Issue type
Bug
Have you reproduced the bug with TensorFlow Nightly?
No
Source
binary
TensorFlow version
2.18.0
Custom code
Yes
OS platform and distribution
Linux 22.04
Mobile device
N\A
Python version
3.9
Bazel version
N\A
GCC/compiler version
N\A
CUDA/cuDNN version
N\A
GPU model and memory
N\A
Current behavior?
I get an invalid shape error when running tf.keras.losses.BinaryCrossentropy
Standalone code to reproduce the issue
def test_binary_crossentropy_invalid_inputs(self):
# Test with invalid inputs
y_true = np.array([0, 1, 0, 1], dtype=np.float32)
y_pred = np.array([0.1, 0.9, 0.2], dtype=np.float32) # Mismatched shape
bce = tf.keras.losses.BinaryCrossentropy()
with self.assertRaises(ValueError):
bce(y_true, y_pred)Relevant log output
Traceback (most recent call last):
File "/home/user/projects/api_guided_testgen/out/bug_detect_gpt4o/exec/basic_rag_apidoc/tf/tf.keras.losses.BinaryCrossentropy.py", line 70, in test_binary_crossentropy_invalid_inputs
bce(y_true, y_pred)
File "/home/user/anaconda3/lib/python3.8/site-packages/keras/losses.py", line 152, in __call__
losses = call_fn(y_true, y_pred)
File "/home/user/anaconda3/lib/python3.8/site-packages/keras/losses.py", line 284, in call
return ag_fn(y_true, y_pred, **self._fn_kwargs)
File "/home/user/anaconda3/lib/python3.8/site-packages/tensorflow/python/util/traceback_utils.py", line 153, in error_handler
raise e.with_traceback(filtered_tb) from None
File "/home/user/anaconda3/lib/python3.8/site-packages/keras/losses.py", line 2176, in binary_crossentropy
backend.binary_crossentropy(y_true, y_pred, from_logits=from_logits),
File "/home/user/anaconda3/lib/python3.8/site-packages/keras/backend.py", line 5688, in binary_crossentropy
bce = target * tf.math.log(output + epsilon())
tensorflow.python.framework.errors_impl.InvalidArgumentError: {{function_node __wrapped__Mul_device_/job:localhost/replica:0/task:0/device:CPU:0}} Incompatible shapes: [4] vs. [3] [Op:Mul]
Issue type
Bug
Have you reproduced the bug with TensorFlow Nightly?
No
Source
binary
TensorFlow version
2.18.0
Custom code
Yes
OS platform and distribution
Linux 22.04
Mobile device
N\A
Python version
3.9
Bazel version
N\A
GCC/compiler version
N\A
CUDA/cuDNN version
N\A
GPU model and memory
N\A
Current behavior?
I get an invalid shape error when running
tf.keras.losses.BinaryCrossentropyStandalone code to reproduce the issue
def test_binary_crossentropy_invalid_inputs(self): # Test with invalid inputs y_true = np.array([0, 1, 0, 1], dtype=np.float32) y_pred = np.array([0.1, 0.9, 0.2], dtype=np.float32) # Mismatched shape bce = tf.keras.losses.BinaryCrossentropy() with self.assertRaises(ValueError): bce(y_true, y_pred)Relevant log output