Setting up a model with
inputs = tf.keras.Input(shape=imageShape)
x = inputs
x = layers.RandomFlip('horizontal')(x)
x = layers.RandomRotation(0.2)(x)
x = layers.Dropout(0.2)(x)
outputs = layers.Dense(1)(x)
Produces this output
2021-12-06 18:19:57.709106: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz
2021-12-06 18:19:57.709239: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
4/4 [==============================] - 1s 99ms/step - loss: 0.6075 - accuracy: 0.6613
Initial loss: 0.61
Initial accuracy: 0.66
Epoch 1/2
Traceback (most recent call last):
<... etc ...>
tensorflow.python.framework.errors_impl.InvalidArgumentError: Cannot assign a device for operation model/random_flip_1/stateful_uniform_full_int/RngReadAndSkip: Could not satisfy explicit device specification '' because the node {{colocation_node model/random_flip_1/stateful_uniform_full_int/RngReadAndSkip}} was colocated with a group of nodes that required incompatible device '/job:localhost/replica:0/task:0/device:GPU:0'. All available devices [/job:localhost/replica:0/task:0/device:CPU:0, /job:localhost/replica:0/task:0/device:GPU:0].
Colocation Debug Info:
Colocation group had the following types and supported devices:
Root Member(assigned_device_name_index_=2 requested_device_name_='/job:localhost/replica:0/task:0/device:GPU:0' assigned_device_name_='/job:localhost/replica:0/task:0/device:GPU:0' resource_device_name_='/job:localhost/replica:0/task:0/device:GPU:0' supported_device_types_=[CPU] possible_devices_=[]
RngReadAndSkip: CPU
_Arg: GPU CPU
Colocation members, user-requested devices, and framework assigned devices, if any:
model_random_flip_1_stateful_uniform_full_int_rngreadandskip_resource (_Arg) framework assigned device=/job:localhost/replica:0/task:0/device:GPU:0
model/random_flip_1/stateful_uniform_full_int/RngReadAndSkip (RngReadAndSkip)
[[{{node model/random_flip_1/stateful_uniform_full_int/RngReadAndSkip}}]] [Op:__inference_train_function_11281]
But, running the model without the following layers:
x = layers.RandomFlip('horizontal')(x)
x = layers.RandomRotation(0.2)(x)
works without any issues. I don't quite understand all the details here in the error message. But looks like it's complaining about GPU problems. Any ideas?
