diff --git a/.gitignore b/.gitignore
index f5e425f752a1de0de0c68036a54e0d19450320bb..04bc0d1b9ec314e6398dcc7907d73686f00655f3 100644
--- a/.gitignore
+++ b/.gitignore
@@ -62,6 +62,7 @@ activate_env.sh
/tmp/
/logging/
/HPC_logging/
+/tmp_downloads/
# test related data #
#####################
diff --git a/mlair/plotting/postprocessing_plotting.py b/mlair/plotting/postprocessing_plotting.py
index c96026736074bc29497073e5669d557fdc6d647d..52b1121e7b8f165476d3c27d9e24b077a731f8e5 100644
--- a/mlair/plotting/postprocessing_plotting.py
+++ b/mlair/plotting/postprocessing_plotting.py
@@ -701,12 +701,23 @@ class PlotCompetitiveSkillScore(AbstractPlotClass):
self._model_setup = model_setup
self._labels = None
self._data = self._prepare_data(data)
+ default_plot_name = self.plot_name
+ # draw full detail plot
+ self.plot_name = default_plot_name + "_full_detail"
self._plot()
self._save()
- # draw also a vertical version
- self.plot_name += "_vertical"
+ # draw also a vertical full detail version
+ self.plot_name = default_plot_name + "_full_detail_vertical"
self._plot_vertical()
self._save()
+ # draw default plot with only model comparison
+ self.plot_name = default_plot_name
+ self._plot(single_model_comparison=True)
+ self._save()
+ # draw also a vertical full detail version
+ self.plot_name = default_plot_name + "_vertical"
+ self._plot_vertical(single_model_comparison=True)
+ self._save()
def _prepare_data(self, data: pd.DataFrame) -> pd.DataFrame:
"""
@@ -724,12 +735,13 @@ class PlotCompetitiveSkillScore(AbstractPlotClass):
self._labels = [str(i) + "d" for i in data.index.levels[1].values]
return data.stack(level=0).reset_index(level=2, drop=True).reset_index(name="data")
- def _plot(self):
+ def _plot(self, single_model_comparison=False):
"""Plot skill scores of the comparisons."""
size = max([len(np.unique(self._data.comparison)), 6])
fig, ax = plt.subplots(figsize=(size, size * 0.8))
- order = self._create_pseudo_order()
- sns.boxplot(x="comparison", y="data", hue="ahead", data=self._data, whis=1., ax=ax, palette="Blues_d",
+ data = self._filter_comparisons(self._data) if single_model_comparison is True else self._data
+ order = self._create_pseudo_order(data)
+ sns.boxplot(x="comparison", y="data", hue="ahead", data=data, whis=1., ax=ax, palette="Blues_d",
showmeans=True, meanprops={"markersize": 3, "markeredgecolor": "k"}, flierprops={"marker": "."},
order=order)
ax.axhline(y=0, color="grey", linewidth=.5)
@@ -740,11 +752,12 @@ class PlotCompetitiveSkillScore(AbstractPlotClass):
ax.legend(handles, self._labels)
plt.tight_layout()
- def _plot_vertical(self):
+ def _plot_vertical(self, single_model_comparison=False):
"""Plot skill scores of the comparisons, but vertically aligned."""
fig, ax = plt.subplots()
- order = self._create_pseudo_order()
- sns.boxplot(y="comparison", x="data", hue="ahead", data=self._data, whis=1., ax=ax, palette="Blues_d",
+ data = self._filter_comparisons(self._data) if single_model_comparison is True else self._data
+ order = self._create_pseudo_order(data)
+ sns.boxplot(y="comparison", x="data", hue="ahead", data=data, whis=1., ax=ax, palette="Blues_d",
showmeans=True, meanprops={"markersize": 3, "markeredgecolor": "k"}, flierprops={"marker": "."},
order=order)
# ax.axhline(x=0, color="grey", linewidth=.5)
@@ -754,12 +767,17 @@ class PlotCompetitiveSkillScore(AbstractPlotClass):
ax.legend(handles, self._labels)
plt.tight_layout()
- def _create_pseudo_order(self):
+ def _create_pseudo_order(self, data):
"""Provide first predefined elements and append all remaining."""
