tf.contrib.linear_optimizer.SDCAOptimizer

class tf.contrib.linear_optimizer.SDCAOptimizer

Wrapper class for SDCA optimizer.

The wrapper is currently meant for use as an optimizer within a tf.learn Estimator.

Example usage: real_feature_column = real_valued_column(...) sparse_feature_column = sparse_column_with_hash_bucket(...) sdca_optimizer = linear.SDCAOptimizer(example_id_column='example_id', num_loss_partitions=1, num_table_shards=1, symmetric_l2_regularization=2.0) classifier = tf.contrib.learn.LinearClassifier( feature_columns=[real_feature_column, sparse_feature_column], weight_column_name=..., optimizer=sdca_optimizer) classifier.fit(input_fn_train, steps=50) classifier.evaluate(input_fn=input_fn_eval)

Here the expectation is that the input_fn_* functions passed to train and evaluate return a pair (dict, label_tensor) where dict has example_id_column as key whose value is a Tensor of shape [batch_size] and dtype string. num_loss_partitions defines the number of partitions of the global loss function and should be set to (#concurrent train ops/per worker) x (#workers). Convergence of (global) loss is guaranteed if num_loss_partitions is larger or equal to the above product. Larger values for num_loss_partitions lead to slower convergence. The recommended value for num_loss_partitions in tf.learn (where currently there is one process per worker) is the number of workers running the train steps. It defaults to 1 (single machine). num_table_shards defines the number of shards for the internal state table, typically set to match the number of parameter servers for large data sets.

Methods

__init__(example_id_column, num_loss_partitions=1, num_table_shards=None, symmetric_l1_regularization=0.0, symmetric_l2_regularization=1.0)

get_name()

get_train_step(columns_to_variables, weight_column_name, loss_type, features, targets, global_step)

Returns the training operation of an SdcaModel optimizer.

Defined in tensorflow/contrib/linear_optimizer/python/sdca_optimizer.py.