index
int64
repo_name
string
branch_name
string
path
string
content
string
import_graph
string
50,206
stevenlandis/SBHACKS_V
refs/heads/master
/PC_Control/Keyboard_Control.py
import keyboard as key def Sleep_Mode(): key.Keyboard.keyDown(key.Keyboard.VK_SLEEP) key.Keyboard.keyUp(key.Keyboard.VK_SLEEP) def Tab(): key.Keyboard.keyDown(key.Keyboard.VK_TAB) key.Keyboard.keyUp(key.Keyboard.VK_TAB) def Shift_Tab(): key.Keyboard.keyDown(key.Keyboard.VK_SHIFT) key.Keyboard.keyDown(key.Keyboard.VK_TAB) key.Keyboard.keyUp(key.Keyboard.VK_TAB) key.Keyboard.keyUp(key.Keyboard.VK_SHIFT) def Enter(): key.Keyboard.keyDown(key.Keyboard.VK_ENTER) key.Keyboard.keyUp(key.Keyboard.VK_ENTER) def Shift(): key.Keyboard.keyDown(key.Keyboard.VK_SHIFT) key.Keyboard.keyUp(key.Keyboard.VK_SHIFT) def Ctrl(): key.Keyboard.keyDown(key.Keyboard.VK_CTRL) key.Keyboard.keyUp(key.Keyboard.VK_CTRL) def BackSpace(): key.Keyboard.keyDown(key.Keyboard.VK_BACKSPACE) key.Keyboard.keyUp(key.Keyboard.VK_BACKSPACE) def B_Back(): key.Keyboard.keyDown(key.Keyboard.VK_BROWSER_BACK) key.Keyboard.keyUp(key.Keyboard.VK_BROWSER_BACK) def B_Favorite(): key.Keyboard.keyDown(key.Keyboard.VK_BROWSER_FAVORITES) key.Keyboard.keyUp(key.Keyboard.VK_BROWSER_FAVORITES) def B_Forward(): key.Keyboard.keyDown(key.Keyboard.VK_BROWSER_FORWARD) key.Keyboard.keyUp(key.Keyboard.VK_BROWSER_FORWARD) def Alt_Tab(): key.Keyboard.keyDown(key.Keyboard.VK_ALT) key.Keyboard.keyDown(key.Keyboard.VK_TAB) key.Keyboard.keyUp(key.Keyboard.VK_TAB) key.Keyboard.keyUp(key.Keyboard.VK_ALT) def Alt_Shift_Tab(): key.Keyboard.keyDown(key.Keyboard.VK_SHIFT) key.Keyboard.keyDown(key.Keyboard.VK_ALT) key.Keyboard.keyDown(key.Keyboard.VK_TAB) key.Keyboard.keyUp(key.Keyboard.VK_TAB) key.Keyboard.keyUp(key.Keyboard.VK_ALT) key.Keyboard.keyUp(key.Keyboard.VK_SHIFT)
{"/dragonIO/testAll/GUI.py": ["/Youtube_Control.py"], "/Receive.py": ["/Youtube_Control.py"]}
50,207
stevenlandis/SBHACKS_V
refs/heads/master
/PC_Control/Volume_Control.py
from sound import Sound def Increase_Volume(amount): for i in range(amount): Sound.volume_up() def Decrease_Volume(amount): for i in range(amount): Sound.volume_down() def Mute(): Sound.mute() def Max_Volume(): Sound.volume_max() def Min_Volume(): Sound.volume_min() def Get_Volume(): return Sound.current_volume()
{"/dragonIO/testAll/GUI.py": ["/Youtube_Control.py"], "/Receive.py": ["/Youtube_Control.py"]}
50,208
stevenlandis/SBHACKS_V
refs/heads/master
/dragonIO/testA0/main.py
import serial ard = serial.Serial('/dev/tty96B0', 9600) if __name__ == '__main__': print('Starting') while True: ardOut = ard.readline() print(ardOut)
{"/dragonIO/testAll/GUI.py": ["/Youtube_Control.py"], "/Receive.py": ["/Youtube_Control.py"]}
50,209
stevenlandis/SBHACKS_V
refs/heads/master
/board.py
import serial ard = serial.Serial('/dev/tty96B0', 115200) if __name__ == '__main__': print('Starting') while True: print(ard.readline()[:-1])
{"/dragonIO/testAll/GUI.py": ["/Youtube_Control.py"], "/Receive.py": ["/Youtube_Control.py"]}
50,210
stevenlandis/SBHACKS_V
refs/heads/master
/Receive.py
from SimpleWebSocketServer import SimpleWebSocketServer, WebSocket import Keyboard_Control as keyboard import Volume_Control as volume import Windows_Control as windows import Youtube_Control as youtube import GUI import sys import wmi import _thread as thread import pythoncom import Speech_Recognition as SR DataString = '' port = 22 def Socket_Function(): pythoncom.CoInitialize() server = SimpleWebSocketServer('', port, SimpleEcho) server.serveforever() class SimpleEcho(WebSocket): def handleMessage(self): # echo message back to client DataString = self.data GUI.Call_Function(DataString) # print(DataString) self.sendMessage(self.data) def handleConnected(self): print(self.address, 'Connected') def handleClose(self): print(self.address, 'Disconnected') thread.start_new_thread(Socket_Function, ()) GUI.GUI()
{"/dragonIO/testAll/GUI.py": ["/Youtube_Control.py"], "/Receive.py": ["/Youtube_Control.py"]}
50,211
stevenlandis/SBHACKS_V
refs/heads/master
/dragonIO/testAll/Speech_Recognition.py
import pip import smtplib import speech_recognition as sr import ssl Sender = 'gareyflee@gmail.com' Recipient = 'gareyflee@gmail.com' pw = "Dew61316131" def Speech_To_Email(): # for index, name in enumerate(sr.Microphone.list_microphone_names()): # print("Microphone with name \"{1}\" found for `Microphone(device_index={0})`".format(index, name)) mic = sr.Microphone() r = sr.Recognizer() with mic as source: r.adjust_for_ambient_noise(source, duration=1) print("Say something!") audio = r.listen(source) voiceHold = r.recognize_google(audio) print(voiceHold) server = smtplib.SMTP('smtp.gmail.com', 587) ## 25 465 587 server.starttls() server.login(Sender, pw) msg = voiceHold + "\n\nSent from Hackathon" server.sendmail(Sender, Recipient, msg) server.quit()
{"/dragonIO/testAll/GUI.py": ["/Youtube_Control.py"], "/Receive.py": ["/Youtube_Control.py"]}
50,212
stevenlandis/SBHACKS_V
refs/heads/master
/dragonIO/testAll/Keyboard_Control.py
import keyboard as key from time import sleep def Scroll_Windows(arg): if arg == 0: Alt_Tab() else: Alt_Shift_Tab() def Scroll_Tabs(arg): if arg == 0: Ctrl_Tab() else: Ctrl_Shift_Tab() def Tab(): key.Keyboard.keyDown(key.Keyboard.VK_TAB) key.Keyboard.keyUp(key.Keyboard.VK_TAB) def Shift_Tab(): key.Keyboard.keyDown(key.Keyboard.VK_SHIFT) key.Keyboard.keyDown(key.Keyboard.VK_TAB) key.Keyboard.keyUp(key.Keyboard.VK_TAB) key.Keyboard.keyUp(key.Keyboard.VK_SHIFT) def Enter(arg): if arg == 0: return key.Keyboard.keyDown(key.Keyboard.VK_ENTER) key.Keyboard.keyUp(key.Keyboard.VK_ENTER) def Shift(): key.Keyboard.keyDown(key.Keyboard.VK_SHIFT) key.Keyboard.keyUp(key.Keyboard.VK_SHIFT) def Ctrl(): key.Keyboard.keyDown(key.Keyboard.VK_CTRL) key.Keyboard.keyUp(key.Keyboard.VK_CTRL) def BackSpace(): key.Keyboard.keyDown(key.Keyboard.VK_BACKSPACE) key.Keyboard.keyUp(key.Keyboard.VK_BACKSPACE) def B_Back(): key.Keyboard.keyDown(key.Keyboard.VK_BROWSER_BACK) key.Keyboard.keyUp(key.Keyboard.VK_BROWSER_BACK) def B_Favorite(): key.Keyboard.keyDown(key.Keyboard.VK_BROWSER_FAVORITES) key.Keyboard.keyUp(key.Keyboard.VK_BROWSER_FAVORITES) def B_Forward(): key.Keyboard.keyDown(key.Keyboard.VK_BROWSER_FORWARD) key.Keyboard.keyUp(key.Keyboard.VK_BROWSER_FORWARD) def Alt_Tab(): key.Keyboard.keyDown(key.Keyboard.VK_ALT) sleep(0.1) key.Keyboard.keyDown(key.Keyboard.VK_TAB) sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_TAB) sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_ALT) def Alt_Shift_Tab(): key.Keyboard.keyDown(key.Keyboard.VK_SHIFT) sleep(0.1) key.Keyboard.keyDown(key.Keyboard.VK_ALT) sleep(0.1) key.Keyboard.keyDown(key.Keyboard.VK_TAB) sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_TAB) sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_ALT) sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_SHIFT) def Ctrl_Tab(): key.Keyboard.keyDown(key.Keyboard.VK_CTRL) # sleep(0.1) key.Keyboard.keyDown(key.Keyboard.VK_TAB) # sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_TAB) # sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_CTRL) def Ctrl_Shift_Tab(): key.Keyboard.keyDown(key.Keyboard.VK_SHIFT) # sleep(0.1) key.Keyboard.keyDown(key.Keyboard.VK_CTRL) # sleep(0.1) key.Keyboard.keyDown(key.Keyboard.VK_TAB) # sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_TAB) # sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_CTRL) # sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_SHIFT) def Ctrl_W(arg): if arg == 0: return key.Keyboard.keyDown(key.Keyboard.VK_CTRL) # sleep(0.1) key.Keyboard.keyDown(key.Keyboard.VK_W) # sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_W) # sleep(0.1) key.Keyboard.keyUp(key.Keyboard.VK_CTRL)
{"/dragonIO/testAll/GUI.py": ["/Youtube_Control.py"], "/Receive.py": ["/Youtube_Control.py"]}
50,213
stevenlandis/SBHACKS_V
refs/heads/master
/dragonIO/testLight/board.py
import serial ard = serial.Serial('/dev/tty96B0', 9600) if __name__ == '__main__': print('Starting') while True: print(ard.readline())
{"/dragonIO/testAll/GUI.py": ["/Youtube_Control.py"], "/Receive.py": ["/Youtube_Control.py"]}
50,215
vsegouin/pic-sorter
refs/heads/master
/files/FileSystemManager.py
# coding=utf-8 import os import re import time import platform from Parameters import Parameters from Reporting import Reporting months = ["Janvier", "Fevrier", "Mars", "Avril", "Mai", "Juin", "Juillet", "Aout", "Septembre", "Octobre", "Novembre", "Decembre"] unauthorizedExtension = ['.ico', '.gif'] class FileSystemManager: root_path = "" database_path = "" duplicate_folder = "" processed_folder = "" def __init__(self, root_path): self.duplicate_file_path = os.path.join(root_path, "duplicate.txt") self.root_path = root_path self.database_path = os.path.join(root_path, "database.txt") self.duplicate_folder = os.path.join(root_path, "duplicate") self.processed_folder = os.path.join(root_path, "processed") self.init_folders() def init_folders(self): """ Create the base folder (processed, duplicate folder) """ self.create_folder_if_not_exists(self.duplicate_folder) self.create_folder_if_not_exists(self.processed_folder) def extract_datetime(self, file_exif): """ try to find the dateTime inside the exif metadata :param file_exif: :return: a string composed of the date if found, an empty string if not """ new_filename = file_exif.get("EXIF DateTimeOriginal") new_filename = str(new_filename).replace(" ", "_") # If DataTimeOriginal doesn't contains data we try another exif meta data if new_filename == " " or new_filename is None or new_filename == "None": new_filename = file_exif.get("Image DateTime") new_filename = str(new_filename).replace(" ", "_") if new_filename == "" or new_filename is None or str(new_filename) == '': return "" else: Reporting.date_by_exif += 1 return new_filename def create_folder_if_not_exists(self, folder): """ Check if the requested folder exists and create it if not :param folder: :return: the folder path created """ if not os.path.exists(folder): if not Parameters.dry_run: os.makedirs(folder) Reporting.path_created.append(folder) return folder def move_file(self, file_directory, filename, dest_directory, dest_filename): """ This method clean the path created, check if a file already exists at this path and rename the file if needed It's possible that due to an error in the filename the programs can't move it :param file_directory: directory of the file to move :param filename: the name of the file to move :param dest_directory: the new directory of the file :param dest_filename: the new filename """ dest_filename = re.sub('[<>:\"/\|\?*]', '_', dest_filename) basename, ext = os.path.splitext(dest_filename) dst_file = os.path.join(dest_directory, dest_filename) # rename if necessary count = 0 while os.path.exists(dst_file): count += 1 dst_file = os.path.join(dest_directory, '%s-%d%s' % (basename, count, ext)) # Reporting.log 'Renaming %s to %s' % (file, dst_file) try: if not Parameters.dry_run: os.rename(os.path.join(file_directory, filename), dst_file) Reporting.file_moved += 1 except WindowsError or FileNotFoundError: Reporting.unmovable_file.append(os.path.join(file_directory, filename)) Reporting.log("CAN'T MOVE THIS FILE !!!!!") def manage_image(self, directory, filename, file_exif): """ Methods which manage the image, prepare the new name, check if it's an authorized extension and then move it :param directory: directory of the file :param filename: name of the file :param file_exif: extracted exif data of the file """ basename, ext = os.path.splitext(filename) try: dest_name = self.extract_datetime(file_exif) except AttributeError: dest_name = "" Reporting.image_found += 1 root_folder="regular" if file_exif == {} or dest_name == "" or dest_name == 'None' or re.search(r'(\d{4}):(\d{2}):(\d{2})', dest_name) == 'None': root_folder = "emptyExif" if ext in unauthorizedExtension: root_folder = "unauthorized" Reporting.increment_unauthorized_extension(ext) else: Reporting.image_without_exif += 1 dest_name, dest_directory = self.detect_file_date(directory, filename, root_folder) else: try: match = re.search(r'(\d{4}):(\d{2}):(\d{2})', dest_name).groups() root_folder = "regular" if ext in unauthorizedExtension: root_folder = "unauthorized" Reporting.increment_unauthorized_extension(ext) else: Reporting.image_with_exif += 1 dest_directory = self.create_folder_if_not_exists(os.path.join(self.processed_folder, root_folder, repr(match[0]).replace("\\", "").replace("'", ""), months[int(match[1]) - 1])) except AttributeError: Reporting.errors_files_details.append(os.path.join(directory,filename)) Reporting.errors_files+=1 root_folder="error" dest_directory = self.create_folder_if_not_exists(os.path.join(self.processed_folder, root_folder, filename)) self.move_file(directory, filename, dest_directory, (dest_name + ext).replace(":", "-")) def manage_non_image(self, directory, filename, file_type): if file_type == "video": dest_directory = self.create_folder_if_not_exists(os.path.join(self.processed_folder, "video")) Reporting.videos_found += 1 elif file_type == "error": dest_directory = self.create_folder_if_not_exists(os.path.join(self.processed_folder, "error")) Reporting.errors_files += 1 Reporting.errors_files_details.append(os.path.join(directory, filename)) else: dest_directory = self.copy_directory_structure(directory, os.path.join(self.processed_folder, "nonImage")) Reporting.other_found += 1 try: self.move_file(directory, filename, dest_directory, filename) except: Reporting.unmovable_file.append(os.path.join(directory,filename)) def manage_duplicate_file(self, directory, filename, file_type): Reporting.duplicate_found += 1 if file_type == "image": dest_directory = self.copy_directory_structure(directory, self.duplicate_folder) elif file_type == "video": dest_directory = self.copy_directory_structure(directory, os.path.join(self.duplicate_folder, "video")) else: dest_directory = self.copy_directory_structure(directory, os.path.join(self.duplicate_folder, "other")) self.move_file(directory, filename, dest_directory, filename) def copy_directory_structure(self, directory_to_copy, directory_destination): """ Copy the structure of a directory based on the root of the destination based on the root_path given in parameter >>> copy_directory_structure('C:/foo/bar/test/bar/foo','C:/foo/bar/process') #with C:/foo/bar as root_parameter 'C:/foo/bar/process/test/bar/foo' :param directory_to_copy: the directory to copy (usually where the file is) :param directory_destination: :return: """ platform.system() new_directory = directory_to_copy.replace(self.root_path, "") pattern = "^" + os.sep if Parameters.is_windows: pattern = "^" + os.sep + os.sep new_directory = re.sub(pattern, "", new_directory) new_directory = os.path.join(directory_destination, new_directory) try: self.create_folder_if_not_exists(new_directory) except FileNotFoundError: Reporting.unmovable_file.append(new_directory) return new_directory def detect_file_date(self, directory, filename, root_folder): """ if the exif doesn't contains the date of creation, this method try to detect the date of the file based on the name if the filename doesn't match any pattern it use the system date of creation :param directory: directory of the file :param filename: the current name of the file :param root_folder: the directory where the file will end up :return: the final name of the file and the new directory """ final_name = "" dest_directory = "" possible_pattern = [ # Annee#mois#jour#serie "[January|February|March|April|May|June|July|August|September|October|November|December]*_[0-9]{2}__[0-9]*", # mois#Annee#jour#serie "([0-9]{4})([0-9]{2})([0-9]{2})[-_]([0-9]*)", # annee mois jour heure minutes secondes "([0-9]{2})([0-9]{2})([0-9]{2})[-_]([0-9]*)", # annee mois jour heure minutes secondes "([0-9]{4})[-_]([0-9]{2})[-_]([0-9]{2})[\-\s]([0-9]{2})[h\:\-\s\.]([0-9]{2})[m\:\-\s\.]([0-9]{2})", "([0-9]{2})[-_]([0-9]{2})[-_]([0-9]{2})[\-\s]([0-9]{2})[h\:\-\s\.]([0-9]{2})[m\:\-\s\.]([0-9]{2})", ] for pattern in possible_pattern: matches = re.match(pattern, final_name) if (matches != None): Reporting.log(matches.groups()) # Last chance : get filesystem creation date if final_name == "": match = time.gmtime(os.path.getmtime(os.path.join(directory, filename))) final_name = repr(match[0]) + ":" + repr(match[1]) + ":" + repr(match[2]) + "_" + repr( match[3]) + ":" + repr(match[4]) + ":" + repr(match[5]) dest_directory = self.create_folder_if_not_exists( os.path.join(self.processed_folder, root_folder, repr(match[0]), months[match[1] - 1], repr(match[2]))) Reporting.date_by_modified += 1 else: Reporting.date_by_name += 1 return final_name, dest_directory
{"/files/FileSystemManager.py": ["/Parameters.py", "/Reporting.py"], "/__main__.py": ["/Parameters.py", "/Reporting.py", "/files/FileBrowser.py", "/files/FileSystemManager.py", "/files/MD5Encoder.py", "/image/ExifReader.py"], "/Reporting.py": ["/Parameters.py"], "/files/MD5Encoder.py": ["/Reporting.py"], "/files/FileBrowser.py": ["/Reporting.py"]}
50,216
vsegouin/pic-sorter
refs/heads/master
/__main__.py
import logging import time from Parameters import Parameters from Reporting import * from files.FileBrowser import FileBrowser from files.FileSystemManager import FileSystemManager from files.MD5Encoder import MD5Encoder from image.ExifReader import ExifReader import os import sys import logging logging.basicConfig(level=logging.INFO) try: Parameters.root_path = sys.argv[1] except IndexError: Reporting.log("PLEASE PROVIDE A PATH AS PARAMETER") exit(0) try: if sys.argv[2] == 'false': Parameters.dry_run = False else: Parameters.dry_run = True except IndexError: Parameters.dry_run = True fsManager = FileSystemManager(Parameters.root_path) md5_encryptor = MD5Encoder(fsManager.root_path) exif_reader = ExifReader() browser = FileBrowser(fsManager.root_path) debut = time.time() browser.count_processed_file() # Crawl the processed folder to recreate database with existing files for root, subdirs, files in browser.crawl_processed_folder(): for file in files: file_path = os.path.join(root, file) Reporting.total_file += 1 # increment total number of files Reporting.showProgress(file,file_path) try: if not md5_encryptor.process_md5(file_path): md5_encryptor.add_file_in_duplicate_list(file_path) except IOError: fsManager.manage_non_image(root, file, "error") browser.count_total_find() Reporting.log(repr(debut)) for root, subdirs, files in browser.crawl_folders(): for file in files: # construct full image path file_path = os.path.join(root, file) # check if it's the database if file_path == fsManager.database_path or file_path == fsManager.duplicate_file_path: continue Reporting.log("-----------------") Reporting.total_file += 1 # increment total number of files Reporting.showProgress(file,file_path) # find the mime type file_type = exif_reader.detect_image_file(os.path.join(root, file)) # check if a duplicate is already found try: if not md5_encryptor.process_md5(file_path): Reporting.log("file is duplicate skip") fsManager.manage_duplicate_file(root, file, file_type) continue except IOError: fsManager.manage_non_image(root, file, "error") # if it's not a duplicated file # we increment the number of unique file found Reporting.total_file_processed += 1 # sort it in the folder if it's not an image if file_type == "video" or file_type == "other": fsManager.manage_non_image(root, file, file_type) # if it's an image elif file_type == "image": try: exif = exif_reader.read_exif(file_path) fsManager.manage_image(root, file, exif) except IOError: fsManager.manage_non_image(root, file, "error") Reporting.doReporting() if Parameters.dry_run: Reporting.log("This was a dry_run test, don't forget to delete the database.txt if it exists") print("--- %s seconds ---" % (time.time() - debut))
{"/files/FileSystemManager.py": ["/Parameters.py", "/Reporting.py"], "/__main__.py": ["/Parameters.py", "/Reporting.py", "/files/FileBrowser.py", "/files/FileSystemManager.py", "/files/MD5Encoder.py", "/image/ExifReader.py"], "/Reporting.py": ["/Parameters.py"], "/files/MD5Encoder.py": ["/Reporting.py"], "/files/FileBrowser.py": ["/Reporting.py"]}
50,217
vsegouin/pic-sorter
refs/heads/master
/Reporting.py
import logging from logging import Logger from Parameters import Parameters import os class Singleton(type): _instances = {} def __call__(cls, *args, **kwargs): if cls not in cls._instances: cls._instances[cls] = super(Singleton, cls).__call__(*args, **kwargs) return cls._instances[cls] # Python2 class Reporting(object): # global data hashed = [] calculated_total_file = 0 total_file = 0 # total_file_processed = 0 # file_moved = 0 # Type image_found = 0 # videos_found = 0 # duplicate_found = 0 # other_found = 0 # # Data on images image_with_exif = 0 image_without_exif = 0 date_by_exif = 0 date_by_name = 0 date_by_modified = 0 # actions unauthorized_extension = {} errors_files = 0 errors_files_details = [] path_created = [] unmovable_file = [] __metaclass__ = Singleton def reporting(self): pass @classmethod def increment_unauthorized_extension(cls, ext): try: cls.unauthorized_extension[ext] += 1 except KeyError: cls.unauthorized_extension[ext] = 0 cls.unauthorized_extension[ext] += 1 pass @classmethod def doReporting(cls): print("\n\r\n\r===========================\n\r========Global infos=======\n\r===========================" + "\n\rTotal File found " + repr(cls.total_file) + "\n\rTotal File processed " + repr(cls.total_file_processed) + "\n\rTotal File Moved " + repr(cls.file_moved) + "\n\r\n\r===========================\n\r=======Type Detected=======\n\r===========================" + "\n\rImages found : " + repr(cls.image_found) + "\n\rVideos found : " + repr(cls.videos_found) + "\n\rOther found : " + repr(cls.other_found) + "\n\rDuplicate found : " + repr(cls.duplicate_found) + "\n\r\n\r===========================\n\r==Informations about image=\n\r===========================" + "\n\rImage with exif : " + repr(cls.image_with_exif) + "\n\rImage without exif : " + repr(cls.image_without_exif) + "\n\rDate found with exif : " + repr(cls.date_by_exif) + "\n\rDate found with filename : " + repr(cls.date_by_name) + "\n\rDate found with modification date : " + repr(cls.date_by_modified) + "\n\r\n\r===========================\n\r==========Errors===========\n\r===========================" + "\n\rErrors file : " + repr(cls.errors_files) + "\n\r\n\r===========================\n\r=Path of problematics file=\n\r===========================" + "\n\rUnmovable file : " + repr(cls.unmovable_file.__len__()) + "\n\r" + "\n\r".join(cls.unmovable_file) + "\n\r\n\r===========================\n\r=======Path created========\n\r===========================" + "\n\rPath Created : " + repr(list(set(cls.path_created)).__len__()) + "\n\r" + "\n\r".join(list(set(cls.path_created))) ) @classmethod def log(cls, string): if Parameters.is_verbose: print(string) @classmethod def showProgress(cls,file,filePath): logging.info(repr(Reporting.total_file) + " / " + repr(Reporting.calculated_total_file)+" | "+file+" | "+repr(os.path.getsize(filePath) >> 20)+"Mo")
{"/files/FileSystemManager.py": ["/Parameters.py", "/Reporting.py"], "/__main__.py": ["/Parameters.py", "/Reporting.py", "/files/FileBrowser.py", "/files/FileSystemManager.py", "/files/MD5Encoder.py", "/image/ExifReader.py"], "/Reporting.py": ["/Parameters.py"], "/files/MD5Encoder.py": ["/Reporting.py"], "/files/FileBrowser.py": ["/Reporting.py"]}
50,218
vsegouin/pic-sorter
refs/heads/master
/image/ExifReader.py
import mimetypes import exifread class ExifReader: def __init__(self): pass def read_exif(self, file): f = open(file, 'rb') # Return Exif tags try: tags = exifread.process_file(f) except TypeError : tags = '' return tags def detect_image_file(self, file): type = mimetypes.guess_type(file)[0] if type == None: return "other" if "image/" in type: return "image" if "image/" in type: return "video" return "other"
{"/files/FileSystemManager.py": ["/Parameters.py", "/Reporting.py"], "/__main__.py": ["/Parameters.py", "/Reporting.py", "/files/FileBrowser.py", "/files/FileSystemManager.py", "/files/MD5Encoder.py", "/image/ExifReader.py"], "/Reporting.py": ["/Parameters.py"], "/files/MD5Encoder.py": ["/Reporting.py"], "/files/FileBrowser.py": ["/Reporting.py"]}
50,219
vsegouin/pic-sorter
refs/heads/master
/files/MD5Encoder.py
