From fa85a5695743536f2660b592889e8583ad0c2894 Mon Sep 17 00:00:00 2001
From: leufen1 <l.leufen@fz-juelich.de>
Date: Mon, 21 Sep 2020 09:58:27 +0200
Subject: [PATCH] trainable is now renamed by train_model
---
mlair/configuration/defaults.py | 2 +-
mlair/run_modules/experiment_setup.py | 20 ++++++++---------
mlair/run_modules/model_setup.py | 6 ++---
mlair/run_modules/training.py | 10 ++++-----
mlair/run_script.py | 4 ++--
mlair/workflows/default_workflow.py | 4 ++--
.../test_run_modules/test_experiment_setup.py | 22 +++++++++----------
test/test_run_modules/test_training.py | 6 ++---
8 files changed, 37 insertions(+), 37 deletions(-)
diff --git a/mlair/configuration/defaults.py b/mlair/configuration/defaults.py
index 31746ec8..b048712f 100644
--- a/mlair/configuration/defaults.py
+++ b/mlair/configuration/defaults.py
@@ -18,7 +18,7 @@ DEFAULT_TRANSFORMATION = {"scope": "data", "method": "standardise"}
DEFAULT_HPC_LOGIN_LIST = ["ju", "hdfmll"] # ju[wels} #hdfmll(ogin)
DEFAULT_HPC_HOST_LIST = ["jw", "hdfmlc"] # first part of node names for Juwels (jw[comp], hdfmlc(ompute).
DEFAULT_CREATE_NEW_MODEL = True
-DEFAULT_TRAINABLE = True
+DEFAULT_TRAIN_MODEL = True
DEFAULT_FRACTION_OF_TRAINING = 0.8
DEFAULT_EXTREME_VALUES = None
DEFAULT_EXTREMES_ON_RIGHT_TAIL_ONLY = False
diff --git a/mlair/run_modules/experiment_setup.py b/mlair/run_modules/experiment_setup.py
index 51e710c2..d66954b0 100644
--- a/mlair/run_modules/experiment_setup.py
+++ b/mlair/run_modules/experiment_setup.py
@@ -10,7 +10,7 @@ from mlair.configuration import path_config
from mlair import helpers
from mlair.configuration.defaults import DEFAULT_STATIONS, DEFAULT_VAR_ALL_DICT, DEFAULT_NETWORK, DEFAULT_STATION_TYPE, \
DEFAULT_START, DEFAULT_END, DEFAULT_WINDOW_HISTORY_SIZE, DEFAULT_OVERWRITE_LOCAL_DATA, DEFAULT_TRANSFORMATION, \
- DEFAULT_HPC_LOGIN_LIST, DEFAULT_HPC_HOST_LIST, DEFAULT_CREATE_NEW_MODEL, DEFAULT_TRAINABLE, \
+ DEFAULT_HPC_LOGIN_LIST, DEFAULT_HPC_HOST_LIST, DEFAULT_CREATE_NEW_MODEL, DEFAULT_TRAIN_MODEL, \
DEFAULT_FRACTION_OF_TRAINING, DEFAULT_EXTREME_VALUES, DEFAULT_EXTREMES_ON_RIGHT_TAIL_ONLY, DEFAULT_PERMUTE_DATA, \
DEFAULT_BATCH_SIZE, DEFAULT_EPOCHS, DEFAULT_TARGET_VAR, DEFAULT_TARGET_DIM, DEFAULT_WINDOW_LEAD_TIME, \
DEFAULT_DIMENSIONS, DEFAULT_TIME_DIM, DEFAULT_INTERPOLATION_METHOD, DEFAULT_INTERPOLATION_LIMIT, \
@@ -39,7 +39,7 @@ class ExperimentSetup(RunEnvironment):
* `data_path` [.]
* `create_new_model` [.]
* `bootstrap_path` [.]
- * `trainable` [.]
+ * `train_model` [.]
* `fraction_of_training` [.]
