def prepare_resume(self):
"""Tries to resume the experiment by using the defined resume path or PytorchExperiment."""
checkpoint_file = ""
base_dir = ""
reset_epochs = self._resume_reset_epochs
if self._resume_path is not None:
if isinstance(self._resume_path, str):
if self._resume_path.endswith(".pth.tar"):
checkpoint_file = self._resume_path
base_dir = os.path.dirname(
os.path.dirname(checkpoint_file))
elif self._resume_path.endswith(
"checkpoint") or self._resume_path.endswith(
"checkpoint/"):
checkpoint_file = get_last_file(self._resume_path)
base_dir = os.path.dirname(
os.path.dirname(checkpoint_file))
elif "checkpoint" in os.listdir(
self._resume_path) and "config" in os.listdir(
self._resume_path):
checkpoint_file = get_last_file(self._resume_path)
base_dir = self._resume_path
else:
warnings.warn(
"You have not selected a valid experiment folder, will search all sub folders",
UserWarning)
if self.elog is not None:
self.elog.text_logger.log_to(
"You have not selected a valid experiment folder, will search all "
"sub folders", "warnings")
checkpoint_file = get_last_file(self._resume_path)
base_dir = os.path.dirname(
os.path.dirname(checkpoint_file))
if base_dir:
if not self._ignore_resume_config:
load_config = Config()
load_config.load(os.path.join(base_dir, "config/config.json"))
self._config_raw = load_config
self.config = Config.init_objects(self._config_raw)
self.print("Loaded existing config from:", base_dir)
if self.n_epochs is None:
self.n_epochs = self._config_raw.get("n_epochs")
if checkpoint_file:
self.load_checkpoint(name="",
path=checkpoint_file,
save_types=self._resume_save_types)
self._resume_path = checkpoint_file
shutil.copyfile(
checkpoint_file,
os.path.join(self.elog.checkpoint_dir, "0_checkpoint.pth.tar"))
self.print("Loaded existing checkpoint from:", checkpoint_file)
self._resume_reset_epochs = reset_epochs
if self._resume_reset_epochs:
self._epoch_idx = 0
def __init__(self,
base_dir,
exp_dir="",
name=None,
decode_config_clean_str=True):
super(ExperimentReader, self).__init__()
self.base_dir = base_dir
self.exp_dir = exp_dir
self.work_dir = os.path.abspath(
os.path.join(self.base_dir, self.exp_dir))
self.config_dir = os.path.join(self.work_dir, "config")
self.log_dir = os.path.join(self.work_dir, "log")
self.checkpoint_dir = os.path.join(self.work_dir, "checkpoint")
self.img_dir = os.path.join(self.work_dir, "img")
self.plot_dir = os.path.join(self.work_dir, "plot")
self.save_dir = os.path.join(self.work_dir, "save")
self.result_dir = os.path.join(self.work_dir, "result")
self.config = Config()
if decode_config_clean_str:
self.config.load(os.path.join(self.config_dir, "config.json"),
decoder_cls_=StringMultiTypeDecoder)
else:
self.config.load(os.path.join(self.config_dir, "config.json"),
decoder_cls_=None)
self.exp_info = Config()
exp_info_file = os.path.join(self.config_dir, "exp.json")
if os.path.exists(exp_info_file):
self.exp_info.load(exp_info_file)
self.__results_dict = None
self.meta_name = None
self.meta_star = False
self.meta_ignore = False
self.read_meta_info()
if name is not None:
self.exp_name = name
elif self.meta_name is not None:
self.exp_name = self.meta_name
elif "name" in self.exp_info:
self.exp_name = self.exp_info['name']
elif "exp_name" in self.config:
self.exp_name = self.config['exp_name']
else:
self.exp_name = "experiments"
self.ignore = self.meta_ignore
self.star = self.meta_star
def get_config():
# Set your own path, if needed.
data_root_dir = os.path.abspath(
'data') # The path where the downloaded dataset is stored.
c = Config(
update_from_argv=
True, # If set 'True', it allows to update each configuration by a cmd/terminal parameter.
# Train parameters
num_classes=2,
in_channels=1,
batch_size=3, # works with 6 on GB GPU
patch_size=512,
n_epochs=1,
learning_rate=0.0002,
fold=
0, # The 'splits.pkl' may contain multiple folds. Here we choose which one we want to use.
device=
"cuda", # 'cuda' is the default CUDA device, you can use also 'cpu'. For more information, see https://pytorch.org/docs/stable/notes/cuda.html
# Logging parameters
name='Basic_Unet',
author='kleina', # Author of this project
plot_freq=10, # How often should stuff be shown in visdom
append_rnd_string=
False, # Appends a random string to the experiment name to make it unique.
start_visdom=
True, # You can either start a visom server manually or have trixi start it for you.
do_instancenorm=
True, # Defines whether or not the UNet does a instance normalization in the contracting path
do_load_checkpoint=False,
checkpoint_dir='',
# Adapt to your own path, if needed.
google_drive_id='1jzeNU1EKnK81PyTsrx0ujfNl-t0Jo8uE', #spleen
dataset_name='Task09_Spleen',
base_dir=os.path.abspath(
'output_experiment'), # Where to log the output of the experiment.
data_root_dir=
data_root_dir, # The path where the downloaded dataset is stored.
data_dir=os.path.join(
data_root_dir, 'Task09_Spleen/preprocessed'
), # This is where your training and validation data is stored
data_test_dir=os.path.join(data_root_dir, 'Task09_Spleen/preprocessed'
), # This is where your test data is stored
split_dir=os.path.join(
data_root_dir, 'Task09_Spleen'
), # This is where the 'splits.pkl' file is located, that holds your splits.
# execute a segmentation process on a specific image using the model
model_dir=os.path.join(
os.path.abspath('output_experiment'),
'20200108-035420_Basic_Unet/checkpoint/checkpoint_current'
), # the model being used for segmentation
)
print(c)
return c
def group_experiments_by(exps, group_by_list):
configs_flat = [e.config.flat()
for e in exps] ### Exclude already combined experiments
config_diff = Config.difference_config_static(*configs_flat)
group_diff_key_list = []
group_diff_val_list = []
for diff_key in group_by_list:
if diff_key in config_diff:
group_diff_key_list.append(diff_key)
group_diff_val_list.append(set(config_diff[diff_key]))
val_combis = itertools.product(*group_diff_val_list)
group_dict = defaultdict(list)
for val_combi in val_combis:
for e in exps:
e_config = e.config.flat()
is_match = True
for key, val in zip(group_diff_key_list, val_combi):
if e_config[key] != val:
is_match = False
if is_match:
group_dict[val_combi].append(e)
return list(group_dict.values())
def get_config(self):
combi_config = copy.deepcopy(self.experiments[0].config)
config_diff = Config.difference_config_static(
*[e.config for e in self.experiments], only_set=True)
combi_config.update(config_diff)
return combi_config
def gen_config(parser):
args = parser.parse_args()
# prepare config dictionary, add all arguments from args
c = Config()
for arg in vars(args):
c[arg] = getattr(args, arg)
return c
def load_config(self, name, **kwargs):
"""
Loads a config from a json file from the experiment config dir
Args:
name: the name of the config file
Returns: A Config/ dict filled with the json file content
"""
if not name.endswith(".json"):
name += ".json"
c = Config()
c.load(os.path.join(self.config_dir, name), **kwargs)
return c
def get_config():
# Set your own path, if needed.
data_root_dir = os.path.abspath(
'data') # The path where the downloaded dataset is stored.
c = Config(
update_from_argv=
True, # If set 'True', it allows to update each configuration by a cmd/terminal parameter.
# Train parameters
num_classes=3,
in_channels=1,
batch_size=8,
patch_size=64,
n_epochs=10,
learning_rate=0.0002,
fold=
0, # The 'splits.pkl' may contain multiple folds. Here we choose which one we want to use.
device=
"cuda", # 'cuda' is the default CUDA device, you can use also 'cpu'. For more information, see https://pytorch.org/docs/stable/notes/cuda.html
# Logging parameters
name='Basic_Unet',
author='kleina', # Author of this project
plot_freq=10, # How often should stuff be shown in visdom
append_rnd_string=
False, # Appends a random string to the experiment name to make it unique.
start_visdom=
True, # You can either start a visom server manually or have trixi start it for you.
do_instancenorm=
True, # Defines whether or not the UNet does a instance normalization in the contracting path
do_load_checkpoint=False,
checkpoint_dir='',
# Adapt to your own path, if needed.
google_drive_id=
'1RzPB1_bqzQhlWvU-YGvZzhx2omcDh38C', # This id is used to download the example dataset.
dataset_name='Task04_Hippocampus',
base_dir=os.path.abspath(
'output_experiment'), # Where to log the output of the experiment.
data_root_dir=
data_root_dir, # The path where the downloaded dataset is stored.
data_dir=os.path.join(
data_root_dir, 'Task04_Hippocampus/preprocessed'
), # This is where your training and validation data is stored
data_test_dir=os.path.join(data_root_dir,
'Task04_Hippocampus/preprocessed'
), # This is where your test data is stored
split_dir=os.path.join(
data_root_dir, 'Task04_Hippocampus'
), # This is where the 'splits.pkl' file is located, that holds your splits.
)
print(c)
return c
def _save_exp_config(self):
if self.elog is not None and not isinstance(self.elog, Mock):
cur_time = time.strftime("%y-%m-%d_%H:%M:%S", time.localtime(time.time()))
self.elog.save_config(Config(**{'name': self.name,
'time': cur_time,
'state': "Stub",
'current_time': cur_time,
'epoch': 0
}),
"exp")
def __init__(self,
base_dir,
exp_dirs=(),
name=None,
decode_config_clean_str=True):
self.base_dir = base_dir
if name is None or name == "":
self.exp_name = "combi-experiments"
else:
self.exp_name = name
self.experiments = []
for exp_dir in exp_dirs:
self.experiments.append(
ExperimentReader(
base_dir=base_dir,
exp_dir=exp_dir,
decode_config_clean_str=decode_config_clean_str))
exp_base_dirs = os.path.commonpath(
[os.path.dirname(e.work_dir) for e in self.experiments])
if exp_base_dirs != "":
self.base_dir = exp_base_dirs
self.exp_info = Config()
self.exp_info["epoch"] = -1
self.exp_info["name"] = self.exp_name
self.exp_info["state"] = "Combined"
self.exp_info["time"] = time.strftime("%y-%m-%d_%H:%M:%S",
time.localtime(time.time()))
self.__results_dict = None
self.work_dir = None
self.config_dir = "not_saved_yet"
self.log_dir = "not_saved_yet"
self.checkpoint_dir = "not_saved_yet"
self.img_dir = "not_saved_yet"
self.plot_dir = "not_saved_yet"
self.save_dir = "not_saved_yet"
self.result_dir = "not_saved_yet"
self.exp_dir = "not_saved_yet"
self.meta_name = None
self.meta_star = False
self.meta_ignore = False
self.config = self.get_config()
self.elog = None
def get_config():
# Set your own path, if needed.
#data_root_dir = os.path.abspath('data') # The path where the downloaded dataset is stored.