import os, hashlib import errno from Reporting import Reporting class MD5Encoder: m_database_path = "" m_current_file = "" m_hashed_value = "" m_duplicate_file = "" def __init__(self, database_location): self.m_database_path = os.path.join(database_location, "database.txt") self.m_duplicate_file = os.path.join(database_location, "duplicate.txt") file = open(self.m_database_path, "a", 1) file.close() def write_all(self,list): file = open("test.txt", "a", 1) for item in list: file.write(item) file.write("\n\r") file.close() def init_file(self, file_path): self.m_current_file = file_path self.m_hashed_value = self.hash_file(file_path) def hash_file(self, file): self.m_current_file = file hash_md5 = hashlib.md5() try: with open(file, "rb") as f: for chunk in iter(lambda: f.read(4096), b""): hash_md5.update(chunk) return hash_md5.hexdigest() except OSError as e: if e.errno == errno.ENOENT: Reporting.log(file + " Not Found continuing") return "" # # Return true if Hash code is not blank and it's not present in database.txt def is_file_already_present(self): f = open(self.m_database_path) contents = f.read() f.close() return True if self.m_hashed_value == "" or contents.count(self.m_hashed_value) > 0 else False def add_file_in_duplicate_list(self,file_path): file = open(self.m_duplicate_file, "a", 1) file.write(file_path) file.write("\n\r") file.close() def add_hash_in_database(self): file = open(self.m_database_path, "a", 1) file.write(self.m_hashed_value) file.write("\n\r") file.close() def process_md5(self, file_path): Reporting.log(file_path) self.init_file(file_path) # if it's a duplicate or the database itself then there is no reason to continue if self.is_file_already_present(): Reporting.log(self.m_hashed_value + " already present") return False self.add_hash_in_database() Reporting.log(self.m_hashed_value + " added") return True
{"/files/FileSystemManager.py": ["/Parameters.py", "/Reporting.py"], "/__main__.py": ["/Parameters.py", "/Reporting.py", "/files/FileBrowser.py", "/files/FileSystemManager.py", "/files/MD5Encoder.py", "/image/ExifReader.py"], "/Reporting.py": ["/Parameters.py"], "/files/MD5Encoder.py": ["/Reporting.py"], "/files/FileBrowser.py": ["/Reporting.py"]}
50,220
vsegouin/pic-sorter
refs/heads/master
/Parameters.py
import platform class Singleton(type): _instances = {} def __call__(cls, *args, **kwargs): if cls not in cls._instances: cls._instances[cls] = super(Singleton, cls).__call__(*args, **kwargs) return cls._instances[cls] # Python2 class Parameters(object): is_verbose = False root_path = "" dry_run = False is_windows = platform.system() == "Windows" __metaclass__ = Singleton def Parameters(self): pass
{"/files/FileSystemManager.py": ["/Parameters.py", "/Reporting.py"], "/__main__.py": ["/Parameters.py", "/Reporting.py", "/files/FileBrowser.py", "/files/FileSystemManager.py", "/files/MD5Encoder.py", "/image/ExifReader.py"], "/Reporting.py": ["/Parameters.py"], "/files/MD5Encoder.py": ["/Reporting.py"], "/files/FileBrowser.py": ["/Reporting.py"]}
50,221
vsegouin/pic-sorter
refs/heads/master
/files/FileBrowser.py
import os from Reporting import Reporting class FileBrowser: root_path = None def __init__(self, root_path): self.root_path = root_path def count_processed_file(self): Reporting.total_file = 0 Reporting.calculated_total_file = 0 for root, subdirs, files in self.crawl_processed_folder(): Reporting.calculated_total_file += files.__len__() def count_total_find(self): Reporting.total_file = 0 Reporting.calculated_total_file = 0 for root, subdirs, files in self.crawl_folders(): Reporting.calculated_total_file += files.__len__() # Will crawl and generate all the files of the root_path def crawl_folders(self): for root, subdirs, files in os.walk(self.root_path): Reporting.log('-- current directory = ' + root + "\n") if ("@eaDir" in root): Reporting.log("it's a eadir folder continue\n") continue if "processed" in root or "duplicate" in root: Reporting.log("it's the processed or the duplicate folder skip") continue yield [root, subdirs, files] # check whether the directory is now empty after deletions, and if so, remove it if len(os.listdir(root)) == 0: os.rmdir(root) def crawl_processed_folder(self): for root, subdirs, files in os.walk(self.root_path): if not "processed" in root: continue yield [root, subdirs, files] def getPath(self): return self.root_path;
{"/files/FileSystemManager.py": ["/Parameters.py", "/Reporting.py"], "/__main__.py": ["/Parameters.py", "/Reporting.py", "/files/FileBrowser.py", "/files/FileSystemManager.py", "/files/MD5Encoder.py", "/image/ExifReader.py"], "/Reporting.py": ["/Parameters.py"], "/files/MD5Encoder.py": ["/Reporting.py"], "/files/FileBrowser.py": ["/Reporting.py"]}
50,222
twolfson/sublime-info
refs/heads/master
/sublime_info/__init__.py
# Load in local modules for extension from __future__ import absolute_import from sublime_info.SublimeInfo import SublimeInfo from sublime_info.errors import * # Define sugar methods def get_sublime_path(): return SublimeInfo.get_sublime_path() def get_sublime_version(): return SublimeInfo.get_sublime_version() def get_package_directory(): return SublimeInfo.get_package_directory()
{"/sublime_info/__init__.py": ["/sublime_info/SublimeInfo.py", "/sublime_info/errors.py"], "/sublime_info/SublimeInfo.py": ["/sublime_info/errors.py"], "/test/sublime-info_test.py": ["/sublime_info/__init__.py"]}
50,223
twolfson/sublime-info
refs/heads/master
/setup.py
from setuptools import setup, find_packages setup( name='sublime_info', version='0.2.0', description='Gather information about Sublime Text', long_description=open('README.rst').read(), keywords=[ 'sublime', 'sublime text', 'info', 'plugin' ], author='Todd Wolfson', author_email='todd@twolfson.com', url='https://github.com/twolfson/sublime-info', download_url='https://github.com/twolfson/sublime-info/archive/master.zip', packages=find_packages(), license='UNLICENSE', install_requires=open('requirements.txt').readlines(), # https://pypi.python.org/pypi?%3Aaction=list_classifiers classifiers=[ 'Development Status :: 3 - Alpha', 'Intended Audience :: Developers', 'License :: Public Domain', 'Operating System :: OS Independent', 'Programming Language :: Python', 'Topic :: Text Editors' ] )
{"/sublime_info/__init__.py": ["/sublime_info/SublimeInfo.py", "/sublime_info/errors.py"], "/sublime_info/SublimeInfo.py": ["/sublime_info/errors.py"], "/test/sublime-info_test.py": ["/sublime_info/__init__.py"]}
50,224
twolfson/sublime-info
refs/heads/master
/sublime_info/SublimeInfo.py
# Load in core, 3rd party, and local dependencies from __future__ import absolute_import import os import re import subprocess from sublime_info.errors import STNotResolvedError, STBadLocationError try: from shutil import which except ImportError: # Python 2 fallback from shutilwhich import which class SublimeInfo(object): # Load in an environment variable constant sublime_path=os.environ.get('SUBLIME_TEXT_PATH', None) # Define init def __init__(self, sublime_path=None): # Allow for customizaton of sublime_path if sublime_path: self.sublime_path = sublime_path # Define internal lookup which ignores @classmethod def _get_sublime_path(cls): # Attempt to resolve Sublime Text path = (which('subl') or which('sublime_text')) # If Sublime is not found, raise our exception if not path: raise STNotResolvedError( 'Sublime Text could not be found via the command "%s" or "%s"' % ('subl', 'sublime_text')) # Otherwise, return the path return path @classmethod def get_sublime_path(cls): """Resolve Sublime Text path (e.g. /usr/bin/subl) If ``SUBLIME_TEXT_PATH`` is provided via environment variables, it will be used. Otherwise, a ``which``-like resolution will be returned. :raises STNotFoundError: If Sublime Text cannot be found, an error will be raised. :returns: ``SUBLIME_TEXT_PATH`` or ``which``-like resolution :rtype: str """ # If sublime_path is provided, verify it exists sublime_path = cls.sublime_path if sublime_path: if not os.path.exists(sublime_path): raise STBadLocationError( 'Sublime Text could not be found at "%s"' % sublime_path) # Otherwise, use the internal lookup else: sublime_path = cls._get_sublime_path() # Return the found path return sublime_path @classmethod def get_sublime_version(cls): """Resolve Sublime Text version (e.g. 2221, 3083) Sublime Text is resolved via ``get_sublime_path`` :raises Exception: If the Sublime Text version cannot be parsed, an error will be raised. :returns: Version of Sublime Text returned by ``sublime_text --version``. :rtype: int """ # Get the path to sublime and grab the version sublime_path = cls.get_sublime_path() child = subprocess.Popen([sublime_path, '--version'], stdout=subprocess.PIPE) version_stdout = str(child.stdout.read()) # Kill the child child.kill() # Parse out build number from stdout # Sublime Text 2 Build 2221 # Sublime Text Build 3083 version_match = re.search(r'\d{4}', version_stdout) if not version_match: raise Exception('Sublime Text version not found in "%s"' % version_stdout) # Coerce and return the version return int(version_match.group(0)) @classmethod def get_package_directory(cls): """Resolve Sublime Text package directory (e.g. /home/todd/.config/sublime-text-2/Packages) :raises Exception: If the Sublime Text version is not recognized, an error will be raised. :returns: Path to Sublime Text's package directory :rtype: str """ # TODO: On Windows, OSX these will not be the same # Get the version version = cls.get_sublime_version() # Run Linux-only logic for pkg_dir pkg_dir = None if version >= 2000 and version < 3000: pkg_dir = os.path.expanduser('~/.config/sublime-text-2/Packages') elif version >= 3000 and version < 4000: pkg_dir = os.path.expanduser('~/.config/sublime-text-3/Packages') # Assert the package dir was found if not pkg_dir: raise Exception('Sublime Text version "%s" not recognized' % version) # Return the package directory return pkg_dir
{"/sublime_info/__init__.py": ["/sublime_info/SublimeInfo.py", "/sublime_info/errors.py"], "/sublime_info/SublimeInfo.py": ["/sublime_info/errors.py"], "/test/sublime-info_test.py": ["/sublime_info/__init__.py"]}
50,225
twolfson/sublime-info
refs/heads/master
/test/sublime-info_test.py
# Load in dependencies import os from unittest import TestCase # Load in local dependencies import sublime_info # Outline tests """ # The majority of these will be satisfied via .travis.yml Sublime Text as subl resolved via `get_sublime_path` has a path of `/usr/bin/subl` Sublime Text as sublime_text resolved via `get_sublime_path` has a path of `/usr/bin/sublime_text` Sublime Text as sublime_texttt specified via SUBLIME_TEXT_PATH resolved via `get_sublime_path` has a path of `/usr/bin/sublime_texttt` sublime_info attempting to resolve Sublime Text when it does not exist raises a STNotFoundError TODO: What about sublime_text.exe TODO: What about sublime_text.app TODO: Create test.sh/cmd for OSX and Windows which do both EXPECT ERROR and not TODO: Installed package dir TODO: Platform TODO: Arch """ # If we are in an error test if os.environ.get('EXPECT_ERROR', None): class TestGetSublimePathError(TestCase): def test_get_sublime_path_raises(self): """Assert that we raise an error when Sublime Text is not on disk.""" self.assertRaises(sublime_info.STNotFoundError, sublime_info.get_sublime_path) # Otherwise, run normal tests else: class TestGetSublimePathNormal(TestCase): def test_get_sublime_path_finds_path(self): """Assert that we find the proper path to Sublime Text.""" expected_path = os.environ['EXPECTED_PATH'] actual_path = sublime_info.get_sublime_path() self.assertEqual(expected_path, actual_path) def test_get_sublime_version_resolves_version(self): """Assert that we find the proper version of Sublime Text.""" expected_version = int(os.environ['EXPECTED_VERSION']) actual_version = sublime_info.get_sublime_version() self.assertEqual(expected_version, actual_version) def test_get_sublime_version_locates_pkg_dir(self): """Assert that we find the proper package directory.""" expected_pkg_dir = os.environ['EXPECTED_PKG_DIR'] actual_pkg_dir = sublime_info.get_package_directory() self.assertEqual(expected_pkg_dir, actual_pkg_dir)
{"/sublime_info/__init__.py": ["/sublime_info/SublimeInfo.py", "/sublime_info/errors.py"], "/sublime_info/SublimeInfo.py": ["/sublime_info/errors.py"], "/test/sublime-info_test.py": ["/sublime_info/__init__.py"]}
50,226
twolfson/sublime-info
refs/heads/master
/sublime_info/errors.py
# Define a custom exception for when Sublime cannot be located class STNotFoundError(Exception): pass class STNotResolvedError(STNotFoundError): pass class STBadLocationError(STNotFoundError): pass
{"/sublime_info/__init__.py": ["/sublime_info/SublimeInfo.py", "/sublime_info/errors.py"], "/sublime_info/SublimeInfo.py": ["/sublime_info/errors.py"], "/test/sublime-info_test.py": ["/sublime_info/__init__.py"]}
50,240
rwsproat/writing_evolution
refs/heads/master
/flags.py
## Licensed under the Apache License, Version 2.0 (the "License"); ## you may not use this file except in compliance with the License. ## You may obtain a copy of the License at ## ## http://www.apache.org/licenses/LICENSE-2.0 ## ## Unless required by applicable law or agreed to in writing, software ## distributed under the License is distributed on an "AS IS" BASIS, ## WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. ## See the License for the specific language governing permissions and ## limitations under the License. ## ## Author: Richard Sproat (rws@xoba.com) """Defines poor man's command-line flag parsing. """ import getopt import sys _FLAGS = [] _DUMMY_STR_FUNCTION_TEMPLATE = """def __x(): global FLAGS_%s FLAGS_%s = "%s" """ _DUMMY_NUM_FUNCTION_TEMPLATE = """def __x(): global FLAGS_%s FLAGS_%s = %s """ def set_dummy_function_template(option, default_value): """Builds the runnable function template. Args: option: name of flag default_value: default value for flag Returns: Function template to be compiled that sets the variables """ try: value = int(default_value) return _DUMMY_NUM_FUNCTION_TEMPLATE % (option, option, value) except ValueError: try: value = float(default_value) return _DUMMY_NUM_FUNCTION_TEMPLATE % (option, option, value) except ValueError: return _DUMMY_STR_FUNCTION_TEMPLATE % (option, option, default_value) def define_flag(option, default_value, documentation): """Defines the flag and sets the default value and documentation. Args: option: name of flag default_value: default value for flag documentation: documentation string Returns: None """ _FLAGS.append((option, default_value, documentation)) function_template = set_dummy_function_template(option, default_value) exec(function_template) __x() def usage(): """Prints usage given the set of supplied flags. Returns: None """ for option, default_value, documentation in _FLAGS: print '\t\t--%s\t"%s" (%s)' % (option, documentation, default_value) def parse_flags(argv): """Parses the flags given the argument list. Args: argv: argument vector Returns: None """ try: optform= map(lambda x: x[0] + '=', _FLAGS) opts, args = getopt.getopt(argv, '', optform) except getopt.GetoptError as err: print str(err) Usage() sys.exit(1) for opt, arg in opts: opt = opt.replace('--', '') function_template = set_dummy_function_template(opt, arg) exec(function_template) __x()
{"/builder.py": ["/pynini_interface.py"]}
50,241
rwsproat/writing_evolution
refs/heads/master
/concepts.py
# -*- coding: utf-8 -*- ## Licensed under the Apache License, Version 2.0 (the "License"); ## you may not use this file except in compliance with the License. ## You may obtain a copy of the License at ## ## http://www.apache.org/licenses/LICENSE-2.0 ## ## Unless required by applicable law or agreed to in writing, software ## distributed under the License is distributed on an "AS IS" BASIS, ## WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. ## See the License for the specific language governing permissions and ## limitations under the License. ## ## Author: Richard Sproat (rws@xoba.com) """Defines the set of concepts, linking them to plausible graphical forms. """ CONCEPTS = { '@PERSON' : '⚥', '@MAN' : '♂', '@WOMAN' : '♀', '@HOUSE' : '☖', '@BRONZE' : '▰', '@GOLD' : 'ꃑ', '@SILVER' : 'ꂢ', '@SWORD' : '⚔', '@MEAT' : '☣', '@SHEEP' : '♈', '@OX' : '♉', '@GOAT' : '♑', '@FISH' : 'ꈊ', '@TREE' : 'ꈔ', '@BARLEY' : 'ꆡ', '@WHEAT' : 'ꎺ', '@WATER' : '♒', '@STONE' : '░', '@CLOTHING' : 'ꆘ', '@FIELD' : 'ꐚ', '@TEMPLE' : '▟', '@GOD' : '☸', '@AXE' : '⁋', '@SCYTHE' : '☭', '@DOG' : 'ꀳ', '@LION' : '♌', '@WOLF' : 'ꉴ', '@DEMON' : '☿', '@SNAKE' : '⚕', '@TURTLE' : 'ꉸ', '@FRUIT' : '☌', '@HILL' : '☶', '@CAVE' : '◠', '@TOWN' : '♖', '@ENCLOSURE' : '☐', '@FLOWER' : '⚜', '@RAIN' : '☔', '@THUNDER' : '⚡', '@CLOUD' : '☁', '@SUN' : '☉', '@MOON' : '☽', '@HEART' : '♡', '@LUNG' : '♺', '@LEG' : '⼅', '@ARM' : 'ꁳ', '@FINGER' : '☚', '@HEAD' : '☺', '@TONGUE' : 'ꇩ', '@EYE' : '◎', '@EAR' : 'ꎙ', '@NOSE' : 'ꎔ', '@GUTS' : '♽', '@PENIS' : 'ꅏ', '@VAGINA' : '⚏', '@HAIR' : 'ꊤ', '@SKIN' : 'ꁸ', '@SHELL' : 'ꌋ', '@BONE' : 'ꀷ', '@BLOOD' : 'ꋓ', '@LIVER' : '❦', '@FARM' : 'ꌆ', '@LOCUST' : 'ꋸ', '@STICK' : '▕', '@STAR' : '☆', '@EARTH' : '☷', '@ASS' : '♘', '@DEATH' : '☠', '@BIRTH' : '〄', '@WOMB' : '☤', '@MILK' : 'ꎍ', '@COAL' : '■', '@SEED' : 'ꄍ', '@LEAF' : '☘', '@CHILD' : 'ꐕ', '@ANTELOPE' : 'ꎢ', '@BEAR' : 'ꄕ', '@BEE' : 'ꎁ', '@MOUSE' : 'ꎃ', '@DUNG' : '♨', '@PLOUGH' : '♠', '@SPROUT' : '♧', '@ICE' : '⼎', '@DAY' : '☀', '@NIGHT' : '☾', '@WINTER' : '☃', '@SUMMER' : '☼', '@AUTUMN' : 'ꏃ', '@SPRING' : 'ꐳ', '@KING' : '♔', '@GOOSE' : 'ꎉ', '@PRIEST' : '♗', '@ROAD' : '⚌', '@CART' : 'ꌴ', '@GRASS' : 'ꍓ', '@FIRE' : '☲', '@WIND' : '☴', '@NAIL' : '丁', '@BREAST' : '⚆', '@BOWL' : '◡', '@CUP' : '☕',} NUM_CONCEPTS = len(CONCEPTS)
{"/builder.py": ["/pynini_interface.py"]}
50,242
rwsproat/writing_evolution
refs/heads/master
/lexicon.py
#!/usr/bin/env python ## Licensed under the Apache License, Version 2.0 (the "License"); ## you may not use this file except in compliance with the License. ## You may obtain a copy of the License at ## ## http://www.apache.org/licenses/LICENSE-2.0 ## ## Unless required by applicable law or agreed to in writing, software ## distributed under the License is distributed on an "AS IS" BASIS, ## WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. ## See the License for the specific language governing permissions and ## limitations under the License. ## ## Author: Richard Sproat (rws@xoba.com) """Defines the lexicon, morphemes, and concepts associated with morphemes. Then runs the simulation. """ # NB: "semantics" and "concept" are used interchangeably # TODO(rws): This seems to generate rather too many morphemes # associated with a particular concept (e.g. 36 for TEMPLE). import builder import concepts import flags import log import os import random import re import sys import time from pynini_interface import sounds_like # Maximum distance that a closest pronunciation can have _MAX_DISTANCE = 0.6 # Probability of reusing an existing spelling # TODO(rws): Make this a paremeter _PROBABILITY_TO_REUSE_SPELLING = 0.01 # Markup colors _BLUE = '\033[34m%s\033[0m' _RED = '\033[31m%s\033[0m' def _clean_colors(string): """Helper to clean up shell color markers from string. Args: string: string with possible color markers Returns: cleaned string """ string = string.replace('\033[34m', '') string = string.replace('\033[31m', '') string = string.replace('\033[0m', '') return string def _phonetic_color(string): """Adds color for phonetic graphemes to string. Args: string: string with symbols Returns: string with phonetic marker (_RED) """ return _RED % (_clean_colors(string)) def _semantic_color(string): """Adds color for semantic graphemes to string. Args: string: string with symbols Returns: string with semantic marker (_BLUE) """ return _BLUE % (_clean_colors(string)) def _uniqify_symbol_list(symbols): """Removes duplicate symbols from list. Args: symbols: list of symbols Returns: uniquified list """ seen = set() new_symbols = [] for symbol in symbols: if symbol.name not in seen: new_symbols.append(symbol) seen.add(symbol.name) return new_symbols def _clean_name(name): """Cleans up symbol name of bracketings for presentation. Args: name: symbol name Returns: cleaned name """ name = re.sub(r'\[[A-Za-z0-9\.]+\]', '', name) name = re.sub(r'\([A-Za-z0-9\.]+\)', '', name) name = name.replace('{', '').replace('}', '') return name # BEGIN: class Lexicon class Lexicon(object): """Holder for morphemes. """ def __init__(self): """Sets up tables to allow lookup of morphemes by sound or meaning. semantics_to_morphemes_primary indicates which morpheme is considered the primary exponent of a concept. """ self._phonology_to_morphemes = {} self._semantics_to_morphemes = {} self._semantics_to_morphemes_primary = {} self._used_spellings = set() self._used_pron_spellings = set() self._used_sem_spellings = set() self._morphemes = [] self._matrix = {} # Distance matrix to be used by PhonologicalDistance self._phonetics_frozen = False self._semantics_frozen = False def add_morpheme(self, morpheme): """Adds a morpheme to the lexicon. Args: morpheme: a Morpheme instance Returns: None """ phonology = morpheme.phonology semantics = morpheme.semantics spelling = morpheme.symbol if phonology in self._phonology_to_morphemes: self._phonology_to_morphemes[phonology].append(morpheme) else: self._phonology_to_morphemes[phonology] = [morpheme] if semantics.name in self._semantics_to_morphemes: self._semantics_to_morphemes[semantics.name].append(morpheme) else: self._semantics_to_morphemes[semantics.name] = [morpheme] if morpheme.is_primary: self._semantics_to_morphemes_primary[semantics.name] = morpheme if spelling: str_spelling = str(spelling) self._used_spellings.add(str_spelling) self._morphemes.append(morpheme) def find_morphemes(self, key): """Finds morphemes by sound or meaning. Args: key: key to search in phonology or semantics tables Returns: morphemes related to key """ try: return self._phonology_to_morphemes[key] except KeyError: pass try: return self._semantics_to_morphemes[key] except KeyError: return [] def apply_ablaut(self): """Applies an ablauting operation to all of the morphs. This results in multiple phonological forms being associated with the same morpheme. Returns: None """ keys = self._phonology_to_morphemes.keys() ablauted = builder.apply_ablaut(keys) i = 0 while i < len(keys): phonology = ablauted[i] for morpheme in self._phonology_to_morphemes[keys[i]]: # We have already ablauted this morpheme if morpheme.marked: continue morpheme.add_alternative_phonology(phonology) morpheme.mark() if phonology in self._phonology_to_morphemes: if morpheme not in self._phonology_to_morphemes[phonology]: self._phonology_to_morphemes[phonology].append(morpheme) else: self._phonology_to_morphemes[phonology] = [morpheme] i += 1 # Finally unmark all the morphemes for key in self._phonology_to_morphemes: for morpheme in self._phonology_to_morphemes[key]: morpheme.unmark() def dump_morphemes(self, outfile = None): """Writes out morphemes to a file, or to stdout. Args: outfile: An output file, or stdout if None Returns: None """ stream = sys.stdout if outfile: stream = open(outfile, 'w') for key in self._phonology_to_morphemes: for morpheme in self._phonology_to_morphemes[key]: stream.write('%s\t%s\t%s\n' % (key, morpheme.symbol_name(), morpheme)) if outfile: stream.close() def pronunciations(self): """Returns all pronunciations. """ return self._phonology_to_morphemes.keys() def useful_pronunciations(self): """Returns useful pronunciations: those associated with written symbols. """ prons = [] for pron in self._phonology_to_morphemes: for morpheme in self._phonology_to_morphemes[pron]: if not morpheme.symbol: continue prons.append(pron) break return prons def dump_morphs(self, outfile = None): """Writes out morphs to a file, or to stdout. Args: outfile: An output file, or stdout if None Returns: None """ stream = sys.stdout if outfile: stream = open(outfile, 'w') for key in self._phonology_to_morphemes: stream.write('%s\n' % key) if outfile: stream.close() def used_spellings(self): """Returns list of spellings that are already used. """ return list(self._used_spellings) def get_symbols_from_pron(self, pron): """Finds and returns all symbols associated with this pronunciation. Converts these to phonological components. """ if pron not in self._phonology_to_morphemes: return [] result = [] for morpheme in self._phonology_to_morphemes[pron]: if morpheme.symbol: symbol = Symbol(morpheme.symbol.name, pron) if (self._phonetics_frozen and str(symbol) not in self._used_pron_spellings): log.log('Disallowing use of {} as phonetic'.format(str(symbol))) continue symbol._colored_name = _phonetic_color(morpheme.symbol.colored_name) result.append(symbol) return _uniqify_symbol_list(result) def get_symbols_from_sem(self, sem): """Finds and returns all symbols associated with this meaning. """ if sem not in self._semantics_to_morphemes: return [] result = [] for morpheme in self._semantics_to_morphemes[sem]: if morpheme.symbol: symbol = Symbol(morpheme.symbol.name, sem) if (self._semantics_frozen and str(symbol) not in self._used_sem_spellings): log.log('Disallowing use of {} as semantic'.format(str(symbol))) continue symbol._colored_name = _semantic_color(morpheme.symbol.colored_name) # TODO(rws): This needs to be reworked since we don't necessarily "use" # this below, so it could be returned to be recycled. self._used_sem_spellings.add(str(symbol)) result.append(symbol) return _uniqify_symbol_list(result) def generate_new_spellings(self): """Generates new spellings with some probability for each morpheme. Works on morphemes that have no spelling. UsefulPronunciations are updated to PhonologicalDistance once each cycle. Returns: None """ useful_pronunciations = self.useful_pronunciations() log.log('# of useful pronunciations = %d' % len(useful_pronunciations)) distance = PhonologicalDistance(useful_pronunciations, self._matrix) morphemes_without_symbols = [] for morpheme in self._morphemes: if not morpheme.symbol: morphemes_without_symbols.append(morpheme) log.log('# of morphemes without symbols = %d' % len(morphemes_without_symbols)) init_time = time.clock() for morpheme in morphemes_without_symbols: init_time = time.clock() if random.random() < flags.FLAGS_probability_to_seek_spelling: ## TODO(rws): at some point we should add in the alternative phonology ## for ablauted forms, otherwise those will never participate: actually ## not completely true since the ablauted forms do inherit from the base ## form so that an ablauted form "work" might be spelled based on the ## phonology of the base form "werk" pron = morpheme.phonology if pron == '': continue # Shouldn't happen close_prons = distance.closest_prons(pron) phonological_spellings = [] spelling_to_pron = {} # Stores pron associated w/ each new spelling for close_pron, unused_cost in close_prons: prons = close_pron.split('.') if len(prons) == 1: # A single pronunciation spellings = self.get_symbols_from_pron(prons[0]) for spelling in spellings: spelling_to_pron[spelling.name] = close_pron phonological_spellings += spellings elif len(prons) == 2: # A telescoped pronunciation phonological_spellings1 = self.get_symbols_from_pron(prons[0]) phonological_spellings2 = self.get_symbols_from_pron(prons[1]) for p1 in phonological_spellings1: for p2 in phonological_spellings2: spelling = p1 + p2 spelling.set_denotation(close_pron) phonological_spellings.append(spelling) spelling_to_pron[spelling.name] = close_pron concept = morpheme.semantics semantic_spellings = [] for sem in concept.name.split(','): semantic_spellings += self.get_symbols_from_sem(sem) # Also tries the whole composite concept: if ',' in concept.name: semantic_spellings += self.get_symbols_from_sem(concept.name) new_spellings = phonological_spellings + semantic_spellings log_string = '\n>>>>>>>>>>>>>>>>>>>>>>>>>\n' log_string += 'For morpheme: %s, %s:\n' % (concept, pron) log_string += 'Phonetic spellings:\n' for phonological_spelling in phonological_spellings: log_string += '%s\n' % phonological_spelling log_string += 'Semantic spellings:\n' for semantic_spelling in semantic_spellings: log_string += '%s\n' % semantic_spelling log_string += '<<<<<<<<<<<<<<<<<<<<<<<<<' log.log(log_string) for phonological_spelling in phonological_spellings: for semantic_spelling in semantic_spellings: combo_spelling = semantic_spelling + phonological_spelling combo_spelling.set_denotation(semantic_spelling.denotation + '+' + phonological_spelling.denotation) spelling_to_pron[ combo_spelling.name] = spelling_to_pron[phonological_spelling.name] new_spellings.append(combo_spelling) # TODO(rws): this is an experiment. Note that with this setting, # eliminating the ridiculously long spellings then picking randomly from # among these, gets a proportion of semantic/phonetic spellings of 0.32 # for the MONOSYLLABLE setting. tmp = [] for sp in new_spellings: if len(sp) < 5: tmp.append(sp) new_spellings = tmp random.shuffle(new_spellings) # Whereas with this setting, commented out for now, always favoring the # absolute shortest, semphon is much lower for 1000, though if you # increase to 5000 it gets to around 0.22. Presumably that is because # with the larger vocab one starts to actually need the semphon # spellings: # # new_spellings.sort(lambda x, y: cmp(len(x), len(y))) for spelling in new_spellings: reuse = str(spelling) in self._used_spellings if (not reuse or random.random() < _PROBABILITY_TO_REUSE_SPELLING): pron = '' if spelling.name in spelling_to_pron: pron = spelling_to_pron[spelling.name] morpheme.set_spelling(spelling) self._used_spellings.add(str(spelling)) log_string = 'Spelling: %s\t' % spelling log_string += 'Morpheme: %s\t' % str(morpheme) if pron: self._used_pron_spellings.add(str(spelling)) log_string += 'Source-pronunciation: %s\t' % pron if reuse: log_string += 'Reuse' log.log(log_string) break def log_pron_to_symbol_map(self): """Adds pron/symbol mapping for the (usually final) lexicon. Returns: None """ for pron in self._phonology_to_morphemes: for morpheme in self._phonology_to_morphemes[pron]: if morpheme.symbol: log.log('SYMBOL:\t{}\t{}'.format(morpheme.symbol, pron)) def freeze_phonetics(self): """Freezes the phonetics. """ self._phonetics_frozen = True def freeze_semantics(self): """Freezes the semantics. """ self._semantics_frozen = True # END: class Lexicon # BEGIN: class Morpheme class Morpheme(object): """Container for a morpheme object consisting of sound paired with meaning. """ def __init__(self, phonology, semantics, symbol, is_primary): self._phonology = phonology self._alternative_phonology = [] # For ablauted forms, etc self._semantics = semantics self._symbol = symbol # set representation of semantics self._semantics_set = set(semantics.name.split(',')) self._is_primary = is_primary # Is the primary exponent of this concept # Book-keeping placeholder to mark whether an operation has applied: self._marked = False def __repr__(self): alternative_phonology = '' if self._alternative_phonology: alternative_phonology = '(%s)' % ','.join(self._alternative_phonology) symbol = '' if self._symbol: symbol = '<%s:%s:%s>' % (str(self._symbol), self._symbol.symbols(), self._symbol.type()) props = '{%s%s:%s:%s:%d}' % (self._phonology, alternative_phonology, str(self._semantics).replace('@', ''), symbol, self._is_primary) return props @property def phonology(self): return self._phonology @property def semantics(self): return self._semantics @property def is_primary(self): return self._is_primary @property def marked(self): return self._marked @property def symbol(self): return self._symbol def symbol_name(self): if self._symbol: return self._symbol.colored_name else: return '<NO_SYMBOL>' def mark(self): self._marked = True def unmark(self): self._marked = False def has_semantics(self, semantics): return semantics in self._semantics_set def add_alternative_phonology(self, phonology): if phonology not in self._alternative_phonology: self._alternative_phonology.append(phonology) def set_spelling(self, spelling): self._symbol = spelling # END: class Morpheme # BEGIN: class Concept class Concept(object): """A representation of meaning. """ def __init__(self, name): """name is a comma-separated set of primitives. """ self._name = name def __repr__(self): return self._name @property def name(self): return self._name # END: class Concept # BEGIN: class Symbol class Symbol(object): """Representation for a symbol and what kind of thing it represents. name is a string of one or more symbols """ def __init__(self, name, denotation=None): """Denotation type defaults to semantic """ self._name = name self._denotation = denotation # _colored_name is the symbol's spelling marked with red or blue according # to whether it is being used as phonetic or semantic. if not denotation or denotation.startswith('@'): self._colored_name = _BLUE % name else: self._colored_name = _RED % name def __repr__(self): if self._denotation.startswith('@'): denotation = '[%s]' % self._denotation[1:] else: denotation = '(%s)' % self._denotation return '{%s}%s' % (self._name, denotation) def __add__(self, other): new_symbol = Symbol(str(self) + str(other)) new_symbol._colored_name = self.colored_name + other.colored_name return new_symbol def __len__(self): """Length is the number of basic symbols comprising this. """ length = 0 for c in unicode(str(self), 'utf8'): if ord(c) > 128: length += 1 return length @property def name(self): return self._name @property def denotation(self): return self._denotation @property def colored_name(self): return self._colored_name def set_denotation(self, denotation): self._denotation = denotation def type(self): """Classification as S(emantic), P(honetic), or SP. """ if '+' in self._denotation: return 'SP' if self._denotation.startswith('@'): return 'S' return 'P' def symbols(self): """Returns string of just the symbols. """ return_value = '' for c in unicode(str(self), 'utf8'): if ord(c) > 128: return_value += c return return_value.encode('utf8') # END: class Symbol # BEGIN: class LexiconGenerator class LexiconGenerator(object): """Generator for lexicon with specified number of morphs and base morph type. """ def __init__(self, nmorphs = 5000, base_morph = 'MONOSYLLABLE'): self._nmorphs = nmorphs self._base_morph = base_morph self._initial = True def select_morphs(self, morphs): """Helper function to select from 1 to 3 morphs from a sequence. Args: morphs: list of morphs Returns: random selection among the morphs """ selections = set() num_selections = random.choice([1, 2, 3]) i = 0 while i < num_selections: morph = random.choice(morphs) selections.add(morph) i += 1 return selections def concept_combinations(self, concepts): """Randomly generates a set of one to three concepts. Args: concepts: list of concepts Returns: random selection among the concepts """ num_concepts = random.choice([1, 2, 3]) return ','.join(random.sample(concepts, num_concepts)) def generate(self, force = False): """Generates and returns a lexicon. Args: force: if True then force building of the grammars. Returns: a Lexicon """ if self._initial or force: builder.build_morphology_grammar() builder.build_soundslike_grammar() morphs = builder.generate_morphs(self._base_morph, self._nmorphs) nth_concept = 0 # Gets the concepts concepts_ = concepts.CONCEPTS lexicon = Lexicon() seen_morphs = set() for concept in concepts_.keys(): selections = self.select_morphs(morphs) is_primary = True for morph in selections: seen_morphs.add(morph) # As this has a basic concept, assign the symbol associated with this # concept as the symbol. # TODO(rws): make it a parameter whether the non-primaries get the # symbol. my_symbol = None if is_primary or flags.FLAGS_initialize_non_primaries_with_symbol: my_symbol = Symbol(concepts_[concept], concept) lexicon.add_morpheme(Morpheme(morph, Concept(concept), my_symbol, is_primary)) is_primary = False # After that the next sets of morphemes are assigned to random combinations # of concepts, one per morph. combinations = set() for morph in morphs: if morph in seen_morphs: continue concept = self.concept_combinations(concepts_.keys()) lexicon.add_morpheme(Morpheme(morph, Concept(concept), None, False if concept in combinations else True)) combinations.add(concept) return lexicon # END: class LexiconGenerator # BEGIN: class PhonologicalDistance class PhonologicalDistance(object): """Computes the phonological distance for a set of terms """ def __init__(self, pronunciations, matrix = {}): self._pronunciations = pronunciations self._matrix = matrix self._telescopings = {} self.compute_cross_product() def __memoize__(self, pron1, pron2): """Memoizes the distance for a particular pair of prons for efficiency. Args: pron1: first pronunciation pron2: second pronunciation Returns: the sounds_like distance between the prons """ if pron1 == pron2: return 0 if (pron1, pron2) in self._matrix: return self._matrix[pron1, pron2] length, cost = sounds_like(pron1, pron2) try: weighted_cost = cost / length except ZeroDivisionError: weighted_cost = float('Infinity') self._matrix[pron1, pron2] = weighted_cost return self._matrix[pron1, pron2] def compute_cross_product(self): """Finds all pairs p1, p2, where p1 ends in a V and p2 starts with a V. Returns: None """ pairs = [] for p1 in self._pronunciations: if not builder.is_vowel(p1[-1]): continue for p2 in self._pronunciations: if p1[-1] == p2[0]: new_pron = p1 + p2[1:] if new_pron not in self._pronunciations: # The only way to get this new pronunciation is via telescoping # so in that case only we add this to the set of new pairs pair = p1 + '.' + p2 # Note that this may be ambiguous: this just gets the last pair # that could produce this pronunciation. self._telescopings[new_pron] = pair pairs.append(new_pron) self._pronunciations += pairs def expand(self, pron): """Possibly expand into a pair of telescoped elements Args: pron Returns: telescoping of pron if in _telescopings, else pron """ if pron in self._telescopings: return self._telescopings[pron] return pron def closest_prons(self, pron1): """Returns an ordered list of closest prons to pron. """ result = [] for pron2 in self._pronunciations: result.append((self.expand(pron2), self.__memoize__(pron1, pron2))) result.sort(lambda x, y: cmp(x[1], y[1])) return [x for x in result if x[1] <= _MAX_DISTANCE] # END: class PhonologicalDistance def main(argv): global _PROBABILITY_TO_SEEK_SPELLING flags.define_flag('ablaut', '0', 'Apply ablaut') flags.define_flag('base_morph', 'MONOSYLLABLE', 'Base morpheme shape to use') flags.define_flag('initialize_non_primaries_with_symbol', '0', 'Sets whether or not non primary morphs get the ' 'symbol initially') flags.define_flag('niter', '5', 'Number of iterations') flags.define_flag('nmorphs', '1000', 'Number of morphs') flags.define_flag('outdir', '/var/tmp/simulation', 'Output directory') # Probability that one will seek a spelling for a morpheme # TODO(rws): make this sensitive to lexical frequency flags.define_flag('probability_to_seek_spelling', '0.3', 'Probability to seek spelling for a form') flags.define_flag('freeze_phonetics_at_iter', '0', 'Do not allow any new phonetic symbols after iteration N') flags.define_flag('freeze_semantics_at_iter', '0', 'Do not allow any new semantic spread after iteration N') flags.parse_flags(argv[1:]) generator = LexiconGenerator(nmorphs=flags.FLAGS_nmorphs, base_morph=flags.FLAGS_base_morph) lexicon = generator.generate() print '{} {}'.format('Probability to seek spelling is', flags.FLAGS_probability_to_seek_spelling) print 'Base morph is', flags.FLAGS_base_morph print '{} {}'.format('initialize_non_primaries_with_symbol is', flags.FLAGS_initialize_non_primaries_with_symbol) print 'Apply ablaut =', flags.FLAGS_ablaut print 'outdir =', flags.FLAGS_outdir print 'niter =', flags.FLAGS_niter print 'nmorphs =', flags.FLAGS_nmorphs if flags.FLAGS_ablaut: lexicon.apply_ablaut() outdir = flags.FLAGS_outdir try: os.makedirs(outdir) except OSError: pass lexicon.dump_morphemes(outdir + '/morphemes_0000.tsv') with open(outdir + '/log.txt', 'w') as stream: log.LOG_STREAM = stream for i in range(1, flags.FLAGS_niter): if flags.FLAGS_freeze_phonetics_at_iter == i: lexicon.freeze_phonetics() if flags.FLAGS_freeze_semantics_at_iter == i: lexicon.freeze_semantics() print 'Iteration %d' % i log.log('Iteration %d' % i) lexicon.generate_new_spellings() lexicon.dump_morphemes(outdir + '/morphemes_%04d.tsv' % i) lexicon.log_pron_to_symbol_map() if __name__ == '__main__': main(sys.argv)
{"/builder.py": ["/pynini_interface.py"]}
50,243
rwsproat/writing_evolution
refs/heads/master
/builder.py
## Licensed under the Apache License, Version 2.0 (the "License"); ## you may not use this file except in compliance with the License. ## You may obtain a copy of the License at ## ## http://www.apache.org/licenses/LICENSE-2.0 ## ## Unless required by applicable law or agreed to in writing, software ## distributed under the License is distributed on an "AS IS" BASIS, ## WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. ## See the License for the specific language governing permissions and ## limitations under the License. ## ## Author: Richard Sproat (rws@xoba.com) """Builds all the needed grammars and lists, placing them in Data directory. """ import os import sys import pynini_interface from base import _BASE _VOWELS = set() # TODO(rws): Remove dependency on Thrax entirely by rewriting the grammars in # Pynini. def build_grammar(name): """Builds the grammars using Thrax, and extracts the relevant fsts. This assumes that thrax utility thraxmakedep is accessible. Args: name: name for the grammar. Returns: None """ os.system('thraxmakedep %s/Grm/%s.grm' % (_BASE, name)) os.system('make') os.system('rm -f Makefile') load_vowel_definitions() def load_vowel_definitions(): """Loads the vowel definitions from phonemes.tsv. Returns: None """ with open('%s/Grm/phonemes.tsv' % _BASE) as stream: for line in stream: try: clas, segment = line.split() except ValueError: continue if clas.startswith('V'): _VOWELS.add(segment) def is_vowel(segment): """Returns true if segment is a vowel. Args: segment: name of the segmeent Returns: Boolean """ return segment in _VOWELS def build_morphology_grammar(): """Builds the morphology grammar. Returns: None """ build_grammar('morphology') def build_soundslike_grammar(): """Builds the sounds-like grammar. Returns: None """ build_grammar('soundslike') def generate_morphs(base_morph='MONOSYLLABLE', n=1000, far=("%s/Grm/morphology.far" % _BASE)): """Generates a set of morphs according to the base_morph template. Args: base_morph: name of the base morph rule, e.g. MONOSYLLABLE n: number of morphs to generate Returns: list of morphs """ pynini_interface.load_rule_from_far(base_morph, far) return pynini_interface.random_paths(base_morph, n) def dump_morphs(morphs, outfile=None): """Dumps the morphs to a file. Args: morphs: list of morphs outfile: output file, or stdout if None Returns: None """ stream = sys.stdout if outfile: stream = open(outfile, 'w') for morph in morphs: stream.write(morph + '\n') if outfile: stream.close() def apply_ablaut(morphs): """Applies the ablaut rule to a set of morphs. Args: morphs: list of morphs Returns: ablauted list of morphs """ ablaut_rule = pynini_interface.load_rule_from_far( 'ABLAUT', '%s/Grm/morphology.far' % _BASE) ablauted = [] for morph in morphs: morph_fst = pynini_interface.to_fst(morph) result = pynini_interface.shortestpath(morph_fst * ablaut_rule) result.project(True) result = pynini_interface.to_string(result) # We do not allow it to delete the morph entirely. if result: ablauted.append(result) else: ablauted.append(morph) return ablauted
{"/builder.py": ["/pynini_interface.py"]}
50,244
rwsproat/writing_evolution
refs/heads/master
/stats.py
#!/usr/bin/env python ## Licensed under the Apache License, Version 2.0 (the "License"); ## you may not use this file except in compliance with the License. ## You may obtain a copy of the License at ## ## http://www.apache.org/licenses/LICENSE-2.0 ## ## Unless required by applicable law or agreed to in writing, software ## distributed under the License is distributed on an "AS IS" BASIS, ## WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. ## See the License for the specific language governing permissions and ## limitations under the License. ## ## Author: Richard Sproat (rws@xoba.com) """Script that computes the basic statistics on the results of a run. Creates an R script to produce plots in plot.R Usage: stats.py simulation_output_directory """ import glob import sys _PLOT = """pdf("plot.pdf") plot(nsyms, xlab="Epoch", ylab=("Prop"), ylim=c(0, 1), type="l", col=1) par(new=TRUE) plot(nsemphon, xlab="Epoch", ylab=("Prop"), ylim=c(0, 1), type="l", col=2) par(new=TRUE) plot(nphon, xlab="Epoch", ylab=("Prop"), ylim=c(0, 1), type="l", col=3) par(new=TRUE) plot(nsem, xlab="Epoch", ylab=("Prop"), ylim=c(0, 1), type="l", col=4) par(new=TRUE) plot(nphon + nsemphon, xlab="Epoch", ylab=("Prop"), ylim=c(0, 1), type="l", col=5) legend(1, 0.9, col=c(1, 2, 3, 4, 5), lty=c(1, 1, 1, 1, 1), legend=c("# spellings", "# sem-phon", "# phon", "# sem", "all phon")) """ def main(argv): print '%10s\t%10s\t%10s\t%10s\t%10s' % ('# morphs', 'prop spell', 'semphon', 'phon', 'sem') nsym_list = [] nsemphon_list = [] nphon_list = [] nsem_list = [] for morph_file in glob.glob(argv[1] + '/morphemes_*.tsv'): nsyms = 0 nsemphon = 0 nphon = 0 nsem = 0 tot = 0.0 with open(morph_file) as strm: for line in strm: tot += 1 if 'NO_SYMBOL' in line: continue nsyms += 1 if ':SP>' in line: nsemphon +=1 if ':P>' in line: nphon +=1 if ':S>' in line: nsem +=1 semphon_str = '%d\t%2.2f' % (nsemphon, nsemphon / float(nsyms)) phon_str = '%d\t%2.2f' % (nphon, nphon / float(nsyms)) sem_str = '%d\t%2.2f' % (nsem, nsem / float(nsyms)) print '%10d\t%10f\t%10s\t%10s\t%10s' % (tot, nsyms/tot, semphon_str, phon_str, sem_str) nsym_list.append(str(nsyms/tot)) nsemphon_list.append(str(nsemphon/tot)) nphon_list.append(str(nphon/tot)) nsem_list.append(str(nsem/tot)) with open('plot.R', 'w') as plot: plot.write('nsyms <- c(%s)\n' % ', '.join(nsym_list)) plot.write('nsemphon <- c(%s)\n' % ', '.join(nsemphon_list)) plot.write('nphon <- c(%s)\n' % ', '.join(nphon_list)) plot.write('nsem <- c(%s)\n' % ', '.join(nsem_list)) plot.write(_PLOT) if __name__ == '__main__': main(sys.argv)
{"/builder.py": ["/pynini_interface.py"]}
50,245
rwsproat/writing_evolution
refs/heads/master
/pynini_interface.py
## Licensed under the Apache License, Version 2.0 (the "License"); ## you may not use this file except in compliance with the License. ## You may obtain a copy of the License at ## ## http://www.apache.org/licenses/LICENSE-2.0 ## ## Unless required by applicable law or agreed to in writing, software ## distributed under the License is distributed on an "AS IS" BASIS, ## WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. ## See the License for the specific language governing permissions and ## limitations under the License. ## ## Author: Richard Sproat (rws@xoba.com) import time import sys from pynini import * from base import _BASE _LOADED_FARS = {} _LOADED_FSTS = {} def load_rule_from_far(rule, far, force=False): """Loads the rule, and caches it, reloading if force=True Args: rule: Rule name far: Far name force: If True, forces reload Returns: loaded fst or fail if no such fst or far """ try: if force or far not in _LOADED_FARS: _LOADED_FARS[far] = Far(far) # TODO(rws): If there is the same rule name in different fars, then this # will just get the first one we encounter. if force or rule not in _LOADED_FSTS: _LOADED_FSTS[rule] = _LOADED_FARS[far][rule] return _LOADED_FSTS[rule] except pywrapfst.FstIOError: sys.stderr.write('Failed loading far from %s\n' % far) sys.exit(1) except KeyError: sys.stderr.write('No rule "%s" in %s\n' % (rule, far)) sys.exit(1) def to_fst(s, syms='byte'): """Constructs an fst from a string. Args: s: string syms: symbol table Returns: fst representing the string """ return acceptor(s, token_type=syms) def to_string(t): """Constructs a string from a fst. Args: t: fst Returns: string representing the fst """ t.rmepsilon() return t.stringify() def random_paths(t, n=1): """Computes a set of random paths from an fst Args: t: fst n: number of paths Returns: list of random path strings """ if type(t) == type('string'): try: t = _LOADED_FSTS[t] except KeyError: sys.stderr.write('Missing transducer %s\n' % t) return [] i = 0 paths = [] while i < n: output = randgen(t, seed=int(time.time() * 1000000), select='uniform') output.rmepsilon() output.topsort() paths.append(output.stringify()) i += 1 return paths _CACHED_COMPOSITIONS = {} def sounds_like(s1, s2, rule='EDIT_DISTANCE', far=('%s/Grm/soundslike.far' % _BASE)): """Computes the distance between two phonetic strings given a grammar. Args: s1: phonetic string 1 s2: phonetic string 2 grm: phonetic similarity grammar Returns: number of arcs in shortest path, shortest distance """ grmfst = load_rule_from_far(rule, far) if (s1, rule) in _CACHED_COMPOSITIONS: fst1 = _CACHED_COMPOSITIONS[s1, rule] else: fst1 = s1 * grmfst _CACHED_COMPOSITIONS[s1, rule] = fst1 result = shortestpath(fst1 * s2) result.rmepsilon() result.topsort() if result.num_states() > 1: dist = shortestdistance(result) dist = (float(str(dist[-1])) + float(str(result.final(result.num_states() - 1)))) return result.num_states() - 1, dist else: return 0, float('inf')
{"/builder.py": ["/pynini_interface.py"]}
50,254
1dayac/magic_bot
refs/heads/master
/candle_factory_5.py