first_elements = [f"{self._model_setup}-persi", "ols-persi", f"{self._model_setup}-ols"]
- uniq, index = np.unique(first_elements + self._data.comparison.unique().tolist(), return_index=True)
+ first_elements = list(filter(lambda x: x in data.comparison.tolist(), first_elements))
+ uniq, index = np.unique(first_elements + data.comparison.unique().tolist(), return_index=True)
return uniq[index.argsort()]
+ def _filter_comparisons(self, data):
+ filtered_headers = list(filter(lambda x: "nn-" in x, data.comparison.unique()))
+ return data[data.comparison.isin(filtered_headers)]
+
def _lim(self) -> Tuple[float, float]:
"""
Calculate axis limits from data (Can be used to set axis extend).
diff --git a/mlair/reference_data_handler/__init__.py b/mlair/reference_models/__init__.py
similarity index 100%
rename from mlair/reference_data_handler/__init__.py
rename to mlair/reference_models/__init__.py
diff --git a/mlair/reference_data_handler/abstract_reference_data_handler.py b/mlair/reference_models/abstract_reference_model.py
similarity index 93%
rename from mlair/reference_data_handler/abstract_reference_data_handler.py
rename to mlair/reference_models/abstract_reference_model.py
index 19c102946f4958992c82c7cf31d1281baa7418b9..e187e7ef62e3fe84f7ba2149a490f63ac718308f 100644
--- a/mlair/reference_data_handler/abstract_reference_data_handler.py
+++ b/mlair/reference_models/abstract_reference_model.py
@@ -36,12 +36,13 @@ class AbstractReferenceModel(ABC):
return res
-class AbstractReferenceb2share(AbstractReferenceModel):
+class AbstractReferenceB2share(AbstractReferenceModel):
"""
Abstract class for reference models located on b2share (eudat or fz-juelich)
See also https://github.com/EUDAT-Training/B2SHARE-Training/blob/master/api/01_Retrieve_existing_record.md
"""
+
def __init__(self, b2share_hosturl: str, b2share_bucket: str, b2share_key: str):
super().__init__()
self.b2share_hosturl = b2share_hosturl
@@ -63,3 +64,6 @@ class AbstractReferenceb2share(AbstractReferenceModel):
out=f"{tmp_download_path}{self.b2share_key}",
bar=self.bar_custom
)
+
+ def make_reference_available_locally(self):
+ raise NotImplementedError
diff --git a/mlair/reference_data_handler/intellio3_v1_reference.py b/mlair/reference_models/reference_model_intellio3_v1.py
similarity index 84%
rename from mlair/reference_data_handler/intellio3_v1_reference.py
rename to mlair/reference_models/reference_model_intellio3_v1.py
index aec05a3bc66a7451088824893abae72af00621fa..5f0bae48a32f54a0e58d0e6d6139370af8106d84 100644
--- a/mlair/reference_data_handler/intellio3_v1_reference.py
+++ b/mlair/reference_models/reference_model_intellio3_v1.py
@@ -11,10 +11,10 @@ import xarray as xr
import shutil
from mlair.configuration.path_config import check_path_and_create
-from mlair.reference_data_handler.abstract_reference_data_handler import AbstractReferenceb2share
+from mlair.reference_models.abstract_reference_model import AbstractReferenceB2share
-class IntelliO3Reference(AbstractReferenceb2share):
+class IntelliO3_ts_v1(AbstractReferenceB2share):
"""
Reference handler that extracts IntelliO3-ts v1.0 forecasts (Kleinert, 2021).