* `extreme_values` [train]
* `extremes_on_right_tail_only` [train]
@@ -94,7 +94,7 @@ class ExperimentSetup(RunEnvironment):
# set post-processing instructions
self._set_param("evaluate_bootstraps", evaluate_bootstraps, scope="general.postprocessing")
- create_new_bootstraps = max([self.data_store.get("trainable", "general"), create_new_bootstraps or False])
+ create_new_bootstraps = max([self.data_store.get("train_model", "general"), create_new_bootstraps or False])
self._set_param("create_new_bootstraps", create_new_bootstraps, scope="general.postprocessing")
self._set_param("number_of_bootstraps", number_of_bootstraps, default=20, scope="general.postprocessing")
self._set_param("plot_list", plot_list, default=DEFAULT_PLOT_LIST, scope="general.postprocessing")
@@ -144,8 +144,8 @@ class ExperimentSetup(RunEnvironment):
:param test_start:
:param test_end:
:param use_all_stations_on_all_data_sets:
- :param trainable: train a new model from scratch or resume training with existing model if `True` (default) or
- freeze loaded model and do not perform any modification on it. ``trainable`` is set to `True` if
+ :param train_model: train a new model from scratch or resume training with existing model if `True` (default) or
+ freeze loaded model and do not perform any modification on it. ``train_model`` is set to `True` if
``create_new_model`` is `True`.
:param fraction_of_train: given value is used to split between test data and train data (including validation data).
The value of ``fraction_of_train`` must be in `(0, 1)` but is recommended to be in the interval `[0.6, 0.9]`.
@@ -166,7 +166,7 @@ class ExperimentSetup(RunEnvironment):
parameter is set to `False`, make sure, that a suitable model already exists in the experiment path. This model
must fit in terms of input and output dimensions as well as ``window_history_size`` and ``window_lead_time`` and
must be implemented as a :py:mod:`model class <src.model_modules.model_class>` and imported in
- :py:mod:`model setup <src.run_modules.model_setup>`. If ``create_new_model`` is `True`, parameter ``trainable``
+ :py:mod:`model setup <src.run_modules.model_setup>`. If ``create_new_model`` is `True`, parameter ``train_model``
is automatically set to `True` too.
:param bootstrap_path:
:param permute_data_on_training: shuffle train data individually for each station if `True`. This is performed each
@@ -215,7 +215,7 @@ class ExperimentSetup(RunEnvironment):
time_dim=None,
interpolation_method=None,
interpolation_limit=None, train_start=None, train_end=None, val_start=None, val_end=None, test_start=None,
- test_end=None, use_all_stations_on_all_data_sets=None, trainable: bool = None, fraction_of_train: float = None,
+ test_end=None, use_all_stations_on_all_data_sets=None, train_model: bool = None, fraction_of_train: float = None,
experiment_path=None, plot_path: str = None, forecast_path: str = None, overwrite_local_data = None, sampling: str = "daily",
create_new_model = None, bootstrap_path=None, permute_data_on_training = None, transformation=None,
train_min_length=None, val_min_length=None, test_min_length=None, extreme_values: list = None,
@@ -233,11 +233,11 @@ class ExperimentSetup(RunEnvironment):
self._set_param("login_nodes", login_nodes, default=DEFAULT_HPC_LOGIN_LIST)
self._set_param("create_new_model", create_new_model, default=DEFAULT_CREATE_NEW_MODEL)
if self.data_store.get("create_new_model"):
- trainable = True
+ train_model = True
data_path = self.data_store.get("data_path")
bootstrap_path = path_config.set_bootstrap_path(bootstrap_path, data_path, sampling)
self._set_param("bootstrap_path", bootstrap_path)
- self._set_param("trainable", trainable, default=DEFAULT_TRAINABLE)
+ self._set_param("train_model", train_model, default=DEFAULT_TRAIN_MODEL)
self._set_param("fraction_of_training", fraction_of_train, default=DEFAULT_FRACTION_OF_TRAINING)
self._set_param("extreme_values", extreme_values, default=DEFAULT_EXTREME_VALUES, scope="train")
self._set_param("extremes_on_right_tail_only", extremes_on_right_tail_only,
@@ -331,7 +331,7 @@ class ExperimentSetup(RunEnvironment):
# set post-processing instructions
self._set_param("evaluate_bootstraps", evaluate_bootstraps, default=DEFAULT_EVALUATE_BOOTSTRAPS,
scope="general.postprocessing")
- create_new_bootstraps = max([self.data_store.get("trainable", "general"),
+ create_new_bootstraps = max([self.data_store.get("train_model", "general"),
create_new_bootstraps or DEFAULT_CREATE_NEW_BOOTSTRAPS])
self._set_param("create_new_bootstraps", create_new_bootstraps, scope="general.postprocessing")
self._set_param("number_of_bootstraps", number_of_bootstraps, default=DEFAULT_NUMBER_OF_BOOTSTRAPS,
diff --git a/mlair/run_modules/model_setup.py b/mlair/run_modules/model_setup.py
index 3dc56f01..e4bff5a6 100644
--- a/mlair/run_modules/model_setup.py
+++ b/mlair/run_modules/model_setup.py
@@ -31,7 +31,7 @@ class ModelSetup(RunEnvironment):
Required objects [scope] from data store:
* `experiment_path` [.]