#clanteam from selenium import webdriver import time import sys from datetime import datetime import re from mtgolibrary import MtgoLibraryParser from card import Card, Price import sqlite3 con = sqlite3.connect('cards.db', isolation_level=None) cursor = con.cursor() tb_exists = "SELECT name FROM sqlite_master WHERE type='table' AND name='records'" if not con.execute(tb_exists).fetchone(): c = cursor.execute("CREATE TABLE records(Id TEXT PRIMARY KEY, CardName TEXT, SetName TEXT, BuyPrice REAL, SellPrice REAL, " "BotNameSeller TEXT, BotNameBuyer TEXT, Time TEXT, Number INT, Foil INT)") else: print("Table exists") def is_basic_land(card): return card.name == "Swamp" or card.name == "Island" or card.name == "Mountain" or card.name == "Plains" or card.name == "Forest" or card.name.startswith("Urza's") import win32api, win32con, win32process def setaffinity(): return pid = win32api.GetCurrentProcessId() mask = 3 # core 7 handle = win32api.OpenProcess(win32con.PROCESS_ALL_ACCESS, True, pid) win32process.SetProcessAffinityMask(handle, mask) setaffinity() import platform if platform.system() == "Windows": chromedriver_path = r"C:\Users\IEUser\Desktop\magic_bot\chromedriver.exe" else: chromedriver_path = "/home/dmm2017/PycharmProjects/candle_factory/chromedriver" class HotlistProcessor(object): def __init__(self): self.start_from = "1" self.set = "1" self.rows = [] self.driver_hotlist = None self.start = None self.i = 0 self.mtgolibrary_parser = MtgoLibraryParser() def restart(self): self.start_from = "1" self.set = "1" self.rows = [] try: self.driver_hotlist.quit() except: pass self.driver_hotlist = None self.start = None self.i = 0 self.mtgolibrary_parser.restart() def openHotlist(self): url = "http://www.mtgotraders.com/hotlist/#/" chrome_options = webdriver.ChromeOptions() chrome_options.add_argument("--headless") self.driver_hotlist = webdriver.Chrome(chromedriver_path, options = chrome_options) self.driver_hotlist.get(url) time.sleep(60) elems = self.driver_hotlist.find_elements_by_class_name('btn') elems[0].click() elems_2 = self.driver_hotlist.find_element_by_xpath( "//*[@id=\"mainContent\"]/div[2]/div[1]/div[2]/div[4]/div[1]/span[2]/span/ul/li[5]") elems_2.click() time.sleep(4) table = self.driver_hotlist.find_element_by_id('main-table') rows = table.find_elements_by_tag_name('tr') return rows def processHotlist(self): self.rows = self.openHotlist() self.start = time.time() while True: try: while self.i < len(self.rows): self.processRow(self.rows[self.i]) self.i += 1 end = time.time() if end - self.start > 600: raise Exception break except: print(sys.exc_info()[1]) while True: try: temp_i = self.i self.restart() self.i = temp_i self.rows = self.openHotlist() print(len(self.rows)) time.sleep(5) self.start = time.time() break except: pass def processRow(self, row): columns = row.find_elements_by_tag_name('td') if len(columns) < 3: return setname = columns[0].text self.set = setname cardname = columns[1].text price = float(columns[3].text) if setname < self.start_from: return if price < 0.05: return foil = cardname.endswith("*") if foil: cardname = cardname[:-7] print(setname + " " + cardname + " " + str(price)) price_struct = Price("", price, 10000, "Hotlistbot3", "", 0) card = Card(cardname, setname, price_struct, foil) if is_basic_land(card) or ((card.set == "MS2" or card.set == "MS3") and card.foil): return p = self.mtgolibrary_parser.get_price(card) if not p: return if price - p.sell_price > 0.025 and p.sell_price != 10000: print("High diff: " + p.bot_name_sell + " " + str(price - p.sell_price)) cursor.execute("INSERT OR REPLACE INTO records VALUES(?,?,?,?,?,?,?,?,?,?)", [setname + cardname, cardname, setname, price, p.sell_price, p.bot_name_sell, "HotListBot3", datetime.now(), min(4, p.number), 1 if foil else 0]) while True: try: processeor = HotlistProcessor() processeor.processHotlist() except: processeor.restart()
{"/candle_factory_5.py": ["/mtgolibrary.py", "/card.py"], "/goatbots_cardhoarder.py": ["/card.py", "/mtgolibrary.py"], "/number_parser.py": ["/card.py", "/queue_with_capacity.py"], "/mtgolibrary.py": ["/card.py", "/number_parser.py"], "/mtgo_controller.py": ["/queue_with_capacity.py"]}
50,255
1dayac/magic_bot
refs/heads/master
/queue_with_capacity.py
import queue class QueueWithMaxCapacity(object): def __init__(self, capacity = 100): self.limit = capacity self.queue = queue.Queue() def add(self, item): if self.queue.qsize() > self.limit: self.queue.get() self.queue.put(item)
{"/candle_factory_5.py": ["/mtgolibrary.py", "/card.py"], "/goatbots_cardhoarder.py": ["/card.py", "/mtgolibrary.py"], "/number_parser.py": ["/card.py", "/queue_with_capacity.py"], "/mtgolibrary.py": ["/card.py", "/number_parser.py"], "/mtgo_controller.py": ["/queue_with_capacity.py"]}
50,256
1dayac/magic_bot
refs/heads/master
/goatbots_cardhoarder.py
from selenium import webdriver import sqlite3 from datetime import datetime from card import Card, Price import time from mtgolibrary import MtgoLibraryParser chromedriver_path = r"C:\Users\dmm2017\Desktop\magic_bot\chromedriver.exe" import win32api, win32con, win32process def setaffinity(): pid = win32api.GetCurrentProcessId() mask = 3 # core 7 handle = win32api.OpenProcess(win32con.PROCESS_ALL_ACCESS, True, pid) win32process.SetProcessAffinityMask(handle, mask) setaffinity() class TopLevelProcessor(object): def __init__(self): self.processors = [] self.main_processor = None self.database_path = "cards.db" con = sqlite3.connect('cards.db', isolation_level=None) self.cursor = con.cursor() tb_exists = "SELECT name FROM sqlite_master WHERE type='table' AND name='records'" if not con.execute(tb_exists).fetchone(): c = self.cursor.execute( "CREATE TABLE records(Id TEXT PRIMARY KEY, CardName TEXT, SetName TEXT, BuyPrice REAL, SellPrice REAL, " "BotNameSeller TEXT, BotNameBuyer TEXT, Time TEXT, Number INT, Foil INT)") else: print("Table exists") def RestartAll(self): for processor in self.processors: processor.restart() self.main_processor.restart() def AddMainProcessor(self, processor): self.main_processor = processor def AddProcessor(self, processor): self.processors.append(processor) def IsHighDiff(self, card): best_buy_price = card.BestBuyPrice() best_sell_price = card.BestSellPrice() print(best_buy_price) print(best_sell_price) return best_buy_price.buy_price - best_sell_price.sell_price >= 0.05 def AddToDatabase(self, card): best_buy_price = card.BestBuyPrice() best_sell_price = card.BestSellPrice() print("High diff: " + best_sell_price.bot_name_sell + "," + str(best_sell_price.sell_price) + " " + best_buy_price.bot_name_buy + "," + str(best_buy_price.buy_price)+": " + str(best_buy_price.buy_price - best_sell_price.sell_price)) self.cursor.execute("INSERT OR REPLACE INTO records VALUES(?,?,?,?,?,?,?,?,?,?)", [card.set + card.name, card.name, card.set, best_buy_price.buy_price, best_sell_price.sell_price, best_sell_price.bot_name_sell, best_buy_price.bot_name_buy, datetime.now(), min(4, best_buy_price.number), 1 if card.foil else 0]) def ParseCards(self): for card in self.main_processor.GetCards(): for processor in self.processors: card.prices.append(processor.get_price(card)) if self.IsHighDiff(card): self.AddToDatabase(card) class GoatBotsParser(object): def __init__(self): self.main_url = "https://www.goatbots.com/prices" chrome_options = webdriver.ChromeOptions() prefs = {'profile.managed_default_content_settings.images': 2} chrome_options.add_experimental_option("prefs", prefs) chrome_options.add_argument("--headless") self.driver = webdriver.Chrome(chromedriver_path, options= chrome_options) self.second_driver = webdriver.Chrome(chromedriver_path, options= chrome_options) self.current_set = None self.card_count = 0 def restart(self): self.second_driver.quit() self.main_url = "https://www.goatbots.com/prices" chrome_options = webdriver.ChromeOptions() chrome_options.add_argument("--headless") prefs = {'profile.managed_default_content_settings.images': 2} chrome_options.add_experimental_option("prefs", prefs) self.second_driver = webdriver.Chrome(chromedriver_path, options= chrome_options) def ProcessRow(self, line): href_cell = line.find_element_by_css_selector('a') if "/search" in str(href_cell.get_attribute("href")): return None card_url = href_cell.get_attribute('href') self.second_driver.get(card_url) card_name = " ".join(self.second_driver.title.split(" ")[:-2]) if "Full Set" in card_name or "Booster" in card_name: return None print(card_name) p = Price("", -5, 1000, "GoatBots3", "GoatBots3", 4) c = Card(card_name, self.current_set, p, 0) #return c price_block = self.second_driver.find_element_by_id("info_" + self.current_set.lower()) all_h3 = price_block.find_elements_by_tag_name('h3') prices = [] for h3 in all_h3: if "price_value" in h3.get_attribute("class"): prices.append(h3.text) try: buy_price = float(prices[0]) except: buy_price = -1.0 try: sell_price = float(prices[1]) except: sell_price = 100000 self.second_driver.find_element_by_class_name("cart_icon").click() self.second_driver.find_element_by_id("header_cart_icon").click() real_sell_price = float(self.second_driver.find_element_by_id("delivery_table").find_elements_by_tag_name("td")[3].find_element_by_tag_name("a").text) self.second_driver.find_element_by_id("form_cart_empty_cart").click() p = Price("", buy_price - (sell_price - real_sell_price), real_sell_price, "GoatBots3", "GoatBots3", 4) c = Card(card_name, self.current_set, p, 0) return c def GetSet(self): set_line = self.driver.find_element_by_xpath("//*[@id=\"text_left\"]/div[1]/h2/a").text return set_line[set_line.find("(") +1 : set_line.find(")")] def GetCards(self): self.driver.get(self.main_url) sets = self.driver.find_elements_by_class_name('cardset') refs = [] min_iter = 40 max_iter = 50 current_iter = 0 for set in sets: current_iter += 1 if current_iter < min_iter: continue if current_iter == max_iter: break try: print(set.find_element_by_tag_name('a').get_attribute('href')) refs.append(set.find_element_by_tag_name('a').get_attribute('href')) except: pass max_ref = 20 current_ref = 0 for ref in refs: current_ref += 1 print("Processing " + ref) max_cards = 20 current = 0 self.driver.get(ref) print(ref) self.current_set = self.GetSet() for line in self.driver.find_element_by_id('pricesTable').find_elements_by_tag_name('tr'): current += 1 if current == max_cards: break if "empty_row" in line.get_attribute('class'): continue if "card" in line.get_attribute('class'): try: card = self.ProcessRow(line) except: continue if card is None: continue if card.MinSellPrice() < 0.3: break else: self.card_count += 1 yield card if self.card_count % 100 == 0: self.restart() class CardHoarderParser(object): def __init__(self): self.main_url = "https://www.cardhoarder.com/cards" chrome_options = webdriver.ChromeOptions() chrome_options.add_argument("--headless") prefs = {'profile.managed_default_content_settings.images': 2} chrome_options.add_experimental_option("prefs", prefs) self.driver = webdriver.Chrome(chromedriver_path, options= chrome_options) self.driver.get(self.main_url) def restart(self): self.driver.quit() self.main_url = "https://www.cardhoarder.com/cards" chrome_options = webdriver.ChromeOptions() prefs = {'profile.managed_default_content_settings.images': 2} chrome_options.add_experimental_option("prefs", prefs) chrome_options.add_argument("--headless") self.driver = webdriver.Chrome(chromedriver_path, options= chrome_options) self.driver.get(self.main_url) def get_price(self, card): input_element = self.driver.find_element_by_id("card-search-input") input_element.clear() input_element.send_keys(card.name) self.driver.find_element_by_class_name("btn-search").click() table = self.driver.find_element_by_id("search-results-table") trs = table.find_elements_by_tag_name("tr") final_href = "" for tr in trs: try: reference = tr.find_element_by_tag_name("a").get_attribute('href') set = reference.split("/")[-1].split("-")[0].upper() if set != card.set or reference.endswith("-foil") or reference.endswith("#"): continue final_href = reference break except: pass self.driver.get(final_href) panel_body = self.driver.find_element_by_class_name('panel-body') try: sell_price = float(panel_body.find_element_by_class_name('card-ordering-details').text.strip().split(" ")[0]) except: sell_price = 100000.0 try: buy_price = float(panel_body.find_element_by_tag_name('h4').text.split(" ")[-2]) except: buy_price = -1.0 return Price("", buy_price, sell_price, "CardBuyingBot3", "CardBot3", 4) import sys, traceback top_level_processor = TopLevelProcessor() top_level_processor.AddMainProcessor(GoatBotsParser()) top_level_processor.AddProcessor(CardHoarderParser()) top_level_processor.AddProcessor(MtgoLibraryParser()) while True: try: top_level_processor.ParseCards() top_level_processor.RestartAll() break except: top_level_processor.RestartAll() print("Unexpected error:", sys.exc_info()[0]) print("Unexpected error:", sys.exc_info()[1]) traceback.print_exc(file=sys.stdout) pass time.sleep(600)
{"/candle_factory_5.py": ["/mtgolibrary.py", "/card.py"], "/goatbots_cardhoarder.py": ["/card.py", "/mtgolibrary.py"], "/number_parser.py": ["/card.py", "/queue_with_capacity.py"], "/mtgolibrary.py": ["/card.py", "/number_parser.py"], "/mtgo_controller.py": ["/queue_with_capacity.py"]}
50,257
1dayac/magic_bot
refs/heads/master
/number_parser.py
import pickle import sys import queue from card import Card, Price from selenium import webdriver import time from queue_with_capacity import QueueWithMaxCapacity import platform if platform.system() == "Windows": chromedriver_path = r"C:\Users\IEUser\Desktop\magic_bot\chromedriver.exe" else: chromedriver_path = "/home/dmm2017/PycharmProjects/candle_factory/chromedriver" class DigitsClassifier(object): def ParseMtgolibraryInternal(self, driver, card, url, botname, sellprice_original): bot_name_original = botname setname = card.set elem = driver.find_elements_by_class_name("sell_row") if len(elem) == 0: return False for e in elem: try: table_setname = e.find_elements_by_class_name("setname")[0].text if table_setname != setname: continue bot_name_sell = e.find_elements_by_class_name("bot_name")[0].text if bot_name_sell != bot_name_original: continue except: continue sell_price_original = str(sellprice_original) images = e.find_element_by_class_name("sell_price_round") images_srcs = [image.get_attribute("src") for image in images.find_elements_by_tag_name('img')] if len(images_srcs) != len(sell_price_original): return for i in range(len(images_srcs)): if images_srcs[i] not in self.super_dict.keys(): self.super_dict[images_srcs[i]] = {} if sell_price_original[i] not in self.super_dict[images_srcs[i]].keys(): self.super_dict[images_srcs[i]][sell_price_original[i]] = 0 self.super_dict[images_srcs[i]][sell_price_original[i]] += 1 def ParseMtgolibrary(self, driver, card, botname, sellprice): setname, url, driver = self.MtgoLibraryGoToCard(driver, card) time.sleep(7) return self.ParseMtgolibraryInternal(driver, card, url, botname, sellprice) def MtgoLibraryGoToCard(self, driver, card): setname = card.set.upper() input_element = driver.find_element_by_id("_cardskeyword") input_element.clear() input_element.send_keys(card.name + " " + setname) driver.find_elements_by_css_selector("button")[1].click() url = driver.current_url return setname, url, driver def __init__(self): try: f = open("obj/dict.pkl", 'rb') self.prices_d = pickle.loads(f.read()) except: self.prices_d = {} self.old_prices = QueueWithMaxCapacity(100) self.super_dict = {} self.index = 0 def smart_way(self, img_src): if self.old_prices.queue.qsize() < 50: return False chrome_options = webdriver.ChromeOptions() chrome_options.add_argument("--headless") driver_library = webdriver.Chrome(chromedriver_path, options=chrome_options) driver_library.get("https://www.mtgowikiprice.com/") index = 1 print("Run smart algorithm for digit recognition") index = 0 for (card, botname, sell_price) in list(self.old_prices.queue.queue): self.ParseMtgolibrary(driver_library, card, botname, sell_price) index += 1 print("Parsing page number " + str(index)) temp_solution = {} for img_src in self.super_dict.keys(): candidate_key = max(self.super_dict[img_src], key=self.super_dict[img_src].get) temp_solution[img_src] = candidate_key if len(temp_solution.keys()) == 11 and len(temp_solution.values()) == len(set(temp_solution.values())): for key in temp_solution.keys(): self.prices_d[key] = temp_solution[key] self.super_dict.clear() pickle.dump(self.prices_d, open("obj/dict.pkl", "wb")) return True self.super_dict.clear() return False def get_symbol(self, img_src): if img_src in self.prices_d.keys(): return self.prices_d[img_src] elif self.smart_way(img_src): return self.prices_d[img_src] else: print(img_src) symbol = sys.stdin.readline() self.prices_d[img_src] = symbol pickle.dump(self.prices_d, open("obj/dict.pkl", "wb")) return symbol def get_price(self, e, botname, card): find_sell_price = False if len(e.find_elements_by_class_name("sell_price_round")): find_sell_price = True images = None if find_sell_price: images = e.find_element_by_class_name("sell_price_round") else: images = e.find_element_by_class_name("buy_price_round") index = 0 res = [] for image in images.find_elements_by_tag_name('img'): index += 1 img_src = image.get_attribute("src") symbol = self.get_symbol(img_src) res.append(symbol) # print("".join(res).replace('\n', '')) if find_sell_price: try: sell_price = float("".join(res).replace('\n', '')) except: sell_price = 100000 self.old_prices.add((card, botname, sell_price)) return sell_price else: try: buy_price = float("".join(res).replace('\n', '')) except: buy_price = -1 return buy_price
{"/candle_factory_5.py": ["/mtgolibrary.py", "/card.py"], "/goatbots_cardhoarder.py": ["/card.py", "/mtgolibrary.py"], "/number_parser.py": ["/card.py", "/queue_with_capacity.py"], "/mtgolibrary.py": ["/card.py", "/number_parser.py"], "/mtgo_controller.py": ["/queue_with_capacity.py"]}
50,258
1dayac/magic_bot
refs/heads/master
/mtgolibrary.py
from selenium import webdriver import sqlite3 from datetime import datetime from card import Card, Price import time import platform import re from number_parser import DigitsClassifier from trusted_bots import * if platform.system() == "Windows": chromedriver_path = r"C:\Users\IEUser\Desktop\magic_bot\chromedriver.exe" else: chromedriver_path = "/home/dmm2017/PycharmProjects/candle_factory/chromedriver" class MtgoLibraryParser(object): digit_clasiffier = DigitsClassifier() def __init__(self): self.main_url = "https://www.mtgowikiprice.com/" chrome_options = webdriver.ChromeOptions() prefs = {'profile.managed_default_content_settings.images': 1} chrome_options.add_experimental_option("prefs", prefs) #chrome_options.add_argument("start-maximized") #chrome_options.add_argument("--headless") self.driver = webdriver.Chrome(chromedriver_path, options= chrome_options) self.driver.get(self.main_url) self.card_count = 0 def restart(self): self.driver.quit() self.main_url = "https://www.mtgowikiprice.com/" chrome_options = webdriver.ChromeOptions() #chrome_options.add_argument("--headless") prefs = {'profile.managed_default_content_settings.images': 2} chrome_options.add_experimental_option("prefs", prefs) self.driver = webdriver.Chrome(chromedriver_path, options= chrome_options) self.driver.get(self.main_url) def ParseMtgolibraryFoil(self, card, parse_buyers = False): setname = card.set.upper() if setname.startswith("BOO") or setname.startswith("PRM"): return False setname, url = self.MtgoLibraryGoToCard(card) try: link = self.driver.find_element_by_link_text('View Foil') time.sleep(0.5) except: return False link.click() time.sleep(4) return self.ParseMtgolibraryInternal(card, url, parse_buyers) def get_price_from_image(self, e, botname, card): return MtgoLibraryParser.digit_clasiffier.get_price(e, botname, card) def MtgoLibraryGoToCard(self, card): setname = card.set.upper() input_element = self.driver.find_element_by_id("_cardskeyword") input_element.clear() input_element.send_keys(card.name + " " + setname) self.driver.find_elements_by_css_selector("button")[1].click() url = self.driver.current_url return setname, url def ParseMtgolibraryInternal(self, card, url, parse_buyers): setname = card.set elem = self.driver.find_elements_by_class_name("sell_row") buy_price = -1 sell_price = 10000 if len(elem) == 0: return None first = True number = 0 bot_name_sell = "" for e in elem: try: table_setname = e.find_elements_by_class_name("setname")[0].text if table_setname != setname: continue number = int(e.find_elements_by_class_name("sell_quantity")[0].text) bot_name_sell = e.find_elements_by_class_name("bot_name")[0].text if bot_name_sell == "": continue if bot_name_sell not in trusted_sell_bots: print("Best price - " + bot_name_sell + " " + str(self.get_price_from_image(e, bot_name_sell, card))) continue except: continue sell_price = self.get_price_from_image(e, bot_name_sell, card) break if not parse_buyers: return Price(url, 0, sell_price, "", bot_name_sell, number) elem2 = self.driver.find_elements_by_class_name("buy_row") bot_name_buy = "" number = 0 for e in elem2: try: bot_name_buy = e.find_elements_by_class_name("bot_name")[0].get_attribute('textContent').strip() number = int(e.find_elements_by_class_name("buy_quantity")[0].get_attribute('textContent').strip().replace("+", "")) if bot_name_buy == "": continue table_setname = e.find_elements_by_class_name("setname")[0].get_attribute('textContent').strip() if table_setname != setname: continue tickets = float(re.split("[+\-]", e.find_elements_by_class_name("tickets")[0].get_attribute('textContent').strip())[0]) except: continue buy_price = self.get_price_from_image(e, bot_name_buy, card) if tickets > buy_price: break return Price(url, buy_price, sell_price, bot_name_buy, bot_name_sell, number) def ParseMtgolibrary(self, card, parse_buyers = False): setname = card.set.upper() if setname.startswith("BOO") or setname.startswith("PRM"): return None setname, url = self.MtgoLibraryGoToCard(card) time.sleep(7) return self.ParseMtgolibraryInternal(card, url, parse_buyers) def get_price(self, card): self.card_count += 1 if self.card_count % 100 == 0: self.restart() if card.foil: return self.ParseMtgolibraryFoil(card) else: return self.ParseMtgolibrary(card)
{"/candle_factory_5.py": ["/mtgolibrary.py", "/card.py"], "/goatbots_cardhoarder.py": ["/card.py", "/mtgolibrary.py"], "/number_parser.py": ["/card.py", "/queue_with_capacity.py"], "/mtgolibrary.py": ["/card.py", "/number_parser.py"], "/mtgo_controller.py": ["/queue_with_capacity.py"]}
50,259
1dayac/magic_bot
refs/heads/master
/mtgo_controller.py
import sqlite3 from transitions import Machine, State from pywinauto.application import Application import pyautogui import sys import os from enum import Enum import math import traceback from trusted_bots import * pyautogui.FAILSAFE = False class TradeStatus(Enum): SUCCESS = 1 BOT_OFFLINE = 2 TRADE_REJECTED = 3 BIG_FAILURE = 4 NONE = 5 import win32api, win32con, win32process def setaffinity(): pid = win32api.GetCurrentProcessId() mask = 5 # core 7 handle = win32api.OpenProcess(win32con.PROCESS_ALL_ACCESS, True, pid) win32process.SetProcessAffinityMask(handle, mask) #setaffinity() class NoConfirmTradeException(Exception): pass set_abbr = {"ZNR" : "Zendikar Rising", "2XM" : "Double Masters", "AER" : "Aether Revolt", "C19" : "Commander (2019 Edition)", "C20" : "Ikoria Commander","AKH" : "Amonkhet", "EXP" : "Zendikar Expeditions", "PZ2" : "Treasure Chest", "MRD" : "Mirrodin", "KLD" : "Kaladesh", "EMN" : "Eldritch Moon", "ISD" : "Innistrad", "CMR" : "Commander Legends", "OGW" : "Oath of the Gatewatch", "DKA" : "Dark Ascension", "CMD" : "Commander (2011 Edition)", "ZEN" : "Zendikar", "XLN" : "Ixalan", "RIX" : "Rivals of Ixalan", "AVR" : "Avacyn Restored", "GTC" : "Gatecrash", "GRN" : "Guilds of Ravnica", "M21" : "Core 2021","BBD" : "Battlebond", "EX" : "Exodus", "MOR" : "Morningtide", "HOU" : "Hour of Devastation", "SOI" : "Shadows over Innistrad", "A25" : "Masters 25", "BFZ" : "Battle for Zendikar", "IKO" : "Ikoria: Lair of Behemoths", "THB" : "Theros Beyond Death", "JOU" : "Journey into Nyx", "IMA" : "Iconic Masters", "ORI" : "Magic Origins", "TPR" : "Tempest Remastered", "WL" : "Weatherlight","DTK" : "Dragons of Tarkir", "FRF" : "Fate Reforged", "M15" : "Magic 2015", "M20" : "Core Set 2020", "M14" : "Magic 2014", "M13" : "Magic 2013", "M12" : "Magic 2012", "M11" : "Magic 2011", "WAR" : "War of the Spark", "MMA" : "Modern Masters (2013 Edition)", "MM2" : "Modern Masters (2015 Edition)", "MM3" : "Modern Masters (2017 Edition)", "RTR" : "Return to Ravnica", "WWK" : "Worldwake", "ARB" : "Alara Reborn", "EVE" : "Eventide", "SHM" : "Shadowmoor", "10E" : "Tenth Edition", "9ED" : "Ninth Edition", "8ED" : "Eighth Edition", "7E" : "Seventh Edition", "LRW" : "Lorwyn", "ELD" : "Throne of Eldraine","PLC" : "Planar Chaos", "VMA" : "Vintage Masters", "TSP" : "Time Spiral", "CSP" : "Coldsnap", "DIS" : "Dissension", "AP" : "Apocalypse", "UMA" : "Ultimate Masters" , "GPT" : "Guild Pact", "VI" : "Visions", "DAR": "Dominaria", "SOK" : "Saviors of Kamigawa", "BOK" : "Betrayers of Kamigawa", "CHK" : "Champions of Kamigawa", "ST" : "Stronghold", "SLD" : "Secret Lair", "THB" : "Theros Beyond Death" ,"TE" : "Tempest", "MI" : "Mirage", "ONS" : "Onslaught", "JUD" : "Judgment", "OD" : "Odyssey", "PS" : "Planeshift", "NE" : "Nemesis", "DGM" : "Dragon's Maze", "MM" : "Mercadian Masques", "THS" : "Theros", "RNA" : "Ravnica Allegiance", "ROE" : "Rise of the Eldrazi", "UZ" : "Urza's Saga", "UL" : "Urza's Legacy", "M10" : "Magic 2010", "SCG" : "Scourge","UD" : "Urza's Destiny", "LGN" : "Legions", "CON" : "Conflux", "M19" : "Core Set 2019", "C14" : "Commander 2014", "ARB" : "Alara Reborn", "ALA" : "Shards of Alara", "DST" : "Darksteel", "FUT" : "Future Sight", "EMA" : "Eternal Masters", "MS2" : "Kaladesh Inventions", "MS3" : "Amonkhet Invocations", "RAV" : "Ravnica: City of Guilds", "5DN" : "Fifth Dawn", "MBS" : "Mirrodin Besieged", "SOM" : "Scars of Mirrodin", "NPH" : "New Phyrexia", "ME4" : "Masters Edition IV", "ME2" : "Masters Edition II", "PR": "Prophecy", "ME3" : "Masters Edition III", "MED" : "Masters Edition I", "IN" : "Invasion", "BNG" : "Born of the Gods", "KTK" : "Khans of Tarkir", "TOR" : "Torment", "TSB" : "Time Spiral Timeshifted", "MH1" : "Modern Horizons"} class Card: def __init__(self): self.name = "" self.set = "" self.prices = [] def __init__(self, name, set, prices): self.name = name self.set = set s self.prices = [prices] def AddPrice(self, price): self.prices.append(price) def __hash__(self): return hash(self.name + self.set) def MaxBuyPrice(self): prices = [price.buy_price for price in self.prices if price.buy_price > 0] try: return max(prices) except: return 0.0 def MinSellPrice(self): prices = [price.sell_price for price in self.prices if