#data_root_dir = "/home/ramesh/Desktop/WS/Implementation/experiment/Data/Filtereddataset"
#data_root_dir ="/home/ramesh/Desktop/IIITB/experiment/data/FilteredDataSet"
data_root_dir = "/home/ramesh/Desktop/IIITB/experiment/data/NucliiData"
taskName = "Nuclii Segmentation"
#taskName = "Task09_Spleen"
c = Config(
update_from_argv=True,
# Train parameters
num_classes=1,
in_channels=1,
batch_size=6,
patch_size=64,
n_epochs=50,
learning_rate=0.00001,
fold=
0, # The 'splits.pkl' may contain multiple folds. Here we choose which one we want to use.
device=
"cuda", # 'cuda' is the default CUDA device, you can use also 'cpu'. For more information, see https://pytorch.org/docs/stable/notes/cuda.html
# Logging parameters
name='Segmentation_Experiment_Unet_Nuclii',
plot_freq=10, # How often should stuff be shown in visdom
append_rnd_string=False,
start_visdom=True,
do_instancenorm=
True, # Defines whether or not the UNet does a instance normalization in the contracting path
do_load_checkpoint=False,
checkpoint_dir=' ',
# Adapt to your own path, if needed.
#google_drive_id='1RzPB1_bqzQhlWvU-YGvZzhx2omcDh38C',
dataset_name=taskName,
base_dir=os.path.abspath(
'output_experiment'), # Where to log the output of the experiment.
data_root_dir=
data_root_dir, # The path where the downloaded dataset is stored.
data_dir=os.path.join(
data_root_dir, taskName, 'imagesTr'
), # This is where your training and validation data is stored
#data_test_dir=os.path.join(data_root_dir, 'Task04_Hippocampus/preprocessed'), # This is where your test data is stored
split_dir=os.path.join(
data_root_dir, taskName, 'preprocessed'
), # This is where the 'splits.pkl' file is located, that holds your splits.
)
print(c)
return c
def get_config():
# Set your own path, if needed.
#data_root_dir = os.path.abspath('data') # The path where the downloaded dataset is stored.
#data_root_dir = "/home/ramesh/Desktop/WS/Implementation/data/MedSegDecathlon"
data_root_dir = "/home/ramesh/Desktop/WS/Implementation/data/Brats17"
c = Config(
update_from_argv=True,
# Train parameters
num_classes=4,
in_channels=1,
batch_size=8,
patch_size=160,
n_epochs=1,
learning_rate=0.0002,
fold=
0, # The 'splits.pkl' may contain multiple folds. Here we choose which one we want to use.
device=
"cuda", # 'cuda' is the default CUDA device, you can use also 'cpu'. For more information, see https://pytorch.org/docs/stable/notes/cuda.html
# Logging parameters
name='Basic_Unet',
plot_freq=1, # How often should stuff be shown in visdom
append_rnd_string=False,
start_visdom=True,
do_instancenorm=
True, # Defines whether or not the UNet does a instance normalization in the contracting path
do_load_checkpoint=False,
checkpoint_dir=
'/home/ramesh/Desktop/WS/Implementation/experiment/basic_unet_example/output_experiment/20190504-180510_Basic_Unet',
# Adapt to your own path, if needed.
google_drive_id='1RzPB1_bqzQhlWvU-YGvZzhx2omcDh38C',
dataset_name='Brats17',
base_dir=os.path.abspath(
'output_experiment'), # Where to log the output of the experiment.
data_root_dir=
data_root_dir, # The path where the downloaded dataset is stored.
data_dir=os.path.join(
data_root_dir, 'Brats17_preprocessed'
), # This is where your training and validation data is stored
data_test_dir=os.path.join(
data_root_dir,
'Brats17_preprocessed'), # This is where your test data is stored
split_dir=os.path.join(
data_root_dir, 'Brats17_preprocessed'
), # This is where the 'splits.pkl' file is located, that holds your splits.
)
print(c)
return c
def _save_exp_config(self):
if self.elog is not None:
cur_time = time.strftime("%y-%m-%d_%H:%M:%S",
time.localtime(time.time()))
self.elog.save_config(
Config(
**{
'name': self.exp_name,
'time': self._time_start,
'state': self._exp_state,
'current_time': cur_time,
'epoch': self._epoch_idx
}), "exp")
def get_add_config():
# Set your own path, if needed.
data_root_dir = os.path.abspath(
'data') # The path where the downloaded dataset is stored.
c = Config(
do_instancenorm=
True, # Defines whether or not the UNet does a instance normalization in the contracting path
do_load_checkpoint=False,
checkpoint_dir='',
# Adapt to your own path, if needed.
google_drive_id='1RzPB1_bqzQhlWvU-YGvZzhx2omcDh38C',
dataset_name='CHD_segmentation_dataset',
base_dir=os.path.abspath(
'output_experiment'), # Where to log the output of the experiment.
data_root_dir=
data_root_dir, # The path where the downloaded dataset is stored.
data_dir=os.path.join(
data_root_dir, 'CHD_interpolation_dataset/preprocessed'
), # This is where your training and validation data is stored
data_test_dir=os.path.join(data_root_dir,
'CHD_interpolation_dataset/preprocessed'
), # This is where your test data is stored
split_dir=os.path.join(
data_root_dir, 'CHD_interpolation_dataset'
), # This is where the 'splits.pkl' file is located, that holds your splits.
scaled_image_16_dir=os.path.join(
data_root_dir, 'CHD_interpolation_dataset/scaled_to_16'),
scaled_image_32_dir=os.path.join(
data_root_dir, 'CHD_interpolation_dataset/scaled_to_32'),
scaled_image_64_dir=os.path.join(
data_root_dir, 'CHD_interpolation_dataset/scaled_to_64'),
stage_1_dir=os.path.join(data_root_dir,
'CHD_interpolation_dataset/stage_1'),
stage_1_dir_32=os.path.join(data_root_dir,
'CHD_interpolation_dataset/stage_1_32'),
# stage_1_dir = os.path.join(data_root_dir, 'CHD_segmentation_dataset/stage_1')
)
print(c)
return c
def get_config():
c = Config()
c.batch_size = 6
c.patch_size = 512
c.n_epochs = 20
c.learning_rate = 0.0002
c.do_ce_weighting = True
c.do_batchnorm = True
if torch.cuda.is_available():
c.use_cuda = True
else:
c.use_cuda = False
c.rnd_seed = 1
c.log_interval = 200
c.base_dir='/media/kleina/Data2/output/meddec'
c.data_dir='/media/kleina/Data2/Data/meddec/Task07_Pancreas_expert_preprocessed'
c.split_dir='/media/kleina/Data2/Data/meddec/Task07_Pancreas_preprocessed'
c.data_file = 'C:/dev/data/Endoviz2018/GIANA/polyp_detection_segmentation/image_gt_data_file_list_all_640x640.csv'
c.additional_slices=5
c.name=''
print(c)
return c
def get_config_heart(fine_tune_type='None',
exp_name='',
checkpoint_filename='',
checkpoint_dir='',
nr_train_samples=0,
download_data_from_drive=False):
# Set your own path, if needed.
data_root_dir = os.path.abspath(
'data') # The path where the downloaded dataset is stored.
c = Config(
update_from_argv=True,
# Train parameters
num_classes=2,
in_channels=1,
batch_size=8,
patch_size=256,
n_epochs=60,
learning_rate=0.0002,
fold=
0, # The 'splits.pkl' may contain multiple folds. Here we choose which one we want to use.
device="cuda",
# 'cuda' is the default CUDA device, you can use also 'cpu'. For more information, see https://pytorch.org/docs/stable/notes/cuda.html
# Logging parameters
name=exp_name,
author='maxi', # Author of this project
plot_freq=10, # How often should stuff be shown in visdom
append_rnd_string=False,
start_visdom=False,
do_instancenorm=
True, # Defines whether or not the UNet does a instance normalization in the contracting path
do_load_checkpoint=True,
checkpoint_filename=checkpoint_filename,
checkpoint_dir=checkpoint_dir,
fine_tune=fine_tune_type,
# Adapt to your own path, if needed.
download_data=download_data_from_drive,
google_drive_id='1RzPB1_bqzQhlWvU-YGvZzhx2omcDh38C',
dataset_name='Task02_Heart',
base_dir=os.path.abspath(
'output_experiment'), # Where to log the output of the experiment.
data_root_dir=
data_root_dir, # The path where the downloaded dataset is stored.
