SkFlow (Tensorflow contrib learn) Teacher

https://www.tensorflow.org/versions/master/api_docs/python/contrib.learn.html#RunConfig

Main skFlow objects and sub objects

tf.contrib.learn.RunConfig

tf.contrib.learn.RunConfig.__init__(tf_master='', num_cores=4, verbose=1, gpu_memory_fraction=1, tf_random_seed=42, keep_checkpoint_max=5, keep_checkpoint_every_n_hours=10000)

tf.contrib.learn.TensorFlowClassifier

tf.contrib.learn.TensorFlowClassifier.__init__(n_classes, batch_size=32, steps=200, optimizer='Adagrad', learning_rate=0.1, class_weight=None, clip_gradients=5.0, continue_training=False, config=None, verbose=1)

SkFlow standard sub-objects

.bias_

.fit(X, y, monitor=None, logdir=None)

.get_params(deep=True)

.get_tensor(name)

.get_tensor_value(name)

.partial_fit(X, y)

.predict(X, axis=1, batch_size=None)

.predict_proba(X, batch_size=None)

.restore(cls, path, config=None)

.save(path)

.score(X, y, sample_weight=None)

.set_params(**params)

.weights_

tf.contrib.learn.TensorFlowDNNClassifier

tf.contrib.learn.TensorFlowDNNClassifier.__init__(hidden_units, n_classes, batch_size=32, steps=200, optimizer='Adagrad', learning_rate=0.1, class_weight=None, clip_gradients=5.0, continue_training=False, config=None, verbose=1, dropout=None)

SkFlow standard sub-objects

.bias_

.fit(X, y, monitor=None, logdir=None)

.get_params(deep=True)

.get_tensor(name)

.get_tensor_value(name)

.partial_fit(X, y)

.predict(X, axis=1, batch_size=None)

.predict_proba(X, batch_size=None)

.restore(cls, path, config=None)

.save(path)

.score(X, y, sample_weight=None)

.set_params(**params)

.weights_

tf.contrib.learn.TensorFlowDNNClassifier

tf.contrib.learn.TensorFlowDNNRegressor.__init__(hidden_units, n_classes=0, batch_size=32, steps=200, optimizer='Adagrad', learning_rate=0.1, clip_gradients=5.0, continue_training=False, config=None, verbose=1, dropout=None)

SkFlow standard sub-objects

.bias_

.fit(X, y, monitor=None, logdir=None)

.get_params(deep=True)

.get_tensor(name)

.get_tensor_value(name)

.partial_fit(X, y)

.predict(X, axis=1, batch_size=None)

.predict_proba(X, batch_size=None)

.restore(cls, path, config=None)

.save(path)

.score(X, y, sample_weight=None)

.set_params(**params)

.weights_

tf.contrib.learn.TensorFlowDNNRegressor

tf.contrib.learn.TensorFlowDNNRegressor.__init__(hidden_units, n_classes=0, batch_size=32, steps=200, optimizer='Adagrad', learning_rate=0.1, clip_gradients=5.0, continue_training=False, config=None, verbose=1, dropout=None)

SkFlow standard sub-objects

.bias_

.fit(X, y, monitor=None, logdir=None)

.get_params(deep=True)

.get_tensor(name)

.get_tensor_value(name)

.partial_fit(X, y)

.predict(X, axis=1, batch_size=None)

.predict_proba(X, batch_size=None)

.restore(cls, path, config=None)

.save(path)

.score(X, y, sample_weight=None)

.set_params(**params)

.weights_

tf.contrib.learn.TensorFlowEstimator

tf.contrib.learn.TensorFlowEstimator.__init__(model_fn, n_classes, batch_size=32, steps=200, optimizer='Adagrad', learning_rate=0.1, clip_gradients=5.0, class_weight=None, continue_training=False, config=None, verbose=1)

SkFlow standard sub-objects

.bias_

.fit(X, y, monitor=None, logdir=None)

.get_params(deep=True)

.get_tensor(name)

.get_tensor_value(name)

.partial_fit(X, y)

.predict(X, axis=1, batch_size=None)

.predict_proba(X, batch_size=None)

.restore(cls, path, config=None)

.save(path)

.score(X, y, sample_weight=None)

.set_params(**params)

