Package pylearn :: Package io :: Package seriestables :: Module series
[hide private]

Module series

source code

Classes [hide private]
Series
Base Series class, with minimal arguments and type checks.
DummySeries
To put in a series dictionary instead of a real series, to do nothing when we don't want a given series to be saved.
ErrorSeries
Most basic Series: saves a single float (called an Error as this is the most common use case I foresee) along with an index (epoch, for example) and timestamp/cpu.clock for each of these floats.
AccumulatorSeriesWrapper
Wraps a Series by accumulating objects passed its Accumulator.append() method and "reducing" (e.g.
BasicStatisticsSeries
SeriesArrayWrapper
Simply redistributes any number of elements to sub-series to respective append()s.
SharedParamsStatisticsWrapper
Save mean, min/max, std of shared parameters place in an array.
Functions [hide private]
 
_get_description_timestamp_cpuclock_columns(store_timestamp, store_cpuclock, pos=0) source code
 
_get_description_n_ints(int_names, int_width=64, pos=0)
Begins construction of a class inheriting from IsDescription to construct an HDF5 table with index columns named with int_names.
source code
 
_get_description_with_n_ints_n_floats(int_names, float_names, int_width=64, float_width=32, store_timestamp=True, store_cpuclock=True)
Constructs a class to be used when constructing a table with PyTables.
source code
 
_index_to_tuple(index) source code
 
_BasicStatisticsSeries_construct_table_toexec(index_names, store_timestamp, store_cpuclock) source code
Variables [hide private]
  _basic_stats_functions = {'mean': lambda(x): numpy.mean(x), 'm...

Imports: tables, numpy, time


Function Details [hide private]

_get_description_n_ints(int_names, int_width=64, pos=0)

source code 

Begins construction of a class inheriting from IsDescription to construct an HDF5 table with index columns named with int_names.

See Series().__init__ to see how those are used.

_get_description_with_n_ints_n_floats(int_names, float_names, int_width=64, float_width=32, store_timestamp=True, store_cpuclock=True)

source code 

Constructs a class to be used when constructing a table with PyTables.

This is useful to construct a series with an index with multiple levels.
E.g. if you want to index your "validation error" with "epoch" first, then
"minibatch_index" second, you'd use two "int_names".

Parameters
----------
int_names : tuple of str
    Names of the int (e.g. index) columns
float_names : tuple of str
    Names of the float (e.g. error) columns
int_width : {'32', '64'}
    Type of ints.
float_width : {'32', '64'}
    Type of floats.
store_timestamp : bool
    See __init__ of Series
store_cpuclock : bool
    See __init__ of Series

Returns
-------
A class object, to pass to createTable()


Variables Details [hide private]

_basic_stats_functions

Value:
{'mean': lambda(x): numpy.mean(x), 'min': lambda(x): numpy.min(x), 'ma\
x': lambda(x): numpy.max(x), 'std': lambda(x): numpy.std(x)}