Source code for spinn_front_end_common.interface.interface_functions.profile_data_gatherer

# Copyright (c) 2017-2019 The University of Manchester
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.

import os
import logging
from spinn_utilities.progress_bar import ProgressBar
from spinn_front_end_common.interface.profiling import AbstractHasProfileData

logger = logging.getLogger(__name__)


[docs]class ProfileDataGatherer(object): __slots__ = [] def __call__( self, transceiver, placements, provenance_file_path, machine_time_step): """ :param transceiver: the SpiNNMan interface object :param placements: The placements of the vertices :param has_ran: token that states that the simulation has ran :param provenance_file_path: The location to store the profile data :param machine_time_step: machine time step in ms """ # pylint: disable=too-many-arguments machine_time_step_ms = float(machine_time_step) / 1000.0 progress = ProgressBar( placements.n_placements, "Getting profile data") # retrieve provenance data from any cores that provide data for placement in progress.over(placements.placements): if isinstance(placement.vertex, AbstractHasProfileData): # get data profile_data = placement.vertex.get_profile_data( transceiver, placement) if profile_data.tags: self._write(placement, profile_data, machine_time_step_ms, provenance_file_path) def _write(self, p, profile_data, machine_time_step_ms, directory): # pylint: disable=too-many-arguments max_tag_len = max(len(tag) for tag in profile_data.tags) # write data file_name = os.path.join( directory, "{}_{}_{}_profile.txt".format(p.x, p.y, p.p)) # set mode of the file based off if the file already exists mode = "w" if os.path.exists(file_name): mode = "a" # write profile data to file with open(file_name, mode) as f: f.write("{: <{}s} {: <7s} {: <14s} {: <14s} {: <14s}\n".format( "tag", max_tag_len, "n_calls", "mean_ms", "n_calls_per_ts", "mean_ms_per_ts")) f.write("{:-<{}s} {:-<7s} {:-<14s} {:-<14s} {:-<14s}\n".format( "", max_tag_len, "", "", "", "")) for tag in profile_data.tags: f.write("{: <{}s} {: >7d} {: >14.6f} {: >14.6f} {: >14.6f}\n" .format( tag, max_tag_len, profile_data.get_n_calls(tag), profile_data.get_mean_ms(tag), profile_data.get_mean_n_calls_per_ts( tag, machine_time_step_ms), profile_data.get_mean_ms_per_ts( tag, machine_time_step_ms)))