price.sell_price > 0] try: return min(prices) except: return 100000.0 def __str__(self): str1 = "" for price in self.prices: str1 += str(price) return self.name + "\t" + self.set + "\t" + str1 + "\t" + str(self.MaxBuyPrice() - self.MinSellPrice()) con = sqlite3.connect(sys.argv[1], isolation_level=None) cursor = con.cursor() def go_to_rectangle(rect, sleep = 0): pyautogui.moveTo((rect.left + rect.right)/2, (rect.top + rect.bottom)/2) time.sleep(sleep) def double_click_multiple(window, times): rect = window.rectangle() for i in range(times): double_click_rectangle(rect) def double_click_rectangle(rect, sleep = 0): pyautogui.click((rect.left + rect.right)/2, (rect.top + rect.bottom)/2, clicks=2, interval=0.1) time.sleep(sleep) def click_rectangle(rect, sleep = 0): pyautogui.click((rect.left + rect.right)/2, (rect.top + rect.bottom)/2) time.sleep(sleep) def right_click_rectangle(rect, sleep = 0): pyautogui.rightClick((rect.left + rect.right)/2, (rect.top + rect.bottom)/2) time.sleep(sleep) def click_collection(app): click_rectangle(app['Magic: The Gathering Online'].window(auto_id="CollectionButton").rectangle(), 5) def click_trade(app): click_rectangle(app['Magic: The Gathering Online'].window(auto_id="TradeButton", found_index = 0).rectangle(), 5) def click_ok_button(app): click_rectangle(app.top_window().window(auto_id="OkButton").rectangle(), 5) def close_chat(app): index = 0 while True: index += 1 try: if index == 100: break print("Try to close chat") click_rectangle(app['Magic: The Gathering Online'].window(auto_id="CloseButtom", found_index=0).rectangle()) time.sleep(1) break except: pass def get_tix_number(app, botname): import io import sys stringio = io.StringIO() previous_stdout = sys.stdout sys.stdout = stringio app.top_window().window(auto_id="ChatItemsControl").print_control_identifiers() sys.stdout = previous_stdout string = stringio.getvalue() num_of_tix = 0 if botname.startswith("Hot") or botname.startswith("CardBuy"): pos = string.rfind("Take") pos1 = string.find(" ", pos + 1) + 1 pos2 = string.find(" ", pos1) num_of_tix = int(string[pos1: pos2]) print("Taking " + str(num_of_tix) + " tix") else: pos = string.rfind("take") pos1 = string.find(" ", pos + 1) + 1 pos2 = string.find(" ", pos1) num_of_tix = math.floor(float(string[pos1: pos2])) print("Taking " + str(num_of_tix) + " tix") return num_of_tix states_my = [State(name = 'initial'), State(name = 'login', on_enter = ['login']), State(name = 'buy', on_enter = ['buy_card']), State(name = 'update_binder_after_buying', on_enter = ['update_binder_after_buy']), State(name = 'sell', on_enter = ['sell_card']), State(name = 'update_binder_after_selling', on_enter = ['update_binder_after_sell']), State(name = 'close', on_enter = ['close_mtgo'])] transitions = [ { 'trigger': 'go_to_login', 'source': 'initial', 'dest': 'login' }, {'trigger': 'go_to_buy', 'source': 'buy', 'dest': 'buy'}, {'trigger': 'go_to_sell', 'source': 'sell', 'dest': 'sell'}, { 'trigger': 'go_to_buy', 'source': 'login', 'dest': 'buy' }, {'trigger': 'go_to_buy', 'source': 'update_binder_after_selling', 'dest': 'buy'}, { 'trigger': 'go_to_update', 'source': 'buy', 'dest': 'update_binder_after_buying'} , { 'trigger': 'go_to_sell', 'source': 'update_binder_after_buying', 'dest': 'sell' }, {'trigger': 'go_to_update', 'source': 'sell', 'dest': 'update_binder_after_selling'}, {'trigger': 'go_to_buy', 'source': 'update_binder_after_selling', 'dest': 'buy'}, {'trigger': 'go_to_restart', 'source': 'update_binder_after_selling', 'dest': 'close'}, {'trigger': 'go_to_restart', 'source': 'sell', 'dest': 'close'}, {'trigger': 'go_to_restart', 'source': 'buy', 'dest': 'close'}, {'trigger': 'go_to_restart', 'source': 'update_binder_after_buying', 'dest': 'close'}, {'trigger': 'go_to_login', 'source': 'close', 'dest': 'login'} ] import subprocess, time start = time.time() from queue_with_capacity import QueueWithMaxCapacity class MTGO_bot(object): def __init__(self): self.last_trades = QueueWithMaxCapacity(10) try: self.app = Application(backend="uia").connect(path='MTGO.exe') self.db_record = "" self.trade_status = TradeStatus.NONE except: subprocess.Popen(['cmd.exe', '/c', r'C:\Users\IEUser\Desktop\mtgo.appref-ms']) time.sleep(5) self.app = Application(backend="uia").connect(path='MTGO.exe') self.db_record = "" self.trade_status = TradeStatus.NONE def close_mtgo(self): os.system("taskkill /f /im MTGO.exe") def login(self): print("Starting...") self.last_trades = QueueWithMaxCapacity(10) try: click_rectangle(self.app.top_window().child_window(auto_id = "CloseButton").rectangle()) except: pass try: self.app['Magic: The Gathering Online'].window(auto_id="UsernameTextBox").type_keys("VerzillaBot") self.app['Magic: The Gathering Online'].window(auto_id="PasswordBox").type_keys("Lastborn220") time.sleep(2.5) self.app['Magic: The Gathering Online'].window(auto_id="PasswordBox").type_keys("{ENTER}") pyautogui.press('enter') time.sleep(20) try: click_rectangle(self.app.top_window().child_window(auto_id="CloseButton").rectangle()) except: pass click_collection(self.app) time.sleep(10) click_trade(self.app) time.sleep(10) except: pass try: click_rectangle(self.app.top_window().child_window(auto_id = "CloseButton").rectangle()) except: pass try: click_collection(self.app) click_rectangle(self.app['Magic: The Gathering Online'].window(title="ABinder", found_index=0).rectangle()) click_collection(self.app) except: pass while True: try: rect = self.app['Magic: The Gathering Online'].child_window(auto_id="DeckPane").child_window(title_re="Item: CardSlot:", found_index=0).rectangle() right_click_rectangle(rect) click_rectangle(self.app['Magic: The Gathering Online'].child_window(title_re="Remove All", found_index=0).rectangle()) except: break try: click_rectangle(self.app['Magic: The Gathering Online'].window(title="Other Products", found_index=1).rectangle()) self.app['Magic: The Gathering Online'].window(auto_id="searchTextBox").type_keys("event{SPACE}tickets{ENTER}") right_click_rectangle( self.app['Magic: The Gathering Online'].child_window(title_re="Item: CardSlot: Event", found_index=0).rectangle()) except: self.close_mtgo() self.trade_status = TradeStatus.BIG_FAILURE try: click_rectangle(self.app['Magic: The Gathering Online'].child_window(title_re="Add All to", found_index=0).rectangle()) except: try: click_rectangle(self.app['Magic: The Gathering Online'].child_window(title_re="Add 1 to", found_index=0).rectangle()) except: pyautogui.moveRel(-10, 0) pyautogui.click() pass def switch_bot(self): if self.db_record[6] == "HotListBot3": self.db_record[6] = "HotListBot4" elif self.db_record[6] == "HotListBot4": self.db_record[6] = "HotListBot" elif self.db_record[6] == "HotListBot": self.db_record[6] = "HotListBot2" elif self.db_record[6] == "HotListBot2": self.db_record[6] = "HotListBot3" if self.db_record[6] == "GoatBots1": self.db_record[6] = "GoatBots2" elif self.db_record[6] == "GoatBots2": self.db_record[6] = "GoatBots3" elif self.db_record[6] == "GoatBots3": self.db_record[6] = "GoatBots1" self.app['Magic: The Gathering Online'].window(auto_id="searchTextBox").type_keys(self.db_record[6] + "{ENTER}") def all_bad_trades(self): if self.last_trades.queue.qsize() < 10: return False for item in self.last_trades.queue: if item == TradeStatus.SUCCESS: return False return True def click_bot_trade(self, botname, binder): index = 0 while True: try: index += 1 if index == 5: return False go_to_rectangle(self.app['Magic: The Gathering Online'].window(title=botname, found_index=1).rectangle()) click_rectangle(self.app['Magic: The Gathering Online'].window(title="Trade", found_index=1).rectangle()) time.sleep(1) click_rectangle(self.app.top_window().window(auto_id=binder, found_index=0).rectangle()) click_ok_button(self.app) return True except: pass def is_trade_cancelled(self): try: self.app.top_window().window(title="Trade Canceled", found_index=1).rectangle() click_rectangle(self.app.top_window().window(auto_id="OkButton", found_index=0).rectangle()) return True except: return False def is_trade_stalled(self): try: click_rectangle(self.app.top_window().window(title="Trade Request", found_index=0).window(title="Cancel", found_index = 0).rectangle()) return True except: return False def get_tix_number_buy(self): try: import io import sys stringio = io.StringIO() previous_stdout = sys.stdout sys.stdout = stringio self.app.top_window().window(auto_id="ChatItemsControl").print_control_identifiers() sys.stdout = previous_stdout string = stringio.getvalue() if self.db_record[5].startswith("Goat"): pos = string.rfind(self.db_record[1]) pos1 = string.find("(", pos) pos2 = string.find(")", pos) price = float(string[pos1 + 1: pos2]) print(price) return price else: if string.rfind(self.db_record[1]) == -1: stringio = io.StringIO() previous_stdout = sys.stdout sys.stdout = stringio self.app.top_window().window(auto_id="ChatItemsControl").print_control_identifiers() sys.stdout = previous_stdout string = stringio.getvalue() pos = string.rfind("YOU RECEIVE ") pos1 = string.find("(", pos) pos2 = string.find(")", pos) if " " in string[pos1:pos2]: pos1 = string.find(" ", pos1) # print(string[pos1 + 1: pos2]) price = float(string[pos1 + 1: pos2]) print(price) return price except: return None def check_inventory(self): click_collection(self.app) print(".") click_rectangle(self.app.top_window().window(title = "Cards", found_index=1).rectangle()) click_rectangle(self.app.top_window().window(auto_id="FilterCards-ResetFilterText").rectangle()) self.app.top_window().window(auto_id="searchTextBox").type_keys( self.db_record[1].replace(" ", "{SPACE}") + "{ENTER}") print("..") try: click_rectangle(self.app.top_window().window(auto_id="FilterCards-Option" + set_abbr[self.db_record[2]]).rectangle()) print("..") except: click_rectangle(self.app.top_window().window(auto_id="FilterCards-HeaderSet-Text").rectangle()) print(".....") time.sleep(0.5) try: click_rectangle(self.app.top_window().window(auto_id="FilterCards-Option" + set_abbr[self.db_record[2]]).rectangle()) except: pass try: time.sleep(0.5) click_rectangle(self.app.top_window().child_window(auto_id="CollectionLayoutView").child_window(title_re="Item: Card", found_index = 0)) print("....") return True except: print(".....") return False pass def buy_card(self): try: click_rectangle(self.app['ToastView'].child_window(auto_id = "CloseButton").rectangle()) except: pass try: self.trade_status = TradeStatus.NONE print("Go to buy card...") cursor = con.cursor() command = "SELECT * FROM records ORDER BY RANDOM() LIMIT 1;" records = cursor.execute(command).fetchall() while len(records) == 0: time.sleep(10) records = cursor.execute(command).fetchall() self.db_record = list(records[0]) while self.db_record[5] not in trusted_sell_bots or (self.db_record[6] not in trusted_buy_bots and self.db_record[6] not in mtgolibrary_buy_bots): command = "DELETE FROM records WHERE Id = ?;" cursor.execute(command, [self.db_record[0]]).fetchall() command = "SELECT * FROM records ORDER BY RANDOM() LIMIT 1;" records = cursor.execute(command).fetchall() while len(records) == 0: time.sleep(10) records = cursor.execute(command).fetchall() self.db_record = list(records[0]) appendix = "(foil)" if (int(self.db_record[9]) == 1) else "(regular)" print("Buying " + str(self.db_record[8]) + "x"+ self.db_record[1] + "(" + self.db_record[2] + ") from " + self.db_record[5] + " " + appendix) if self.db_record[5] == "Applegrove": self.db_record[5] = "AppleGrove" if self.db_record[5] == "VRTStorebot3": self.db_record[5] = "VRTStoreBot3" if self.db_record[5] == "VRTStorebot2": self.db_record[5] = "VRTStoreBot2" if self.db_record[5] == "VRTStorebot": self.db_record[5] = "VRTStoreBot" if self.db_record[5] == "VRTSToreBot": self.db_record[5] = "VRTStoreBot" if self.db_record[5] == "VRTSToreBot2": self.db_record[5] = "VRTStoreBot2" if self.db_record[5] == "VRTSToreBot3": self.db_record[5] = "VRTStoreBot3" if self.db_record[5] == "Manatraders_booster1": self.db_record[5] = "ManaTraders_Booster1" if self.db_record[5] == "Manatraders_seller1": self.db_record[5] = "ManaTraders_Seller1" if self.db_record[5] == "Manatraders_seller2": self.db_record[5] = "ManaTraders_Seller2" if self.db_record[5] == "Manatraders_seller3": self.db_record[5] = "ManaTraders_Seller3" if self.db_record[5] == "Manatraders_seller4": self.db_record[5] = "ManaTraders_Seller4" if self.db_record[5] == "Manatraders_seller5": self.db_record[5] = "ManaTraders_Seller5" if self.db_record[5] == "Vintage-Cardbot2": self.db_record[5] = "Vintage-cardbot2" #if self.check_inventory(): # command = "DELETE FROM records WHERE Id = ?;" # cursor.execute(command, [self.db_record[0]]).fetchall() # return time.sleep(1) try: click_trade(self.app) self.app['Magic: The Gathering Online'].window(auto_id="searchTextBox").type_keys(self.db_record[5] + "{ENTER}") except: return if not self.click_bot_trade(self.db_record[5], "ABinder"): print("Bot is offline") self.is_trade_cancelled() self.last_trades.add(TradeStatus.BOT_OFFLINE) return time.sleep(5) number_of_cancelled_trades = 0 while self.is_trade_cancelled(): number_of_cancelled_trades += 1 if number_of_cancelled_trades == 5: self.last_trades.add(TradeStatus.BOT_OFFLINE) self.trade_status = TradeStatus.BOT_OFFLINE return self.click_bot_trade(self.db_record[5], "ABinder") time.sleep(3) if self.is_trade_stalled(): return print(2) try: time.sleep(2) click_rectangle(self.app.top_window().window(auto_id="FilterCards-ResetFilterText").rectangle()) time.sleep(0.1) if int(self.db_record[9]) == 1: self.db_record[8] = str(min(int(self.db_record[8]), 2)) click_rectangle(self.app.top_window().window(title="Versions", found_index = 0).rectangle()) click_rectangle(self.app.top_window().window(title="Show Foils", found_index = 0).rectangle()) time.sleep(0.1) self.app.top_window().window(auto_id="searchTextBox").type_keys(self.db_record[1].replace(" ", "{SPACE}") + "{ENTER}") except: print("Unexpected error:", sys.exc_info()[0]) traceback.print_exc(file=sys.stdout) return print(3) try: click_rectangle(self.app.top_window().window(auto_id="FilterCards-HeaderSet-Text").rectangle()) click_rectangle(self.app.top_window().window(auto_id="FilterCards-Option" + set_abbr[self.db_record[2]]).rectangle()) time.sleep(1.5) print(4) double_click_multiple(self.app.top_window().child_window(title_re="Item: CardSlot: " + self.db_record[1].split(",")[0], found_index = 0), int(self.db_record[8])) except: print("Unexpected error:", sys.exc_info()[0]) traceback.print_exc(file=sys.stdout) command = "DELETE FROM records WHERE Id = ?;" cursor.execute(command, [self.db_record[0]]).fetchall() click_rectangle(self.app.top_window().window(title="Cancel Trade", found_index=1).rectangle()) close_chat(self.app) return print(5) time.sleep(8) price = self.get_tix_number_buy() if price is not None and price > float(self.db_record[3]): command = "DELETE FROM records WHERE Id = ?;" cursor.execute(command, [self.db_record[0]]).fetchall() click_rectangle(self.app.top_window().window(title="Cancel Trade", found_index=1).rectangle()) close_chat(self.app) return time.sleep(2) click_rectangle(self.app.top_window().window(title="Submit", found_index=1).rectangle()) time.sleep(5) try: click_rectangle(self.app.top_window().window(title="Submit", found_index=1).rectangle()) except: pass print(6) time.sleep(1) index = 0 while True: try: index += 1 click_rectangle(self.app.top_window().window(title="Confirm Trade", found_index=1).rectangle()) break except: time.sleep(1) if index >= 10: self.trade_status = TradeStatus.BIG_FAILURE return pass print(4) close_chat(self.app) print(5) index = 0 while True: try: click_rectangle(self.app['ToastView'].child_window(auto_id="CloseButton").rectangle()) except: pass try: index += 1 time.sleep(2) click_rectangle(self.app.top_window().window(title="Added to your Collection:", found_index = 0).window(auto_id="TitleBarCloseButton").rectangle()) break except: try: click_rectangle(self.app.top_window().window(auto_id="OkButton", found_index=0).rectangle()) break except: if index >= 20: self.trade_status = TradeStatus.BIG_FAILURE return pass command = "DELETE FROM records WHERE Id = ?;" cursor.execute(command, [self.db_record[0]]).fetchall() self.trade_status = TradeStatus.SUCCESS self.last_trades.add(TradeStatus.SUCCESS) except: print("Unexpected error:", sys.exc_info()[0]) traceback.print_exc(file=sys.stdout) command = "DELETE FROM records WHERE Id = ?;" cursor.execute(command, [self.db_record[0]]).fetchall() self.trade_status = TradeStatus.BIG_FAILURE def sell_card(self): try: click_rectangle(self.app.top_window().child_window(auto_id = "CloseButton").rectangle()) except: pass try: self.trade_status = TradeStatus.NONE print("Go to sell card...") print("Selling " + self.db_record[0] + " to " + self.db_record[6]) try: click_trade(self.app) self.app.top_window().window(auto_id="searchTextBox").type_keys(self.db_record[6] + "{ENTER}") except: return while not self.click_bot_trade(self.db_record[6], "Full Trade List") or self.is_trade_cancelled() or self.is_trade_stalled(): self.switch_bot() time.sleep(6) window_sell_name = "Trade: " + self.db_record[6] if self.db_record[6] in mtgolibrary_buy_bots: try: self.app.top_window().window(auto_id="ChatSendEditBox").type_keys("sell{ENTER}") except: pass self.app.top_window().window(auto_id="ChatSendEditBox").type_keys("{ENTER}") try: num_of_tix = get_tix_number(self.app, self.db_record[6]) except: raise Exception try: if num_of_tix != 0: click_rectangle(self.app[window_sell_name].window(title="Other Products", found_index=1).rectangle()) if self.db_record[6].startswith("Goat"): self.app[window_sell_name].window(auto_id="searchTextBox").type_keys("event{SPACE}tickets{ENTER}") double_click_multiple(self.app[window_sell_name].child_window(title_re="Item: CardSlot: Event", found_index=0), num_of_tix) except: pass click_rectangle(self.app[window_sell_name].window(title="Submit", found_index=1).rectangle()) time.sleep(5) try: click_rectangle(self.app[window_sell_name].window(title="Submit", found_index=1).rectangle()) except: pass time.sleep(3) index = 0 while True: try: index += 1 click_rectangle(self.app[window_sell_name].window(title="Confirm Trade", found_index=1).rectangle()) time.sleep(1) break except: if index == 10: raise NoConfirmTradeException() pass close_chat(self.app) index = 0 while True: try: index += 1 print("Trying to close window with stuff") click_rectangle(self.app.top_window().window(title="Added to your Collection:", found_index = 0).window(auto_id="TitleBarCloseButton").rectangle()) time.sleep(1) break except: if index == 20: self.trade_status = TradeStatus.BIG_FAILURE return try: print("Trying to close window without stuff") click_rectangle(self.app.top_window().window(auto_id="OkButton", found_index=0).rectangle()) break except: pass self.trade_status = TradeStatus.SUCCESS except NoConfirmTradeException: try: click_rectangle(self.app.top_window().window(title="Cancel Trade", found_index=0).rectangle()) close_chat(self.app) except: print(sys.exc_info()[0]) print(sys.exc_info()[1]) traceback.print_exc(file=sys.stdout) pass except: print("Unexpected error:", sys.exc_info()[0]) print("Unexpected error:", sys.exc_info()[1]) traceback.print_exc(file=sys.stdout) self.trade_status = TradeStatus.BIG_FAILURE def update_binder_after_buy(self): try: click_rectangle(self.app.top_window().child_window(auto_id = "CloseButton").rectangle()) except: pass self.trade_status = TradeStatus.NONE print("Go to update values...") return while True: try: click_collection(self.app) break except: time.sleep(1) time.sleep(1) click_rectangle(self.app.top_window().window(title="Cards", found_index=1).rectangle()) time.sleep(1) self.app.top_window().window(auto_id="searchTextBox").type_keys(self.db_record[1].replace(" ", "{SPACE}") + "{ENTER}") time.sleep(1) try: double_click_multiple(self.app.top_window().child_window(auto_id="CollectionLayoutView").child_window(title_re="Item: Card", found_index = 0), int(self.db_record[8])) except: pass def update_binder_after_sell(self): try: click_rectangle(self.app.top_window().child_window(auto_id = "CloseButton").rectangle()) except: pass self.trade_status = TradeStatus.NONE print("Go to update values...") click_collection(self.app) time.sleep(1) click_rectangle(self.app.top_window().window(title="Other Products", found_index=1).rectangle()) self.app.top_window().window(auto_id="searchTextBox").type_keys("event{SPACE}tickets{ENTER}") right_click_rectangle(self.app.top_window().child_window(auto_id="CollectionLayoutView").child_window(title_re="Item: CardSlot: Event", found_index=0).rectangle()) try: click_rectangle(self.app.top_window().child_window(title_re="Add All to", found_index=0).rectangle()) except: try: click_rectangle(self.app.top_window().child_window(title_re="Add 1 to", found_index=0).rectangle()) except: pyautogui.moveRel(-10, 0) pyautogui.click() pass def close(self): os.system("taskkill /f /im MTGO.exe") while True: try: my_bot = MTGO_bot() my_MTGO_bot_Machine = Machine(model=my_bot, states=states_my, transitions=transitions, initial='initial') print(my_bot.state) my_bot.go_to_login() while True: while True: if my_bot.trade_status == TradeStatus.BIG_FAILURE or my_bot.all_bad_trades(): my_bot.go_to_restart() my_bot.__init__() my_bot.go_to_login() if my_bot.trade_status == TradeStatus.SUCCESS: break if my_bot.trade_status == TradeStatus.NONE or my_bot.trade_status == TradeStatus.BOT_OFFLINE: my_bot.go_to_buy() my_bot.go_to_update() if my_bot.trade_status == TradeStatus.NONE or my_bot.trade_status == TradeStatus.BOT_OFFLINE: my_bot.go_to_sell() if my_bot.trade_status == TradeStatus.BIG_FAILURE: my_bot.go_to_restart() my_bot.__init__() my_bot.go_to_login() continue my_bot.go_to_update() except: pass #app = Application(backend="uia").start("notepad.exe")
{"/candle_factory_5.py": ["/mtgolibrary.py", "/card.py"], "/goatbots_cardhoarder.py": ["/card.py", "/mtgolibrary.py"], "/number_parser.py": ["/card.py", "/queue_with_capacity.py"], "/mtgolibrary.py": ["/card.py", "/number_parser.py"], "/mtgo_controller.py": ["/queue_with_capacity.py"]}
50,260
1dayac/magic_bot
refs/heads/master
/card.py
class Price: def __init__(self): self.url = "" self.buy_price = 0.0 self.sell_price = 0.0 self.bot_name_buy = "" self.bot_name_sell = "" self.number = 0 def __init__(self, url, buy_price, sell_price, bot_name_buy, bot_name_sell, number): self.url = url self.buy_price = buy_price self.sell_price = sell_price self.bot_name_buy = bot_name_buy self.bot_name_sell = bot_name_sell self.number = number def __str__(self): return str(self.buy_price) + "\t" + self.bot_name_buy + "\t" + str(self.sell_price) + "\t" + self.bot_name_sell +"\t" class Card: def __init__(self): self.name = "" self.set = "" self.foil = False self.prices = [] def __init__(self, name, set, prices, foil): self.name = name self.set = set self.prices = [prices] self.foil = foil def AddPrice(self, price): self.prices.append(price) def __hash__(self): return hash(self.name + self.set) def MaxBuyPrice(self): prices = [price.buy_price for price in self.prices if price.buy_price > 0] try: return max(prices) except: return 0.0 def MinSellPrice(self): prices = [price.sell_price for price in self.prices if price.sell_price > 0] try: return min(prices) except: return 100000.0 def __str__(self): str1 = "" for price in self.prices: str1 += str(price) return self.name + "\t" + self.set + "\t" + str(self.foil) + "\t" + str1 + "\t" + str(self.MaxBuyPrice() - self.MinSellPrice()) def BestSellPrice(self): best_price = None for price in self.prices: if price == None: continue if best_price == None: best_price = price elif price.sell_price < best_price.sell_price: best_price = price return best_price def BestBuyPrice(self): best_price = None for price in self.prices: if price == None: continue if best_price == None: best_price = price elif price.buy_price > best_price.buy_price: best_price = price return best_price
{"/candle_factory_5.py": ["/mtgolibrary.py", "/card.py"], "/goatbots_cardhoarder.py": ["/card.py", "/mtgolibrary.py"], "/number_parser.py": ["/card.py", "/queue_with_capacity.py"], "/mtgolibrary.py": ["/card.py", "/number_parser.py"], "/mtgo_controller.py": ["/queue_with_capacity.py"]}
50,266
Jcfurtado/PythonProjects
refs/heads/master
/geckoshouse/main.py
from time import sleep from .gled import clear, cock_shut, cockcrow, overnight def morning(): cockcrow() print("\n Cockcrow!! ON!") sleep(300) clear() print("\n Cockcrow!!! OFF!") def evening(): from time import sleep cock_shut() print("\n Cockshut, ON!") sleep(300) clear() print("\n Cockshut, OFF!") def light_ctrl(): if KEY == 'cockcrow': # wawr LED morning() elif KEY == 'day': # day lights day_ctrl() elif KEY == 'cockshut': # cockshut LED cock_shut() elif KEY == 'overnight': # night lights night_ctrl() else: print('error... esp_light_controller') ''' def day_ctrl(): # control day lamps if lpd_hot_day == 'cold_day': # day heat lamp lpd_day.on() return print("\n Heat lamp, ON!") elif lpd_hot_day == 'hot_day': lpd_day.off() return print("\n Heat lamp, OFF!") elif uv_on == 'rooster': # UV light lpd_uv.on() return print("\n UV, ON!") elif uv_off == 'rooster': lpd_uv.off() return print("\n UV, OFF!") elif uva_on == 'rooster': # UVA-B light lpd_uva.on() return print("\n UVA-B, ON!") elif uva_off == 'rooster': lpd_uva.off() return print("\n UVA-B, OFF!") else: print('error... esp_day_controller') def night_ctrl(): # control overnight lights if hot_night == 'cold_night': # heat lamp night lpd_night.on() return print("\n Overnight heat lamp, ON!") elif hot_night == 'hot_night': lpd_night.off() print("\n Overnight heat lamp, OFF!") overnight() # overnight LED print("\n Overnight LED light, ON!") else: clear() print("\n Overnight LED light, OFF!") '''
