`generate_data.py` for processing and fitting statistical distributions to data.

'distributions.py' to create new dists to fit

'analysis_data.ipynb" notebook for data analysis
This commit is contained in:
2026-04-10 15:01:07 -03:00
parent aacfe3f977
commit 9aa97fc3d4
8 changed files with 973 additions and 3 deletions

2
.gitignore vendored
View File

@@ -33,3 +33,5 @@ data/*
# Claude Code
.claude/
etc/tools/read_raw_data.py

15
.vscode/launch.json vendored Normal file
View File

@@ -0,0 +1,15 @@
{
// Use IntelliSense to learn about possible attributes.
// Hover to view descriptions of existing attributes.
// For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
"version": "0.2.0",
"configurations": [
{
"name": "Python Debugger: Current File",
"type": "debugpy",
"request": "launch",
"program": "${file}",
"console": "integratedTerminal"
}
]
}

View File

@@ -1,3 +1 @@
from .plots import stacked_plot, noise_mean, calculate_cdf, plot_cdfs
__all__ = ["stacked_plot", "noise_mean", "calculate_cdf", "plot_cdfs"]

View File

@@ -0,0 +1,43 @@
from scipy.stats import rv_continuous
from scipy.special import gamma, kv, gammaln
from scipy.stats._distn_infrastructure import _ShapeInfo
import numpy as np
import pandas as pd
class k_gen(rv_continuous):
"""K distribution for radar clutter modeling.
The probability density function is::
f(x; nu, b) = 2/Gamma(nu) * b^((nu+1)/2) * x^((nu-1)/2) * K_{nu-1}(2*sqrt(b*x))
for x >= 0, where K_{nu-1} is the modified Bessel function of the second
kind of order nu-1, nu > 0 is the shape parameter and b > 0 is the rate
parameter.
"""
def _shape_info(self):
return [_ShapeInfo("nu", domain=(0, np.inf), inclusive=(False, True)),
_ShapeInfo("b", domain=(0, np.inf), inclusive=(False, True))]
def _argcheck(self, nu, b):
return (nu > 0) & (b > 0)
def _pdf(self, x, nu, b):
# k.pdf(x, nu, b) = 2/Gamma(nu) * b^((nu+1)/2) * x^((nu-1)/2) * K_{nu-1}(2*sqrt(b*x))
return np.exp(self._logpdf(x, nu, b))
def _logpdf(self, x, nu, b):
return (
np.log(2.0)
- gammaln(nu)
+ (nu + 1) / 2.0 * np.log(b)
+ (nu - 1) / 2.0 * np.log(x)
+ np.log(kv(nu - 1, 2.0 * np.sqrt(b * x)))
)
k_dist = k_gen(a=0.0, name="k_distribution", shapes="nu, b")

View File

@@ -17,6 +17,7 @@ dependencies = [
"matplotlib",
"seaborn",
"ruff>=0.15.9",
"python-dotenv>=1.2.2",
]
classifiers = ["Private :: Do Not Upload"]

