clean up

tests/test_utils.py

@@ -0,0 +1,110 @@
+# Name: Wenqi Wang
+# Github username: acse-ww721
+
+import pytest
+import os
+import numpy as np
+import pandas as pd
+from utils.folder_utils import get_current_directory, find_folder, create_folder
+from utils.model_utils_tf import get_initial_weights
+from utils.time_utils import time_select, is_leap_year, days_check
+
+
+def test_get_current_directory():
+ current_dir = get_current_directory()
+ assert os.path.isabs(
+ current_dir
+ ) # Ensure that the returned path is an absolute path
+ assert os.path.exists(current_dir) # Ensure that the returned path exists
+
+
+def test_find_folder():
+ folder_path = find_folder("GB", "data", "raw_data", "ASOS_DATA")
+ assert os.path.isabs(
+ folder_path
+ ) # Ensure that the returned path is an absolute path
+ assert os.path.exists(folder_path) # Ensure that the returned path exists
+
+
+def test_create_folder(tmpdir):
+ # Use a temporary directory for testing
+ temp_dir = str(tmpdir)
+ data_folder = os.path.join(temp_dir, "data")
+ data_category = "raw_data"
+ output_folder = "ASOS_DATA"
+ c = "GB"
+
+ folder_path = create_folder(c, data_folder, data_category, output_folder)
+
+ assert os.path.isabs(
+ folder_path
+ ) # Ensure that the returned path is an absolute path
+ assert os.path.exists(folder_path) # Ensure that the created folder exists
+ assert os.path.isdir(folder_path) # Ensure that it is a directory
+
+
+def test_get_initial_weights():
+ output_size = 4
+ weights = get_initial_weights(output_size)
+ assert len(weights) == 2 # Ensure that the returned list contains two elements
+ assert isinstance(
+ weights[0], np.ndarray
+ ) # Ensure that the first element is a numpy array
+ assert isinstance(
+ weights[1], np.ndarray
+ ) # Ensure that the second element is a numpy array
+ assert weights[0].shape == (
+ output_size,
+ 6,
+ ) # Ensure that the shape of the weight matrix is as expected
+ assert weights[1].shape == (
+ 6,
+ ) # Ensure that the shape of the bias vector is as expected
+ assert np.all(
+ weights[1][:2] == np.array([1.0, 0.0])
+ ) # Ensure that the bias vector is initialized correctly
+
+
+def test_time_select():
+ data = {"Date": ["20220101", "20220102", "20220103"], "Value": [10, 15, 20]}
+ df = pd.DataFrame(data)
+ start_date = "20220102"
+ end_date = "20220103"
+ filtered_df = time_select(df, "Date", start_date, end_date)
+ assert (
+ len(filtered_df) == 2
+ ) # Ensure that the filtered DataFrame has the expected number of rows
+ assert (
+ "Date" in filtered_df.columns
+ ) # Ensure that the 'Date' column is present in the filtered DataFrame
+ assert (
+ "Value" in filtered_df.columns
+ ) # Ensure that the 'Value' column is present in the filtered DataFrame
+ assert (
+ filtered_df["Date"] >= start_date
+ ).all() # Ensure that all dates are greater than or equal to the start_date
+ assert (
+ filtered_df["Date"] <= end_date
+ ).all() # Ensure that all dates are less than or equal to the end_date
+
+
+def test_is_leap_year():
+ assert is_leap_year(2020) is True # 2020 is a leap year
+ assert is_leap_year(2100) is False # 2100 is not a leap year
+ assert is_leap_year(2000) is True # 2000 is a leap year (divisible by 400)
+ assert (
+ is_leap_year(1900) is False
+ ) # 1900 is not a leap year (divisible by 4 and 100 but not by 400)
+
+
+def test_days_check():
+ assert days_check("2023", "02") == [
+ str(day).zfill(2) for day in range(1, 29)
+ ] # February 2023 has 28 days (not a leap year)
+ assert days_check("2020", "02") == [
+ str(day).zfill(2) for day in range(1, 30)
+ ] # February 2020 has 29 days (leap year)
+ assert days_check("2023", "04") == [
+ str(day).zfill(2) for day in range(1, 31)
+ ] # April has 30 days
+ assert days_check("2023", "13") == [] # Invalid month should return an empty list