{"/geckoshouse/main.py": ["/geckoshouse/gled.py"], "/geckoshouse/gpublish.py": ["/geckoshouse/gdht.py", "/geckoshouse/glamp.py"]}
50,267
Jcfurtado/PythonProjects
refs/heads/master
/geckoshouse/gjson.py
import ujson def w_dht(area, temp, humid): dht_json = { "dht": { "area": area, "temperature": temp, "humidity": humid }, } with open("gdht.json", "w") as write_file: ujson.dump(dht_json, write_file) def r_dht(): with open("gdht.json", "r") as read_file: measurement = ujson.load(read_file) return measurement def w_lamp(uv, uva, rgb, day, night): lamp_json = { "lamps": { "uv": uv, "uva": uva, "led": rgb, "h_day": day, "h_night": night } } with open("glamp.json", "w") as write_file: ujson.dump(lamp_json, write_file) def r_lamp(): with open("glamp.json", "r") as read_file: lamps = ujson.load(read_file) return lamps
{"/geckoshouse/main.py": ["/geckoshouse/gled.py"], "/geckoshouse/gpublish.py": ["/geckoshouse/gdht.py", "/geckoshouse/glamp.py"]}
50,268
Jcfurtado/PythonProjects
refs/heads/master
/geckoshouse/gpublish.py
from time import sleep from umqtt.simple import MQTTClient from .gdht import d_one, d_two from .glamp import state_lamp def data_pub(server="192.168.0.102"): c = MQTTClient("umqtt_client", server) c.connect() # hot side temp1, humid1 = d_one() # cold side temp2, humid2 = d_two() # Measurements all_measurements = [temp1, humid1, temp2, humid2] while True: for measurement in all_measurements: c.publish(b"tempone", "Temp: {}°C\nHumid: {}%\n".format(*all_measurements).encode("utf-8")) c.publish(b"tempone", "Light: {}".format(list(state_lamp())).encode("utf-8")) sleep(2) c.disconnect()
{"/geckoshouse/main.py": ["/geckoshouse/gled.py"], "/geckoshouse/gpublish.py": ["/geckoshouse/gdht.py", "/geckoshouse/glamp.py"]}
50,269
Jcfurtado/PythonProjects
refs/heads/master
/geckoshouse/glamp.py
from machine import Pin def state_lamp(): lpd_uv = Pin(5, Pin.OUT, Pin.PULL_UP) # (D1) UV light lpd_uva = Pin(5, Pin.OUT, Pin.PULL_UP) # (D1) UVA-B light led_rgb = Pin(5, Pin.OUT, Pin.PULL_UP) # (D1) RGB LED overnight lpd_day = Pin(5, Pin.OUT, Pin.PULL_UP) # (D1) day heat lamp lpd_night = Pin(5, Pin.OUT, Pin.PULL_UP) # (D1) night heat lamp uv = lpd_uv.value() uva = lpd_uva.value() rgb = led_rgb.value() day = lpd_day.value() night = lpd_night.value() return uv, uva, rgb, day, night
{"/geckoshouse/main.py": ["/geckoshouse/gled.py"], "/geckoshouse/gpublish.py": ["/geckoshouse/gdht.py", "/geckoshouse/glamp.py"]}
50,270
Jcfurtado/PythonProjects
refs/heads/master
/geckoshouse/gled.py
from machine import Pin from neopixel import NeoPixel # --- rgb led --- # # the number 12 is number of led rgb_led = NeoPixel(Pin(5), 12) # (D1) overnight RGB LED lpd_uv = Pin(5, Pin.OUT, Pin.PULL_UP) # (D1) UV light lpd_uva = Pin(5, Pin.OUT, Pin.PULL_UP) # (D1) UVA-B light lpd_day = Pin(5, Pin.OUT, Pin.PULL_UP) # (D1) day heat lamp lpd_night = Pin(5, Pin.OUT, Pin.PULL_UP) # (D1) night heat lamp def clear(): for i in range(12): rgb_led[i] = (0, 0, 0) rgb_led.write() def cockcrow(): sunrise = 64, 35, 0 for i in range(4 * 12): for j in range(12): rgb_led[j] = sunrise rgb_led.write() def cock_shut(): sunset = 128, 88, 0 for i in range(4 * 12): for j in range(12): rgb_led[j] = sunset rgb_led.write() def overnight(): night = 15, 0, 45 for i in range(4 * 12): for j in range(12): rgb_led[j] = night rgb_led.write()
{"/geckoshouse/main.py": ["/geckoshouse/gled.py"], "/geckoshouse/gpublish.py": ["/geckoshouse/gdht.py", "/geckoshouse/glamp.py"]}
50,271
Jcfurtado/PythonProjects
refs/heads/master
/geckoshouse/gdht.py
from time import sleep import dht from machine import Pin def d_one(): # ( Sensor DHT hot side) sensor = dht.DHT22(Pin(4, Pin.IN, Pin.PULL_UP)) sensor.measure() temp = sensor.temperature() # eg. 30.6 (°C) humid = sensor.humidity() # eg. 25.3 (% RH) area = "hot" return area, temp, humid def d_two(): # ( Sensor DHT cold side) sensor = dht.DHT22(Pin(4, Pin.IN, Pin.PULL_UP)) sensor.measure() temp = sensor.temperature() # eg. 30.6 (°C) humid = sensor.humidity() # eg. 25.3 (% RH) area = "cold" return area, temp, humid def d_main(): while True: d_one() d_two() sleep(2) if __name__ == '__main__': d_main()
{"/geckoshouse/main.py": ["/geckoshouse/gled.py"], "/geckoshouse/gpublish.py": ["/geckoshouse/gdht.py", "/geckoshouse/glamp.py"]}
50,272
Jcfurtado/PythonProjects
refs/heads/master
/geckoshouse/gconnect.py
import network import ujson def key(): with open("key.json", "r") as read_file: k = ujson.load(read_file) return k.ssid, k.pswd def do_connect(): lan = network.WLAN(network.STA_IF) lan.active(True) if not lan.isconnected(): print('connecting to network...') lan.connect('AG45', 'fortyfive') while not lan.isconnected(): pass print('network config:', lan.ifconfig()) print('connected is ', lan.isconnected()) if __name__ == '__main__': do_connect()
{"/geckoshouse/main.py": ["/geckoshouse/gled.py"], "/geckoshouse/gpublish.py": ["/geckoshouse/gdht.py", "/geckoshouse/glamp.py"]}
50,273
Jcfurtado/PythonProjects
refs/heads/master
/geckoshouse/gfirebase.py
import .firebase.ufirebase as firebase URL = 'lucid-lychee' # see note on URLs at the bottom of documentation print firebase.get(URL) # this is an empty Firebase None firebase.put(URL, 'tell me everything') # can take a string print firebase.get(URL) tell me everything firebase.put(URL, {'lucidity': 9001}) # or a dictionary print firebase.get(URL) {u'lucidity': 9001} firebase.put(URL, {'color': 'red'}) # replaces old value print firebase.get(URL) {u'color': u'red'} print firebase.get(URL + '/color') red
{"/geckoshouse/main.py": ["/geckoshouse/gled.py"], "/geckoshouse/gpublish.py": ["/geckoshouse/gdht.py", "/geckoshouse/glamp.py"]}
50,275
andycasey/velociraptor
refs/heads/master
/scripts/npm_collect_the_battery_stars_comparison.py
import pickle import yaml import numpy as np from astropy.io import fits from astropy.table import Table with open("data/the-battery-stars-comparison.indices.pkl", "rb") as fp: comp_indices = pickle.load(fp) with open("results/the-battery-stars-comparison.astrometric_unit_weight_error.pkl", "rb") as fp: astrometric_results = pickle.load(fp) with open("results/the-battery-stars-comparison.rv_single_epoch_scatter.pkl", "rb") as fp: rv_results = pickle.load(fp) with open("the-battery-stars.astrometry.yaml", "r") as fp: config = yaml.load(fp) # Load in the data. data = fits.open(config["data_path"])[1].data subset = Table(data[comp_indices]) keys = ("theta", "mu_single", "sigma_single", "mu_multi", "sigma_multi") for i, key in enumerate(keys): subset["rv_{}".format(key)] = rv_results[comp_indices, i] subset["astrometric_{}".format(key)] = astrometric_results[comp_indices, i] finite = np.isfinite(subset["astrometric_theta"] * subset["rv_theta"]) subset = subset[finite] print(len(subset)) subset.write("results/the-battery-stars-comparison.results.fits", overwrite=True)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,276
andycasey/velociraptor
refs/heads/master
/article/figures/make.py
""" Produce figures for the Velociraptor project. Science verification -------------------- [X] 1. PDF of known single stars (Soubiran) and binary stars (SB9). [X] 2. SB9 mass versus period, coloured by tau_single_rv. [X] 3. SB9 mass versus period, coloured by tau_single_ast. [X] 4. SB9 RV semi-amplitude vs our estimated RV semi-amplitude, w.r.t alpha, coloured by period. [X] 5. Like #4, but coloured by other things: eccentricity, etc. Astrophysics ------------ 1. Main-sequence (bp rp vs absolute g-mag), coloured by median single star probability for each panel: RV, astrometry, joint. 2. """ import os from astropy.io import fits from astropy.table import Table import matplotlib.pyplot as plt import numpy as np from matplotlib.ticker import MaxNLocator from mpl_utils import mpl_style plt.style.use(mpl_style) BASE_PATH = "../../" try: velociraptor except NameError: velociraptor = fits.open("../../results/die-hard-subset.fits")[1].data else: print("Warning: using pre-loaded velociraptor catalog") cmap_binary = matplotlib.cm.coolwarm def cross_match(A_source_ids, B_source_ids): A = np.array(A_source_ids, dtype=np.long) B = np.array(B_source_ids, dtype=np.long) ai = np.where(np.in1d(A, B))[0] bi = np.where(np.in1d(B, A))[0] return (ai[np.argsort(A[ai])], bi[np.argsort(B[bi])]) def _literature_multiplicity_pdfs(source_ids, pdfs, soubiran_catalog, sb9_catalog, pdf_idx): # Cross-match our source ids with literature. vl_sb9_ids, sb9_ids = cross_match(source_ids, sb9_catalog["source_id"]) vl_soubiran_ids, soubiran_ids = cross_match(source_ids, soubiran_catalog["source_id"]) # Build up PDF density. pdf_soubiran = pdfs[pdf_idx, vl_soubiran_ids] pdf_sb9 = pdfs[pdf_idx, vl_sb9_ids] return (pdf_soubiran, pdf_sb9) def plot_literature_multiplicity_pdf(source_ids, pdfs, soubiran_catalog, sb9_catalog, pdf_idx=0, colors=None, **kwargs): pdf_soubiran, pdf_sb9 = _literature_multiplicity_pdfs(source_ids, pdfs, soubiran_catalog, sb9_catalog, pdf_idx=pdf_idx) bins = np.linspace(0, 1, 21) if colors is None: colors = [cmap_binary(255), cmap_binary(0)] fig, ax = plt.subplots(figsize=(4.5, 4.5)) ax.hist([pdf_soubiran, pdf_sb9][::-1], bins=bins, label=[ r"$\textrm{Single stars (Soubiran et al. 2009)}$", r"$\textrm{Binary stars (Pourbaix et al. 2004)}$" ][::-1], histtype="barstacked", color=colors[::-1]) plt.legend(frameon=False) ax.set_xlabel(r"$p(\textrm{single}|\sigma_{vr})$") ax.set_yticks([]) ax.xaxis.set_major_locator(MaxNLocator(6)) fig.tight_layout() return fig def plot_literature_multiplicity_classifications(source_ids, pdfs, soubiran_catalog, sb9_catalog, colors=None, pdf_idx=0, **kwargs): pdf_soubiran, pdf_sb9 = _literature_multiplicity_pdfs(source_ids, pdfs, soubiran_catalog, sb9_catalog, pdf_idx=pdf_idx) soubiran_class = np.round(np.mean(pdf_soubiran, axis=1)).astype(int) sb9_class = np.round(np.mean(pdf_sb9, axis=1)).astype(int) fig, ax = plt.subplots(figsize=(5, 2.5)) S_single = np.sum(soubiran_class == 1) S_binary = np.sum(soubiran_class == 0) SB9_single = np.sum(sb9_class == 1) SB9_binary = np.sum(sb9_class == 0) if colors is None: colors = [cmap_binary(255), cmap_binary(0)] ax.barh(0, S_single, facecolor=colors[0], label=r"$\textrm{Classified as single}$") ax.barh(0, S_binary, facecolor=colors[1], left=S_single) ax.barh(1, SB9_binary, facecolor=colors[1], label=r"$\textrm{Classified as binary}$") ax.barh(1, SB9_single, facecolor=colors[0], left=SB9_binary) plt.legend(frameon=False) ax.set_yticks([0, 1]) ax.tick_params(axis="y", pad=60) ax.set_yticklabels([ r"$\textrm{Known single stars}$" + "\n" + \ r"$\textrm{(Soubiran et al. 2009)}$", r"$\textrm{Known binary stars}$" + "\n" + \ r"$\textrm{(Pourbaix et al. 2004)}$" ], horizontalalignment="center") ax.xaxis.set_major_locator(MaxNLocator(6)) ax.set_xlabel(r"$\textrm{Classifications}$") fig.tight_layout() return fig def literature_binary_properties(velociraptor, sb9_catalog, x_label="Per", y_label="K1", z_label="joint_confidence", errors=True, log_axes=True, latex_labels=None): vl_sb9_ids, sb9_ids = cross_match(velociraptor["source_id"], sb9_catalog["source_id"]) latex_labels_ = dict(Per=r"$\textrm{period}$ $\textrm{/\,days}$", K1=r"$\textrm{radial velocity semi-amplitude}$ $\textrm{/\,km\,s}^{-1}$", e=r"$\textrm{eccentricity}$", rv_confidence=r"$\textrm{Single star confidence (rv only)}$", ast_confidence=r"$\textrm{Single star confidence (ast only)}$", joint_confidence=r"$\textrm{Single star confidence (joint)}$") latex_labels_.update(latex_labels or dict()) x = sb9_catalog[x_label][sb9_ids] y = sb9_catalog[y_label][sb9_ids] if z_label in velociraptor.dtype.names: z = velociraptor[z_label][vl_sb9_ids] else: z = sb9_catalog[z_label][sb_9_ids] fig, ax = plt.subplots(figsize=(6.18, 5.0)) kwds = dict(s=15, cmap=cmap_binary) scat = ax.scatter(x, y, c=z, **kwds) if errors: ax.errorbar(x, y, xerr=sb9_catalog[f"e_{x_label}"][sb9_ids], yerr=sb9_catalog[f"e_{y_label}"][sb9_ids], fmt=None, zorder=-1, c="#666666", linewidth=0.5) if log_axes is True: ax.loglog() elif isinstance(log_axes, (tuple, list)): if log_axes[0]: ax.semilogx() if log_axes[1]: ax.semilogy() ax.set_xlabel(latex_labels_.get(x_label, x_label)) ax.set_ylabel(latex_labels_.get(y_label, y_label)) cbar = plt.colorbar(scat) cbar.set_label(latex_labels_.get(z_label, z_label)) fig.tight_layout() return fig def plot_semi_amplitude_wrt_literature(velociraptor, sb9_catalog, scale=True, loglog=True, z_log=False, z_label="joint_confidence"): vl_sb9_ids, sb9_ids = cross_match(velociraptor["source_id"], sb9_catalog["source_id"]) K_est = velociraptor["rv_excess"][vl_sb9_ids] K_err = velociraptor["rv_excess_var"][vl_sb9_ids]**0.5 if scale: scalar = (1.0 / (sb9_catalog["Per"] * (1 - sb9_catalog["e"]**2)**0.5))[sb9_ids] else: scalar = 1.0 y = K_est * scalar yerr = K_err * scalar xp = np.array([ sb9_catalog["K1"][sb9_ids] * scalar, sb9_catalog["K2"][sb9_ids] * scalar ]) xperr = np.array([ sb9_catalog["e_K1"][sb9_ids] * scalar, sb9_catalog["e_K2"][sb9_ids] * scalar ]) """ diff = np.abs(xp - y) diff[~np.isfinite(diff)] = np.inf idx = np.nanargmin(diff, axis=0) x = xp[idx, np.arange(y.size)] xerr = xperr[idx, np.arange(y.size)] """ x, xerr = xp[0], xperr[0] if z_label in velociraptor.dtype.names: c = velociraptor[z_label][vl_sb9_ids] else: c = sb9_catalog[z_label][sb9_ids] if z_log: c = np.log10(c) fig, ax = plt.subplots(1, 1, figsize=(8, 7)) scat = ax.scatter(x, y, c=c, cmap=cmap_binary, s=15, rasterized=True) ax.errorbar(x, y, xerr=xerr, yerr=yerr, fmt="none", ecolor="#CCCCCC", zorder=-1, linewidth=1, capsize=0) if loglog: ax.loglog() limits = np.array([ax.get_xlim(), ax.get_ylim()]) limits = (np.min(limits), np.max(limits)) kwds = dict(c="#666666", linestyle=":", zorder=-1, linewidth=0.5) ax.plot(limits, limits, **kwds) ax.set_xlim(limits) ax.set_ylim(limits) latex_labels_ = dict(Per=r"$\textrm{period}$ $\textrm{/\,days}$", K1=r"$\textrm{radial velocity semi-amplitude}$ $\textrm{/\,km\,s}^{-1}$", e=r"$\textrm{eccentricity}$", rv_confidence=r"$\textrm{Single star confidence (rv only)}$", ast_confidence=r"$\textrm{Single star confidence (ast only)}$", joint_confidence=r"$\textrm{Single star confidence (joint)}$", rv_nb_transits=r"$\textrm{number of radial velocity transits}$") cbar = plt.colorbar(scat) cbar.set_label(latex_labels_.get(z_label, z_label)) fig.tight_layout() if scale: ax.set_xlabel(r"$K_{1} / \sqrt{P(1-e^2)}$ \textrm{(Pourbaix et al. 2004)}") ax.set_ylabel(r"$K_{1} / \sqrt{P(1-e^2)}$ \textrm{(this work)}") else: ax.set_xlabel(r"$K_{1}\,/\,\textrm{km\,s}^{-1}$ \textrm{(Pourbaix et al. 2004)}") ax.set_ylabel(r"$K_{1}\,/\,\textrm{km\,s}^{-1}$ \textrm{(this work)}") fig.tight_layout() return fig sb9 = Table.read(os.path.join(BASE_PATH, "data", "sb9_xm_gaia.fits")) sb9 = sb9.group_by("source_id") sb9 = sb9[sb9.groups.indices[:-1]] sb9_mask = (sb9["f_K1"] != ">") \ * (sb9["f_T0"] == 0) \ * (sb9["Grade"] > 0) \ * (sb9["f_omega"] != "a") \ * (sb9["o_K1"] > 0) sb9 = sb9[sb9_mask] soubiran = Table.read(os.path.join( BASE_PATH, "data", "soubiran-2013-xm-gaia.fits")) soubiran = soubiran.group_by("source_id") soubiran = soubiran[soubiran.groups.indices[:-1]] velociraptor_source_ids = np.memmap("../../results/die-hard-subset.sources.memmap", mode="r", dtype=">i8") velociraptor_pdf = np.memmap("../../results/die-hard-subset.pdf.memmap", mode="r", dtype=np.float32, shape=(3, len(velociraptor_source_ids), 100)) model_names = ["ast", "rv", "joint"] fig = plot_semi_amplitude_wrt_literature(velociraptor, sb9, z_label="Per", scale=False, loglog=False, z_log=True) fig = plot_semi_amplitude_wrt_literature(velociraptor, sb9, z_label="e", scale=False, loglog=False) fig = plot_semi_amplitude_wrt_literature(velociraptor, sb9, scale=False, loglog=False) fig = plot_semi_amplitude_wrt_literature(velociraptor, sb9, z_label="rv_nb_transits", scale=False, loglog=False) fig = literature_binary_properties(velociraptor, sb9) for i, model_name in enumerate(model_names): fig = plot_literature_multiplicity_pdf( velociraptor_source_ids, velociraptor_pdf, soubiran, sb9, pdf_idx=i) for i, model_name in enumerate(model_names): fig_classes = plot_literature_multiplicity_classifications( velociraptor_source_ids, velociraptor_pdf, soubiran, sb9,pdf_idx=i)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,277
andycasey/velociraptor
refs/heads/master
/scripts/npm_run_elastic_ball_test.py
""" Set up and run the non-parametric model. """ import numpy as np import os import multiprocessing as mp import pickle import yaml import tqdm import logging from time import sleep from astropy.io import fits import npm_utils as npm import stan_utils as stan with open(npm.CONFIG_PATH, "r") as fp: config = yaml.load(fp) # Load in the data. data = fits.open(config["data_path"])[1].data all_label_names = list(config["kdtree_label_names"]) \ + list(config["predictor_label_names"]) # Set up a KD-tree. X = np.vstack([data[ln] for ln in config["kdtree_label_names"]]).T finite = np.all([np.isfinite(data[ln]) for ln in all_label_names], axis=0) finite_indices = np.where(finite)[0] N, D = X.shape F = finite_indices.size L = 4 * len(config["predictor_label_names"]) + 1 C = config.get("share_optimised_result_with_nearest", 0) kdt, scales, offsets = npm.build_kdtree(X[finite], relative_scales=config.get("kdtree_relative_scales", None)) kdt_kwds = dict(offsets=offsets, scales=scales, full_output=False) kdt_kwds.update( minimum_radius=config.get("kdtree_minimum_radius", None), # DEFAULT minimum_points=config.get("kdtree_minimum_points", 1024), # DEFAULT maximum_points=config.get("kdtree_maximum_points", 8192), minimum_density=config.get("kdtree_minimum_density", None) ) # DEFAULT logging.info("k-d tree keywords: {}".format(kdt_kwds)) model = stan.load_stan_model(config["model_path"], verbose=False) default_opt_kwds = dict( verbose=False, tol_obj=7./3 - 4./3 - 1, # machine precision tol_grad=7./3 - 4./3 - 1, # machine precision tol_rel_grad=1e3, tol_rel_obj=1e4, iter=10000) default_opt_kwds.update(config.get("optimisation_kwds", {})) # Make sure that some entries have the right units. for key in ("tol_obj", "tol_grad", "tol_rel_grad", "tol_rel_obj"): if key in default_opt_kwds: default_opt_kwds[key] = float(default_opt_kwds[key]) done = np.zeros(N, dtype=bool) queued = np.zeros(N, dtype=bool) results = np.nan * np.ones((N, L), dtype=float) # Set everything done except a box in parameter space. do = (3 > data["bp_rp"]) \ * (data["bp_rp"] > 2.5) \ * (data["absolute_rp_mag"] > -7.5) \ * (data["absolute_rp_mag"] < -2.5) \ * finite done[~do] = True default_init = np.array([0.6985507, 3.0525788, 1.1474566, 1.8999833, 0.6495420]) def optimize_mixture_model(index, init=None): # Select indices and get data. indices = finite_indices[npm.query_around_point(kdt, X[index], **kdt_kwds)] y = np.array([data[ln][indices] for ln in config["predictor_label_names"]]).T if init is None: init = npm.get_initialization_point(y) opt_kwds = dict(init=init, data=dict(y=y, N=y.shape[0], D=y.shape[1])) opt_kwds.update(default_opt_kwds) # Do optimization. with stan.suppress_output(): try: p_opt = npm._check_params_dict(model.optimizing(**opt_kwds)) except: p_opt = None return (index, p_opt, indices) def swarm(*indices, max_random_starts=3, in_queue=None, candidate_queue=None, out_queue=None): def _random_index(): yield from np.random.choice(indices, max_random_starts, replace=False) _ri = _random_index() random_start = lambda *_: (_ri.__next__(), default_init) swarm = True while swarm: for func in (in_queue.get_nowait, random_start): try: index, init = func() except mp.queues.Empty: logging.info("Using a random index to start") continue except StopIteration: logging.warning("Swarm is bored") sleep(5) except: logging.exception("Unexpected exception:") swarm = False else: if index is None and init is False: swarm = False break try: _, result, kdt_indices = optimize_mixture_model(index, init) except: logging.exception("Exception when optimizing on {} from {}"\ .format(index, init)) break out_queue.put((index, result)) if result is not None: if C > 0: # Assign the closest points to have the same result. # (On the other end of the out_qeue we will deal with # multiple results.) out_queue.put((kdt_indices[:C + 1], result)) # Candidate next K points K = 10 candidate_queue.put((kdt_indices[C + 1:C + 1 + K], result)) break return None while not all(done): P = 20 # mp.cpu_count() # Only do stars in the box. do_indices = np.where(~done)[0] D = do_indices.size with mp.Pool(processes=P) as pool: manager = mp.Manager() in_queue = manager.Queue() candidate_queue = manager.Queue() out_queue = manager.Queue() swarm_kwds = dict(max_random_starts=10, in_queue=in_queue, out_queue=out_queue, candidate_queue=candidate_queue) j = [] for _ in range(P): j.append(pool.apply_async(swarm, do_indices, kwds=swarm_kwds)) # The swarm will just run at random initial points until we communicate # back that the candidates are good. with tqdm.tqdm(total=D) as pbar: while True: has_candidates, has_results = (True, True) # Check for candidates. try: r = candidate_queue.get_nowait() except mp.queues.Empty: has_candidates = False else: candidate_indices, init = r candidate_indices = np.atleast_1d(candidate_indices) for index in candidate_indices: if not done[index] and not queued[index] and finite[index]: in_queue.put((index, init)) queued[index] = True # Check for output. try: r = out_queue.get(timeout=5) except mp.queues.Empty: has_results = False else: index, result = r index = np.atleast_1d(index) updated = index.size - sum(done[index]) done[index] = True if result is not None: results[index] = npm._pack_params(**result) pbar.update(updated) if not has_candidates and not has_results: break raise a # Clean up any difficult cases. with mp.Pool(processes=P) as pool: manager = mp.Manager() in_queue = manager.Queue() candidate_queue = manager.Queue() out_queue = manager.Queue() swarm_kwds = dict(max_random_starts=0, in_queue=in_queue, out_queue=out_queue, candidate_queue=candidate_queue) # Do a check for entries that are not done. not_done = np.where((~done * finite))[0] ND = not_done.size logging.info("{} not done".format(ND)) for index in not_done: # Get nearest points that are done. indices = finite_indices[npm.query_around_point(kdt, X[index], **kdt_kwds)] in_queue.put((index, results[indices[done[indices]][0]])) j = [] for _ in range(P): j.append(pool.apply_async(swarm, finite_indices, kwds=swarm_kwds)) with tqdm.tqdm(total=ND) as pbar: while True: has_candidates, has_results = (True, True) # Check for candidates. try: r = candidate_queue.get_nowait() except mp.queues.Empty: has_candidates = False else: candidate_indices, init = r candidate_indices = np.atleast_1d(candidate_indices) for index in candidate_indices: if not done[index] and not queued[index] and finite[index]: in_queue.put((index, init)) queued[index] = True # Check for output. try: r = out_queue.get(timeout=5) except mp.queues.Empty: has_results = False else: index, result = r index = np.atleast_1d(index) updated = index.size - sum(done[index]) done[index] = True if result is not None: results[index] = npm._pack_params(**result) pbar.update(updated) if not has_candidates and not has_results: sleep(1) # Do a check for entries that are not done. not_done = np.where((~done * finite))[0] ND = not_done.size if ND == 0: break results_path = config.get("results_path", "results.pkl") with open(results_path, "wb") as fp: pickle.dump(results, fp, -1) logging.info("Results written to {}".format(results_path))
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,278
andycasey/velociraptor
refs/heads/master
/npm.py