707
scripts/analysis_data.ipynb Normal file

File diff suppressed because one or more lines are too long

193
scripts/generate_data.py Normal file
View File

@@ -0,0 +1,193 @@
from pathlib import Path
import sys
# Add project root to path
sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
from etc.tools.read_raw_data import load_data, RC_PREFIX, DP_PREFIX
from etc.fitting import Fitter
from etc.tools.statistics import aic_statistic
import pandas as pd
import plotly.graph_objects as go
import plotly.io as pio
import numpy as np
from plotly.subplots import make_subplots
import itertools
import os
import seaborn as sns
import matplotlib.pyplot as plt
from scipy.special import gamma as gamma_func
from scipy.special import kv as modified_bessel_second_kind
from scipy.special import gammaln
from scipy.stats import gamma,gumbel_r,gumbel_l,gompertz ,norm,weibull_min, lognorm, genextreme, genpareto, rayleigh, kstest, rv_continuous, goodness_of_fit
import statsmodels.api as sm
from dotenv import load_dotenv
pio.renderers.default = "browser"
# LOAD env variables in dir above
load_dotenv(os.path.join(os.path.dirname(__file__), "..", ".env"))
DATA_PATH = os.path.join(os.path.dirname(__file__), "..", "data")
if __name__ == "__main__":
aberturas = ['12','13','14','15']
voltas = ['0','2','4','6','8','10']
ads = ['0','3']
bursts = ['0','3']
# if not exists, create the directory to save the dataframes and the figure
if not os.path.exists(DATA_PATH):
os.makedirs(DATA_PATH)
dist_list = [gumbel_r, weibull_min, lognorm, rayleigh]
dist_list_log = [gumbel_l, weibull_min, genextreme, gamma]
statistics_dataframe = pd.DataFrame(columns=[dist.name for dist in dist_list_log])
# create dataframes for each dist in list
# weibull_dataframe = pd.DataFrame(columns=[f'weibull_{param}' for param in ['shape', 'loc', 'scale']])
# rayleigh_dataframe = pd.DataFrame(columns=[f'rayleigh_{param}' for param in ['loc', 'scale']])
# normal_dataframe = pd.DataFrame(columns=[f'normal_{param}' for param in ['loc', 'scale']])
# gumbel_l_dataframe = pd.DataFrame(columns=[f'gumbel_l_{param}' for param in ['loc','scale']])
# genextreme_dataframe = pd.DataFrame(columns=[f'genextreme_{param}' for param in ['shape','loc','scale']])
# gamma_dataframe = pd.DataFrame(columns=[f'gamma_{param}' for param in ['shape','loc','scale']])
# gompertz_dataframe = pd.DataFrame(columns=[f'gompertz_{param}' for param in ['shape','loc','scale']])
i = 1
for abertura, volta, ad, burst in itertools.product(aberturas, voltas, ads, bursts):
print(f"########### ITERATION {i} ###########")
print(f"Processing ad{ad} burst{burst} abertura{abertura} volta{volta}")
_data_rc = load_data(RC_PREFIX, ad_num=ad, burst_num=burst, feixe_num=0, abertura_num=abertura, volta_num=volta)
_data_rc = np.squeeze(_data_rc)
_data_dp = load_data(DP_PREFIX, ad_num=ad, burst_num=burst, feixe_num=0, abertura_num=abertura, volta_num=volta)
_data_dp = np.squeeze(_data_dp)
_noise_data_dp = _data_dp[5000:6000,:]
# values of _noise_data_dp should be greater than 0, take only values greater than 0
_noise_data_dp = _noise_data_dp[np.abs(_noise_data_dp) > 0]
_mean__noise_dp = np.abs(_noise_data_dp).mean()
factor_quartile_99 = 3.424
factor_quantile_pfa_1e_6 = 4.194
_data_dp_threshold = _mean__noise_dp * factor_quantile_pfa_1e_6
_dp_above_noise_idx = np.where(np.abs(_data_dp).mean(axis=1) > _data_dp_threshold)[0]
# plot dp mean and threshold to check if the threshold is correct, save the figure
# fig = go.Figure()
# fig.add_trace(go.Scatter(y=np.abs(_data_dp).mean(axis=1), name='Mean DP Power'))
# fig.add_trace(go.Scatter(y=[_data_dp_threshold]*len(_data_dp), name='Threshold', line=dict(dash='dash')))
# fig.update_layout(title=f'Mean DP Power and Threshold for ad{ad} burst{burst} abertura{abertura} volta{volta}', autosize=True)
# # plot dots on _dp_above_noise_idx
# fig.add_trace(go.Scatter(x=_dp_above_noise_idx, y=np.abs(_data_dp).mean(axis=1)[_dp_above_noise_idx], mode='markers', name='Above Threshold', marker=dict(color='red', size=5)))
# fig.show()
_dp_above_noise_idx = _dp_above_noise_idx[_dp_above_noise_idx >2250] # remove the first 1500 samples to avoid noise
_rain_data = _data_rc[_dp_above_noise_idx,:]
# get rain data above zero
_rain_data = _rain_data[np.abs(_rain_data) > 0]
# plot rain data to check if it is correct, plot the figure
# fig = go.Figure()
# fig.add_trace(go.Scatter(y=np.abs(_data_rc).mean(axis=1), name='Rain Data'))
# fig.update_layout(title=f'Rain Data for ad{ad} burst{burst} abertura{abertura} volta{volta}', autosize=True)
# # add dots on _dp_above_noise_idx
# fig.add_trace(go.Scatter(x=_dp_above_noise_idx, y=np.abs(_data_rc).mean(axis=1)[_dp_above_noise_idx], mode='markers', name='Above Threshold', marker=dict(color='red', size=5)))
# fig.show()
# check if _rain_data is empty, if it is, skip this iteration
if _rain_data.size == 0:
continue
### LOG DATA
_rain_data_log = np.log(_rain_data)