data_dir=os.path.join(data_root_dir, 'Task02_Heart/preprocessed'),
# This is where your training and validation data is stored
data_test_dir=os.path.join(data_root_dir, 'Task02_Heart/preprocessed'),
# This is where your test data is stored
split_dir=os.path.join(data_root_dir, 'Task02_Heart'),
# This is where the 'splits.pkl' file is located, that holds your splits.
train_samples=nr_train_samples,
# This is the amount of samples used in the train set. Use 0 for original split (1/2 train, 1/4 val, 1/4 test)
# The validation set will be the same size, and the test set is the rest of the images
# Testing
visualize_segm=True)
print(c)
return c
def run_experiment(experiment, configs, args, mods=None, **kwargs):
# set a few defaults
if "explogger_kwargs" not in kwargs:
kwargs["explogger_kwargs"] = dict(
folder_format="{experiment_name}_%Y%m%d-%H%M%S")
if "explogger_freq" not in kwargs:
kwargs["explogger_freq"] = 1
if "resume_save_types" not in kwargs:
kwargs["resume_save_types"] = ("model", "simple", "th_vars", "results")
config = Config(file_=args.config) if args.config is not None else Config()
config.update_missing(configs[args.default_config].deepcopy())
if args.mods is not None and mods is not None:
for mod in args.mods:
config.update(mods[mod])
config = Config(config=config, update_from_argv=True)
# GET EXISTING EXPERIMENTS TO BE ABLE TO SKIP CERTAIN CONFIGS
if args.skip_existing:
existing_configs = []
for exp in os.listdir(args.base_dir):
try:
existing_configs.append(
Config(file_=os.path.join(args.base_dir, exp, "config",
"config.json")))
except Exception as e:
pass
if args.grid is not None:
grid = GridSearch().read(args.grid)
else:
grid = [{}]
for combi in grid:
config.update(combi)
if args.skip_existing:
skip_this = False
for existing_config in existing_configs:
if existing_config.contains(config):
skip_this = True
break
if skip_this:
continue
if "backup_every" in config:
kwargs["save_checkpoint_every_epoch"] = config["backup_every"]
loggers = {}
if args.visdomlogger:
loggers["v"] = ("visdom", {}, 1)
if args.tensorboardxlogger is not None:
if args.tensorboardxlogger == "same":
loggers["tx"] = ("tensorboard", {}, 1)
else:
loggers["tx"] = ("tensorboard", {
"target_dir": args.tensorboardxlogger
}, 1)
if args.telegramlogger:
kwargs["use_telegram"] = True
if args.automatic_description:
difference_to_default = Config.difference_config_static(
config, configs["DEFAULTS"]).flat(keep_lists=True,
max_split_size=0,
flatten_int=True)
description_str = ""
for key, val in difference_to_default.items():
val = val[0]
description_str = "{} = {}\n{}".format(key, val,
description_str)
config.description = description_str
exp = experiment(config=config,
base_dir=args.base_dir,
resume=args.resume,
ignore_resume_config=args.ignore_resume_config,
loggers=loggers,
**kwargs)
trained = False
if args.resume is None or args.test is False:
exp.run()
trained = True
if args.test:
exp.run_test(setup=not trained)
if isinstance(args.resume,
str) and exp.elog is not None and args.copy_test:
for f in glob.glob(os.path.join(exp.elog.save_dir, "test*")):
if os.path.isdir(f):
shutil.copytree(
f,
os.path.join(args.resume, "save",
os.path.basename(f)))
else:
shutil.copy(f, os.path.join(args.resume, "save"))
def experiment(base_dir):
experiment_paths = request.args.getlist('exp')
name = request.args.get('name', "")
do_save = request.args.get('save', "")
combi = request.args.get('combi', 'false')
experiments = []
if combi == "true":
combi_exp = CombiExperimentReader(base_dir,
experiment_paths,
name=name)
if do_save == "true":
combi_exp.save()
experiments = [combi_exp]
else:
# Get all Experiments
for experiment_path in sorted(experiment_paths):
exp = ExperimentReader(base_dir, experiment_path)
experiments.append(exp)
# Assign unique names
exp_names = [exp.exp_name for exp in experiments]
if len(exp_names) > len(set(exp_names)):
for i, exp in enumerate(experiments):
exp.exp_name += str(i)
exp_names = [exp.exp_name for exp in experiments]
# Site Content
content = {}
# Get config
default_val = "-"
combi_config = {}
exp_configs = [exp.config.flat(False) for exp in experiments]
diff_config_keys = list(
Config.difference_config_static(*exp_configs).keys())
config_keys = set([k for c in exp_configs for k in c.keys()])
for k in sorted(config_keys):
combi_config[k] = []
for conf in exp_configs:
combi_config[k].append(conf.get(k, default_val))
config_keys = list(sorted(list(config_keys)))
# Get results
default_val = "-"
combi_results = {}
exp_results = [exp.get_results() for exp in experiments]
result_keys = set([k for r in exp_results for k in r.keys()])
for k in sorted(result_keys):
combi_results[k] = []
for res in exp_results:
combi_results[k].append(res.get(k, default_val))
result_keys = list(sorted(list(result_keys)))
# Get images
images = OrderedDict({})
image_keys = set()
image_path = {}
for exp in experiments:
exp_images = exp.get_images()
img_groups = group_images(exp_images)
images[exp.exp_name] = img_groups
image_path[exp.exp_name] = exp.img_dir
image_keys.update(list(img_groups.keys()))
image_keys = list(image_keys)
image_keys.sort()
# Get Plots
plots = OrderedDict({})
for exp in experiments:
exp_plots = exp.get_plots()
plots[exp.exp_name] = exp_plots
# Get logs
logs_dict = OrderedDict({})
for exp in experiments:
exp_logs = [(os.path.basename(l), exp.exp_dir) for l in exp.get_logs()]
logs_dict[exp.exp_name] = exp_logs
content["title"] = experiments
content["images"] = {
"img_path": image_path,
"imgs": images,
"img_keys": image_keys
}
content["plots"] = {"plots": plots}
content["config"] = {
"exps": experiments,
"configs": combi_config,
"keys": config_keys,
"diff_keys": diff_config_keys
}
content["results"] = {
"exps": exp_names,
"results": combi_results,
"keys": result_keys
}
content["logs"] = {"logs_dict": logs_dict}
return render_template('experiment.html', **content)
def make_defaults(patch_size=112,
in_channels=4,
latent_size=3,
labels=[0, 1, 2, 3]):
if hasattr(patch_size, "__iter__"):
if len(patch_size) > 1:
patch_size = tuple(patch_size)
else:
patch_size = patch_size[0]
if not hasattr(patch_size, "__iter__"):
patch_size = tuple([
patch_size,
] * 3)
DEFAULTS = Config(
# Base
name=os.path.basename(__file__).split(".")[0],
description=DESCRIPTION,
n_epochs=50000,
batch_size=2,
batch_size_val=1,
patch_size=patch_size,
in_channels=in_channels,
out_channels=len(labels),
latent_size=latent_size,
seed=1,
device="cuda",
# Data
split_val=3,
split_test=4,
data_module=data,
data_dir=None, # we're setting data_module.data_dir if this is given
mmap_mode="r",
npz=False,
debug=0, # 1 selects (10, 5, 5) patients, 2 a single batch
train_on_all=False, # adds val and test to training set
generator_train=data.RandomBatchGenerator,
generator_val=data.LinearBatchGenerator,
transforms_train={
0: {
"type": SpatialTransform,
"kwargs": {
"patch_size": patch_size,
"patch_center_dist_from_border": patch_size[0] // 2,
"do_elastic_deform": False,
"p_el_per_sample": 0.2,
"p_rot_per_sample": 0.3,
"p_scale_per_sample": 0.3
},
"active": True
},
1: {
"type": MirrorTransform,
"kwargs": {
"axes": (0, 1, 2)
},
"active": True
},
2: {
"type": SegLabelSelectionBinarizeTransform,
"kwargs": {
"label": [1, 2, 3]
},
"active": False
}
},
transforms_val={
0: {
"type": CenterCropTransform,
"kwargs": {
"crop_size": patch_size
},
"active": False
},
1: {
"type": SegLabelSelectionBinarizeTransform,
"kwargs": {
"label": [1, 2, 3]
},
"active": False
},
2: {
"type": SpatialTransform,
"kwargs": {
"patch_size": patch_size,
"patch_center_dist_from_border": patch_size[0] // 2,
"do_elastic_deform": False,
"do_rotation": False,
"do_scale": True,
"p_scale_per_sample": 1,
"scale": (1.25, 1.25)
},
"active": False
}
},
augmenter_train=MultiThreadedAugmenter,
augmenter_train_kwargs={
"num_processes": 11,
"num_cached_per_queue": 6,
"pin_memory": True
},
augmenter_val=MultiThreadedAugmenter,
augmenter_val_kwargs={
"num_processes": 2,
"pin_memory": True
},
# Model
model=ProbabilisticSegmentationNet,
model_kwargs={
"in_channels": in_channels,
"out_channels": len(labels),
"num_feature_maps": 24,
"latent_size": latent_size,
"depth": 5,
"latent_distribution": distributions.Normal,
"task_op": InjectionUNet3D,
"task_kwargs": {
"output_activation_op": nn.LogSoftmax,
"output_activation_kwargs": {
"dim": 1
},
"activation_kwargs": {
"inplace": True
}
},
"prior_op": InjectionConvEncoder3D,
"prior_kwargs": {
"in_channels": in_channels,
"out_channels": latent_size * 2,
"depth": 5,
"block_depth": 2,
"num_feature_maps": 24,
"feature_map_multiplier": 2,
"activation_kwargs": {
"inplace": True
},
"norm_depth": 2,
},
"posterior_op": InjectionConvEncoder3D,
"posterior_kwargs": {
"in_channels": in_channels + len(labels),
"out_channels": latent_size * 2,
"depth": 5,
"block_depth": 2,
"num_feature_maps": 24,
"feature_map_multiplier": 2,
"activation_kwargs": {
"inplace": True
},
"norm_depth": 2,
},
},
model_init_weights_args=[nn.init.kaiming_uniform_, 0],
model_init_bias_args=[nn.init.constant_, 0],
# Learning
optimizer=optim.Adam,
optimizer_kwargs={"lr": 1e-4},
scheduler=optim.lr_scheduler.StepLR,
scheduler_kwargs={
"step_size": 200,
"gamma": 0.985
},
criterion_segmentation=nn.NLLLoss,
criterion_segmentation_kwargs={"reduction": "sum"},
criterion_latent=distributions.kl_divergence,
criterion_latent_kwargs={},
criterion_latent_init=False,
criterion_segmentation_seg_onehot=False,
criterion_segmentation_weight=1.0,
criterion_latent_weight=1.0,
criterion_segmentation_seg_dtype=torch.long,
# Logging
backup_every=1000,
validate_every=1000,
validate_subset=0.1, # validate only this percentage randomly
show_every=10,
validate_metrics=["Dice"],
labels=labels,
evaluator=Evaluator,
evaluator_kwargs={
"label_values": list(labels) + [tuple(labels[1:])],
"label_names": {
0: "Background",
1: "Edema",
2: "Enhancing",
3: "Necrosis",
tuple(labels[1:]): "Whole Tumor"
},
"nan_for_nonexisting": True
},
val_save_output=False,
val_example_samples=10,
val_save_images=False,
latent_plot_range=[-5, 5],
test_on_val=True,
test_save_output=False,
test_future=True,
test_std_factor=3,
test_std_scale=1.)