""" Non-parametric model of binarity and single stars across the H-R diagram. """ import numpy as np import matplotlib import matplotlib.pyplot as plt import pickle from astropy.table import Table from astropy.io import fits from scipy import (spatial, optimize as op) from sklearn import neighbors as neighbours from time import time import velociraptor import stan_utils as stan import npm_utils from corner import corner np.random.seed(123) DEBUG_PLOTS = False data_path = "data/rv-all-subset-1e4.fits" #data = fits.open(data_path)[1].data data = Table.read(data_path) data["rv_single_epoch_scatter"] = data["rv_single_epoch_variance"]**0.5 # [1] Construct KD-Tree # [2] Select a random X number of stars # [3] Construct some way to get nearest neighbour results that are not Null??? # --> select the next nearest star? or one star in the volume that does not have results? # [3] Run the optimization etc. # what labels are we going to build the KD-tree in? kdt_label_names = ("bp_rp", "absolute_rp_mag", "phot_rp_mean_mag") predictor_label_names = ( "rv_single_epoch_scatter", "astrometric_unit_weight_error", "phot_bp_rp_excess_factor", "rv_abs_diff_template_teff", ) parameter_names = ["theta", "mu_single", "sigma_single", "mu_multiple", "sigma_multiple"] X_kdt = np.vstack([data[ln] for ln in kdt_label_names]).T # TODO: Right now I am *requiring* that all predictor labels and KD-Tree # labels are finite, but we may want to change this in the future. all_label_names = tuple(list(kdt_label_names) + list(predictor_label_names)) subset = np.all(np.isfinite(np.vstack([data[ln] for ln in all_label_names])), axis=0) X_kdt = X_kdt[subset] X_scale = np.ptp(X_kdt, axis=0) X_mean = np.mean(X_kdt, axis=0) _scale = lambda a: (a - X_mean)/X_scale _descale = lambda a: a * X_scale + X_mean # Normalise the array for the KD-tree X_norm = _scale(X_kdt) # Construct the KD-Tree kdt = npm_utils.build_kdt(X_norm) data = data[subset] model = stan.load_stan_model("npm.stan") # Calculate the total number of parameters M = len(data) L = len(predictor_label_names) K = 1 + 4 * L opt_params = np.empty((M, K)) #subset_points = np.random.choice(M, size=1000, replace=False) #for i in range(M): #for j, i in enumerate(subset_points): for j, i in enumerate(range(M)): print("At point {}/{}: {}".format(j, M, i)) k_indices, dist = npm_utils.get_ball_around_point(kdt, X_norm[[i]], full_output=True) y = np.array([data[ln][k_indices] for ln in predictor_label_names]).T N, D = y.shape init_values = npm_utils.get_initialization_point(y) init_dict = dict(zip(parameter_names, npm_utils._unpack_params(init_values, D))) init_dict["mu_multiple_uv"] = 0.5 * np.ones(D) data_dict = dict(y=y, N=N, D=D) opt_kwds = dict( data=data_dict, init=init_dict, verbose=False, tol_obj=7./3 - 4./3 - 1, # machine precision tol_grad=7./3 - 4./3 - 1, # machine precision tol_rel_grad=1e3, tol_rel_obj=1e4, iter=10000) t_init = time() p_opt = model.optimizing(**opt_kwds) t_opt = time() - t_init print("Optimization took {:.2f} seconds".format(t_opt)) del p_opt["mu_multiple_uv"] for k in p_opt.keys(): if k == "theta": continue p_opt[k] = np.atleast_1d(p_opt[k]) opt_params[i] = npm_utils._pack_params(**p_opt) print("Single star fraction at this point: {:.2f}".format(opt_params[i, 0])) if DEBUG_PLOTS: print("Running DEBUG plots") for l, predictor_label_name in enumerate(predictor_label_names): fig, ax = plt.subplots() xi = np.linspace( np.min(data_dict["y"].T[l].flatten()), np.max(data_dict["y"].T[l].flatten()), 1000) ax.hist(data_dict["y"].T[l].flatten(), bins=50, facecolor="#cccccc", zorder=-1) show = init_dict ax.plot(xi, N * npm_utils.norm_pdf(xi, show["mu_single"][l], show["sigma_single"][l], show["theta"]), c='r') ax.plot(xi, N * npm_utils.lognorm_pdf(xi, show["mu_multiple"][l], show["sigma_multiple"][l], show["theta"]), c='b') show = p_opt ax.plot(xi, N * npm_utils.norm_pdf(xi, show["mu_single"][l], show["sigma_single"][l], show["theta"]), c='m') ax.plot(xi, N * npm_utils.lognorm_pdf(xi, show["mu_multiple"][l], show["sigma_multiple"][l], show["theta"]), c='y') samples = model.sampling(**stan.sampling_kwds( data=opt_kwds["data"], init=p_opt, iter=2000, chains=2)) chains_dict = samples.extract() if L == 1: chains = np.vstack([ chains_dict["theta"], chains_dict["mu_single"], chains_dict["sigma_single"], chains_dict["mu_multiple"], chains_dict["sigma_multiple"] ]).T else: chains = np.hstack([ np.atleast_2d(chains_dict["theta"]).T, chains_dict["mu_single"], chains_dict["sigma_single"], chains_dict["mu_multiple"], chains_dict["sigma_multiple"] ]) fig = corner(chains) # Make plots of the pdf of each distribution. for l, predictor_label_name in enumerate(predictor_label_names): xi = np.linspace(np.min(y.T[l]), np.max(y.T[l]), 1000) indices = np.random.choice(len(chains), 100, replace=False) fig, ax = plt.subplots() for index in indices: idx_theta = parameter_names.index("theta") idx_norm_mu = parameter_names.index("mu_single") idx_norm_sigma = parameter_names.index("sigma_single") idx_lognorm_mu = parameter_names.index("mu_multiple") idx_lognorm_sigma = parameter_names.index("sigma_multiple") theta = chains[index, 0] norm_mu = chains[index, 1 + l] norm_sigma = chains[index, 1 + L + l] lognorm_mu = chains[index, 1 + 2*L + l] lognorm_sigma = chains[index, 1 + 3*L + l] ax.plot(xi, npm_utils.norm_pdf(xi, norm_mu, norm_sigma, theta), c='r', alpha=0.1) ax.plot(xi, npm_utils.lognorm_pdf(xi, lognorm_mu, lognorm_sigma, theta), c='b', alpha=0.1) _ = ax.hist(y.T[l], bins=500, facecolor="#000000", zorder=-1, normed=True) print(y.size) ax.set_title(predictor_label_name.replace("_", " ")) raise a with open("rv-all-subset-1e4-optimized-unconstrained.pickle", "wb") as fp: pickle.dump(opt_params, fp, -1) def normal_lpdf(y, mu, sigma): ivar = sigma**(-2) return 0.5 * (np.log(ivar) - np.log(2 * np.pi) - (y - mu)**2 * ivar) def lognormal_lpdf(y, mu, sigma): ivar = sigma**(-2) return - 0.5 * np.log(2 * np.pi) - np.log(y * sigma) \ - 0.5 * (np.log(y) - mu)**2 * ivar from scipy.special import logsumexp # Calculate log-probabilities for all of the stars we considered. def membership_probability(y, p_opt): y = np.atleast_1d(y) theta, s_mu, s_sigma, m_mu, m_sigma = npm_utils._unpack_params(p_opt) assert s_mu.size == y.size, "The size of y should match the size of mu" D = y.size ln_prob = np.zeros((D, 2)) for d in range(D): ln_prob[d] = [ normal_lpdf(y[d], s_mu[d], s_sigma[d]), lognormal_lpdf(y[d], m_mu[d], m_sigma[d]) ] # TODO: I am not certain that I am summing these log probabilities correctly sum_ln_prob = np.sum(ln_prob, axis=0) # per mixture ln_likelihood = logsumexp(sum_ln_prob) with np.errstate(under="ignore"): ln_membership = sum_ln_prob - ln_likelihood return np.exp(ln_membership) y = np.array([data[ln] for ln in predictor_label_names]).T N, D = y.shape p_single = np.array([membership_probability(y[i], opt_params[i])[0] for i in range(N)]) # Now calculate the amount of excess RV variance def rv_excess_scatter(y, p_opt, label_index): y = np.atleast_1d(y) _, s_mu, s_sigma, __, ___ = npm_utils._unpack_params(p_opt) assert s_mu.size == y.size, "The size of y should match the size of mu" rv_single_epoch_excess = np.sqrt(y[label_index]**2 - s_mu[label_index]**2) rv_single_epoch_significance = rv_single_epoch_excess/s_sigma[label_index] return (rv_single_epoch_excess, rv_single_epoch_significance) li = list(predictor_label_names).index("rv_single_epoch_scatter") rv_excess = np.array([ rv_excess_scatter(y[i], opt_params[i], li) for i in range(N)]) # Remember that the probabilities of binarity take into acocunt more than just # the radial velocity excess! # Make a corner plot of the various properties? Coloured by probabilities? with open("rv-all-subset-1e4-results.pickle", "wb") as fp: pickle.dump((subset, opt_params, predictor_label_names, p_single, rv_excess), fp, -1)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,279
andycasey/velociraptor
refs/heads/master
/run_probabilities.py
""" Run this after run_analysis.py """ import logging import numpy as np import sys import pickle import tqdm from scipy.special import logsumexp from astropy.io import fits def ps(y, theta, s_mu, s_sigma, b_mu, b_sigma): s_ivar = s_sigma**-2 b_ivar = b_sigma**-2 hl2p = 0.5 * np.log(2*np.pi) s_lpdf = -hl2p + 0.5 * np.log(s_ivar) \ - 0.5 * (y - s_mu)**2 * s_ivar b_lpdf = -np.log(y*b_sigma) - hl2p \ - 0.5 * (np.log(y) - b_mu)**2 * b_ivar b_lpdf[~np.isfinite(b_lpdf)] = -np.inf foo = np.vstack([s_lpdf, b_lpdf]).T + np.log([theta, 1-theta]) p_single = np.exp(foo.T[0] - logsumexp(foo, axis=1)) assert np.all(np.isfinite(p_single)) return p_single def lnprob_single(y, theta, s_mu, s_sigma, b_mu, b_sigma): s_ivar = s_sigma**-2 b_ivar = b_sigma**-2 hl2p = 0.5 * np.log(2*np.pi) s_lpdf = -hl2p + 0.5 * np.log(s_ivar) \ - 0.5 * (y - s_mu)**2 * s_ivar b_lpdf = -np.log(y*b_sigma) - hl2p \ - 0.5 * (np.log(y) - b_mu)**2 * b_ivar b_lpdf[~np.isfinite(b_lpdf)] = -np.inf return np.vstack([s_lpdf, b_lpdf]).T + np.log([theta, 1-theta]) if __name__ == "__main__": result_paths = sys.argv[1:] if len(result_paths) < 1: raise ValueError("no result paths given") all_results = [] for result_path in result_paths: logging.info(f"Loading results from {result_path}") with open(result_path, "rb") as fp: r = pickle.load(fp) all_results.append(r) # Check seeds are the same. initial_seed = all_results[0]["seed"] if initial_seed is None: raise ValueError("seed is not known") for r in all_results: if int(r["seed"]) != int(initial_seed): raise ValueError(f"seeds differ: {initial_seed} in {result_paths[0]} != {r['seed']}") # TODO: check that the data files etc are the same between results paths # TODO: this is so hacky just re-factor the model format and run_analysis.py results = all_results[0] config = results["config"] data = fits.open(config["data_path"])[1].data all_label_names = list(config["kdtree_label_names"]) \ + list(config["require_finite_label_names"]) \ + list(config["predictor_label_names"]) all_label_names = list(np.unique(all_label_names)) # Set up a KD-tree. finite = np.all([np.isfinite(data[ln]) for ln in all_label_names], axis=0) y = np.array([data[ln][finite] for ln in config["predictor_label_names"]]).flatten() K = 100 M = len(result_paths) lnprobs = np.zeros((M, y.size, K, 2)) p_single = np.zeros((M + 1, y.size, K)) for m, (result_path, results) in enumerate(zip(result_paths, all_results)): mu_single, mu_single_var, sigma_single, sigma_single_var, \ sigma_multiple, sigma_multiple_var = results["gp_predictions"].T # Calculate probabilities. # TODO MAGIC HACK ALL DIRECTLY BELOW mean_theta = np.mean(results["results"][:, 0]) mask = np.array([1, 2, 4]) rhos = np.corrcoef(results["results"][:, mask].T) logging.info(f"Calculating probabilities for {result_path}") for d in tqdm.tqdm(range(y.size)): diag = np.atleast_2d(np.array([ mu_single_var[d], sigma_single_var[d], sigma_multiple_var[d] ]))**0.5 cov = diag * rhos * diag.T mu = np.array([mu_single[d], sigma_single[d], sigma_multiple[d]]) ms, ss, sm = np.random.multivariate_normal(mu, cov, size=K).T # Calculate the mu_multiple. TODO HACK MAGIC mm = np.log(ms + ss) + sm**2 lnprobs[m, d] = lnprob_single(y[[d]], mean_theta, ms, ss, mm, sm) p_single[m, d] = np.exp(lnprobs[m, d, :, 0] - logsumexp(lnprobs[m, d], axis=1)) # Calculate joint probabilities. for d in tqdm.tqdm(range(y.size)): for k in range(K): foo = lnprobs[:, d, k] numerator = logsumexp(lnprobs[:, d, k, 0]) denominator = logsumexp(lnprobs[:, d, k, 1]) p_single[-1, d, k] = np.exp(numerator - logsumexp([numerator, denominator])) # Save percentiles of the probability distributions. # Save the draws from the probability distributions. raise a # Need: # result_paths # Check seed for each one and ensure they are the same. # Calculate the mu_multiple (as the lower value). # Calculate probabilities for each source, after propagating variance.
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,280
andycasey/velociraptor
refs/heads/master
/_convert_results.py
""" Convert results files. """ import os import sys import pickle import yaml if __name__ == "__main__": output_path = sys.argv[1] joint_config_path = sys.argv[2] result_paths = sys.argv[3:] with open(joint_config_path, "r") as fp: config = yaml.load(fp) output = dict(config=config) for i, result_path in enumerate(result_paths): model_name = os.path.basename(result_path).split(".")[0] with open(result_path, "rb") as fp: results = pickle.load(fp) output[model_name] = [ results["results"], results["gp_parameters"], results["gp_predictions"] ] with open(output_path, "wb") as fp: pickle.dump(output, fp)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,281
andycasey/velociraptor
refs/heads/master
/npm_utils.py
import numpy as np import os from scipy import (optimize as op, stats) from sklearn import neighbors as neighbours from scipy.special import logsumexp from time import time import warnings import stan_utils as stan CONFIG_PATH = "npm-config.yaml" def get_output_path(source_id, config, check_path_exists=True): results_suffix = config.get("results_suffix", None) path = os.path.join(config["results_path"], "{0}/{1}{2}".format( split_source_id(source_id), "+".join(config["predictor_label_names"]), ".{}".format(results_suffix) if results_suffix is not None else "")) if check_path_exists: os.makedirs(os.path.dirname(path), exist_ok=True) return path def build_kdtree(X, relative_scales=None,**kwargs): """ Build a KD-tree from the finite values in the given array. """ offset = np.mean(X, axis=0) if relative_scales is None: # Whiten the data. relative_scales = np.ptp(X, axis=0) X_norm = (X - offset)/relative_scales kdt_kwds = dict(leaf_size=40, metric="minkowski") kdt_kwds.update(kwargs) kdt = neighbours.KDTree(X_norm, **kdt_kwds) return (kdt, relative_scales, offset) def query_around_point(kdtree, point, offsets=0, scales=1, minimum_radius=None, maximum_radius=None, minimum_points=1, maximum_points=None, minimum_density=None, dualtree=False, full_output=False, **kwargs): """ Query around a point in the KD-Tree until certain conditions are met (e.g., the number of points in the ball, and the minimum radius that the ball expands out to). :param kdtree: The pre-computed KD-Tree. :param point: The (unscaled) point to query around. :param offsets: [optional] The offsets to apply to the query point. :param scales: [optional] The scaling to apply to the query point, after subtracting the offsets. :param minimum_radius: [optional] The minimum radius (or radii) that the ball must extend to. :param minimum_points: [optional] The minimum number of points to return in the ball. :param maximum_points: [optional] The maximum number of points to return in the ball. If the number of points returned exceeds this value, then a random subset of the points will be returned. :param minimum_density: [optional] The minimum average density of points per dimension for the ball. This can be useful to ensure that points that are in the edge of the k-d tree parameter space will be compared against points that are representative of the underlying space, and not just compared against nearest outliers. :param dualtree: [optional] Use the dual tree formalism for the query: a tree is built for the query points, and the pair of trees is used to efficiently search this space. This can lead to better performance as the number of points grows large. :param full_output: [optional] If `True`, return a two length tuple of the distances to each point and the indicies, otherwise just return the indices. """ #print("querying k-d tree") offsets = np.atleast_1d(offsets) scales = np.atleast_1d(scales) point_orig = np.atleast_1d(point).reshape(1, -1) point = (point_orig - offsets)/scales # Simple case. if minimum_radius is None and minimum_density is None: # We can just query the nearest number of points. d, indices = kdtree.query(point, k=minimum_points, sort_results=True, return_distance=True, dualtree=dualtree) else: # We need to find the minimum radius that meets our constraints. if minimum_radius is None: minimum_radius = 0 if minimum_density is None: minimum_density = 0 minimum_radius = np.atleast_1d(minimum_radius) minimum_density = np.atleast_1d(minimum_density) # Need to scale the minimum radius from the label space to the normalised # k-d tree space. minimum_radius_norm = np.max(minimum_radius / np.atleast_1d(scales)) K = kdtree.two_point_correlation(point, minimum_radius_norm)[0] # "density" = N/(2*R) # if N > 2 * R * density then our density constraint is met K_min = np.max(np.hstack([ minimum_points, 2 * minimum_density * minimum_radius ])) # Check that the minimum radius norm will also meet our minimum number # of points constraint. Otherwise, we need to use two point # auto-correlation functions to see how far to go out to. if K >= K_min: # All constraints met. radius_norm = minimum_radius_norm else: #print("Using k-dtree to step out in radius because we have {} points within {} but need {}".format( # K, minimum_radius_norm, K_min)) # We need to use the k-d tree to step out until our constraints are # met. maximum_radius_norm = 2 * np.max(np.ptp(kdtree.data, axis=0)) # This is the initial coarse search. N, D = kdtree.data.shape left, right = (minimum_radius_norm, maximum_radius_norm) Q = kwargs.get("Q", 10) # MAGIC HACK # MAGIC HACK tolerance = maximum_points if maximum_points is not None \ else 2 * minimum_points while True: # Shrink it. ri = np.logspace(np.log10(left), np.log10(right), Q) counts = kdtree.two_point_correlation(point, ri) #print("tolerance {}: {} {} {} {}".format(tolerance, left, right, ri, counts)) minimum_counts = np.clip(2 * np.max(np.dot(ri.reshape(-1, 1), (minimum_density * scales).reshape(1, -1)), axis=1), minimum_points, N) indices = np.arange(Q)[counts >= minimum_counts] left, right = (ri[indices[0] - 1], ri[indices[0] + 1]) #print("new: {} {}".format(left, right)) if np.diff(counts[indices]).max() < tolerance: radius_norm = ri[indices[0]] break # two_point_correlation(point, minimum_radius_norm) # is eequivalent to # query_radius(point, minimum_radius_norm, count_only=True) # but in my tests two_point_correlation was a little faster. # kdtree.query_radius returns indices, d # kdtree.query returns d, indices # .... are you serious? indices, d = kdtree.query_radius(point, radius_norm, return_distance=True, sort_results=True) d, indices = (d[0], indices[0]) L = len(indices) if maximum_points is not None and L > maximum_points: if maximum_points < minimum_points: raise ValueError("minimum_points must be smaller than maximum_points") if maximum_radius is not None: L = np.where(np.all(np.abs(point[0] - np.asarray(kdtree.data)[indices]) <= maximum_radius, axis=1))[0] maximum_points = min(maximum_points, L.size) # Sub-sample a random number. sub_idx = np.random.choice(L, maximum_points, replace=False) d, indices = (d[sub_idx], indices[sub_idx]) elif maximum_radius is not None: L = np.where(np.all(np.abs(point[0] - np.asarray(kdtree.data)[indices]) <= maximum_radius, axis=1))[0] maximum_points = min(maximum_points, L.size) # Sub-sample a random number. sub_idx = np.random.choice(L, maximum_points, replace=False) d, indices = (d[sub_idx], indices[sub_idx]) # Meta should include the PTP values of points in the ball. meta = dict() #assert minimum_points is None or indices.size >= minimum_points return (d, indices, meta) if full_output else indices def normal_lpdf(y, mu, sigma): ivar = sigma**(-2) return 0.5 * (np.log(ivar) - np.log(2 * np.pi) - (y - mu)**2 * ivar) def lognormal_lpdf(y, mu, sigma): ivar = sigma**(-2) return - 0.5 * np.log(2 * np.pi) - np.log(y * sigma) \ - 0.5 * (np.log(y) - mu)**2 * ivar # Calculate log-probabilities for all of the stars we considered. def membership_probability(y, p_opt): y = np.atleast_1d(y) theta, s_mu, s_sigma, m_mu, m_sigma = _unpack_params(_pack_params(**p_opt)) assert s_mu.size == y.size, "The size of y should match the size of mu" D = y.size ln_prob = np.zeros((D, 2)) for d in range(D): ln_prob[d] = [ normal_lpdf(y[d], s_mu[d], s_sigma[d]), lognormal_lpdf(y[d], m_mu[d], m_sigma[d]) ] # TODO: I am not certain that I am summing these log probabilities correctly sum_ln_prob = np.sum(ln_prob, axis=0) # per mixture ln_likelihood = logsumexp(sum_ln_prob) with np.errstate(under="ignore"): ln_membership = sum_ln_prob - ln_likelihood return np.exp(ln_membership) def label_excess(y, p_opt, label_index): y = np.atleast_1d(y) _, s_mu, s_sigma, __, ___ = _unpack_params(_pack_params(**p_opt)) assert s_mu.size == y.size, "The size of y should match the size of mu" excess = np.sqrt(y[label_index]**2 - s_mu[label_index]**2) significance = excess/s_sigma[label_index] return (excess, significance) def build_kdt(X_norm, **kwargs): kdt_kwds = dict(leaf_size=40, metric="minkowski") kdt_kwds.update(kwargs) kdt = neighbours.KDTree(X_norm, **kdt_kwds) return kdt def get_ball_around_point(kdt, point, K=1000, scale=1, offset=0, full_output=False): dist, k_indices = kdt.query((point - offset)/scale, K) dist, k_indices = (dist[0], k_indices[0]) return (k_indices, dist) if full_output else k_indices # Stan needs a finite value to initialize correctly, so we will use a dumb (more # robust) optimizer to get an initialization value. def norm_pdf(x, norm_mu, norm_sigma, theta): return theta * (2 * np.pi * norm_sigma**2)**(-0.5) * np.exp(-(x - norm_mu)**2/(2*norm_sigma**2)) def lognorm_pdf(x, lognorm_mu, lognorm_sigma, theta): return (1.0 - theta)/(x * lognorm_sigma * np.sqrt(2*np.pi)) \ * np.exp(-0.5 * ((np.log(x) - lognorm_mu)/lognorm_sigma)**2) def ln_likelihood(y, theta, s_mu, s_sigma, b_mu, b_sigma): s_ivar = s_sigma**-2 b_ivar = b_sigma**-2 hl2p = 0.5 * np.log(2*np.pi) s_lpdf = -hl2p + 0.5 * np.log(s_ivar) \ - 0.5 * (y - s_mu)**2 * s_ivar b_lpdf = -np.log(y*b_sigma) - hl2p \ - 0.5 * (np.log(y) - b_mu)**2 * b_ivar foo = np.vstack([s_lpdf, b_lpdf]).T + np.log([theta, 1-theta]) ll = np.sum(logsumexp(foo, axis=1)) #ll = np.sum(s_lpdf) + np.sum(b_lpdf) ##print(lpdf) #assert np.isfinite(ll) return ll def ln_prior(theta, s_mu, s_sigma, b_mu, b_sigma): # Ensure that the *mode* of the log-normal distribution is larger than the # mean of the normal distribution with warnings.catch_warnings(): warnings.simplefilter("ignore") min_mu_multiple = np.log(s_mu + s_sigma) + b_sigma**2 if not (1 >= theta >= 0) \ or not (15 >= s_mu >= 0.5) \ or not (10 >= s_sigma >= 0.05) \ or not (1.6 >= b_sigma >= 0.20) \ or np.any(b_mu < min_mu_multiple): return -np.inf # Beta prior on theta. return stats.beta.logpdf(theta, 5, 5) def ln_prob(y, L, *params): theta, s_mu, s_sigma, b_mu, b_sigma = _unpack_params(params, L=L) lp = ln_prior(theta, s_mu, s_sigma, b_mu, b_sigma) if np.isfinite(lp): return lp + ln_likelihood(y, theta, s_mu, s_sigma, b_mu, b_sigma) return lp def _unpack_params(params, L=None): # unpack the multdimensional values. if L is None: L = int((len(params) - 1)/4) theta = params[0] mu_single = np.array(params[1:1 + L]) sigma_single = np.array(params[1 + L:1 + 2 * L]) mu_multiple = np.array(params[1 + 2 * L:1 + 3 * L]) sigma_multiple = np.array(params[1 + 3 * L:1 + 4 * L]) return (theta, mu_single, sigma_single, mu_multiple, sigma_multiple) def _pack_params(theta, mu_single, sigma_single, mu_multiple, sigma_multiple, mu_multiple_uv=None, **kwargs): if mu_multiple_uv is None: return np.hstack([theta, mu_single, sigma_single, mu_multiple, sigma_multiple]) else: return np.hstack([theta, mu_single, sigma_single, mu_multiple, sigma_multiple, mu_multiple_uv]) def _check_params_dict(d, bounds_dict=None, fail_on_bounds=True, tolerance=0.01): if d is None: return d dc = {**d} for k in ("mu_single", "sigma_single", "mu_multiple", "sigma_multiple"): dc[k] = np.atleast_1d(dc[k]).flatten()[0] if bounds_dict is not None and k in bounds_dict: lower, upper = bounds_dict[k] if (not np.all(upper >= dc[k]) or not np.all(dc[k] >= lower)): if fail_on_bounds: raise ValueError("bounds not met: {} = {} not within ({} {})"\ .format(k, dc[k], lower, upper)) else: print("Clipping initial {} to be within bounds ({}, {}): {}"\ .format(k, lower, upper, dc[k])) dc[k] = np.clip(dc[k], lower + tolerance, upper - tolerance) return dc def nlp(params, y, L): return -ln_prob(y, L, *params) def _get_1d_initialisation_point(y, scalar=5, bounds=None): N= y.size init = dict( theta=0.75, mu_single=np.min([np.median(y, axis=0), 10]), sigma_single=0.2, sigma_multiple=0.5) if bounds is not None: for k, (lower, upper) in bounds.items(): if not (upper >= init[k] >= lower): init[k] = np.mean([upper, lower]) lower_mu_multiple = np.log(init["mu_single"] + scalar * init["sigma_single"]) \ + init["sigma_multiple"]**2 init["mu_multiple"] = 1.1 * lower_mu_multiple op_kwds = dict(x0=_pack_params(**init), args=(y, 1)) nlp = lambda params, y, L: -ln_prob(y, L, *params) p_opt = op.minimize(nlp, **op_kwds) keys = ("theta", "mu_single", "sigma_single", "mu_multiple", "sigma_multiple") init_dict = _check_params_dict(init) op_dict = _check_params_dict(dict(zip(keys, _unpack_params(p_opt.x)))) # Only return valid init values. valid_inits = [] for init in (init_dict, op_dict): if np.isfinite(nlp(_pack_params(**init), y, 1)): valid_inits.append(init) valid_inits.append("random") return valid_inits def get_initialization_point(y): N, D = y.shape ok = y <= np.mean(y) init_dict = dict( theta=0.5, mu_single=np.median(y[ok], axis=0), sigma_single=0.1 * np.median(y[ok], axis=0), sigma_multiple=0.1 * np.ones(D), ) # mu_multiple is *highly* constrained. Select the mid-point between what is # OK: mu_multiple_ranges = np.array([ np.log(init_dict["mu_single"] + 1 * init_dict["sigma_single"]) + init_dict["sigma_multiple"]**2, np.log(init_dict["mu_single"] + 5 * init_dict["sigma_single"]) + pow(init_dict["sigma_multiple"], 2) ]) init_dict["mu_multiple"] = np.mean(mu_multiple_ranges, axis=0) #init_dict["mu_multiple_uv"] = 0.5 * np.ones(D) x0 = _pack_params(**init_dict) op_kwds = dict(x0=x0, args=(y, D)) p_opt = op.minimize(nlp, **op_kwds) init_dict = dict(zip( ("theta", "mu_single", "sigma_single", "mu_multiple", "sigma_multiple"), _unpack_params(p_opt.x))) init_dict["mu_multiple_uv"] = 0.5 * np.ones(D) init_dict = _check_params_dict(init_dict) return init_dict
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,282
andycasey/velociraptor
refs/heads/master
/attic/validation.py
