Skip to content

Commit

Permalink
Fix: Update doc strings to Google format; add (optional) to optional …
Browse files Browse the repository at this point in the history 
…args
  • Loading branch information
@josephwkania
josephwkania committed Mar 6, 2024
1 parent 182b4c7 commit 5cd74f5
Show file tree
Hide file tree
Showing 4 changed files with 392 additions and 312 deletions.
141 changes: 78 additions & 63 deletions will/calculate.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -24,14 +24,14 @@ def std_min_func(sigma: float, mu: float, std: float) -> float:
have standard deviation `std`
Args:
sigma - lognormal sigma
sigma: lognormal sigma.
mu - lognormal mu
mu: lognormal mu.
std - The desired standard deviation
std: The desired standard deviation.
Returns:
desired variance - Actual variance(sigma, mu)
Desired variance - Actual variance(sigma, mu).
"""
return std**2 - (np.exp(sigma**2) - 1) * np.exp(2 * mu + sigma**2)

Expand All @@ -42,12 +42,12 @@ def log_normal_from_stats(median: float, std: float, size: int) -> np.ndarray:
and standard deviation.
Args:
median - median of resulting distribution
median: median of resulting distribution.
std - Standard deviation of returned samples
std: Standard deviation of returned samples.
Returns:
`size` numbers from the sampled from lognormal distribution
`size` numbers from the sampled from lognormal distribution.
"""
mu = np.log(median)
sigma_guess = np.sqrt(mu - 0.5 * np.log(std**2))
Expand Down Expand Up @@ -75,19 +75,20 @@ def quicksort(
Quicksort in place.
Args:
array - Array to be sorted
array: Array to be sorted.
left - Left most element of sort
left: Left most element of sort (Optional).
right - Right most element of sort
right: Right most element of sort (Optional).
sort_fraction - The fraction of the array to
stop sorting.
sort_fraction: The fraction of the array to
stop sorting, default is to sort
entire array (Optional).
sort_ascend - Sort increasing
sort_ascend: Sort increasing (Optional).
Returns:
None - Sort in place
None: Sort in place (Optional).
"""
if left is None:
left = 0
Expand Down Expand Up @@ -133,15 +134,16 @@ def sort_subarrays(
type shape.
Args:
array - array to sort
array: Array to sort.
num_subarrays - Number of subarrays to sort
num_subarrays: Number of subarrays to sort.
sort_fraction - The fraction of the subarray to
stop sorting.
sort_fraction: The fraction of the subarray to
stop sorting, default is to sort
the entire array (Optional).
Returns:
Subarray sorted (also is sorted in place)
Subarray sorted (also is sorted in place).
"""
splits = np.array_split(array, num_subarrays)
for j, split in enumerate(splits):
Expand All @@ -159,14 +161,14 @@ def calculate_dm_widths(
Eqn 2 in seconds.
Args:
dm - The Dispersion Measure in pc/cm^3.
dm: The Dispersion Measure [pc/cm^3].
channel_width - The channel width in MHz.
channel_width: The channel width [MHz].
chan_freqs - The channel frequencies in MHz.
chan_freqs: The channel frequencies [MHz].
Returns:
inter-channel dm smearing in seconds
inter-channel dm smearing in seconds.
"""
return 8.3 / 10e6 * dm * channel_width * (1000 / chan_freqs) ** 3