TASKMEAN = Config(
criterion_segmentation_kwargs={"reduction": "elementwise_mean"})
ELASTIC = Config(
transforms_train={0: {
"kwargs": {
"do_elastic_deform": True
}
}})
NONORM = Config(model_kwargs={
"prior_kwargs": {
"norm_depth": 0
},
"posterior_kwargs": {
"norm_depth": 0
}
})
FULLNORM = Config(
model_kwargs={
"prior_kwargs": {
"norm_depth": "full"
},
"posterior_kwargs": {
"norm_depth": "full"
}
})
BATCHNORM = Config(
model_kwargs={
"prior_kwargs": {
"norm_op": nn.BatchNorm3d
},
"posterior_kwargs": {
"norm_op": nn.BatchNorm3d
},
"task_kwargs": {
"norm_op": nn.BatchNorm3d
}
})
WHOLETUMOR = Config(transforms_train={2: {
"active": True
}},
transforms_val={1: {
"active": True
}},
out_channels=2,
labels=[0, 1],
model_kwargs={
"out_channels": 2,
"posterior_kwargs": {
"in_channels": in_channels + 2
}
},
evaluator_kwargs={
"label_values": [0, 1],
"label_names": {
0: "Background",
1: "Whole Tumor"
}
})
ENHANCING = Config(
transforms_train={2: {
"kwargs": {
"label": 2
},
"active": True
}},
transforms_val={1: {
"kwargs": {
"label": 2
},
"active": True
}},
out_channels=2,
labels=[0, 1],
model_kwargs={
"out_channels": 2,
"posterior_kwargs": {
"in_channels": in_channels + 2
}
},
evaluator_kwargs={
"label_values": [0, 1],
"label_names": {
0: "Background",
1: "Whole Tumor"
}
})
NOAUGMENT = Config(
transforms_train={
0: {
"kwargs": {
"p_el_per_sample": 0,
"p_rot_per_sample": 0,
"p_scale_per_sample": 0
}
},
1: {
"active": False
}
})
LOWAUGMENT = Config(
transforms_train={
0: {
"kwargs": {
"p_el_per_sample": 0.,
"p_rot_per_sample": 0.15,
"p_scale_per_sample": 0.15
}
}
})
NOBG = Config(criterion_segmentation_kwargs={"ignore_index": 0})
VALIDATEPATCHED = Config(transforms_val={2: {"active": True}})
MODS = {
"TASKMEAN": TASKMEAN,
"ELASTIC": ELASTIC,
"NONORM": NONORM,
"FULLNORM": FULLNORM,
"BATCHNORM": BATCHNORM,
"WHOLETUMOR": WHOLETUMOR,
"ENHANCING": ENHANCING,
"NOAUGMENT": NOAUGMENT,
"LOWAUGMENT": LOWAUGMENT,
"NOBG": NOBG,
"VALIDATEPATCHED": VALIDATEPATCHED
}
return {"DEFAULTS": DEFAULTS}, MODS
def test_2Dexperiment():
c = Config()
c.batch_size = 200
c.n_epochs = 40
c.learning_rate = 0.001
if torch.cuda.is_available():
c.use_cuda = True
else:
c.use_cuda = False
c.rnd_seed = 1
c.log_interval = 200
# model-specific
c.n_coupling = 8
c.prior = 'gauss'
exp = SmileyExperiment(
c,
name='gauss',
n_epochs=c.n_epochs,
seed=42,
base_dir='experiment_dir',
loggers={'visdom': ['visdom', {
"exp_name": "myenv"
}]})
exp.run()
# sampling
samples = exp.model.sample(1000).cpu().numpy()
sns.jointplot(samples[:, 0], samples[:, 1])
plt.show()
class ExperimentReader(object):
"""Reader class to read out experiments created by :class:`trixi.experimentlogger.ExperimentLogger`.
Args:
work_dir (str): Directory with the structure defined by
:class:`trixi.experimentlogger.ExperimentLogger`.
name (str): Optional name for the experiment. If None, will try
to read name from experiment config.
"""
def __init__(self,
base_dir,
exp_dir="",
name=None,
decode_config_clean_str=True):
super(ExperimentReader, self).__init__()
self.base_dir = base_dir
self.exp_dir = exp_dir
self.work_dir = os.path.abspath(
os.path.join(self.base_dir, self.exp_dir))
self.config_dir = os.path.join(self.work_dir, "config")
self.log_dir = os.path.join(self.work_dir, "log")
self.checkpoint_dir = os.path.join(self.work_dir, "checkpoint")
self.img_dir = os.path.join(self.work_dir, "img")
self.plot_dir = os.path.join(self.work_dir, "plot")
self.save_dir = os.path.join(self.work_dir, "save")
self.result_dir = os.path.join(self.work_dir, "result")
self.config = Config()
if decode_config_clean_str:
self.config.load(os.path.join(self.config_dir, "config.json"),
decoder_cls_=StringMultiTypeDecoder)
else:
self.config.load(os.path.join(self.config_dir, "config.json"),
decoder_cls_=None)
self.exp_info = Config()
exp_info_file = os.path.join(self.config_dir, "exp.json")
if os.path.exists(exp_info_file):
self.exp_info.load(exp_info_file)
self.__results_dict = None
self.meta_name = None
self.meta_star = False
self.meta_ignore = False
self.read_meta_info()
if name is not None:
self.exp_name = name
elif self.meta_name is not None:
self.exp_name = self.meta_name
elif "name" in self.exp_info:
self.exp_name = self.exp_info['name']
elif "exp_name" in self.config:
self.exp_name = self.config['exp_name']
else:
self.exp_name = "experiments"
self.ignore = self.meta_ignore
self.star = self.meta_star
@staticmethod
def get_file_contents(folder):
"""Get all files in a folder.
Returns:
list: All files joined with folder path.
"""
if os.path.isdir(folder):
list_ = map(lambda x: os.path.join(folder, x),
sorted(os.listdir(folder)))
return list(filter(lambda x: os.path.isfile(x), list_))
else:
return []
def get_images(self):
imgs = []
imgs += ExperimentReader.get_file_contents(self.img_dir)
if os.path.isdir(self.img_dir):
for f in os.listdir(self.img_dir):
f = os.path.join(self.img_dir, f)
if os.path.isdir(f):
imgs += ExperimentReader.get_file_contents(f)
return imgs
def get_plots(self):
return ExperimentReader.get_file_contents(self.plot_dir)
def get_checkpoints(self):
return ExperimentReader.get_file_contents(self.checkpoint_dir)
def get_logs(self):
return ExperimentReader.get_file_contents(self.log_dir)
def get_log_file_content(self, file_name):
"""Read out log file and HTMLify.
Args:
file_name (str): Name of the log file.
Returns:
str: Log file contents as HTML ready string.
"""
content = ""
log_file = os.path.join(self.log_dir, file_name)
if os.path.exists(log_file):
with open(log_file, 'r') as f:
content = f.read()
content = content.replace("\n", "<br>")
return content
def get_results_log(self):
"""Build result dictionary.
During the experiment result items are
written out as a stream of quasi-atomic units. This reads the stream and
builds arrays of corresponding items.
The resulting dict looks like this::
{
"result group": {
"result": {
"counter": x-array,
"data": y-array
}
}
}
Returns:
dict: Result dictionary.
"""
results_merged = {}
results = []
try:
with open(os.path.join(self.result_dir, "results-log.json"),
"r") as results_file:
results = json.load(results_file)
except Exception as e:
try:
with open(os.path.join(self.result_dir, "results-log.json"),
"r") as results_file:
results_str = results_file.readlines()
if "]" in "]" in "".join(results_str):
results_str = [
rs.replace("]", ",") for rs in results_str
]
results_str.append("{}]")
results_json = "".join(results_str)
results = json.loads(results_json)
except Exception as ee:
print("Could not load result log from", self.result_dir)
print(ee)
for result in results:
for key in result.keys():
counter = result[key]["counter"]
data = result[key]["data"]
label = str(result[key]["label"])
if label not in results_merged:
results_merged[label] = {}
if key not in results_merged[label]:
results_merged[label][key] = defaultdict(list)
results_merged[label][key]["data"].append(data)
results_merged[label][key]["counter"].append(counter)
if "max" in result[key] and "min" in result[key]:
results_merged[label][key]["max"].append(
result[key]["max"])
results_merged[label][key]["min"].append(
result[key]["min"])
return results_merged
def get_results(self):
"""Get the last result item.
Returns:
dict: The last result item in the experiment.
"""
if self.__results_dict is None:
self.__results_dict = {}
results_file = os.path.join(self.result_dir, "results.json")
if os.path.exists(results_file):
try:
with open(results_file, "r") as f:
self.__results_dict = json.load(f)
except Exception as e:
pass
return self.__results_dict
def ignore_experiment(self):
"""Create a flag file, so the browser ignores this experiment."""
