def test_load_bad_file():
with pytest.raises(AssertionError) as AE:
pproc = Preprocessor()
gen = pproc.load_df_from_directory_gen('dir_with_corrupt_file')
for t in gen:
print(t)
assert 'Could not load' in str(AE)
def test_remove_baseline_bad_threshold():
with pytest.raises(AssertionError) as AE:
pproc = Preprocessor()
gen = pproc.load_df_from_directory_gen('good_data')
for t in gen:
pproc.remove_baseline(t.position_data, threshold=0)
assert 'Threshold' in str(AE)
def test_source_dir_with_bad_file_name():
with pytest.raises(AssertionError) as AE:
pproc = Preprocessor()
gen = pproc.load_df_from_directory_gen('dir_with_bad_file_name')
for t in gen:
print(t)
assert 'No source files' in str(AE)
def test_invalid_source_dir():
with pytest.raises(AssertionError) as AE:
pproc = Preprocessor()
gen = pproc.lopad_df_from_directory_gen('$')
for t in gen:
print(t)
assert 'Destination directory does not exist' in str(AE)
def test_remove_baseline_empty_df():
with pytest.raises(AssertionError) as AE:
pproc = Preprocessor()
gen = pproc.load_df_from_directory_gen('good_data')
df = pd.DataFrame()
for t in gen:
pproc.remove_baseline(df)
assert 'No data' in str(AE)
def test_data_processing_and_save_csv():
path_to_raw_data = 'good_data/'
path_to_trials = 'good_data/csv/'
pproc = Preprocessor()
raw_data_gen = pproc.load_df_from_directory_gen(dir_path=path_to_raw_data,
cols=('x', 'y'))
for trial in raw_data_gen:
trial.preprocess(pproc)
trial.save_df(trial.data, path_to_trials)
list_of_raws = glob(os.path.join(path_to_raw_data, 'li_*.csv'))
list_of_results = glob(os.path.join(path_to_trials, '*.csv'))
assert list_of_raws == list_of_results
def test_preprocessor_instance():
pproc = Preprocessor()
assert isinstance(pproc, Preprocessor)
# -*- coding: utf-8 -*-
from DataProcessing import Preprocessor, Trial
path_to_raw_data = '..\\data\\results\\simonvisual\\1'
path_to_database = '..\\data\\processed_data.mat'
sample_rate = 240 #S/s
pproc = Preprocessor()
raw_data_gen = pproc.load_df_from_directory_gen(dir_path=path_to_raw_data)
for trial in raw_data_gen:
trial.preprocess(pproc, axes=['x', 'y'])
pproc.plot(trial.velocity_data[['x', 'y']], mode='pandas')
pproc.plot(trial.filtered_velocity_data[['x', 'y']], mode='pandas')
print(1)
break
# TODO: calculate bounds
# bounds_low = [0, 0.1, -50, -50]
# bounds_high = [t.max(), t.max()-t.min(), 200, 200]
popt_vx, pcov_vx = curve_fit(func, t, Vx, p0=guess_vx)
perr_vx = np.sqrt(np.diag(pcov_vx))
popt_vy, pcov_vy = curve_fit(func, t, Vy, p0=guess_vy)
perr_vy = np.sqrt(np.diag(pcov_vy))
fit_vx = func(t, *popt_vx)
fit_vy = func(t, *popt_vy)
return fit_vx, fit_vy, popt_vx, perr_vx, popt_vy, perr_vy
if __name__ == '__main__':
path_to_raw_data = '..\\data\\results\\simonvisual\\1'
path_to_trials = '..\\data\\trials'
pproc = Preprocessor()
raw_data_gen = pproc.load_df_from_directory_gen(dir_path=path_to_raw_data, cols=('x', 'y'))
# demo for a single trial:
for trial in raw_data_gen:
trial.preprocess(pproc)
trial.save_df(trial.data, path_to_trials)
res = optimize_jerk(trial, True)
print(f'Parameters: {res[1]}')
print(f'Parameter errors: {res[2]}')
break