weights[i] * y_true[:, i] * tf.math.log(y_pred[:, i] + epsilon) +
pos_weights[i] * (1 - y_true[:, i]) * tf.math.log(1 - y_pred[:, i] + epsilon))
loss = tf.math.reduce_sum(loss)
return loss
return weighted_loss
self._model.compile(
loss=get_weighted_loss(self._neg_weights, self._pos_weights),
optimizer=Adam(),
)
self._model.load_weights("models/pretrained_model.h5")
preprocessed_input = self._load_image_normalize(img_path, self._mean, self._std)
layer_name='conv5_block16_concat'
with tf.GradientTape() as tape:
predictions = self._model.predict(preprocessed_input)
spatial_map_layer = self._model.get_layer(layer_name).output
# Watch the intermediate layer's output
tape.watch(spatial_map_layer)