Expand All @@ -177,16 +179,16 @@ def calculate_dm_boxcar_widths(
chan_freqs: np.ndarray,
) -> np.ndarray:
"""
This calculates the boxcar widths in samples that correspond to the
This calculates the boxcar widths in samples that correspond to the
inter-channel dispersion delays in seconds. If delay is less than
than the sample time of one, return a boxcar width of one.
than the sample time of one, return a boxcar width of one.
Args:
dm - Dispersion Measure in pc/cm^3.
dm: Dispersion Measure [pc/cm^3].
sampling_time - Sampling time of data.
sampling_time: Sampling time of data [second].
chan_freqs - The channel frequencies in MHz.
chan_freqs: The channel frequencies [MHz].
Returns:
Array with boxcar lengths in samples.
Expand All @@ -207,12 +209,12 @@ def generate_boxcar_array(
Make a 2D array of boxcars for the given boxcar_lengths.
Args:
boxcar_lengths - Array of boxcar lengths.
boxcar_lengths: Array of boxcar lengths.
normalization_func - Function to normalize the boxcar, default
is to use sqrt.
normalization_func: Function to normalize the boxcar, default
is to use sqrt (Optional).
return_max - Return max boxcar length
return_max: Return max boxcar length (Optional).
Returns:
Array of boxcars, with the boxcars stacked horizontally.
Expand All @@ -235,13 +237,13 @@ def convolve_multi_boxcar(profile: np.ndarray, boxcar_array: np.ndarray) -> np.n
Convolve an profile array with an array that contains multiple boxcars.
Args:
profile - 1D or 2D array that contains the profile. For 1D array,
will be convolved with all boxcars. If 2D, time should
be on the vertical axis and number of profiles and
boxcars should match.
profile: 1D or 2D array that contains the profile. For 1D array,
will be convolved with all boxcars. If 2D, time should
be on the vertical axis and number of profiles and
boxcars should match.
boxcar_array - Array containing boxcars, boxcars should be stacked along
axis=1, see `generate_boxcar_array`.
boxcar_array: Array containing boxcars, boxcars should be stacked along
axis=1, see `generate_boxcar_array`.
Returns:
Profile convolved with the boxcar_array.
Expand Down Expand Up @@ -269,12 +271,12 @@ def boxcar_convolved(time_profile: np.ndarray, boxcar_widths: np.ndarray) -> np.
Calculate the pulse profile convolved with a boxcar width.
Args:
time_profile - Time profile of the pulse
time_profile: Time profile of the pulse (Optional).
boxcar_widths - Array of boxcar widths
boxcar_widths: Array of boxcar widths (Optional).
Returns:
Pulse profile convolved with a boxcar
Pulse profile convolved with a boxcar.
"""
boxcar_widths = np.array(boxcar_widths, ndmin=1)
powers = np.zeros(boxcar_widths.shape, dtype=np.float64)
Expand All @@ -295,12 +297,12 @@ def median_line_detrend(array: np.ndarray, num_sample: int) -> np.ndarray:
to these medians, then subtracting this line.
Args:
array - Array to detrend.
array: Array to detrend.
num_samples - Number of samples per median block.
num_samples: Number of samples per median block.
Return:
array with trend subtracted.
Array with trend subtracted.
Inspired by 'Advanced Architectures for Astrophysical Supercomputing',
section 4.2.3
Expand Down Expand Up @@ -347,18 +349,18 @@ def calculate_noises_multi(
Then use `scale_func` to calculate the noise level.
Args:
time_series - The dedispersed time series.
time_series: The dedispersed time series.
box_car_length - Length of the boxcars.
box_car_length: Length of the boxcars.
sigma - Return pulses with significance above
sigma: Return pulses with significance above
this
smoothing_factor - Detrend blocks are
smoothing_factor*box_car_length long.
smoothing_factor: Detrend blocks are
smoothing_factor*box_car_length long (Optional).
scale_func - The function used to calculate the measure of scale.
must accept `axis=0` argument.
scale_func: The function used to calculate the measure of scale.
must accept `axis=0` argument (Optional).
Returns:
Measures of scale for each boxcar in `boxcar array`.
Expand All @@ -379,12 +381,13 @@ class NoiseInfoResult:
"""
The noise levels for a file.
noise_levels - The noise level at each location (row) and
each boxcar, column.
Args:
noise_levels: The noise level at each location (row) and
each boxcar, column.
boxcar_lengths - Length of the boxcars.
boxcar_lengths: Length of the boxcars.
num_chans - Number of channels.
num_chans: Number of channels.
"""

noise_levels: np.ndarray
Expand All @@ -395,13 +398,25 @@ class NoiseInfoResult:
def mean_noise_levels(self):
"""
The mean noise levels across the file.
Args:
None.
Returns:
Average noise level in the file.
"""
return self.noise_levels.mean(axis=0)

@property
def median_noise_levels(self):
"""
The mean noise levels across the file.
The median noise levels across the file.
Args:
None.
Returns:
Median noise level across the file.
"""
return np.median(self.noise_levels, axis=0)

Expand Down Expand Up @@ -446,18 +461,18 @@ def noise_info(
convolved with boxcars of the given length. The noise is then calculated my MAD.
Args:
file_path - String that is the path to a file.
file_path: String that is the path to a file.
dm - Dispersion Measure in pc/cm^3.
dm: Dispersion Measure [pc/cm^3].
boxcar_lengths - Numpy array of boxcar lengths to consider.
boxcar_lengths: Numpy array of boxcar lengths to consider.
num_locations - Number of locations in the file to consider.
num_locations: Number of locations in the file to consider (Optional).
num_samples - Number of sample to consider, must be 4 times the
largest boxcar.
num_samples: Number of sample to consider, must be 4 times the
largest boxcar (Optional).
chan_mask - Mask these channels if given. True=value to mask
chan_mask: Mask these channels if given. True=value to mask (Optional).
Returns:
NoiseInfoResult
Expand Down
Loading

0 comments on commit 5cd74f5

Please sign in to comment.