@@ -57,25 +57,24 @@ class IntelliO3Reference(AbstractReferenceb2share):
:return: base dir of tmp path and list of forecast files
:rtype: tuple(str, list(str))
"""
- for root, dirs, file_names in os.walk(self.tmp_extract_path+self.orig_forecast_path):
+ for base_dir, dirs, file_names in os.walk(self.tmp_extract_path + self.orig_forecast_path):
pass
- return root, file_names
+ return base_dir, file_names
def read_and_drop(self, sel_coords: dict = None):
"""
Reads original forecast files, renames coord type and store forecasts as NetCdf4 files
:param sel_coords:
- :type sel_coords:
"""
if sel_coords is None:
- sel_coords = {'type': 'CNN'}
+ sel_coords = {'type': ['CNN']}
in_path, files = self.file_list()
check_path_and_create(self.ref_store_path)
for infile in files:
- data = xr.open_dataarray(f"{in_path}{infile}")
+ data = xr.open_dataarray(os.path.join(in_path, infile))
data = data.sel(**sel_coords)
- data.coords['type'] = (self.ref_name)
- data.to_netcdf(f"{self.ref_store_path}{infile}")
+ data.coords['type'] = ['nn']
+ data.to_netcdf(os.path.join(self.ref_store_path, infile))
def make_reference_available_locally(self, remove_tmp_dir: bool = True):
"""
@@ -93,6 +92,6 @@ class IntelliO3Reference(AbstractReferenceb2share):
if __name__ == '__main__':
- io3 = IntelliO3Reference('IntelliO3-ts')
+ io3 = IntelliO3_ts_v1('IntelliO3-ts')
io3.make_reference_available_locally()
diff --git a/mlair/run_modules/pre_processing.py b/mlair/run_modules/pre_processing.py
index 21aebd62bab490363797c0ef0624daa1d488097b..bf54b0619f94d21524edc95a52c2ad49dab788c5 100644
--- a/mlair/run_modules/pre_processing.py
+++ b/mlair/run_modules/pre_processing.py
@@ -13,7 +13,7 @@ import numpy as np
import pandas as pd
from mlair.data_handler import DataCollection, AbstractDataHandler
-from mlair.helpers import TimeTracking
+from mlair.helpers import TimeTracking, to_list
from mlair.configuration import path_config
from mlair.helpers.join import EmptyQueryResult
from mlair.run_modules.run_environment import RunEnvironment
@@ -65,6 +65,7 @@ class PreProcessing(RunEnvironment):
self.data_store.set("stations", valid_stations)
self.split_train_val_test()
self.report_pre_processing()
+ self.prepare_competitors()
def report_pre_processing(self):
"""Log some metrics on data and create latex report."""
@@ -325,6 +326,25 @@ class PreProcessing(RunEnvironment):
if transformation_dict is not None:
self.data_store.set("transformation", transformation_dict)
+ def prepare_competitors(self):
+ """
+ Prepare competitor models already in the preprocessing stage. This is performed here, because some models might
+ need to have internet access, which is depending on the operating system not possible during postprocessing.
+ This method checks currently only, if the Intelli03-ts-v1 model is requested as competitor and downloads the
+ data if required.
+ """
+ logging.info("Searching for competitors to be prepared for use.")
+ competitors = to_list(self.data_store.get_default("competitors", default=[]))
+ if len(competitors) > 0:
+ for competitor_name in competitors:
+ if competitor_name.lower() == "IntelliO3-ts-v1".lower():
+ logging.info("Prepare IntelliO3-ts-v1 model")
+ from mlair.reference_models.reference_model_intellio3_v1 import IntelliO3_ts_v1
+ path = os.path.join(self.data_store.get("competitor_path"), competitor_name)
+ IntelliO3_ts_v1("IntelliO3-ts-v1", path).make_reference_available_locally(remove_tmp_dir=False)
+ else:
+ logging.info("No preparation required because no competitor was provided to the workflow.")