self.update_meta_info(ignore=True)
def read_meta_info(self):
"""Reads the meta info of the experiment i.e. new name, stared or ignored"""
meta_dict = {}
meta_file = os.path.join(self.work_dir, ".exp_info")
if os.path.exists(meta_file):
with open(meta_file, "r") as mf:
meta_dict = json.load(mf)
self.meta_name = meta_dict.get("name")
self.meta_star = meta_dict.get("star", False)
self.meta_ignore = meta_dict.get("ignore", False)
def update_meta_info(self, name=None, star=None, ignore=None):
"""
Updates the meta info i.e. new name, stared or ignored and saves it in the experiment folder
Args:
name (str): New name of the experiment
star (bool): Flag, if experiment is starred/ favorited
ignore (boll): Flag, if experiment should be ignored
"""
if name is not None:
self.meta_name = name
if star is not None:
self.meta_star = star
if ignore is not None:
self.meta_ignore = ignore
meta_dict = {
"name": self.meta_name,
"star": self.meta_star,
"ignore": self.meta_ignore
}
meta_file = os.path.join(self.work_dir, ".exp_info")
with open(meta_file, "w") as mf:
json.dump(meta_dict, mf)
def __init__(self,
config=None,
name=None,
n_epochs=None,
seed=None,
base_dir=None,
globs=None,
resume=None,
ignore_resume_config=False,
resume_save_types=("model", "optimizer", "simple", "th_vars",
"results"),
resume_reset_epochs=True,
parse_sys_argv=False,
parse_config_sys_argv=True,
checkpoint_to_cpu=True,
safe_checkpoint_every_epoch=1,
use_visdomlogger=True,
visdomlogger_kwargs=None,
visdomlogger_c_freq=1,
use_explogger=True,
explogger_kwargs=None,
explogger_c_freq=100,
use_telegrammessagelogger=False,
telegrammessagelogger_kwargs=None,
telegrammessagelogger_c_freq=1000,
append_rnd_to_name=False):
# super(PytorchExperiment, self).__init__()
Experiment.__init__(self)
if parse_sys_argv:
config_path, resume_path = get_vars_from_sys_argv()
if config_path:
config = config_path
if resume_path:
resume = resume_path
self._config_raw = None
if isinstance(config, str):
self._config_raw = Config(file_=config,
update_from_argv=parse_config_sys_argv)
elif isinstance(config, Config):
self._config_raw = Config(config=config,
update_from_argv=parse_config_sys_argv)
elif isinstance(config, dict):
self._config_raw = Config(config=config,
update_from_argv=parse_config_sys_argv)
else:
self._config_raw = Config(update_from_argv=parse_config_sys_argv)
self.n_epochs = n_epochs
if 'n_epochs' in self._config_raw:
self.n_epochs = self._config_raw["n_epochs"]
if self.n_epochs is None:
self.n_epochs = 0
self._seed = seed
if 'seed' in self._config_raw:
self._seed = self._config_raw.seed
if self._seed is None:
random_data = os.urandom(4)
seed = int.from_bytes(random_data, byteorder="big")
self._config_raw.seed = seed
self._seed = seed
self.exp_name = name
if 'name' in self._config_raw:
self.exp_name = self._config_raw["name"]
if append_rnd_to_name:
rnd_str = ''.join(
random.choice(string.ascii_letters + string.digits)
for _ in range(5))
self.exp_name += "_" + rnd_str
if 'base_dir' in self._config_raw:
base_dir = self._config_raw["base_dir"]
self._checkpoint_to_cpu = checkpoint_to_cpu
self._safe_checkpoint_every_epoch = safe_checkpoint_every_epoch
self.results = dict()
# Init loggers
logger_list = []
self.vlog = None
if use_visdomlogger:
if visdomlogger_kwargs is None:
visdomlogger_kwargs = {}
self.vlog = PytorchVisdomLogger(name=self.exp_name,
**visdomlogger_kwargs)
if visdomlogger_c_freq is not None and visdomlogger_c_freq > 0:
logger_list.append((self.vlog, visdomlogger_c_freq))
self.elog = None
if use_explogger:
if explogger_kwargs is None:
explogger_kwargs = {}
self.elog = PytorchExperimentLogger(base_dir=base_dir,
experiment_name=self.exp_name,
**explogger_kwargs)
if explogger_c_freq is not None and explogger_c_freq > 0:
logger_list.append((self.elog, explogger_c_freq))
# Set results log dict to the right path
self.results = ResultLogDict("results-log.json",
base_dir=self.elog.result_dir)
self.tlog = None
if use_telegrammessagelogger:
if telegrammessagelogger_kwargs is None:
telegrammessagelogger_kwargs = {}
self.tlog = TelegramMessageLogger(**telegrammessagelogger_kwargs,
exp_name=self.exp_name)
if telegrammessagelogger_c_freq is not None and telegrammessagelogger_c_freq > 0:
logger_list.append((self.tlog, telegrammessagelogger_c_freq))
self.clog = CombinedLogger(*logger_list)
set_seed(self._seed)
# Do the resume stuff
self._resume_path = None
self._resume_save_types = resume_save_types
self._ignore_resume_config = ignore_resume_config
self._resume_reset_epochs = resume_reset_epochs
if resume is not None:
if isinstance(resume, str):
if resume == "last":
self._resume_path = os.path.join(
base_dir,
sorted(os.listdir(base_dir))[-1])
else:
self._resume_path = resume
elif isinstance(resume, PytorchExperiment):
self._resume_path = resume.elog.base_dir
if self._resume_path is not None and not self._ignore_resume_config:
self._config_raw.update(Config(file_=os.path.join(
self._resume_path, "config", "config.json")),
ignore=list(
map(lambda x: re.sub("^-+", "", x),
sys.argv)))
# self.elog.save_config(self.config, "config_pre")
if globs is not None:
zip_name = os.path.join(self.elog.save_dir, "sources.zip")
SourcePacker.zip_sources(globs, zip_name)
# Init objects in config
self.config = Config.init_objects(self._config_raw)
atexit.register(self.at_exit_func)
class PytorchExperiment(Experiment):
"""
A PytorchExperiment extends the basic
functionality of the :class:`.Experiment` class with
convenience features for PyTorch (and general logging) such as creating a folder structure,
saving, plotting results and checkpointing your experiment.
The basic life cycle of a PytorchExperiment is the same as
:class:`.Experiment`::
setup()
prepare()
for epoch in n_epochs:
train()
validate()
end()
where the distinction between the first two is that between them
PytorchExperiment will automatically restore checkpoints and save the
:attr:`_config_raw` in :meth:`._setup_internal`. Please see below for more
information on this.
To get your own experiment simply inherit from the PytorchExperiment and
overwrite the :meth:`.setup`, :meth:`.prepare`, :meth:`.train`,
:meth:`.validate` method (or you can use the `very` experimental decorator
:func:`.experimentify` to convert your class into a experiment).
Then you can run your own experiment by calling the :meth:`.run` method.
Internally PytorchExperiment will provide a number of member variables which
you can access.
- n_epochs
Number of epochs.
- exp_name
Name of your experiment.
- config
The (initialized) :class:`.Config` of your experiment. You can
access the uninitialized one via :attr:`_config_raw`.
- result
A dict in which you can store your result values. If a
:class:`.PytorchExperimentLogger` is used, results will be a
:class:`.ResultLogDict` that directly automatically writes to a file
and also stores the N last entries for each key for quick access
(e.g. to quickly get the running mean).
- vlog (if use_visdomlogger is True)
A :class:`.PytorchVisdomLogger` instance which can log your results
to a running visdom server. Start the server via
:code:`python -m visdom.server` or pass :data:`auto_start=True` in
the :attr:`visdomlogger_kwargs`.
- elog (if use_explogger is True)
A :class:`.PytorchExperimentLogger` instance which can log your
results to a given folder.
- tlog (if use_telegrammessagelogger is True)
A :class:`.TelegramMessageLogger` instance which can send the results to
your telegram account
- clog
A :class:`.CombinedLogger` instance which logs to all loggers with
different frequencies (specified with the :attr:`_c_freq` for each
logger where 1 means every time and N means every Nth time,
e.g. if you only want to send stuff to Visdom every 10th time).
The most important attribute is certainly :attr:`.config`, which is the
initialized :class:`.Config` for the experiment. To understand how it needs
to be structured to allow for automatic instantiation of types, please refer
to its documentation. If you decide not to use this functionality,
:attr:`config` and :attr:`_config_raw` are identical. **Beware however that
by default the Pytorchexperiment only saves the raw config** after
:meth:`.setup`. If you modify :attr:`config` during setup, make sure
to implement :meth:`._setup_internal` yourself should you want the modified
config to be saved::
def _setup_internal(self):
super(YourExperiment, self)._setup_internal() # calls .prepare_resume()
self.elog.save_config(self.config, "config")
Args:
config (dict or Config): Configures your experiment. If :attr:`name`,
:attr:`n_epochs`, :attr:`seed`, :attr:`base_dir` are given in the
config, it will automatically
overwrite the other args/kwargs with the values from the config.
In addition (defined by :attr:`parse_config_sys_argv`) the config
automatically parses the argv arguments and updates its values if a
key matches a console argument.
name (str):
The name of the PytorchExperiment.
n_epochs (int): The number of epochs (number of times the training
cycle will be executed).
seed (int): A random seed (which will set the random, numpy and
torch seed).
base_dir (str): A base directory in which the experiment result folder
will be created.
globs: The :func:`globals` of the script which is run. This is necessary
to get and save the executed files in the experiment folder.
resume (str or PytorchExperiment): Another PytorchExperiment or path to
the result dir from another PytorchExperiment from which it will
load the PyTorch modules and other member variables and resume
the experiment.
ignore_resume_config (bool): If :obj:`True` it will not resume with the
config from the resume experiment but take the current/own config.
resume_save_types (list or tuple): A list which can define which values
to restore when resuming. Choices are:
- "model" <-- Pytorch models
- "optimizer" <-- Optimizers
- "simple" <-- Simple python variables (basic types and lists/tuples
- "th_vars" <-- torch tensors/variables
- "results" <-- The result dict
parse_sys_argv (bool): Parsing the console arguments (argv) to get a
:attr:`config path` and/or :attr:`resume_path`.
parse_config_sys_argv (bool): Parse argv to update the config
(if the keys match).
checkpoint_to_cpu (bool): When checkpointing, transfer all tensors to
the CPU beforehand.
safe_checkpoint_every_epoch (int): Determines after how many epochs a
checkpoint is stored.
use_visdomlogger (bool): Use a :class:`.PytorchVisdomLogger`. Is
accessible via the :attr:`vlog` attribute.
visdomlogger_kwargs (dict): Keyword arguments for :attr:`vlog`
instantiation.
visdomlogger_c_freq (int): The frequency x (meaning one in x) with which
the :attr:`clog` will call the :attr:`vlog`.
use_explogger (bool): Use a :class:`.PytorchExperimentLogger`. Is
accessible via the :attr:`elog` attribute. This will create the
experiment folder structure.
explogger_kwargs (dict): Keyword arguments for :attr:`elog`
instantiation.
explogger_c_freq (int): The frequency x (meaning one in x) with which
the :attr:`clog` will call the :attr:`elog`.
use_telegrammessagelogger (bool): Use a :class:`.TelegramMessageLogger`. Is
accessible via the :attr:`tlog` attribute.
telegrammessagelogger_kwargs (dict): Keyword arguments for :attr:`tlog`
instantiation.
telegrammessagelogger_c_freq (int): The frequency x (meaning one in x) with which
the :attr:`clog` will call the :attr:`tlog`.
append_rnd_to_name (bool): If :obj:`True`, will append a random six
digit string to the experiment name.
