def test_Fs_is_persisted():
Fs = 1234.0
with UnlinkingTempfile(".f32") as tmp_fn:
eeg = F32(tmp_fn, "w+", shape=(1,1), Fs=Fs)
eeg.close()
eeg2 = F32(tmp_fn)
assert_equal(Fs, eeg2.Fs)
def _internal_write_and_read_back(data):
with UnlinkingTempfile(".f32") as tmp_fn:
eeg = F32(tmp_fn, "w+", shape=data.shape)
eeg[:,:] = data
eeg.close()
eeg2 = F32(tmp_fn)
read_data = np.array(eeg2[:])
assert_array_almost_equal(data, read_data)
def test_filtered_reading():
with UnlinkingTempfile(".f32") as tmp_fn:
data = np.random.random((100,1))
eeg = F32(tmp_fn, "w+", shape=data.shape)
eeg[:,:] = data
eeg.close()
eeg_filtered = F32filtered(tmp_fn, lambda x: 3*x)
read_data = eeg_filtered[:]
assert_array_almost_equal(data, read_data/3)
# -*- coding: utf-8 -*-
"""Calculate ERP-data on some generated data.
..hint::
Generation of the data might take some time..."""
import matplotlib.pyplot as plt
from eegpy.example_data.generated import simple_erp_like_data
from eegpy import F32
from eegpy.analysis.timefreq import ERPAnalyzer
from eegpy.temp import UnlinkingTempfile
with UnlinkingTempfile(".f32") as temp_fn:
# create some example data
data, events = simple_erp_like_data()
eeg = F32(temp_fn, "w+", cNames=["A", "B", "C"], shape=data.shape, Fs=250.)
eeg[:, :] = data
# plot whole 'eeg' with events
plt.subplot(211)
plt.plot(eeg[:, :], "k-")
for event in events.all_events:
plt.axvline(event, color="r", lw=2., alpha=0.5)
# analyze ERPs
analyzer = ERPAnalyzer(eeg, (-300, +600), baseline=(-200, 0))
for condition in events.keys():
analyzer.add_condition(condition, events[condition])
# plot ERPs
relative_times = analyzer.times_to_trigger
# -*- coding: utf-8 -*-
import os
import numpy as np
from eegpy import F32
from eegpy.temp import UnlinkingTempfile
with UnlinkingTempfile("f32") as filename:
channel_names = ["F3", "Fz", "F3", "C3", "Cz", "C4", "P3", "Pz", "P4"]
eeg = F32(filename,
"w+",
shape=(10000, len(channel_names)),
cNames=channel_names,
Fs=250.)
eeg[:, :] = np.random.random((10000, len(channel_names)))
eeg[:, 0] *= 5
print eeg.channel_names
print eeg.shape
print eeg.num_datapoints, eeg.num_channels
print eeg.Fs
eeg.close()
def test_tempfile_is_unlinked(self):
with UnlinkingTempfile() as fn:
with open(fn, "w") as fh:
fh.write("Test")
assert_true(os.path.exists(fn))
assert_false(os.path.exists(fn))
def test_dir_is_used(self):
dir_ = "/path/to/somewhere"
with UnlinkingTempfile(dir=dir_) as fn:
assert_equal(os.path.normpath(dir_),
os.path.normpath(os.path.split(fn)[0]))
def test_prefix_is_used(self):
with UnlinkingTempfile(prefix="xyz") as fn:
file_part = os.path.split(fn)[1]
assert_true(file_part.startswith("xyz"))
def test_suffix_is_used(self):
with UnlinkingTempfile(suffix=".xyz") as fn:
assert_true(fn.endswith(".xyz"))
def test_file_is_never_created(self):
with UnlinkingTempfile() as fn:
assert_false(os.path.exists(fn))
assert_false(os.path.exists(fn))
def test_channel_names_are_automatically_created():
with UnlinkingTempfile(".f32") as tmp_fn:
eeg = F32(tmp_fn, "w+", shape=(1,10))
assert_equal(10, len(eeg.channel_names))
for ch in eeg.channel_names:
assert_false(ch=="")
def _internal_write(data):
with UnlinkingTempfile(".f32") as tmp_fn:
eeg = F32(tmp_fn, "w+", shape=data.shape)
eeg[:,:] = data
eeg.close()
assert_true(os.path.getsize(tmp_fn) > 1024)