_rain_data_log = _rain_data_log[np.isfinite(_rain_data_log)]
fitter = Fitter(dist_list_log, statistic_method=aic_statistic)
print(f"Fitting distributions for ad{ad} burst{burst} abertura{abertura} volta{volta}")
fitter.fit(_rain_data_log)
print(f"Validating fits for ad{ad} burst{burst} abertura{abertura} volta{volta}")
fitter.validate(n_mc_samples=1,)
# create a pandas dataframe to store the statistics of the fitted distributions and concatenate it with the existing dataframe
dist_stats = {dist.name: fitter[dist.name].pvalue for dist in dist_list_log}
print(f"dist stats: {dist_stats}")
dist_stats_df = pd.DataFrame(dist_stats,index=[0])
statistics_dataframe = pd.concat([statistics_dataframe, dist_stats_df], ignore_index=True)
# check if any p-value is greater than 0.05, if it isnt, break the loop and do not save the dataframes and the figure
if not any(stat > 0.05 for stat in dist_stats.values()):
print(f"No distribution passed the GoF test for ad{ad} burst{burst} abertura{abertura} volta{volta}. Skipping saving results.")
break
# create histogram and save .html file with the figure
fig_1 = fitter.histogram_with_fits()
fig_1.write_html(f'{DATA_PATH}/histogram_with_fits_ad{ad}_burst{burst}_abertura{abertura}_volta{volta}.html')
# save qqplot and save .html file with the figure
fig_2 = fitter.plot_qq_plots(method='filliben')
fig_2.write_html(f'{DATA_PATH}/qq_plots_ad{ad}_burst{burst}_abertura{abertura}_volta{volta}.html')
# # create pandas dataframe to store weibull parameters and concatenate it with the existing dataframe
# weibull_params = {f'weibull_{param}': fitter._dist['weibull_min'].fit_result_params[i] for i, param in enumerate(['shape', 'loc', 'scale'])}
# weibull_params_df = pd.DataFrame(weibull_params,index=[0])
# weibull_dataframe = pd.concat([weibull_dataframe, weibull_params_df], ignore_index=True)
# # create pandas dataframe for rayleigh parameters and concatenate it with the existing dataframe
# rayleigh_params = {f'rayleigh_{param}': fitter._dist['rayleigh'].fit_result_params[i] for i, param in enumerate(['loc', 'scale'])}
# rayleigh_params_df = pd.DataFrame(rayleigh_params,index=[0])
# rayleigh_dataframe = pd.concat([rayleigh_dataframe, rayleigh_params_df], ignore_index=True)
# # create pandas dataframe for normal parameters and concatenate it with the existing dataframe
# normal_params = {f'normal_{param}': fitter._dist['norm'].fit_result_params[i] for i, param in enumerate(['loc', 'scale'])}
# normal_params_df = pd.DataFrame(normal_params,index=[0])
# normal_dataframe = pd.concat([normal_dataframe, normal_params_df], ignore_index=True)
# create pandas dataframe for gumbel_r parameters and concatenate it with the existing dataframe
# gumbel_l_params = {f'gumbel_l_{param}': fitter._dist['gumbel_l'].fit_result_params[i] for i, param in enumerate(['loc', 'scale'])}
# gumbel_l_params_df = pd.DataFrame(gumbel_l_params,index=[0])
# gumbel_l_dataframe = pd.concat([gumbel_l_dataframe, gumbel_l_params_df], ignore_index=True)
# create pandas dataframe for genextreme parameters and concatenate it with the existing dataframe
# genextreme_params = {f'genextreme_{param}': fitter._dist['genextreme'].fit_result_params[i] for i, param in enumerate(['shape', 'loc', 'scale'])}
# genextreme_params_df = pd.DataFrame(genextreme_params,index=[0])
# genextreme_dataframe = pd.concat([genextreme_dataframe, genextreme_params_df], ignore_index=True)
# create pandas dataframe for gamma parameters and concatenate it with the existing dataframe
# gamma_params = {f'gamma_{param}': fitter._dist['gamma'].fit_result_params[i] for i, param in enumerate(['shape', 'loc', 'scale'])}
# gamma_params_df = pd.DataFrame(gamma_params,index=[0])
# gamma_dataframe = pd.concat([gamma_dataframe, gamma_params_df], ignore_index=True)
# create pandas dataframe for gompertz parameters and concatenate it with the existing dataframe
# gompertz_params = {f'gompertz_{param}': fitter._dist['gompertz'].fit_result_params[i] for i, param in enumerate(['shape', 'loc', 'scale'])}
# gompertz_params_df = pd.DataFrame(gompertz_params,index=[0])
# gompertz_dataframe = pd.concat([gompertz_dataframe, gompertz_params_df], ignore_index=True)
i += 1
# path to save the dataframes and the figure
statistics_dataframe.to_csv(f'{DATA_PATH}/distribution_fit_pvalue_statistics.csv', index=False)
# rayleigh_dataframe.to_csv(f'{DATA_PATH}/rayleigh_parameters.csv', index=False)
#weibull_dataframe.to_csv(f'{DATA_PATH}/weibull_parameters.csv', index=False)
# normal_dataframe.to_csv(f'{DATA_PATH}/normal_parameters.csv', index=False)
# gumbel_l_dataframe.to_csv(f'{DATA_PATH}/gumbel_l_parameters.csv', index=False)
# genextreme_dataframe.to_csv(f'{DATA_PATH}/genextreme_parameters.csv', index=False)
# gamma_dataframe.to_csv(f'{DATA_PATH}/gamma_parameters.csv', index=False)
# gompertz_dataframe.to_csv(f'{DATA_PATH}/gompertz_parameters.csv', index=False)