"""
def __init__(self,
config=None,
name=None,
n_epochs=None,
seed=None,
base_dir=None,
globs=None,
resume=None,
ignore_resume_config=False,
resume_save_types=("model", "optimizer", "simple", "th_vars",
"results"),
resume_reset_epochs=True,
parse_sys_argv=False,
parse_config_sys_argv=True,
checkpoint_to_cpu=True,
safe_checkpoint_every_epoch=1,
use_visdomlogger=True,
visdomlogger_kwargs=None,
visdomlogger_c_freq=1,
use_explogger=True,
explogger_kwargs=None,
explogger_c_freq=100,
use_telegrammessagelogger=False,
telegrammessagelogger_kwargs=None,
telegrammessagelogger_c_freq=1000,
append_rnd_to_name=False):
# super(PytorchExperiment, self).__init__()
Experiment.__init__(self)
if parse_sys_argv:
config_path, resume_path = get_vars_from_sys_argv()
if config_path:
config = config_path
if resume_path:
resume = resume_path
self._config_raw = None
if isinstance(config, str):
self._config_raw = Config(file_=config,
update_from_argv=parse_config_sys_argv)
elif isinstance(config, Config):
self._config_raw = Config(config=config,
update_from_argv=parse_config_sys_argv)
elif isinstance(config, dict):
self._config_raw = Config(config=config,
update_from_argv=parse_config_sys_argv)
else:
self._config_raw = Config(update_from_argv=parse_config_sys_argv)
self.n_epochs = n_epochs
if 'n_epochs' in self._config_raw:
self.n_epochs = self._config_raw["n_epochs"]
if self.n_epochs is None:
self.n_epochs = 0
self._seed = seed
if 'seed' in self._config_raw:
self._seed = self._config_raw.seed
if self._seed is None:
random_data = os.urandom(4)
seed = int.from_bytes(random_data, byteorder="big")
self._config_raw.seed = seed
self._seed = seed
self.exp_name = name
if 'name' in self._config_raw:
self.exp_name = self._config_raw["name"]
if append_rnd_to_name:
rnd_str = ''.join(
random.choice(string.ascii_letters + string.digits)
for _ in range(5))
self.exp_name += "_" + rnd_str
if 'base_dir' in self._config_raw:
base_dir = self._config_raw["base_dir"]
self._checkpoint_to_cpu = checkpoint_to_cpu
self._safe_checkpoint_every_epoch = safe_checkpoint_every_epoch
self.results = dict()
# Init loggers
logger_list = []
self.vlog = None
if use_visdomlogger:
if visdomlogger_kwargs is None:
visdomlogger_kwargs = {}
self.vlog = PytorchVisdomLogger(name=self.exp_name,
**visdomlogger_kwargs)
if visdomlogger_c_freq is not None and visdomlogger_c_freq > 0:
logger_list.append((self.vlog, visdomlogger_c_freq))
self.elog = None
if use_explogger:
if explogger_kwargs is None:
explogger_kwargs = {}
self.elog = PytorchExperimentLogger(base_dir=base_dir,
experiment_name=self.exp_name,
**explogger_kwargs)
if explogger_c_freq is not None and explogger_c_freq > 0:
logger_list.append((self.elog, explogger_c_freq))
# Set results log dict to the right path
self.results = ResultLogDict("results-log.json",
base_dir=self.elog.result_dir)
self.tlog = None
if use_telegrammessagelogger:
if telegrammessagelogger_kwargs is None:
telegrammessagelogger_kwargs = {}
self.tlog = TelegramMessageLogger(**telegrammessagelogger_kwargs,
exp_name=self.exp_name)
if telegrammessagelogger_c_freq is not None and telegrammessagelogger_c_freq > 0:
logger_list.append((self.tlog, telegrammessagelogger_c_freq))
self.clog = CombinedLogger(*logger_list)
set_seed(self._seed)
# Do the resume stuff
self._resume_path = None
self._resume_save_types = resume_save_types
self._ignore_resume_config = ignore_resume_config
self._resume_reset_epochs = resume_reset_epochs
if resume is not None:
if isinstance(resume, str):
if resume == "last":
self._resume_path = os.path.join(
base_dir,
sorted(os.listdir(base_dir))[-1])
else:
self._resume_path = resume
elif isinstance(resume, PytorchExperiment):
self._resume_path = resume.elog.base_dir
if self._resume_path is not None and not self._ignore_resume_config:
self._config_raw.update(Config(file_=os.path.join(
self._resume_path, "config", "config.json")),
ignore=list(
map(lambda x: re.sub("^-+", "", x),
sys.argv)))
# self.elog.save_config(self.config, "config_pre")
if globs is not None:
zip_name = os.path.join(self.elog.save_dir, "sources.zip")
SourcePacker.zip_sources(globs, zip_name)
# Init objects in config
self.config = Config.init_objects(self._config_raw)
atexit.register(self.at_exit_func)
def process_err(self, e):
if self.elog is not None:
self.elog.text_logger.log_to(
"\n".join(traceback.format_tb(e.__traceback__)), "err")
def update_attributes(self, var_dict, ignore=()):
"""
Updates the member attributes with the attributes given in the var_dict
Args:
var_dict (dict): dict in which the update values stored. If a key matches a member attribute name
the member attribute will be updated
ignore (list or tuple): iterable of keys to ignore
"""
for key, val in var_dict.items():
if key == "results":
self.results.load(val)
continue
if key in ignore:
continue
if hasattr(self, key):
setattr(self, key, val)
def get_pytorch_modules(self, from_config=True):
"""
Returns all torch.nn.Modules stored in the experiment in a dict.
Args:
from_config (bool): Also get modules that are stored in the :attr:`.config` attribute.
Returns:
dict: Dictionary of PyTorch modules
"""
pyth_modules = dict()
for key, val in self.__dict__.items():
if isinstance(val, torch.nn.Module):
pyth_modules[key] = val
if from_config:
for key, val in self.config.items():
if isinstance(val, torch.nn.Module):
if type(key) == str:
key = "config." + key
pyth_modules[key] = val
return pyth_modules
def get_pytorch_optimizers(self, from_config=True):
"""
Returns all torch.optim.Optimizers stored in the experiment in a dict.
Args:
from_config (bool): Also get optimizers that are stored in the :attr:`.config`
attribute.
Returns:
dict: Dictionary of PyTorch optimizers
"""
pyth_optimizers = dict()
for key, val in self.__dict__.items():
if isinstance(val, torch.optim.Optimizer):
pyth_optimizers[key] = val
if from_config:
for key, val in self.config.items():
if isinstance(val, torch.optim.Optimizer):
if type(key) == str:
key = "config." + key
pyth_optimizers[key] = val
return pyth_optimizers
def get_simple_variables(self, ignore=()):
"""
Returns all standard variables in the experiment in a dict.
Specifically, this looks for types :class:`int`, :class:`float`, :class:`bytes`,
:class:`bool`, :class:`str`, :class:`set`, :class:`list`, :class:`tuple`.
Args:
ignore (list or tuple): Iterable of names which will be ignored
Returns:
dict: Dictionary of variables
"""
simple_vars = dict()
for key, val in self.__dict__.items():
if key in ignore:
continue
if isinstance(val,
(int, float, bytes, bool, str, set, list, tuple)):
simple_vars[key] = val
return simple_vars
def get_pytorch_tensors(self, ignore=()):
"""
Returns all torch.tensors in the experiment in a dict.
Args:
ignore (list or tuple): Iterable of names which will be ignored
Returns:
dict: Dictionary of PyTorch tensor
"""
pytorch_vars = dict()
for key, val in self.__dict__.items():
if key in ignore:
continue
if torch.is_tensor(val):
pytorch_vars[key] = val
return pytorch_vars
def get_pytorch_variables(self, ignore=()):
"""Same as :meth:`.get_pytorch_tensors`."""
return self.get_pytorch_tensors(ignore)
def save_results(self, name="results.json"):
"""
Saves the result dict as a json file in the result dir of the experiment logger.
Args:
name (str): The name of the json file in which the results are written.
"""
if self.elog is None:
return
with open(os.path.join(self.elog.result_dir, name), "w") as file_:
json.dump(self.results, file_, indent=4)
def save_pytorch_models(self):
"""Saves all torch.nn.Modules as model files in the experiment checkpoint folder."""
if self.elog is None:
return
pyth_modules = self.get_pytorch_modules()
for key, val in pyth_modules.items():
self.elog.save_model(val, key)
def load_pytorch_models(self):
"""Loads all model files from the experiment checkpoint folder."""
if self.elog is None:
return
pyth_modules = self.get_pytorch_modules()
for key, val in pyth_modules.items():
self.elog.load_model(val, key)
def log_simple_vars(self):
"""
Logs all simple python member variables as a json file in the experiment log folder.
The file will be names 'simple_vars.json'.
"""
if self.elog is None:
return
simple_vars = self.get_simple_variables()
with open(os.path.join(self.elog.log_dir, "simple_vars.json"),
"w") as file_:
json.dump(simple_vars, file_)
def load_simple_vars(self):
"""
Restores all simple python member variables from the 'simple_vars.json' file in the log
folder.
"""
if self.elog is None:
return
simple_vars = {}
with open(os.path.join(self.elog.log_dir, "simple_vars.json"),
"r") as file_:
simple_vars = json.load(file_)
self.update_attributes(simple_vars)
def save_checkpoint(self,
name="checkpoint",
save_types=("model", "optimizer", "simple", "th_vars",
"results"),
n_iter=None,
iter_format="{:05d}",
prefix=False):
"""
Saves a current model checkpoint from the experiment.
Args:
name (str): The name of the checkpoint file
save_types (list or tuple): What kind of member variables should be stored? Choices are:
"model" <-- Pytorch models,
"optimizer" <-- Optimizers,
"simple" <-- Simple python variables (basic types and lists/tuples),
"th_vars" <-- torch tensors,
"results" <-- The result dict
n_iter (int): Number of iterations. Together with the name, defined by the iter_format,
a file name will be created.
iter_format (str): Defines how the name and the n_iter will be combined.
prefix (bool): If True, the formatted n_iter will be prepended, otherwise appended.