* `experiment_name` [.]
- * `trainable` [.]
+ * `train_model` [.]
* `create_new_model` [.]
* `generator` [train]
* `model_class` [.]
@@ -64,7 +64,7 @@ class ModelSetup(RunEnvironment):
self.model_name = self.path % "%s.h5"
self.checkpoint_name = self.path % "model-best.h5"
self.callbacks_name = self.path % "model-best-callbacks-%s.pickle"
- self._trainable = self.data_store.get("trainable")
+ self._train_model = self.data_store.get("train_model")
self._create_new_model = self.data_store.get("create_new_model")
self._run()
@@ -80,7 +80,7 @@ class ModelSetup(RunEnvironment):
self.plot_model()
# load weights if no training shall be performed
- if not self._trainable and not self._create_new_model:
+ if not self._train_model and not self._create_new_model:
self.load_weights()
# create checkpoint
diff --git a/mlair/run_modules/training.py b/mlair/run_modules/training.py
index f8909e15..113765e0 100644
--- a/mlair/run_modules/training.py
+++ b/mlair/run_modules/training.py
@@ -23,7 +23,7 @@ class Training(RunEnvironment):
Train your model with this module.
This module isn't required to run, if only a fresh post-processing is preformed. Either remove training call from
- your run script or set create_new_model and trainable both to false.
+ your run script or set create_new_model and train_model both to false.
Schedule of training:
#. set_generators(): set generators for training, validation and testing and distribute according to batch size
@@ -40,7 +40,7 @@ class Training(RunEnvironment):
* `model_name` [model]
* `experiment_name` [.]
* `experiment_path` [.]
- * `trainable` [.]
+ * `train_model` [.]
* `create_new_model` [.]
* `generator` [train, val, test]
* `plot_path` [.]
@@ -72,7 +72,7 @@ class Training(RunEnvironment):
self.epochs = self.data_store.get("epochs")
self.callbacks: CallbackHandler = self.data_store.get("callbacks", "model")
self.experiment_name = self.data_store.get("experiment_name")
- self._trainable = self.data_store.get("trainable")
+ self._train_model = self.data_store.get("train_model")
self._create_new_model = self.data_store.get("create_new_model")
self._run()
@@ -80,12 +80,12 @@ class Training(RunEnvironment):
"""Run training. Details in class description."""
self.set_generators()
self.make_predict_function()
- if self._trainable:
+ if self._train_model:
self.train()
self.save_model()
self.report_training()
else:
- logging.info("No training has started, because trainable parameter was false.")
+ logging.info("No training has started, because train_model parameter was false.")
def make_predict_function(self) -> None:
"""
diff --git a/mlair/run_script.py b/mlair/run_script.py
index 6dea98ba..10851da1 100644
--- a/mlair/run_script.py
+++ b/mlair/run_script.py
@@ -6,7 +6,7 @@ import inspect
def run(stations=None,
- trainable=None, create_new_model=None,
+ train_model=None, create_new_model=None,
window_history_size=None,
experiment_date="testrun",
variables=None, statistics_per_var=None,
@@ -39,5 +39,5 @@ def run(stations=None,
if __name__ == "__main__":
from mlair.model_modules.model_class import MyBranchedModel
- run(statistics_per_var={'o3': 'dma8eu', "temp": "maximum"}, trainable=True,
+ run(statistics_per_var={'o3': 'dma8eu', "temp": "maximum"}, train_model=True,
create_new_model=True, model=MyBranchedModel, station_type="background")
diff --git a/mlair/workflows/default_workflow.py b/mlair/workflows/default_workflow.py
index 3dba7e6c..85d6726b 100644
--- a/mlair/workflows/default_workflow.py
+++ b/mlair/workflows/default_workflow.py
@@ -14,7 +14,7 @@ class DefaultWorkflow(Workflow):
the mentioned ordering."""