+
def f_proc(data_handler, station, name_affix, store, **kwargs):
"""
diff --git a/requirements.txt b/requirements.txt
index 77f300df6a8b97395fd5c6771310f7b7b9537a59..47995404d7c8c202071197bb2aab9dba81e18689 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -24,6 +24,7 @@ Markdown==3.2.1
matplotlib==3.2.0
mock==4.0.1
more-itertools==8.2.0
+netcdf4==1.5.5.1
numpy==1.18.1
packaging==20.3
pandas==1.0.1
diff --git a/run.py b/run.py
index 9f5d0f083081d097e24b86f5dfc2b3b380e28e9b..f2bb336e8a886a3c0c4d60736c77b5ebc27cad67 100644
--- a/run.py
+++ b/run.py
@@ -5,6 +5,7 @@ import argparse
from mlair.workflows import DefaultWorkflow
from mlair.helpers import remove_items
from mlair.configuration.defaults import DEFAULT_PLOT_LIST
+import os
def load_stations():
@@ -20,9 +21,14 @@ def load_stations():
def main(parser_args):
plots = remove_items(DEFAULT_PLOT_LIST, "PlotConditionalQuantiles")
- workflow = DefaultWorkflow(stations=load_stations(),
- train_model=False, create_new_model=False, network="UBA",
- evaluate_bootstraps=False, plot_list=["PlotStationMap"], **parser_args.__dict__)
+ workflow = DefaultWorkflow( # stations=load_stations(),
+ # stations=["DEBW087","DEBW013", "DEBW107", "DEBW076"],
+ stations=["DEBW013", "DEBW087", "DEBW107", "DEBW076"],
+ train_model=False, create_new_model=True, network="UBA",
+ evaluate_bootstraps=False, # plot_list=["PlotCompetitiveSkillScore"],
+ competitors=["test_model", "test_model2"],
+ competitor_path=os.path.join(os.getcwd(), "data", "comp_test"),
+ **parser_args.__dict__)
workflow.run()
diff --git a/run_mixed_sampling.py b/run_mixed_sampling.py
index 04683a17ede641a5370aaeef741d2f4546f966b7..6ffb659953157060c39afb5960821e729df555dd 100644
--- a/run_mixed_sampling.py
+++ b/run_mixed_sampling.py
@@ -25,9 +25,9 @@ def main(parser_args):
interpolation_limit=(3, 1), overwrite_local_data=False,
sampling=("hourly", "daily"),
statistics_per_var=stats,
- create_new_model=False, train_model=False, epochs=1,
- window_history_size=48,
- window_history_offset=17,
+ create_new_model=True, train_model=False, epochs=1,
+ window_history_size=6 * 24 + 16,
+ window_history_offset=16,
kz_filter_length=[100 * 24, 15 * 24],
kz_filter_iter=[4, 5],
start="2006-01-01",
diff --git a/test/test_reference_data_handler/test_abstract_reference_data_handler.py b/test/test_reference_models/test_abstract_reference_model.py
similarity index 80%
rename from test/test_reference_data_handler/test_abstract_reference_data_handler.py
rename to test/test_reference_models/test_abstract_reference_model.py
index 98267fc913df299647c0fe8a1d95074e5d22bf04..4c455219be88a97e4f7f13fec663bd95bc8cf7d8 100644
--- a/test/test_reference_data_handler/test_abstract_reference_data_handler.py
+++ b/test/test_reference_models/test_abstract_reference_model.py
@@ -1,8 +1,8 @@
import pytest
import mock
-from mlair.reference_data_handler.abstract_reference_data_handler import AbstractReferenceModel
-from mlair.reference_data_handler.abstract_reference_data_handler import AbstractReferenceb2share
+from mlair.reference_models.abstract_reference_model import AbstractReferenceModel
+from mlair.reference_models.abstract_reference_model import AbstractReferenceB2share
class TestAbstractReferenceDataHandler:
@@ -32,13 +32,13 @@ class TestAbstractReferenceb2share:
b2share_hosturl = "b2share.url"
b2share_bucket = "1111-2222-3333"
b2share_key = "b2share_key.tar.gz"
- return AbstractReferenceb2share(b2share_hosturl, b2share_bucket, b2share_key)
+ return AbstractReferenceB2share(b2share_hosturl, b2share_bucket, b2share_key)
def test_inheritance(self):
- assert issubclass(AbstractReferenceb2share, AbstractReferenceModel)
+ assert issubclass(AbstractReferenceB2share, AbstractReferenceModel)
def test_init(self, ar):
- assert isinstance(ar, AbstractReferenceb2share)
+ assert isinstance(ar, AbstractReferenceB2share)
assert ar.b2share_hosturl == "b2share.url"