"""
if self.elog is None:
return
model_dict = {}
optimizer_dict = {}
simple_dict = {}
th_vars_dict = {}
results_dict = {}
if "model" in save_types:
model_dict = self.get_pytorch_modules()
if "optimizer" in save_types:
optimizer_dict = self.get_pytorch_optimizers()
if "simple" in save_types:
simple_dict = self.get_simple_variables()
if "th_vars" in save_types:
th_vars_dict = self.get_pytorch_variables()
if "results" in save_types:
results_dict = {"results": self.results}
checkpoint_dict = {
**model_dict,
**optimizer_dict,
**simple_dict,
**th_vars_dict,
**results_dict
}
self.elog.save_checkpoint(name=name,
n_iter=n_iter,
iter_format=iter_format,
prefix=prefix,
move_to_cpu=self._checkpoint_to_cpu,
**checkpoint_dict)
def load_checkpoint(self,
name="checkpoint",
save_types=("model", "optimizer", "simple", "th_vars",
"results"),
n_iter=None,
iter_format="{:05d}",
prefix=False,
path=None):
"""
Loads a checkpoint and restores the experiment.
Make sure you have your torch stuff already on the right devices beforehand,
otherwise this could lead to errors e.g. when making a optimizer step
(and for some reason the Adam states are not already on the GPU:
https://discuss.pytorch.org/t/loading-a-saved-model-for-continue-training/17244/3 )
Args:
name (str): The name of the checkpoint file
save_types (list or tuple): What kind of member variables should be loaded? Choices are:
"model" <-- Pytorch models,
"optimizer" <-- Optimizers,
"simple" <-- Simple python variables (basic types and lists/tuples),
"th_vars" <-- torch tensors,
"results" <-- The result dict
n_iter (int): Number of iterations. Together with the name, defined by the iter_format,
a file name will be created and searched for.
iter_format (str): Defines how the name and the n_iter will be combined.
prefix (bool): If True, the formatted n_iter will be prepended, otherwise appended.
path (str): If no path is given then it will take the current experiment dir and formatted
name, otherwise it will simply use the path and the formatted name to define the
checkpoint file.
"""
if self.elog is None:
return
model_dict = {}
optimizer_dict = {}
simple_dict = {}
th_vars_dict = {}
results_dict = {}
if "model" in save_types:
model_dict = self.get_pytorch_modules()
if "optimizer" in save_types:
optimizer_dict = self.get_pytorch_optimizers()
if "simple" in save_types:
simple_dict = self.get_simple_variables()
if "th_vars" in save_types:
th_vars_dict = self.get_pytorch_variables()
if "results" in save_types:
results_dict = {"results": self.results}
checkpoint_dict = {
**model_dict,
**optimizer_dict,
**simple_dict,
**th_vars_dict,
**results_dict
}
if n_iter is not None:
name = name_and_iter_to_filename(name,
n_iter,
".pth.tar",
iter_format=iter_format,
prefix=prefix)
if path is None:
restore_dict = self.elog.load_checkpoint(name=name,
**checkpoint_dict)
else:
checkpoint_path = os.path.join(path, name)
if checkpoint_path.endswith("/"):
checkpoint_path = checkpoint_path[:-1]
restore_dict = self.elog.load_checkpoint_static(
checkpoint_file=checkpoint_path, **checkpoint_dict)
self.update_attributes(restore_dict)
def _end_internal(self):
"""Ends the experiment and stores the final results/checkpoint"""
if isinstance(self.results, ResultLogDict):
self.results.close()
self.save_results()
self.save_end_checkpoint()
self._save_exp_config()
self.print("Experiment ended. Checkpoints stored =)")
def _end_test_internal(self):
"""Ends the experiment after test and stores the final results and config"""
self.save_results()
self._save_exp_config()
self.print("Testing ended. Results stored =)")
def at_exit_func(self):
"""
Stores the results and checkpoint at the end (if not already stored).
This method is also called if an error occurs.
"""
if self._exp_state not in ("Ended", "Tested"):
if isinstance(self.results, ResultLogDict):
self.results.print_to_file("]")
self.save_checkpoint(name="checkpoint_exit-" + self._exp_state)
self.save_results()
self._save_exp_config()
self.print("Experiment exited. Checkpoints stored =)")
time.sleep(
2
) # allow checkpoint saving to finish. We need a better solution for this :D
def _setup_internal(self):
self.prepare_resume()
if self.elog is not None:
self.elog.save_config(self._config_raw, "config")
self._save_exp_config()
def _start_internal(self):
self._save_exp_config()
def prepare_resume(self):
"""Tries to resume the experiment by using the defined resume path or PytorchExperiment."""
checkpoint_file = ""
base_dir = ""
reset_epochs = self._resume_reset_epochs
if self._resume_path is not None:
if isinstance(self._resume_path, str):
if self._resume_path.endswith(".pth.tar"):
checkpoint_file = self._resume_path
base_dir = os.path.dirname(
os.path.dirname(checkpoint_file))
elif self._resume_path.endswith(
"checkpoint") or self._resume_path.endswith(
"checkpoint/"):
checkpoint_file = get_last_file(self._resume_path)
base_dir = os.path.dirname(
os.path.dirname(checkpoint_file))
elif "checkpoint" in os.listdir(
self._resume_path) and "config" in os.listdir(
self._resume_path):
checkpoint_file = get_last_file(self._resume_path)
base_dir = self._resume_path
else:
warnings.warn(
"You have not selected a valid experiment folder, will search all sub folders",
UserWarning)
if self.elog is not None:
self.elog.text_logger.log_to(
"You have not selected a valid experiment folder, will search all "
"sub folders", "warnings")
checkpoint_file = get_last_file(self._resume_path)
base_dir = os.path.dirname(
os.path.dirname(checkpoint_file))
if base_dir:
if not self._ignore_resume_config:
load_config = Config()
load_config.load(os.path.join(base_dir, "config/config.json"))
self._config_raw = load_config
self.config = Config.init_objects(self._config_raw)
self.print("Loaded existing config from:", base_dir)
if checkpoint_file:
self.load_checkpoint(name="",
path=checkpoint_file,
save_types=self._resume_save_types)
self._resume_path = checkpoint_file
shutil.copyfile(
checkpoint_file,
os.path.join(self.elog.checkpoint_dir, "0_checkpoint.pth.tar"))
self.print("Loaded existing checkpoint from:", checkpoint_file)
self._resume_reset_epochs = reset_epochs
if self._resume_reset_epochs:
self._epoch_idx = 0
def _end_epoch_internal(self, epoch):
self.save_results()
if epoch % self._safe_checkpoint_every_epoch == 0:
self.save_temp_checkpoint()
self._save_exp_config()
def _save_exp_config(self):
if self.elog is not None:
cur_time = time.strftime("%y-%m-%d_%H:%M:%S",
time.localtime(time.time()))
self.elog.save_config(
Config(
**{
'name': self.exp_name,
'time': self._time_start,
'state': self._exp_state,
'current_time': cur_time,
'epoch': self._epoch_idx
}), "exp")
def save_temp_checkpoint(self):
"""Saves the current checkpoint as checkpoint_current."""
self.save_checkpoint(name="checkpoint_current")
def save_end_checkpoint(self):
"""Saves the current checkpoint as checkpoint_last."""
self.save_checkpoint(name="checkpoint_last")
def add_result(self,
value,
name,
counter=None,
tag=None,
label=None,
plot_result=True,
plot_running_mean=False):
"""
Saves a results and add it to the result dict, this is similar to results[key] = val,
but in addition also logs the value to the combined logger
(it also stores in the results-logs file).
**This should be your preferred method to log your numeric values**
Args:
value: The value of your variable
name (str): The name/key of your variable
counter (int or float): A counter which can be seen as the x-axis of your value.
Normally you would just use the current epoch for this.
tag (str): A label/tag which can group similar values and will plot values with the same
label in the same plot
label: deprecated label
plot_result (bool): By default True, will also log all your values to the combined
logger (with show_value).
"""
if label is not None:
warnings.warn(
"label in add_result is deprecated, please use tag instead")
if tag is None:
tag = label
tag_name = tag
if tag_name is None:
tag_name = name
r_elem = ResultElement(data=value,
label=tag_name,
epoch=self._epoch_idx,
counter=counter)
self.results[name] = r_elem
if plot_result:
if tag is None:
legend = False
else:
legend = True
if plot_running_mean:
value = np.mean(self.results.running_mean_dict[name])
self.clog.show_value(value=value,
name=name,
tag=tag_name,
counter=counter,
show_legend=legend)
def get_result(self, name):
"""
Similar to result[key] this will return the values in the results dictionary with the given
name/key.
Args:
name (str): the name/key for which a value is stored.
Returns:
The value with the key 'name' in the results dict.
"""
return self.results.get(name)
def add_result_without_epoch(self, val, name):
"""
A faster method to store your results, has less overhead and does not call the combined
logger. Will only store to the results dictionary.
Args:
val: the value you want to add.
name (str): the name/key of your value.
"""
self.results[name] = val
def get_result_without_epoch(self, name):
"""
Similar to result[key] this will return the values in result with the given name/key.
Args:
name (str): the name/ key for which a value is stores.
Returns:
The value with the key 'name' in the results dict.
"""
return self.results.get(name)
def print(self, *args):
"""
Calls 'print' on the experiment logger or uses builtin 'print' if former is not
available.
