class TrainingRuns(Mapping):
"""A map from integers to TrainingRuns."""
def __init__(self, root_dir, src_dir, run_factory, check_commit=True):
"""Create TrainingRuns object.
Args:
root_dir (str): directory where all training run data will be stored
src_dir (str): a Git repository path (used to check commits)
run_factory (Callable[[Config, str], TrainingRun]): a Callable, which takes a Config and a save_dir
as arguments, and creates a new TrainingRun.
check_commit (bool): if True, checks that current working directory is on same commit as when the run
was originally created.
"""
self._int_dirs = IntegerDirectories(root_dir)
self._src_dir = src_dir
self._run_factory = run_factory
self._check_commit = check_commit
def _config_path(self, save_dir):
return join(save_dir, 'config.txt')
def __getitem__(self, i):
"""Reload an existing TrainingRun."""
save_dir = self._int_dirs[i]
config = Config.from_file(self._config_path(save_dir))
run = self._run_factory(config, save_dir)
if self._check_commit:
run.match_commit(self._src_dir)
logging.info('Reloaded TrainingRun #{}'.format(i))
return run
def new(self, config, name=None):
"""Create a new TrainingRun."""
print('TrainingRun configuration:\n{}'.format(config))
save_dir = self._int_dirs.new_dir(name=name)
cfg_path = self._config_path(save_dir)
config.to_file(cfg_path) # save the config
run = self._run_factory(config, save_dir)
try:
run.record_commit(self._src_dir)
run.dump_diff(self._src_dir)
except InvalidGitRepositoryError:
print('WARNING: could not obtain Git information.')
run.metadata[
'config'] = config._config_tree # save config in metadata, for programmatic access
print('New TrainingRun created at: {}'.format(run.workspace.root))
return run
def __iter__(self):
return iter(self._int_dirs)
def __len__(self):
return len(self._int_dirs)
def paths(self):
return list(self._int_dirs.values())
def __init__(self, root_dir, src_dir, run_factory, check_commit=True):
"""Create TrainingRuns object.
Args:
root_dir (str): directory where all training run data will be stored
src_dir (str): a Git repository path (used to check commits)
run_factory (Callable[[Config, str], TrainingRun]): a Callable, which takes a Config and a save_dir
as arguments, and creates a new TrainingRun.
check_commit (bool): if True, checks that current working directory is on same commit as when the run
was originally created.
"""
self._int_dirs = IntegerDirectories(root_dir)
self._src_dir = src_dir
self._run_factory = run_factory
self._check_commit = check_commit
def load_trace_groups(run_num):
"""Load traces for a particular TrainingRun.
Returns:
trace_groups (dict[str, Traces]): map from trace type to Traces
"""
run_dirs = IntegerDirectories(data.workspace.experiments)
traces_dir = join(run_dirs[run_num], 'traces')
trace_groups = {}
for trace_type in TRACE_TYPES:
trace_dir = join(traces_dir, trace_type)
filenames = os.listdir(trace_dir)
train_step_to_trace = {}
for full_name in filenames:
name, ext = splitext(full_name)
if ext != '.json':
continue
full_path = join(trace_dir, full_name)
train_step = int(name)
with open(full_path, 'r') as f:
trace = json.load(f)
train_step_to_trace[train_step] = trace
trace_groups[trace_type] = Traces(train_step_to_trace)
return trace_groups
def int_dirs(self, tmpdir):
tmpdir.mkdir('152_blah')
tmpdir.mkdir('153_woo')
tmpdir.mkdir('1_') # no suffix, should still match
tmpdir.mkdir('-1') # no suffix, should still match
tmpdir.mkdir('_10') # prefix is not integer, ignore
tmpdir.mkdir('.DS_Store')
tmpdir.mkdir('other')
return IntegerDirectories(str(tmpdir))
def __init__(self,
root_dir,
src_dir,
experiment_factory,
default_config_path,
check_commit=True):
"""Create Experiments object.
Args:
root_dir (str): directory where all experiment data will be stored
src_dir (str): a Git repository path (used to check commits)
experiment_factory (Callable[[Config, str], Experiment]): a Callable, which takes a Config and a save_dir
as arguments, and creates a new Experiment.
default_config_path (str): path to a default config, to be used when no config is specified
check_commit (bool): if True, checks that current working directory is on same commit as when the experiment
was originally created.
"""
self._int_dirs = IntegerDirectories(root_dir)
self._src_dir = src_dir
self._exp_factory = experiment_factory
self._check_commit = check_commit
self._default_config_path = default_config_path
parser = argparse.ArgumentParser()
parser.add_argument('-p', '--port', type=int, default=6006)
parser.add_argument('-m', '--model-file', default='')
parser.add_argument('-s', '--config-strings', action='append', default=[])
parser.add_argument('config_paths', nargs='+')
args = parser.parse_args()
set_log_level('WARNING')
# create run
config_strings = []
config_paths = args.config_paths
if len(config_paths) == 1 and config_paths[0].isdigit():
# Get configs from a run
run_dirname = IntegerDirectories(data.workspace.experiments)[int(
config_paths[0])]
with open(join(run_dirname, 'config.txt')) as fin:
config_strings.append(fin.read())
else:
# Merge strings to allow object overwites
run_dirname = None
for filename in config_paths:
with open(filename) as fin:
config_strings.append(fin.read())
for config_string in args.config_strings:
config_strings.append(config_string)
config = Config.from_str('\n'.join(config_strings))
eval_run = PhraseNodeEvalRun(config)
# load model
class Experiments(Mapping):
"""A map from integers to Experiments."""
def __init__(self,
root_dir,
src_dir,
experiment_factory,
default_config_path,
check_commit=True):
"""Create Experiments object.
Args:
root_dir (str): directory where all experiment data will be stored
src_dir (str): a Git repository path (used to check commits)
experiment_factory (Callable[[Config, str], Experiment]): a Callable, which takes a Config and a save_dir
as arguments, and creates a new Experiment.
default_config_path (str): path to a default config, to be used when no config is specified
check_commit (bool): if True, checks that current working directory is on same commit as when the experiment
was originally created.
"""
self._int_dirs = IntegerDirectories(root_dir)
self._src_dir = src_dir
self._exp_factory = experiment_factory
self._check_commit = check_commit
self._default_config_path = default_config_path
def _config_path(self, save_dir):
return join(save_dir, 'config.txt')
def __getitem__(self, i):
"""Reload an existing Experiment."""
save_dir = self._int_dirs[i]
config = Config.from_file(self._config_path(save_dir))
exp = self._exp_factory(config, save_dir)
if self._check_commit:
exp.match_commit(self._src_dir)
logging.info('Reloaded experiment #{}'.format(i))
return exp
def new(self, config=None, name=None):
"""Create a new Experiment."""
if config is None:
config = Config.from_file(self._default_config_path)
save_dir = self._int_dirs.new_dir(name=name)
cfg_path = self._config_path(save_dir)
config.to_file(cfg_path) # save the config
exp = self._exp_factory(config, save_dir)
exp.record_commit(self._src_dir)
logging.info('New experiment created at: {}'.format(
exp.workspace.root))
logging.info('Experiment configuration:\n{}'.format(config))
return exp
def __iter__(self):
return iter(self._int_dirs)
def __len__(self):
return len(self._int_dirs)
def paths(self):
return self._int_dirs.values()