13
uv.lock generated
View File

@@ -93,6 +93,7 @@ dependencies = [
{ name = "numpy" },
{ name = "pathlib2" },
{ name = "plotly" },
{ name = "python-dotenv" },
{ name = "ruff" },
{ name = "scipy" },
{ name = "seaborn" },
@@ -109,6 +110,7 @@ requires-dist = [
{ name = "numpy" },
{ name = "pathlib2" },
{ name = "plotly" },
{ name = "python-dotenv", specifier = ">=1.2.2" },
{ name = "ruff", specifier = ">=0.15.9" },
{ name = "scipy" },
{ name = "seaborn" },
@@ -687,7 +689,7 @@ name = "pexpect"
version = "4.9.0"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "ptyprocess" },
{ name = "ptyprocess", marker = "sys_platform != 'emscripten' and sys_platform != 'win32'" },
]
sdist = { url = "https://files.pythonhosted.org/packages/42/92/cc564bf6381ff43ce1f4d06852fc19a2f11d180f23dc32d9588bee2f149d/pexpect-4.9.0.tar.gz", hash = "sha256:ee7d41123f3c9911050ea2c2dac107568dc43b2d3b0c7557a33212c398ead30f", size = 166450, upload-time = "2023-11-25T09:07:26.339Z" }
wheels = [
@@ -877,6 +879,15 @@ wheels = [
{ url = "https://files.pythonhosted.org/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl", hash = "sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427", size = 229892, upload-time = "2024-03-01T18:36:18.57Z" },
]
[[package]]
name = "python-dotenv"
version = "1.2.2"
source = { registry = "https://pypi.org/simple" }
sdist = { url = "https://files.pythonhosted.org/packages/82/ed/0301aeeac3e5353ef3d94b6ec08bbcabd04a72018415dcb29e588514bba8/python_dotenv-1.2.2.tar.gz", hash = "sha256:2c371a91fbd7ba082c2c1dc1f8bf89ca22564a087c2c287cd9b662adde799cf3", size = 50135, upload-time = "2026-03-01T16:00:26.196Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/0b/d7/1959b9648791274998a9c3526f6d0ec8fd2233e4d4acce81bbae76b44b2a/python_dotenv-1.2.2-py3-none-any.whl", hash = "sha256:1d8214789a24de455a8b8bd8ae6fe3c6b69a5e3d64aa8a8e5d68e694bbcb285a", size = 22101, upload-time = "2026-03-01T16:00:25.09Z" },
]
[[package]]
name = "pyzmq"
version = "27.1.0"