TFM/01_descarga-Copy1.py at master · lytex/TFM · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
# ---
# jupyter:
#   jupytext:
#     formats: ipynb,py:percent
#     text_representation:
#       extension: .py
#       format_name: percent
#       format_version: '1.3'
#       jupytext_version: 1.16.2
#   kernelspec:
#     display_name: TFM
#     language: python
#     name: tfm
# ---

# %%
# %pdb off
import pandas as pd
import lightkurve as lk
import numpy as np
import pywt
import pickle
import os
import matplotlib.pyplot as plt
from LCWavelet import *
from binning import bin_and_aggregate
from tqdm import tqdm
import pickle

df_path = 'cumulative_2024.06.01_09.08.01.csv'
df = pd.read_csv(df_path ,skiprows=144)
failure = pd.read_csv('all_data_2024-07-17/failure_1721424039.csv')
df = pd.merge(failure[['kepoi_name']], df, on="kepoi_name")


def process_light_curve(row, mission="Kepler", download_dir="data3/",
                        binning_parameters=None,
                        sigma=20, sigma_upper=4,
                        wavelet_window=None,
                        wavelet_family=None, levels=None, cut_border_percent=0.1,
                        plot = False, plot_comparative=False,save=False, path=""):

    kic = f'KIC {row.kepid}'
    lc_search = lk.search_lightcurve(kic, mission=mission)

    light_curve_collection = lc_search.download_all(download_dir=download_dir)

    # lc_collection = lk.LightCurveCollection([lc.remove_outliers(sigma=sigma, sigma_upper=sigma_upper) for lc in light_curve_collection])
    lc_collection = lk.LightCurveCollection([lc for lc in light_curve_collection])

    lc_ro = lc_collection.stitch()

    lc_nonans = lc_ro.remove_nans()

    lc_fold = lc_nonans.fold(period = row.koi_period,epoch_time = row.koi_time0bk)
    lc_odd = lc_fold[lc_fold.odd_mask]
    lc_even = lc_fold[lc_fold.even_mask]

    if binning_parameters is not None and "lambda" in binning_parameters.keys() and "delta" in binning_parameters.keys():
        lmb = binning_parameters["lambda"]
        delta = binning_parameters["delta"]
        lc_odd =  bin_and_aggregate(np.arange(lc_odd), lc_odd)
        lc_odd =  bin_and_aggregate(np.arange(lc_odd), lc_odd)
        pass

    if wavelet_window is not None:
        print('Aplicando ventana ...')
        lc_impar = cut_wavelet(lc_odd, wavelet_window)
        lc_par = cut_wavelet(lc_even, wavelet_window)
    else:
        lc_impar = lc_odd
        lc_par = lc_even

    # para quitar oscilaciones en los bordes (quizás mejor no guardar los datos con esto quitado)
    lc_w_par = apply_wavelet(lc_par, wavelet_family, levels, cut_border_percent=cut_border_percent)
    lc_w_impar = apply_wavelet(lc_impar, wavelet_family, levels, cut_border_percent=cut_border_percent)

    headers = {
        "period": row.koi_period,
        "koi_period_err1": row.koi_period_err1,
        "koi_period_err2": row.koi_period_err2,
        "depth": row.koi_depth,
        "depth_err1": row.koi_depth_err1,
        "depth_err2": row.koi_depth_err2,
        "duration": row.koi_duration,
        "duration_err1": row.koi_duration_err1,
        "duration_err2": row.koi_duration_err2,
        "steff": row.koi_steff,
        "steff_err1": row.koi_steff_err1,
        "steff_err2": row.koi_steff_err2,
        "impact": row.koi_impact,
        "impact_err1": row.koi_impact_err1,
        "impact_err2": row.koi_impact_err2,
        "class": row.koi_disposition,
        "wavelet_family":wavelet_family,
        "levels":levels,
        "window":wavelet_window,
        "border_cut":cut_border_percent,
        "Kepler_name":row.kepoi_name
    }
    lc_wavelet_collection = LightCurveWaveletCollection(row.kepid, headers, lc_w_par, lc_w_impar)

    if(plot):
        print('graficando wavelets obtenidas...')
        lc_w_par.plot()
        lc_w_inpar.plot()
    if(plot_comparative):
        print('graficando wavelets obtenidas...')
        lc_wavelet_collection.plot_comparative()
    if(save):
        # print('guardando wavelets obtenidas...')
        lc_wavelet_collection.save(path)
    return lc_wavelet_collection

# for _, row in tqdm(df.iterrows(), total=len(df)):
#     process_light_curve(row, levels=[1, 2, 3, 4], wavelet_family="sym5", plot=False, plot_comparative=False, save=True, path="all_data_2024-06-01/"
#                        binning_parameters={"delta": , "lambda": })

# process_light_curve


# %%
np.arange(len([1, 2, 3, 4]))

# %%
import os
kep_id = list('KIC ' + df.kepid.astype(str))
import traceback
def download_one(kic):
    try:
        lc_search = lk.search_lightcurve(kic, mission="Kepler")
        light_curve_collection = lc_search.download_all(download_dir="data3/")
        file_name = "data3/"+kic+".pickle"
        with open(file_name, "wb") as f:
            pickle.dump(lc_search, f)
    except Exception:
        print(traceback.format_exc())
    return None


# for row in tqdm(kep_id):
#     download_one(row)


# from prpl import prpl
# from atpbar import flushing
# prpl(target_list=kep_id, target_function=download_one, list_sep=16, timer=True)
from concurrent import futures
from parallelbar import progress_map
# with flushing(), futures.ThreadPoolExecutor(max_workers=32) as executor:
#     res = executor.map(download_one, kep_id)
for kepid in kep_id:
    download_one(kepid)
# progress_map(download_one, kep_id, n_cpu=16, error_behavior='coerce')

# pd.concat([lc.table.to_pandas() for lc in lc_search]).to_csv("table.csv", index=None)
# lc_search[0].table.to_pandas()
# light_curve_collection

# %%