search

tensorflow / models

Models and examples built with TensorFlow
Other
77k stars 45.78k forks source link

Models can not be loaded #5649

Closed Svdvoort closed 5 years ago

Svdvoort commented 5 years ago

System information

Minimal working example code (Running with Python 3.6.6):

from tensorflow.keras.layers import Conv2D, MaxPooling2D, ReLU, Flatten, Dense, Input
from tensorflow.keras.models import Model
import numpy as np
import tensorflow as tf
from tensorflow.keras.models import load_model

def cnn_model(rows, cols, channels):
    model = Model()
    inputs = Input(shape=(rows, cols, channels), dtype='float32')

    conv1 = Conv2D(64, (3, 3),activation='linear',kernel_initializer='he_uniform')(inputs)
    relu1 = ReLU()(conv1)
    pooling1 = MaxPooling2D(pool_size=(5, 5))(relu1)

    flatten = Flatten()(pooling1)

    dense1 = Dense(512)(flatten)

    relu2 = ReLU()(dense1)
    predictions = Dense(4, activation='softmax')(relu2)

    model = Model(inputs=inputs, outputs=predictions)

    return model

batch_size = 16
rows = 128
cols = 128
channels = 1
model_input = np.random.randint(0, 255, (batch_size, rows, cols, channels))
model_labels = np.random.randint(0, 1, (batch_size, 4))

model = cnn_model(rows, cols,channels)

adam = tf.train.AdamOptimizer(learning_rate=0.01, beta1=0.9, beta2=0.999, epsilon=1e-8)

model.compile(loss='categorical_crossentropy',
              optimizer=adam,
              metrics=['categorical_accuracy'])
model.fit(x=model_input,
          y=model_labels,
          epochs=2)

model.save("/home/svdvoort/test_model.hdf5")
load_model("/home/svdvoort/test_model.hdf5")

Describe the problem

Returns the following output (error included):

Epoch 1/2
2018-10-31 19:09:12.162738: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1411] Found device 0 with properties: 
name: GeForce GTX 1080 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.683
pciBusID: 0000:01:00.0
totalMemory: 10.91GiB freeMemory: 10.76GiB
2018-10-31 19:09:12.290346: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1411] Found device 1 with properties: 
name: GeForce GTX 1080 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.683
pciBusID: 0000:02:00.0
totalMemory: 10.92GiB freeMemory: 10.76GiB
2018-10-31 19:09:12.291317: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1490] Adding visible gpu devices: 0, 1
2018-10-31 19:09:12.750668: I tensorflow/core/common_runtime/gpu/gpu_device.cc:971] Device interconnect StreamExecutor with strength 1 edge matrix:
2018-10-31 19:09:12.750702: I tensorflow/core/common_runtime/gpu/gpu_device.cc:977]      0 1 
2018-10-31 19:09:12.750707: I tensorflow/core/common_runtime/gpu/gpu_device.cc:990] 0:   N Y 
2018-10-31 19:09:12.750710: I tensorflow/core/common_runtime/gpu/gpu_device.cc:990] 1:   Y N 
2018-10-31 19:09:12.751100: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1103] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10404 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1080 Ti, pci bus id: 0000:01:00.0, compute capability: 6.1)
2018-10-31 19:09:12.855072: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1103] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:1 with 10407 MB memory) -> physical GPU (device: 1, name: GeForce GTX 1080 Ti, pci bus id: 0000:02:00.0, compute capability: 6.1)
16/16 [==============================] - 3s 201ms/step - loss: 0.0000e+00 - categorical_accuracy: 0.0000e+00
Epoch 2/2
16/16 [==============================] - 0s 1ms/step - loss: 0.0000e+00 - categorical_accuracy: 0.0000e+00
WARNING:tensorflow:TensorFlow optimizers do not make it possible to access optimizer attributes or optimizer state after instantiation. As a result, we cannot save the optimizer as part of the model save file.You will have to compile your model again after loading it. Prefer using a Keras optimizer instead (see keras.io/optimizers).
Traceback (most recent call last):
  File "Loading_error_example.py", line 47, in <module>
    load_model("/home/svdvoort/test_model.hdf5")
  File "/packages/tensorflow/1.11.0/Python-3.6.6/tensorflow/python/keras/engine/saving.py", line 230, in load_model
    model = model_from_config(model_config, custom_objects=custom_objects)
  File "/packages/tensorflow/1.11.0/Python-3.6.6/tensorflow/python/keras/engine/saving.py", line 310, in model_from_config
    return deserialize(config, custom_objects=custom_objects)
  File "/packages/tensorflow/1.11.0/Python-3.6.6/tensorflow/python/keras/layers/serialization.py", line 64, in deserialize
    printable_module_name='layer')
  File "/packages/tensorflow/1.11.0/Python-3.6.6/tensorflow/python/keras/utils/generic_utils.py", line 173, in deserialize_keras_object
    list(custom_objects.items())))
  File "/packages/tensorflow/1.11.0/Python-3.6.6/tensorflow/python/keras/engine/network.py", line 1292, in from_config
    process_layer(layer_data)
  File "/packages/tensorflow/1.11.0/Python-3.6.6/tensorflow/python/keras/engine/network.py", line 1278, in process_layer
    layer = deserialize_layer(layer_data, custom_objects=custom_objects)
  File "/packages/tensorflow/1.11.0/Python-3.6.6/tensorflow/python/keras/layers/serialization.py", line 64, in deserialize
    printable_module_name='layer')
  File "/packages/tensorflow/1.11.0/Python-3.6.6/tensorflow/python/keras/utils/generic_utils.py", line 175, in deserialize_keras_object
    return cls.from_config(config['config'])
  File "/packages/tensorflow/1.11.0/Python-3.6.6/tensorflow/python/keras/engine/base_layer.py", line 1617, in from_config
    return cls(**config)
  File "/packages/tensorflow/1.11.0/Python-3.6.6/tensorflow/python/keras/layers/advanced_activations.py", line 310, in __init__
    if max_value is not None and max_value < 0.:
TypeError: '<' not supported between instances of 'dict' and 'float'

Same is true for all other models that are being trained

Source code / logs

Output of tf_env attached tf_env.txt

Svdvoort commented 5 years ago

Submitted to wrong repo, resubmitted to tensorflow repo: https://github.com/tensorflow/tensorflow/issues/23413