add testset, script to make + test

quartz_solar_forecast/dataset/make_test_set.py

@@ -0,0 +1,50 @@
+"""
+Make a random test set
+
+This takes a random subset of times and for various pv ids and makes a test set
+"""
+from typing import Optional
+import pandas as pd
+import numpy as np
+import os
+
+test_start_date = pd.Timestamp("2021-01-01")
+test_end_date = pd.Timestamp("2021-02-01")
+
+# this have been chosen from the entire training set. This ideas
+pv_ids = [ 8215,  8229,  8253,  8266,  8267,  8281, 16474, 16477, 16480,
+       16483,  8376, 16570, 16597,  8411,  8412,  8419,  8420,  8453,
+        8464,  8505, 16715, 16717,  8551,  8558, 16769, 16770,  8587,
+        8591,  8612,  8648,  8670,  8708,  8713, 16920, 16921,  8801,
+       17034, 17035,  8856, 17062,  8914, 17166,  9069,  9070,  9101,
+        9107, 17333,  9153,  9171,  9173]
+
+np.random.seed(42)
+
+
+def make_test_set(
+    output_file_name: Optional[str] = None, number_of_samples_per_system: int = 50
+):
+    """
+    Make a test set of random times and pv ids
+
+    :param output_file_name: the name of the file to write the test set to
+    :param number_of_samples_per_system: the number of samples to take per pv id
+    """
+
+    if output_file_name is None:
+        # get the folder where this file is
+        output_file_name = os.path.dirname(os.path.abspath(__file__)) + "/testset.csv"
+
+    test_set = []
+    for pv_id in pv_ids:
+        ts = pd.date_range(start=test_start_date, end=test_end_date, freq="15min")
+        ts = ts[np.random.choice(len(ts), size=number_of_samples_per_system, replace=False)]
+        test_set.append(pd.DataFrame({"pv_id": pv_id, "datetime": ts}))
+    test_set = pd.concat(test_set)
+    test_set.to_csv(output_file_name, index=False)
+
+    return test_set
+
+# To run the script, un comment the following line and run this file
+make_test_set()