hadifar

def neural_net_model(inputs, mode):
    with tf.variable_scope('ConvModel'):
        inputs = inputs / 255
        input_layer = tf.reshape(inputs, [-1, 28, 28, 1])
        conv1 = tf.layers.conv2d(
            inputs=input_layer,
            filters=20,
            kernel_size=[5, 5],
            padding='valid',
            activation=tf.nn.relu)

hadifar

def model_fn(features, labels, mode):
    logits = neural_net_model(features, mode)
    class_prediction = tf.argmax(logits, axis=-1)
    preds = class_prediction

    loss = None
    train_op = None
    eval_metric_ops = {}

    if mode in (tf.estimator.ModeKeys.EVAL, tf.estimator.ModeKeys.TRAIN):

hadifar

def model_fn(features, labels, mode, params, config)

hadifar

iris = datasets.load_iris()
X = iris.data[:, :2]
Y = iris.target

clf = MLPClassifier(solver='lbfgs', hidden_layer_sizes=(10,10))

# Create an instance of Logistic Regression Classifier and fit the data.
clf.fit(X, Y)

# Prediction phase

hadifar

# Build 2 hidden layer DNN with 10, 10 units respectively.
classifier = tf.estimator.DNNClassifier(
    feature_columns=my_feature_columns,
    # Two hidden layers of 10 nodes each.
    hidden_units=[10, 10],
    # The model must choose between 3 classes.
    n_classes=3,
    # The directory which model to be saved
    model_dir='./tmp'
)

hadifar

_NumericColumn(key='SepalLength', shape=(1,), default_value=None, dtype=tf.float32, normalizer_fn=None)
_NumericColumn(key='SepalWidth', shape=(1,), default_value=None, dtype=tf.float32, normalizer_fn=None)
_NumericColumn(key='PetalLength', shape=(1,), default_value=None, dtype=tf.float32, normalizer_fn=None)
_NumericColumn(key='PetalWidth', shape=(1,), default_value=None, dtype=tf.float32, normalizer_fn=None)

hadifar

def input_fn():
   ...  # manipulate dataset, extracting the feature dict and the label
   return feature_dict, label

hadifar

# Feature columns describe how to use the input.
my_feature_columns = []
for key in iris_data.train_x.keys():
    my_feature_columns.append(tf.feature_column.numeric_column(key=key))

# Build 2 hidden layer DNN with 10, 10 units respectively.
classifier = tf.estimator.DNNClassifier(
    feature_columns=my_feature_columns,
    # Two hidden layers of 10 nodes each.
    hidden_units=[10, 10],

hadifar

elems = np.array(['H', 'e', 'l', 'l', 'o', ' ', 'T', 'e', 'n', 's', 'o', 'r'])
sum = tf.scan(lambda a, x: a + x, elems)

>>> ['H' 'He' 'Hel' 'Hell' 'Hello' 'Hello ' 'Hello T' 'Hello Te'
     'Hello Ten' 'Hello Tens' 'Hello Tenso' 'Hello Tensor']

hadifar

training_data = np.random.rand(3,20)
training_labels = np.random.rand(3,1)
with tf.Session():
  input_data = tf.constant(training_data)
  input_labels = tf.constant(training_labels)
  ...