I praktisk videnskabelig databehandling er signaler ofte forurenet med støj, hvilket gør det vanskeligt at udtrække meningsfulde egenskaber. Filtrering og toppunktdetektion er vigtige værktøjer til analyse af sådanne støjfyldte data. I denne udfordring skal du bruge `scipy.signal` til at behandle en tidsserie ved at fjerne støj og derefter identificere væsentlige toppe, som ofte er interessante i ingeniørmæssige og videnskabelige anvendelser.


import unittest
import user_code
import ast
import re   
import unittest
import numpy as np
from scipy import signal
import importlib

class TestTask(unittest.TestCase):
    def setUp(self):
        self.time = np.linspace(0, 10, 500)
        np.random.seed(0)
        self.clean_signal = np.sin(2 * np.pi * 1.5 * self.time)
        self.noise = np.random.normal(0, 0.5, self.time.shape)
        self.noisy_signal = self.clean_signal + self.noise

    def test_returns_indices(self):
        import user_code
        importlib.reload(user_code)
        indices = user_code.filter_and_find_peaks(self.time, self.noisy_signal)
        _dynamic_test(
            self,
            isinstance(indices, (list, np.ndarray)),
            "Function returns a list or array of indices.",
            f"Expected a list or array, got {type(indices)}."
        )
        _dynamic_test(
            self,
            all(isinstance(i, (int, np.integer)) for i in indices),
            "All returned values are integer indices.",
            "Returned indices are not all integers."
        )

    def test_filter_is_used(self):
        import user_code
        importlib.reload(user_code)
        # Get filtered signal by re-implementing expected filter
        b, a = signal.butter(N=4, Wn=0.1, btype='low')
        filtered = signal.filtfilt(b, a, self.noisy_signal)
        # Find peaks in filtered signal
        expected_peaks, _ = signal.find_peaks(filtered)
        result_peaks = user_code.filter_and_find_peaks(self.time, self.noisy_signal)
        # Test that result is not just peaks from original noisy data
        noisy_peaks, _ = signal.find_peaks(self.noisy_signal)
        # Compare overlap with expected
        overlap = len(set(result_peaks).intersection(set(expected_peaks)))
        _dynamic_test(
            self,
            overlap >= len(expected_peaks) * 0.8,
            "Detected peaks match those found after correct filtering.",
            f"Detected peaks do not match expected peaks after filtering. Overlap: {overlap} / {len(expected_peaks)}"
        )
        # Should not just return peaks from noisy data
        overlap_noisy = len(set(result_peaks).intersection(set(noisy_peaks)))
        _dynamic_test(
            self,
            overlap > overlap_noisy,
            "Peak detection is performed on filtered signal, not raw data.",
            "Peak detection appears to be performed on raw data, not filtered signal."
        )

    def test_indices_are_local_maxima(self):
        import user_code
        importlib.reload(user_code)
        b, a = signal.butter(N=4, Wn=0.1, btype='low')
        filtered = signal.filtfilt(b, a, self.noisy_signal)
        peaks = user_code.filter_and_find_peaks(self.time, self.noisy_signal)
        for idx in peaks:
            if 1 <= idx < len(filtered) - 1:
                cond = filtered[idx] > filtered[idx-1] and filtered[idx] > filtered[idx+1]
                _dynamic_test(
                    self,
                    cond,
                    f"Index {idx} is a local maximum in the filtered signal.",
                    f"Index {idx} is not a local maximum in the filtered signal."
                )

def _dynamic_test(test_case, condition, success_message, failure_message):
    if condition:
        test_case._testMethodName = success_message
        test_case.assertTrue(True, success_message)
    else:
        test_case._testMethodName = failure_message
        test_case.fail(failure_message)

def normalize_text(text):
    text = text.lower()
    text = re.sub(r"\s{2,}", " ", text)
    text = re.sub(r"\s*([,:?])\s*", r"\1 ", text)
    return text.strip()

def change_var(code: str, var_name: str, value: str) -> str:
    tree = ast.parse(code)
    lines = code.splitlines()
    for i, node in enumerate(tree.body):
        if isinstance(node, ast.Assign):
            for target in node.targets:
                if isinstance(target, ast.Name) and target.id == var_name:
                    start_line = node.lineno - 1
                    line = lines[start_line]
                    indent = ' ' * (len(line) - len(line.lstrip()))
                    lines[start_line] = f"{indent}{var_name} = {value}"
                    next_line = len(lines)
                    for next_node in tree.body[i+1:]:
                        if hasattr(next_node, 'lineno'):
                            next_line = next_node.lineno - 1
                            break
                    if next_line > start_line + 1:
                        lines[start_line+1:next_line] = []
                    
                    return '\\n'.join(lines)
    return code

if __name__ == "__main__":
    unittest.main()


test_main.py

Et omfattende kursus designet til studerende med stærke matematiske forudsætninger og mellemliggende til avancerede Python-færdigheder, med fokus på at udnytte SciPy til videnskabelig databehandling, optimering, integration og avancerede matematiske operationer.

Udforsk strukturen, kernemodulerne og de væsentlige funktioner i SciPy til videnskabelig databehandling.

Dyk ned i SciPys kraftfulde lineære algebrafunktioner til løsning af virkelige matematiske problemer.

Behersk optimeringsteknikker og rodfindingsalgoritmer ved hjælp af SciPy til videnskabelige og ingeniørmæssige problemer.

Udforsk avancerede matematiske operationer, herunder integration, interpolation og grundlæggende signalbehandling med SciPy.

Challenge: Signalfiltrering og Analyse

Løsning