assert ar.b2share_bucket == "1111-2222-3333"
assert ar.b2share_key == "b2share_key.tar.gz"
@@ -51,4 +51,4 @@ class TestAbstractReferenceb2share:
pass
def test_download_from_b2share(self):
- pass
\ No newline at end of file
+ pass
diff --git a/test/test_reference_data_handler/test_intellio3_v1_reference.py b/test/test_reference_models/test_reference_model_intellio3_v1.py
similarity index 61%
rename from test/test_reference_data_handler/test_intellio3_v1_reference.py
rename to test/test_reference_models/test_reference_model_intellio3_v1.py
index 6582b99ec197f282a561bcdecd4d331cfc964327..bfd4cae3f04121ab0dd6a25e5d88ff88d73b54d3 100644
--- a/test/test_reference_data_handler/test_intellio3_v1_reference.py
+++ b/test/test_reference_models/test_reference_model_intellio3_v1.py
@@ -1,19 +1,18 @@
import pytest
-
-from mlair.reference_data_handler.intellio3_v1_reference import IntelliO3Reference
-from mlair.reference_data_handler.abstract_reference_data_handler import AbstractReferenceb2share
+from mlair.reference_models.reference_model_intellio3_v1 import IntelliO3_ts_v1
+from mlair.reference_models.abstract_reference_model import AbstractReferenceB2share
class TestIntelliO3Reference:
@pytest.fixture
def io3(self):
- return IntelliO3Reference("IntelliO3-test")
+ return IntelliO3_ts_v1("IntelliO3-test")
def test_init_none_path(self):
- io3 = IntelliO3Reference("IntelliO3-test")
- assert isinstance(io3, IntelliO3Reference)
+ io3 = IntelliO3_ts_v1("IntelliO3-test")
+ assert isinstance(io3, IntelliO3_ts_v1)
assert io3.ref_name == "IntelliO3-test"
assert io3.ref_store_path == "IntelliO3-test/"
assert io3.tmp_extract_path == "tmp_downloads/"
@@ -24,12 +23,12 @@ class TestIntelliO3Reference:
assert io3.b2share_hosturl == "https://b2share.eudat.eu"
def test_init_extra_path(self):
- io3 = IntelliO3Reference("IntelliO3-test", "DummyExtraPath/")
- assert isinstance(io3, IntelliO3Reference)
+ io3 = IntelliO3_ts_v1("IntelliO3-test", "DummyExtraPath/")
+ assert isinstance(io3, IntelliO3_ts_v1)
assert io3.ref_store_path == "DummyExtraPath/"
def test_inheritance(self):
- assert issubclass(IntelliO3Reference, AbstractReferenceb2share)
+ assert issubclass(IntelliO3_ts_v1, AbstractReferenceB2share)
def test_untar_forecasts(self, io3):
pass
@@ -41,5 +40,4 @@ class TestIntelliO3Reference:
pass
def test_make_reference_available_locally(self):
- io3 = IntelliO3Reference("IntelliO3-test", "DummyExtraPath/")
-
+ pass
diff --git a/test/test_reference_data_handler/tmp_downloads/IntelliO3-ts.tar.gz b/test/test_reference_models/tmp_downloads/IntelliO3-ts.tar.gz
similarity index 100%
rename from test/test_reference_data_handler/tmp_downloads/IntelliO3-ts.tar.gz
rename to test/test_reference_models/tmp_downloads/IntelliO3-ts.tar.gz
diff --git a/test/test_run_modules/test_pre_processing.py b/test/test_run_modules/test_pre_processing.py
index 11c46e99fb38489f5cbb26a8a87032049c96c7ca..b5a1914e6b2aacd238f244d304184d9754326db7 100644
--- a/test/test_run_modules/test_pre_processing.py
+++ b/test/test_run_modules/test_pre_processing.py
@@ -46,8 +46,12 @@ class TestPreProcessing:
with PreProcessing():
assert caplog.record_tuples[0] == ('root', 20, 'PreProcessing started')
assert caplog.record_tuples[1] == ('root', 20, 'check valid stations started (preprocessing)')
- assert caplog.record_tuples[-1] == ('root', 20, PyTestRegex(r'run for \d+:\d+:\d+ \(hh:mm:ss\) to check 5 '
+ assert caplog.record_tuples[-3] == ('root', 20, PyTestRegex(r'run for \d+:\d+:\d+ \(hh:mm:ss\) to check 5 '
r'station\(s\). Found 5/5 valid stations.'))
+ assert caplog.record_tuples[-2] == ('root', 20, "Searching for competitors to be prepared for use.")
+ assert caplog.record_tuples[-1] == (
+ 'root', 20, "No preparation required because no competitor was provided "
+ "to the workflow.")
RunEnvironment().__del__()
def test_run(self, obj_with_exp_setup):