CODEWITHSUNDEEP
pythontensorflowautoencoder
Lily
Lily

Reputation: 229

Failed copying input tensor from /job:localhost/replica:0/task:0/device:CPU:0 to /job:localhost/replica:0/task:0/device:GPU:0 in order to run Cast

I have an autoencoder class where I try to feed normal_test_data with shape (2933314, 600) to the encoder. I get an Failed copying input tensor from /job:localhost/replica:0/task:0/device:CPU:0 to /job:localhost/replica:0/task:0/device:GPU:0 in order to run Cast: Dst tensor is not initialized. [Op:Cast] out-of-memory error. How do I resolve this error using the Dataset API method? Any help will be greatly appreciated.

class AutoEncoder(Model):
  def _init_(self):
    super(AutoEncoder, self)._init_()
    #############
    ## ENCODER ##
    #############
    self.encoder = tensorflow1.keras.Sequential([
      tensorflow1.keras.layers.Dense(512, activation="relu"),
      tensorflow1.keras.layers.Dense(256, activation="relu"),
      tensorflow1.keras.layers.Dense(128, activation="relu"),
      tensorflow1.keras.layers.Dense(64, activation="relu"),
      tensorflow1.keras.layers.Dense(32, activation="relu"),
      tensorflow1.keras.layers.Dense(16, activation="relu")])

    #############
    ## DECODER ##
    #############
    self.decoder = tensorflow1.keras.Sequential([
      tensorflow1.keras.layers.Dense(32, activation="relu"),
      tensorflow1.keras.layers.Dense(64, activation="relu"),
      tensorflow1.keras.layers.Dense(128, activation="relu"),
      tensorflow1.keras.layers.Dense(256, activation="relu"),
      tensorflow1.keras.layers.Dense(512, activation="relu"),
      tensorflow1.keras.layers.Dense(600, activation="sigmoid")])
    
  def call(self, x):
    encoded = self.encoder(x)
    decoded = self.decoder(encoded)
    return decoded

model = AutoEncoder()

encoder_out = model.encoder(normal_test_data).numpy()

Upvotes: 1

Views: 9788

Answers (1)

PanDe
PanDe

Reputation: 974

This might be an old post but I faced the same issue, therefore, if others who are searching for this issue, please follow, comment#2: Failed copying input tensor from CPU to GPU in order to run GatherVe: Dst tensor is not initialized. [Op:GatherV2]

Upvotes: 0

Related Questions