def __init__(self, stations=None,
- trainable=None, create_new_model=None,
+ train_model=None, create_new_model=None,
window_history_size=None,
experiment_date="testrun",
variables=None, statistics_per_var=None,
@@ -58,7 +58,7 @@ class DefaultWorkflowHPC(Workflow):
Training and PostProcessing in exact the mentioned ordering."""
def __init__(self, stations=None,
- trainable=None, create_new_model=None,
+ train_model=None, create_new_model=None,
window_history_size=None,
experiment_date="testrun",
variables=None, statistics_per_var=None,
diff --git a/test/test_run_modules/test_experiment_setup.py b/test/test_run_modules/test_experiment_setup.py
index abd265f5..ff355085 100644
--- a/test/test_run_modules/test_experiment_setup.py
+++ b/test/test_run_modules/test_experiment_setup.py
@@ -38,7 +38,7 @@ class TestExperimentSetup:
data_store = exp_setup.data_store
# experiment setup
assert data_store.get("data_path", "general") == prepare_host()
- assert data_store.get("trainable", "general") is True
+ assert data_store.get("train_model", "general") is True
assert data_store.get("create_new_model", "general") is True
assert data_store.get("fraction_of_training", "general") == 0.8
# set experiment name
@@ -99,7 +99,7 @@ class TestExperimentSetup:
data_store = exp_setup.data_store
# experiment setup
assert data_store.get("data_path", "general") == prepare_host()
- assert data_store.get("trainable", "general") is True
+ assert data_store.get("train_model", "general") is True
assert data_store.get("create_new_model", "general") is True
assert data_store.get("fraction_of_training", "general") == 0.5
# set experiment name
@@ -145,22 +145,22 @@ class TestExperimentSetup:
# use all stations on all data sets (train, val, test)
assert data_store.get("use_all_stations_on_all_data_sets", "general.test") is False
- def test_init_trainable_behaviour(self):
- exp_setup = ExperimentSetup(trainable=False, create_new_model=True)
+ def test_init_train_model_behaviour(self):
+ exp_setup = ExperimentSetup(train_model=False, create_new_model=True)
data_store = exp_setup.data_store
- assert data_store.get("trainable", "general") is True
+ assert data_store.get("train_model", "general") is True
assert data_store.get("create_new_model", "general") is True
- exp_setup = ExperimentSetup(trainable=False, create_new_model=False)
+ exp_setup = ExperimentSetup(train_model=False, create_new_model=False)
data_store = exp_setup.data_store
- assert data_store.get("trainable", "general") is False
+ assert data_store.get("train_model", "general") is False
assert data_store.get("create_new_model", "general") is False
- exp_setup = ExperimentSetup(trainable=True, create_new_model=True)
+ exp_setup = ExperimentSetup(train_model=True, create_new_model=True)
data_store = exp_setup.data_store
- assert data_store.get("trainable", "general") is True
+ assert data_store.get("train_model", "general") is True
assert data_store.get("create_new_model", "general") is True
- exp_setup = ExperimentSetup(trainable=True, create_new_model=False)
+ exp_setup = ExperimentSetup(train_model=True, create_new_model=False)
data_store = exp_setup.data_store
- assert data_store.get("trainable", "general") is True
+ assert data_store.get("train_model", "general") is True
assert data_store.get("create_new_model", "general") is False
def test_compare_variables_and_statistics(self):
diff --git a/test/test_run_modules/test_training.py b/test/test_run_modules/test_training.py
index c5f1ba9d..d6a7f829 100644
--- a/test/test_run_modules/test_training.py
+++ b/test/test_run_modules/test_training.py
@@ -73,7 +73,7 @@ class TestTraining:
path_plot = os.path.join(path, "plots")
os.makedirs(path_plot)
obj.data_store.set("plot_path", path_plot, "general")
- obj._trainable = True
+ obj._train_model = True
obj._create_new_model = False
yield obj
if os.path.exists(path):
@@ -203,9 +203,9 @@ class TestTraining:
def test_no_training(self, ready_to_init, caplog):
caplog.set_level(logging.INFO)
- ready_to_init.data_store.set("trainable", False)
+ ready_to_init.data_store.set("train_model", False)
Training()
- message = "No training has started, because trainable parameter was false."
+ message = "No training has started, because train_model parameter was false."
assert caplog.record_tuples[-2] == ("root", 20, message)
def test_run(self, ready_to_run):
--
GitLab