.weights_

tf.contrib.learn.TensorFlowLinearClassifier

tf.contrib.learn.TensorFlowLinearClassifier.__init__(n_classes, batch_size=32, steps=200, optimizer='Adagrad', learning_rate=0.1, class_weight=None, clip_gradients=5.0, continue_training=False, config=None, verbose=1)

SkFlow standard sub-objects

.bias_

.fit(X, y, monitor=None, logdir=None)

.get_params(deep=True)

.get_tensor(name)

.get_tensor_value(name)

.partial_fit(X, y)

.predict(X, axis=1, batch_size=None)

.predict_proba(X, batch_size=None)

.restore(cls, path, config=None)

.save(path)

.score(X, y, sample_weight=None)

.set_params(**params)

.weights_

tf.contrib.learn.TensorFlowLinearRegressor

tf.contrib.learn.TensorFlowLinearRegressor.__init__(n_classes=0, batch_size=32, steps=200, optimizer='Adagrad', learning_rate=0.1, clip_gradients=5.0, continue_training=False, config=None, verbose=1)

SkFlow standard sub-objects

.bias_

.fit(X, y, monitor=None, logdir=None)

.get_params(deep=True)

.get_tensor(name)

.get_tensor_value(name)

.partial_fit(X, y)

.predict(X, axis=1, batch_size=None)

.predict_proba(X, batch_size=None)

.restore(cls, path, config=None)

.save(path)

.score(X, y, sample_weight=None)

.set_params(**params)

.weights_

tf.contrib.learn.TensorFlowRNNClassifier

tf.contrib.learn.TensorFlowRNNClassifier.__init__(rnn_size, n_classes, cell_type='gru', num_layers=1, input_op_fn=null_input_op_fn, initial_state=None, bidirectional=False, sequence_length=None, batch_size=32, steps=50, optimizer='Adagrad', learning_rate=0.1, class_weight=None, clip_gradients=5.0, continue_training=False, config=None, verbose=1)

SkFlow standard sub-objects

.bias_

.fit(X, y, monitor=None, logdir=None)

.get_params(deep=True)

.get_tensor(name)

.get_tensor_value(name)

.partial_fit(X, y)

.predict(X, axis=1, batch_size=None)

.predict_proba(X, batch_size=None)

.restore(cls, path, config=None)

.save(path)

.score(X, y, sample_weight=None)

.set_params(**params)

.weights_

tf.contrib.learn.TensorFlowRNNRegressor

tf.contrib.learn.TensorFlowRNNRegressor.__init__(rnn_size, cell_type='gru', num_layers=1, input_op_fn=null_input_op_fn, initial_state=None, bidirectional=False, sequence_length=None, n_classes=0, batch_size=32, steps=50, optimizer='Adagrad', learning_rate=0.1, clip_gradients=5.0, continue_training=False, config=None, verbose=1)

SkFlow standard sub-objects

.bias_

.fit(X, y, monitor=None, logdir=None)

.get_params(deep=True)

.get_tensor(name)

.get_tensor_value(name)

.partial_fit(X, y)

.predict(X, axis=1, batch_size=None)

.predict_proba(X, batch_size=None)

.restore(cls, path, config=None)

.save(path)

.score(X, y, sample_weight=None)

.set_params(**params)

.weights_

tf.contrib.learn.TensorFlowRegressor

tf.contrib.learn.TensorFlowRegressor.__init__(n_classes=0, batch_size=32, steps=200, optimizer='Adagrad', learning_rate=0.1, clip_gradients=5.0, continue_training=False, config=None, verbose=1)

SkFlow standard sub-objects

.bias_

.fit(X, y, monitor=None, logdir=None)

.get_params(deep=True)

.get_tensor(name)

.get_tensor_value(name)

.partial_fit(X, y)

.predict(X, axis=1, batch_size=None)

.predict_proba(X, batch_size=None)

.restore(cls, path, config=None)

.save(path)

.score(X, y, sample_weight=None)

.set_params(**params)

.weights_

Extract Data Helper Objects

• tf.contrib.learn.extract_dask_data(data)

Extract data from dask.Series or dask.DataFrame for predictors

• tf.contrib.learn.extract_dask_labels(labels)

Extract data from dask.Series for labels

• tf.contrib.learn.extract_pandas_data(data)

Extract data from pandas.DataFrame for predictors

• tf.contrib.learn.extract_pandas_labels(labels)

Extract data from pandas.DataFrame for labels

• tf.contrib.learn.extract_pandas_matrix(data)

Extracts numpy matrix from pandas DataFrame.