"""
if self.elog is None:
print(*args)
else:
self.elog.print(*args)
def test_MNIST_experiment():
c = Config()
c.batch_size = 64
c.n_epochs = 50
c.learning_rate = 0.001
c.weight_decay = 5e-5
if torch.cuda.is_available():
c.use_cuda = True
else:
c.use_cuda = False
c.rnd_seed = 1
c.log_interval = 100
c.subset_size = 10
# model-specific
c.n_coupling = 8
c.n_filters = 64
exp = MNISTExperiment(
c,
name='mnist_test',
n_epochs=c.n_epochs,
seed=42,
base_dir='experiment_dir',
loggers={'visdom': ['visdom', {
"exp_name": "myenv"
}]})
exp.run()
exp.model.eval()
exp.model.to('cpu')
with torch.no_grad():
samples = exp.model.sample(16, device='cpu')
img_grid = make_grid(samples).permute((1, 2, 0))
plt.imshow(img_grid)
plt.show()
return exp.model
def run_experiment(experiment, configs, args, mods=None, **kwargs):
config = Config(file_=args.config) if args.config is not None else Config()
config.update_missing(configs[args.default_config])
if args.mods is not None:
for mod in args.mods:
config.update(mods[mod])
config = Config(config=config, update_from_argv=True)
# GET EXISTING EXPERIMENTS TO BE ABLE TO SKIP CERTAIN CONFIGS
if args.skip_existing:
existing_configs = []
for exp in os.listdir(args.base_dir):
try:
existing_configs.append(
Config(file_=os.path.join(args.base_dir, exp, "config",
"config.json")))
except Exception as e:
pass
if args.grid is not None:
grid = GridSearch().read(args.grid)
else:
grid = [{}]
for combi in grid:
config.update(combi)
if args.skip_existing:
skip_this = False
for existing_config in existing_configs:
if existing_config.contains(config):
skip_this = True
break
if skip_this:
continue
loggers = {}
if args.visdomlogger:
loggers["visdom"] = ("visdom", {}, 1)
exp = experiment(config=config,
base_dir=args.base_dir,
resume=args.resume,
ignore_resume_config=args.ignore_resume_config,
loggers=loggers,
**kwargs)
if not args.test:
exp.run()
else:
exp.run_test()
import numpy as np
import torch
from trixi.util import Config
from experiment import MNISTexperiment
from util import plot_dependency_map
import matplotlib.pyplot as plt
c = Config()
c.batch_size = 128
c.n_epochs = 10
c.learning_rate = 0.001
if torch.cuda.is_available():
c.use_cuda = True
else:
c.use_cuda = False
c.rnd_seed = 1
c.log_interval = 100
exp = MNISTexperiment(config=c,
name='test',
n_epochs=c.n_epochs,
seed=c.rnd_seed,
base_dir='./experiment_dir',
loggers={"visdom": ["visdom", {
"exp_name": "myenv"
}]})
# # run backpropagation for each dimension to compute what other
# # dimensions it depends on.
# exp.setup()
# d = 28
def test_Resnet():
c = Config()
c.batch_size = 64
c.batch_size_test = 1000
c.n_epochs = 10
c.learning_rate = 0.01
c.momentum = 0.9
if torch.cuda.is_available():
c.use_cuda = True
else:
c.use_cuda = False
c.rnd_seed = 1
c.log_interval = 200
exp = MNIST_classification(config=c,
name='experiment',
n_epochs=c.n_epochs,
seed=42,
base_dir='./experiment_dir',
loggers={"visdom": "visdom"})
exp.run()
def process_base_dir(base_dir,
view_dir="",
default_val="-",
short_len=25,
ignore_keys=IGNORE_KEYS):
"""Create an overview table of all experiments in the given directory.
Args:
directory (str): A directory containing experiment folders.
default_val (str): Default value if an entry is missing.
short_len (int): Cut strings to this length. Full string in alt-text.
Returns:
dict: {"ccols": Columns for config entries,
"rcols": Columns for result entries,
"rows": The actual data}
"""
full_dir = os.path.join(base_dir, view_dir)
config_keys = set()
result_keys = set()
exps = []
non_exps = []
### Load Experiments with keys / different param values
for sub_dir in sorted(os.listdir(full_dir)):
dir_path = os.path.join(full_dir, sub_dir)
if os.path.isdir(dir_path):
try:
exp = ExperimentReader(full_dir, sub_dir)
if exp.ignore:
continue
config_keys.update(list(exp.config.flat().keys()))
result_keys.update(list(exp.get_results().keys()))
exps.append(exp)
except Exception as e:
print("Could not load experiment: ", dir_path)
print(e)
print("-" * 20)
non_exps.append(os.path.join(view_dir, sub_dir))
### Get not common val keys
diff_keys = list(
Config.difference_config_static(*[xp.config for xp in exps]).flat())
### Remove unwanted keys
config_keys -= set(ignore_keys)
result_keys -= set(ignore_keys)
### Generate table rows
sorted_c_keys1 = sorted([c for c in config_keys if c in diff_keys],
key=lambda x: str(x).lower())
sorted_c_keys2 = sorted([c for c in config_keys if c not in diff_keys],
key=lambda x: str(x).lower())
sorted_r_keys = sorted(result_keys, key=lambda x: str(x).lower())
rows = []
for exp in exps:
config_row = []
for key in sorted_c_keys1:
attr_strng = str(exp.config.flat().get(key, default_val))
config_row.append((attr_strng, attr_strng[:short_len]))
for key in sorted_c_keys2:
attr_strng = str(exp.config.flat().get(key, default_val))
config_row.append((attr_strng, attr_strng[:short_len]))
result_row = []
for key in sorted_r_keys:
attr_strng = str(exp.get_results().get(key, default_val))
result_row.append((attr_strng, attr_strng[:short_len]))
name = exp.exp_name
time = exp.exp_info.get(
"time", default_val) if "time" in exp.exp_info else exp.config.get(
"time", default_val)
state = exp.exp_info.get(
"state",
default_val) if "state" in exp.exp_info else exp.config.get(
"state", default_val)
epoch = exp.exp_info.get(
"epoch",
default_val) if "epoch" in exp.exp_info else exp.config.get(
"epoch", default_val)
rows.append((os.path.relpath(exp.work_dir,
base_dir), exp.star, str(name), str(time),
str(state), str(epoch), config_row, result_row))
return {
"ccols1": sorted_c_keys1,
"ccols2": sorted_c_keys2,
"rcols": sorted_r_keys,
"rows": rows,
"noexp": non_exps
}
def make_defaults():
DEFAULTS = Config(
# Base
name="gqn",
description=DESCRIPTION,
n_epochs=1000000,
batch_size=36,
batch_size_val=36,
seed=1,
device="cuda",
# Data
split_val=3, # index for set of 5
split_test=4, # index for set of 5
data_module=loader,
dataset="shepard_metzler_5_parts",
data_dir=None, # will be set for data_module if not None
debug=
0, # 1 for single repeating batch, 2 for single viewpoint (i.e. reconstruct known images)
generator_train=loader.RandomBatchGenerator,
generator_val=loader.LinearBatchGenerator,
num_viewpoints_val=8, # use this many viewpoints in validation
shuffle_viewpoints_val=False,
augmenter=MultiThreadedAugmenter,
augmenter_kwargs={"num_processes": 8},
# Model
model=GenerativeQueryNetwork,
model_kwargs={
"in_channels": 3,
"query_channels": 7,
"r_channels": 256,
"encoder_kwargs": {
"activation_op": nn.ReLU
},
"decoder_kwargs": {
"z_channels": 64,
"h_channels": 128,
"scale": 4,
"core_repeat": 12
}
},
model_init_weights_args=None, # e.g. [nn.init.kaiming_normal_, 1e-2],
model_init_bias_args=None, # e.g. [nn.init.constant_, 0],
# Learning
optimizer=optim.Adam,
optimizer_kwargs={"weight_decay": 1e-5},
lr_initial=5e-4,
lr_final=5e-5,
lr_cutoff=16e4, # lr is increased linearly in cutoff epochs
sigma_initial=2.0,
sigma_final=0.7,
sigma_cutoff=2e4, # sigma is increased linearly in cutoff epochs
kl_weight_initial=0.05,
kl_weight_final=1.0,
kl_weight_cutoff=1e5, # kl_weight is increased linearly in cutoff epochs
nll_weight=1.0,
# Logging
backup_every=10000,
validate_every=1000,
validate_subset=0.01, # validate only this percentage randomly
show_every=100,
val_example_samples=
10, # draw this many random samples for last validation item
test_on_val=True, # test on the validation set
)
SHAREDCORES = Config(
model_kwargs={"decoder_kwargs": {
"core_shared": True
}})
MODS = {"SHAREDCORES": SHAREDCORES}
return {"DEFAULTS": DEFAULTS}, MODS
def get_config():
# Set your own path, if needed.
data_root_dir = os.path.abspath(
'data') # The path where the downloaded dataset is stored.
c = Config(
update_from_argv=True,
# Train parameters
num_classes=8,
in_channels=1,
batch_size=8,
patch_size=64,
n_epochs=50,
learning_rate=0.0002,
fold=
0, # The 'splits.pkl' may contain multiple folds. Here we choose which one we want to use.
device=
"cuda", # 'cuda' is the default CUDA device, you can use also 'cpu'. For more information, see https://pytorch.org/docs/stable/notes/cuda.html
# Logging parameters
name='Basic_Unet',
plot_freq=10, # How often should stuff be shown in visdom
append_rnd_string=False,
start_visdom=True,
do_instancenorm=
True, # Defines whether or not the UNet does a instance normalization in the contracting path
do_load_checkpoint=False,
checkpoint_dir='',
# Adapt to your own path, if needed.
google_drive_id='1RzPB1_bqzQhlWvU-YGvZzhx2omcDh38C',
dataset_name='CHD_segmentation_dataset',
base_dir=os.path.abspath(
'output_experiment'), # Where to log the output of the experiment.
data_root_dir=
data_root_dir, # The path where the downloaded dataset is stored.
data_dir=os.path.join(
data_root_dir, 'CHD_segmentation_dataset/preprocessed'
), # This is where your training and validation data is stored
data_test_dir=os.path.join(data_root_dir,
'CHD_segmentation_dataset/preprocessed'
), # This is where your test data is stored
split_dir=os.path.join(
data_root_dir, 'CHD_segmentation_dataset'
), # This is where the 'splits.pkl' file is located, that holds your splits.
scaled_image_16_dir=os.path.join(
data_root_dir, 'CHD_segmentation_dataset/scaled_to_16'),
scaled_image_32_dir=os.path.join(
data_root_dir, 'CHD_segmentation_dataset/scaled_to_32'),
scaled_image_64_dir=os.path.join(
data_root_dir, 'CHD_segmentation_dataset/scaled_to_64'),
stage_1_dir=os.path.join(data_root_dir,
'CHD_segmentation_dataset/stage_1'),
stage_1_dir_32=os.path.join(data_root_dir,
'CHD_segmentation_dataset/stage_1_32'),
# stage_1_dir = os.path.join(data_root_dir, 'CHD_segmentation_dataset/stage_1')
)
print(c)
return c