""" Validation of the radial velocity calibration model. """ import numpy as np import matplotlib.pyplot as plt import corner import pickle from scipy.special import logsumexp import velociraptor # Load the data and make some plots. data_path = "data/rv-all.fits" sources = velociraptor.load_gaia_sources(data_path) # Calculate temperature from bp-rp colour and use the distance from the # template as a predictor for the error in radial velocity (under the assumption # that the wrong template was used, giving a higher velocity error) use_in_fit = np.isfinite(sources["radial_velocity"]) \ * (sources["phot_bp_rp_excess_factor"] < 1.5) \ * np.isfinite(sources["bp_rp"]) \ * np.isfinite(sources["teff_val"]) \ * (sources["bp_rp"] < 2.5) \ * (sources["bp_rp"] > 0.5) x = sources["bp_rp"][use_in_fit] y = sources["teff_val"][use_in_fit] coeff = np.polyfit(1.0/x, y, 2) xi = np.linspace(x.min(), x.max(), 100) fig, ax = plt.subplots() ax.scatter(x, y, s=1, alpha=0.05, facecolor="k") ax.plot(xi, np.polyval(coeff, 1.0/xi), c='r', lw=2) #sources["approx_teff"] = np.polyval(coeff, 1.0/sources["bp_rp"]) sources["teff_from_rv_template"] = np.abs( np.polyval(coeff, 1.0/sources["bp_rp"]) - sources["rv_template_teff"]) N_bins = 100 H_all, xedges, yedges = np.histogram2d( sources["teff_from_rv_template"][use_in_fit], sources["rv_excess_variance"][use_in_fit], bins=( np.linspace(0, 3000, N_bins), np.linspace(0, 10000, N_bins) )) kwds = dict( aspect=np.ptp(xedges)/np.ptp(yedges), extent=(xedges[0], xedges[-1], yedges[-1], yedges[0]), ) from matplotlib.colors import LogNorm fig, ax = plt.subplots(figsize=(7.0, 5.5)) image = ax.imshow(H_all.T, norm=LogNorm(), cmap="Blues", **kwds) cbar = plt.colorbar(image, ax=ax) cbar.set_label(r"\textrm{count}") ax.set_ylim(ax.get_ylim()[::-1]) ax.set_xlabel(r"$|T_{\rm eff} - T_{\rm eff,template}|$ $({\rm K})$") ax.set_ylabel(r"\textrm{radial velocity excess variance} $({\rm km}^2\,{\rm s}^{-2})$") fig.tight_layout() fig.savefig("figures/rv_excess_variance_wrt_teff_diff.pdf", dpi=300) N_bins = 30 x = sources["teff_from_rv_template"][use_in_fit] fig, axes = plt.subplots(1, 2, figsize=(8, 4)) for i, (ax, equidensity) in enumerate(zip(axes, (True, False))): if equidensity: bins = np.percentile(x, np.linspace(0, 100, N_bins)) else: bins = np.linspace(np.min(x), np.max(x), N_bins) binary_fraction = np.zeros(N_bins - 1, dtype=float) indices = np.digitize(sources["teff_from_rv_template"][use_in_fit], bins) - 1 from collections import Counter counts = Counter(indices) N_per_bin = np.array([counts.get(k, 0) for k in range(N_bins - 1)], dtype=float) counts_finite = Counter(indices[(sources["p_sb_50"] > 0.5)[use_in_fit]]) N_rv_per_bin = np.array([counts_finite.get(k, 0) for k in range(N_bins - 1)], dtype=float) f_rv_per_bin = N_rv_per_bin / N_per_bin f_rv_per_bin_err = f_rv_per_bin * np.sqrt( (np.sqrt(N_rv_per_bin)/N_rv_per_bin)**2 + \ (np.sqrt(N_per_bin)/N_per_bin)**2) centroids = bins[:-1] + 0.5 * np.diff(bins) #fig, ax = plt.subplots() ax.plot(centroids, f_rv_per_bin, drawstyle="steps-mid") ax.errorbar(centroids, f_rv_per_bin, yerr=f_rv_per_bin_err, fmt=None, ) raise a model, data_dict, init_dict, used_in_fit = velociraptor.prepare_model( S=1e4, **sources) # This *works*, but you could start from any random position. init_dict = dict([ ('theta', 0.20), ('mu_coefficients', np.array([0.3, 1e-4, 1e12, 1, 1])), ('sigma_coefficients', np.array([0.3, 1e-4, 4e11, 1, 1])), ]) p_opt = model.optimizing(data=data_dict, init=init_dict) print(p_opt) fig, ax = plt.subplots() ax.scatter(sources["phot_rp_mean_flux"][used_in_fit], data_dict["rv_variance"], s=1, alpha=0.5, facecolor="k") x = np.logspace( np.log10(np.nanmin(sources["phot_rp_mean_flux"])), np.log10(np.nanmax(sources["phot_rp_mean_flux"])), 1000) bp_rp = np.nanmean(sources["bp_rp"]) * np.ones(x.size) mu_opt = np.dot( p_opt["mu_coefficients"], velociraptor._rvf_design_matrix(x, bp_rp=bp_rp)) sigma_opt = np.dot( p_opt["sigma_coefficients"], velociraptor._rvf_design_matrix(x, bp_rp=bp_rp)) mu_init = np.dot( init_dict["mu_coefficients"], velociraptor._rvf_design_matrix(x, bp_rp=bp_rp)) sigma_init = np.dot( init_dict["sigma_coefficients"], velociraptor._rvf_design_matrix(x, bp_rp=bp_rp)) ax.plot(x, mu_init, c='r') ax.fill_between(x, mu_init - sigma_init, mu_init + sigma_init, facecolor="r", alpha=0.3, edgecolor="none") ax.plot(x, mu_opt, c='b') ax.fill_between(x, mu_opt - sigma_opt, mu_opt + sigma_opt, facecolor="b", alpha=0.3, edgecolor="none") ax.semilogx() ax.set_ylim(0, 400) iterations = 2000 p_samples = model.sampling(**velociraptor.stan.sampling_kwds( data=data_dict, chains=2, iter=iterations, init=p_opt)) print(p_samples) parameter_names = ("theta", "mu_coefficients", "sigma_coefficients") label_names = (r"$\theta$", r"$\mu_0$", r"$\mu_1$", r"$\mu_2$", r"$\mu_3$", r"$\mu_4$", r"$\sigma_0$", r"$\sigma_1$", r"$\sigma_2$", r"$\sigma_3$", r"$\sigma_4$") chains_dict = p_samples.extract(parameter_names, permuted=True) chains = np.hstack([ chains_dict["theta"].reshape((iterations, 1)), chains_dict["mu_coefficients"], chains_dict["sigma_coefficients"] ]) """ fig = corner.corner(chains, labels=label_names) # Make many draws. x = np.logspace( np.log10(np.nanmin(sources["phot_rp_mean_flux"])), np.log10(np.nanmax(sources["phot_rp_mean_flux"])), 1000) bp_rp = np.nanmean(sources["bp_rp"]) * np.ones(x.size) dm = velociraptor._rvf_design_matrix(x, bp_rp=bp_rp) fig, ax = plt.subplots() ax.scatter(sources["phot_rp_mean_flux"][used_in_fit], data_dict["rv_variance"], s=1, alpha=0.5, facecolor="#000000") N_mu, N_sigma = 5, 5 for index in np.random.choice(iterations, min(iterations, 250), replace=False): y = np.dot(chains[index][1:1 + N_mu], dm) \ + np.random.normal(0, 1) * np.dot(chains[index][1 + N_mu:], dm) ax.plot(x, y, "r-", alpha=0.05) ax.semilogx() ax.set_ylim(0, ax.get_ylim()[1]) """ # Calculate point estimates of binary probability for all stars. has_rv = np.isfinite(sources["radial_velocity"]) J = sum(has_rv) percentiles = [16, 50, 84] for p in percentiles: sources["p_sb_{:.0f}".format(p)] = np.nan * np.ones(len(sources)) sources["rv_excess_variance"] = np.nan * np.ones(len(sources)) for j, index in enumerate(np.where(has_rv)[0]): dm = velociraptor._rvf_design_matrix(**sources[[index]]) mu = np.dot(dm.T, chains_dict["mu_coefficients"].T) ivar = np.dot(dm.T, chains_dict["sigma_coefficients"].T)**-2 log_pb = np.log(chains_dict["theta"]) - np.log(np.max(data_dict["rv_variance"])) log_ps = np.log(1 - chains_dict["theta"]) \ - 0.5 * np.log(2 * np.pi) + 0.5 * np.log(ivar) \ - 0.5 * (sources["rv_single_epoch_variance"][index, np.newaxis] - mu)**2 * ivar log_p_sb = log_pb - logsumexp([log_pb * np.ones_like(log_ps), log_ps], axis=0) p_sb = np.exp(log_p_sb).flatten() vs = np.percentile(p_sb, percentiles) for p, v in zip(percentiles, vs): sources["p_sb_{:.0f}".format(p)][index] = v print(j, J, vs) # Calculate excess variance. rv_max_single_star_variance = np.percentile( np.random.normal(0, 1, size=mu.shape) * ivar**-0.5, 99, axis=1) rv_excess_variance = np.clip( sources["rv_single_epoch_variance"][index] - rv_max_single_star_variance, 0, np.inf) sources["rv_excess_variance"][index] = rv_excess_variance # TODO: Save the PDFs? sources.write(data_path, overwrite=True) #with open("data/binary-pdfs.pkl", "wb") as fp: # pickle.dump((sources["source_id"][has_rv], p_sb), fp, -1) raise a # Plot the model against various properties of interest. shorthand_parameters = [ ("phot_rp_mean_flux", "mean rp flux", True), ("phot_bp_mean_flux", "mean bp flux", True), ("phot_g_mean_flux", "mean g flux", True), ("bp_rp", "bp-rp colour", False), ("phot_g_mean_mag", "mean g magnitude", False), ("phot_rp_mean_mag", "mean rp magnitude", False), ("phot_bp_mean_mag", "mean bp magnitude", False), ("teff_val", "inferred temperature", False), ("ra", "right ascension", False), ("dec", "declination", False), ("radial_velocity", "radial velocity", False), ("rv_nb_transits", "number of radial velocity transits", False), ("absolute_g_mag", "absolute g magnitude", False), ("absolute_bp_mag", "absolute bp magnitude", False), ("absolute_rp_mag", "absolute rp magnitude", False), ("rv_template_teff", "rv template teff", False), ("rv_template_logg", "rv template logg", False), ("rv_template_fe_h", "rv template [Fe/H]", False), ] for label_name, description, is_semilogx in shorthand_parameters: x = sources[label_name] y = sources["rv_single_epoch_variance"] - model_rv_sev_mu idx = np.argsort(x) fig, axes = plt.subplots(1, 2, figsize=(12, 6)) for ax in axes: ax.scatter(x, y, **model_scatter_kwds) if is_semilogx: ax.semilogx() ax.set_xlabel(r"\textrm{{{0}}}".format(description)) ax.set_ylabel(r"\textrm{residual single epoch radial velocity variance} $(\textrm{km}^2\,\textrm{s}^{-2})$") axes[0].set_title(r"\textrm{all points}") axes[1].set_title(r"\textrm{5th-95th percentile in} $y$\textrm{-axis}") axes[1].set_ylim(*np.nanpercentile(y, [5, 95])) fig.tight_layout() # Calculate excess given number of transits. # TODO: Do this somewhere else? scalar = 2.0 / (np.pi * sources["rv_nb_transits"]) model_rv_error_mu = np.sqrt(model_rv_sev_mu * scalar) model_rv_error_sigma_pos = \ np.sqrt((model_rv_sev_mu + model_rv_sev_sigma) * scalar) - model_rv_error_mu model_rv_error_sigma_neg = \ np.sqrt((model_rv_sev_mu - model_rv_sev_sigma) * scalar) - model_rv_error_mu x = sources["phot_rp_mean_flux"] y = sources["rv_single_epoch_variance"] idx = np.argsort(x) fig, ax = plt.subplots() ax.scatter(x, y, **scatter_kwds) ax.plot(x[idx], model_rv_sev_mu[idx], "r-") ax.fill_between( x[idx], (model_rv_sev_mu - model_rv_sev_sigma)[idx], (model_rv_sev_mu + model_rv_sev_sigma)[idx], facecolor="r", alpha=0.3, edgecolor="none") ax.set_xlim(np.nanmin(x), np.nanmax(x)) ax.set_ylim(-0.5, 2 * np.nanmax(model_rv_sev_mu + model_rv_sev_sigma)) ax.semilogx() ax.set_xlabel(r"\textrm{rp mean flux}") ax.set_ylabel(r"\textrm{single epoch radial velocity variance} $(\textrm{km}^2\,\textrm{s}^{-2})$") fig.tight_layout() x = sources["phot_rp_mean_flux"] y = sources["radial_velocity_error"] idx = np.argsort(x) fig, ax = plt.subplots() ax.scatter(x, y, **scatter_kwds) ax.plot(x[idx], model_rv_error_mu[idx], "r-") ax.fill_between( x[idx], (model_rv_error_mu + model_rv_error_sigma_pos)[idx], (model_rv_error_mu + model_rv_error_sigma_neg)[idx], facecolor="r", alpha=0.3, edgecolor="none") ax.set_xlim(np.nanmin(x), np.nanmax(x)) ax.set_ylim(-0.5, 2 * np.nanmax(model_rv_error_mu + model_rv_error_sigma_pos)) ax.semilogx() ax.set_xlabel(r"\textrm{rp mean flux}") ax.set_ylabel(r"\textrm{radial velocity error} $(\textrm{km\,s}^{-1})$") fig.tight_layout() # Plot the model residuals against different stellar properties. y = sources["rv_single_epoch_variance"] mu = model_rv_sev_mu sigma = model_rv_sev_sigma for label_name, description, is_semilogx in shorthand_parameters: fig, ax = plt.subplots() x = sources[label_name] idx = np.argsort(x) ax.scatter(x, y - mu, **scatter_kwds) ax.axhline(0, c="r") ax.fill_between(x[idx], -sigma[idx], +sigma[idx], facecolor="r", alpha=0.3, edgecolor="none") ax.set_xlim(np.nanmin(x), np.nanmax(x)) # TODO: set ylim abs_ylim = 2 * np.nanmax(sigma) ax.set_ylim(-abs_ylim, +abs_ylim) if is_semilogx: ax.semilogx() ax.set_xlabel(r"\textrm{{{0}}}".format(description)) ax.set_ylabel(r"\textrm{residual single epoch radial valocity variance} $(\textrm{km}^2\,\textrm{s}^{-2})$") fig.tight_layout() # Plot the binary fraction with different parameters. N_bins = 15 prob_binarity = np.mean(np.exp(p_samples["log_membership_probability"]), axis=0) for label_name, description, is_semilogx in shorthand_parameters: x = sources[label_name][used_in_fit] y = prob_binarity finite = np.isfinite(x * y) x, y = (x[finite], y[finite]) if is_semilogx: es_bins = np.logspace( np.log10(x.min()), np.log10(x.max()), N_bins) else: es_bins = np.linspace(x.min(), x.max(), N_bins) es_bin_centers = es_bins[:-1] + np.diff(es_bins)/2. es_binarity = np.zeros(es_bins.size - 1, dtype=float) es_binarity_error = np.zeros_like(es_binarity) for i, left_edge in enumerate(es_bins[:-1]): right_edge = es_bins[i + 1] in_bin = (right_edge > x) * (x >= left_edge) es_binarity[i] += np.sum((y > 0.5) * in_bin) \ / np.sum(in_bin) es_binarity_error[i] += np.sqrt(np.sum((y > 0.5) * in_bin)) \ / np.sum(in_bin) if is_semilogx: ed_bins = 10**np.percentile(np.log10(x), np.linspace(0, 100, N_bins)) else: ed_bins = np.percentile(x, np.linspace(0, 100, N_bins)) ed_bin_centers = ed_bins[:-1] + np.diff(ed_bins)/2. ed_binarity = np.zeros(ed_bins.size - 1, dtype=float) ed_binarity_error = np.zeros_like(ed_binarity) for i, left_edge in enumerate(ed_bins[:-1]): right_edge = ed_bins[i + 1] in_bin = (right_edge > x) * (x >= left_edge) ed_binarity[i] += np.sum((y > 0.5) * in_bin) \ / np.sum(in_bin) ed_binarity_error[i] += np.sqrt(np.sum((y > 0.5) * in_bin))\ / np.sum(in_bin) fig, axes = plt.subplots(1, 2, figsize=(8, 4)) xx, yy = velociraptor.mpl_utils.plot_histogram_steps( axes[0], es_bin_centers, es_binarity, es_binarity_error) axes[0].set_title(r"\textrm{equi-spaced bins}") xx, yy = velociraptor.mpl_utils.plot_histogram_steps( axes[1], ed_bin_centers, ed_binarity, ed_binarity_error) axes[1].set_title(r"\textrm{equi-density bins}") ylims = np.hstack([ax.get_ylim() for ax in axes]) for ax in axes: ax.set_xlim(xx[0], xx[-1]) ax.set_ylim(np.min(ylims), np.max(ylims)) if is_semilogx: ax.semilogx() ax.set_ylabel(r"\textrm{binary fraction}") ax.set_xlabel(r"\textrm{{{0}}}".format(description)) fig.tight_layout() # OK let's do this for a sample of giants and a sample of dwarfs and then # compare. qc = (sources["parallax"] > 0) \ * ((sources["parallax"]/sources["parallax_error"]) > 5) fig, ax = plt.subplots() ax.scatter(sources["bp_rp"], sources["absolute_g_mag"], **scatter_kwds) ax.set_ylim(ax.get_ylim()[::-1]) ax.set_xlabel(r"\textrm{bp - rp}") ax.set_ylabel(r"\textrm{absolute g magnitude}") fig.tight_layout() qc_giants = qc * (3 > sources["bp_rp"]) * (sources["bp_rp"] > 1.0) \ * (sources["absolute_g_mag"] < 3) qc_dwarfs = qc * (3 > sources["bp_rp"]) * (sources["bp_rp"] > 1.0) \ * (sources["absolute_g_mag"] > 4) fig, axes = plt.subplots(1, 2, figsize=(8, 4)) for ax in axes: ax.scatter( sources["bp_rp"][qc], sources["absolute_g_mag"][qc], **scatter_kwds) ax.set_ylim(ax.get_ylim()[::-1]) ax.set_xlabel(r"\textrm{bp - rp}") ax.set_ylabel(r"\textrm{absolute g magnitude}") axes[0].scatter( sources["bp_rp"][qc_dwarfs], sources["absolute_g_mag"][qc_dwarfs], s=5, zorder=10) axes[1].scatter( sources["bp_rp"][qc_giants], sources["absolute_g_mag"][qc_giants], s=5, zorder=10) axes[0].set_title(r"\textrm{main-sequence stars}") axes[1].set_title(r"\textrm{giant stars}") fig.tight_layout() print("Number of stars in dwarf model: {}".format(sum(qc_dwarfs))) print("Number of stars in giant model: {}".format(sum(qc_giants))) # Run the model for each subset. dwarf_model, dwarf_data_dict, dwarf_init_dict, dwarf_used_in_fit \ = velociraptor.prepare_model(**sources[qc_dwarfs]) dwarf_p_opt = dwarf_model.optimizing(data=dwarf_data_dict, init=dwarf_init_dict) dwarf_samples = dwarf_model.sampling(**velociraptor.stan.sampling_kwds( data=dwarf_data_dict, chains=2, iter=2000, init=dwarf_p_opt)) giant_model, giant_data_dict, giant_init_dict, giant_used_in_fit \ = velociraptor.prepare_model(**sources[qc_giants]) giant_p_opt = giant_model.optimizing(data=giant_data_dict, init=giant_init_dict) giant_samples = giant_model.sampling(**velociraptor.stan.sampling_kwds( data=giant_data_dict, chains=2, iter=2000, init=giant_p_opt)) # Plot the performance of the two models on the same figure. giant_model_rv_sev_mu, giant_model_rv_sev_sigma = \ velociraptor.predict_map_rv_single_epoch_variance(giant_samples, **sources[qc_giants]) dwarf_model_rv_sev_mu, dwarf_model_rv_sev_sigma = \ velociraptor.predict_map_rv_single_epoch_variance(dwarf_samples, **sources[qc_dwarfs]) shorthand_parameters = [ ("phot_rp_mean_flux", "mean rp flux", True, True), ("phot_bp_mean_flux", "mean bp flux", True, True), ("phot_g_mean_flux", "mean g flux", True, True), ("bp_rp", "bp-rp colour", False, True), ("phot_g_mean_mag", "mean g magnitude", False, True), ("phot_rp_mean_mag", "mean rp magnitude", False, True), ("phot_bp_mean_mag", "mean bp magnitude", False, True), ("teff_val", "inferred temperature", False, True), ("rv_template_teff", "rv template teff", False, False), ("rv_template_logg", "rv template logg", False, False), ("rv_template_fe_h", "rv template [Fe/H]", False, False), ] def _show_text_upper_right(ax, text): return ax.text(0.95, 0.95, text, transform=ax.transAxes, horizontalalignment="right", verticalalignment="top") def _show_number_of_data_points(ax, N): return _show_text_upper_right(ax, r"${0:.0f}$".format(N)) def _smooth_model(x, y, yerr, is_semilogx, average_function=np.mean, equidensity=True, N_smooth_points=30): N = min(N_smooth_points, len(set(x))) if equidensity: if is_semilogx: xb = 10**np.percentile(np.log10(x), np.linspace(0, 100, N)) else: xb = np.percentile(x, np.linspace(0, 100, N)) else: if is_semilogx: xb = np.logspace(np.log10(x.min()), np.log10(x.max()), N) else: xb = np.linspace(x.min(), x.max(), N) xi = xb[:-1] + np.diff(xb)/2. yi = np.zeros_like(xi) yerri = np.zeros_like(xi) for i, left_edge in enumerate(xb[:-1]): right_edge = xb[i + 1] in_bin = (right_edge > x) * (x >= left_edge) yi[i] = average_function(y[in_bin]) yerri[i] = average_function(yerr[in_bin]) return (xi, yi, yerri) for label_name, description, is_semilogx, equidensity in shorthand_parameters: fig, axes = plt.subplots(1, 3, figsize=(18, 6)) x = sources[label_name][qc_dwarfs][dwarf_used_in_fit] y = dwarf_model_rv_sev_mu idx = np.argsort(x) axes[0].scatter( x, sources["rv_single_epoch_variance"][qc_dwarfs][dwarf_used_in_fit], facecolor="r", s=1, alpha=0.05, rasterized=True) _show_number_of_data_points(axes[0], len(x)) # Smooth out the effects. xi, yi, yerri = _smooth_model( x, dwarf_model_rv_sev_mu, dwarf_model_rv_sev_sigma, is_semilogx, equidensity=equidensity) for ax in (axes[0], axes[2]): ax.plot(xi, yi, "r-") ax.fill_between(xi, yi - yerri, yi + yerri, facecolor="r", alpha=0.3, edgecolor="none") x = sources[label_name][qc_giants][giant_used_in_fit] y = giant_model_rv_sev_mu idx = np.argsort(x) axes[1].scatter( x, sources["rv_single_epoch_variance"][qc_giants][giant_used_in_fit], facecolor="b", s=1, alpha=0.05, rasterized=True) _show_number_of_data_points(axes[1], len(x)) xi, yi, yerri = _smooth_model( x, giant_model_rv_sev_mu, giant_model_rv_sev_sigma, is_semilogx, equidensity=equidensity) for ax in (axes[1], axes[2]): ax.plot(xi, yi, "b-") ax.fill_between( xi, yi - yerri, yi + yerri, facecolor="b", alpha=0.3, edgecolor="none") axes[0].set_title(r"\textrm{main-sequence stars}") axes[1].set_title(r"\textrm{giant stars}") x = sources[label_name][qc] xlims = (np.nanmin(x), np.nanmax(x)) for ax in axes: ax.set_xlabel(r"\textrm{{{0}}}".format(description)) ax.set_ylabel(r"\textrm{single epoch radial velocity variance} $(\textrm{km}^2\,\textrm{s}^{-2})$") if is_semilogx: ax.semilogx() ax.set_xlim(xlims) ax.set_ylim(axes[2].get_ylim()) _show_text_upper_right( axes[2], r"$N_\textrm{{model bins}} = {0}$ \textrm{{(equi-{1})}}".format( len(xi) + 1, "density" if equidensity else "spaced")) fig.tight_layout() fig.savefig( "figures/giant-vs-dwarf-{}.pdf".format(label_name.replace("_", "-")), dpi=150)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,283
andycasey/velociraptor
refs/heads/master
/run_analysis.py
""" Analysis script for the Velociraptor project. """ import logging import multiprocessing as mp import numpy as np import os import pickle import sys import tqdm import yaml from time import (sleep, time) from astropy.io import fits from scipy import optimize as op from scipy.special import logsumexp import george import npm_utils as npm import stan_utils as stan USE_SV_MASK = True if __name__ == "__main__": config_path = sys.argv[1] with open(config_path, "r") as fp: config = yaml.load(fp) random_seed = int(config["random_seed"]) np.random.seed(random_seed) logging.info(f"Config path: {config_path} with seed {random_seed}") # Check results path now so we don't die later. results_path = config["results_path"] # Load data. data = fits.open(config["data_path"])[1].data # Get a list of all relevant label names all_label_names = [] for model_name, model_config in config["models"].items(): all_label_names.append(model_config["predictor_label_name"]) all_label_names.extend(model_config["kdtree_label_names"]) all_label_names = list(np.unique(all_label_names)) # Mask for finite data points. finite = np.all([np.isfinite(data[ln]) for ln in all_label_names], axis=0) USE_SV_MASK = config["sv_mask"] if USE_SV_MASK: # Mask for science verifiation with open("sv.mask", "rb") as fp: sv_mask = pickle.load(fp) sv_mask = sv_mask[finite] # Load the model. model = stan.load_stan_model(config["model_path"], verbose=False) # Make sure that some entries have the right type. default_opt_kwds = config.get("optimisation_kwds", {}) for key in ("tol_obj", "tol_grad", "tol_rel_grad", "tol_rel_obj"): if key in default_opt_kwds: default_opt_kwds[key] = float(default_opt_kwds[key]) logging.info("Optimization keywords: {}".format(default_opt_kwds)) default_bounds = dict(bound_theta=[0.5, 1], bound_mu_single=[0.5, 15], bound_sigma_single=[0.05, 10], bound_sigma_multiple=[0.2, 1.6]) M = config["number_of_sources"] indices = np.random.choice(sum(finite), M, replace=False) model_results = dict() for model_name, model_config in config["models"].items(): logging.info(f"Running model {model_name} with config:\n{model_config}") bounds = default_bounds.copy() for k, (lower, upper) in model_config["bounds"].items(): bounds[f"bound_{k}"] = [lower, upper] # Set up a KD-tree. X = np.vstack([data[ln][finite] for ln in model_config["kdtree_label_names"]]).T Y = np.array(data[model_config["predictor_label_name"]])[finite] N, D = X.shape kdt, scales, offsets = npm.build_kdtree(X, relative_scales=model_config["kdtree_relative_scales"]) kdt_kwds = dict(offsets=offsets, scales=scales, full_output=True) kdt_kwds.update( minimum_radius=model_config["kdtree_minimum_radius"], maximum_radius=model_config.get("kdtree_maximum_radius", None), minimum_points=model_config["kdtree_minimum_points"], maximum_points=model_config["kdtree_maximum_points"], minimum_density=model_config.get("kdtree_minimum_density", None)) # Optimize the non-parametric model for those sources. results = np.zeros((M, 5)) done = np.zeros(M, dtype=bool) def optimize_mixture_model(index, inits=None, scalar=5): # Select indices and get data. d, nearby_idx, meta = npm.query_around_point(kdt, X[index], **kdt_kwds) y = Y[nearby_idx] ball = X[nearby_idx] if inits is None: inits = npm._get_1d_initialisation_point( y, scalar=scalar, bounds=model_config["bounds"]) # Update meta dictionary with things about the data. meta = dict(max_log_y=np.log(np.max(y)), N=nearby_idx.size, y_percentiles=np.percentile(y, [16, 50, 84]), ball_ptps=np.ptp(ball, axis=0), ball_medians=np.median(ball, axis=0), init_points=inits, kdt_indices=nearby_idx) data_dict = dict(y=y, N=y.size, scalar=scalar) data_dict.update(bounds) #for k, v in model_config["parameter_bounds"].items(): # data_dict["{}_bounds".format(k)] = v p_opts = [] ln_probs = [] for j, init_dict in enumerate(inits): opt_kwds = dict( init=init_dict, data=data_dict) opt_kwds.update(default_opt_kwds) # Do optimization. # TODO: Suppressing output is always dangerous. with stan.suppress_output(config.get("suppress_stan_output", True)) as sm: try: p_opt = model.optimizing(**opt_kwds) except: logging.exception(f"Exception occurred when optimizing index {index}"\ f" from {init_dict}:") else: if p_opt is not None: p_opts.append(p_opt) ln_probs.append(npm.ln_prob(y, 1, *npm._pack_params(**p_opt))) try: p_opt except UnboundLocalError: logging.warning("Stan failed. STDOUT & STDERR:") logging.warning("\n".join(sm.outputs)) if p_opt is None: stdout, stderr = sm.outputs logging.warning(f"STDOUT:\n{stdout}\nSTDERR:\n{stderr}") if len(p_opts) < 1: logging.warning("Optimization on index {} did not converge from any "\ "initial point trialled. Consider relaxing the "\ "optimization tolerances! If this occurs regularly "\ "then something is very wrong!".format(index)) return (index, None, meta) # evaluate best. else: idx = np.argmax(ln_probs) p_opt = p_opts[idx] meta["init_idx"] = idx """ if sum(done) > 550 and sum(done) < 570: theta, mu_single, sigma_single, mu_multiple, sigma_multiple = npm._pack_params(**p_opt) fig, ax = plt.subplots() xi = np.linspace(0, 20, 1000) y_s = npm.norm_pdf(xi, mu_single, sigma_single, theta) y_m = npm.lognorm_pdf(xi, mu_multiple, sigma_multiple, theta) ax.plot(xi, y_s, c="tab:blue") ax.plot(xi, y_m, c="tab:red") p_single = np.exp(np.log(y_s) - logsumexp([np.log(y_s), np.log(y_m)], axis=0)) ax.plot(xi, p_single, c="k") ax.set_title(f"{index}: {theta:.1e} {mu_single:.2f} {sigma_single:.2f} {sigma_multiple:.2f}") ax.hist(y, bins=np.linspace(0, 20, 20), alpha=0.5, facecolor="#666666", normed=True) if sum(done) > 570: raise a """ return (index, p_opt, meta) def sp_swarm(*sp_indices, **kwargs): logging.info("Running single processor swarm") with tqdm.tqdm(sp_indices, total=len(sp_indices)) as pbar: for j, index in enumerate(sp_indices): if done[j]: continue _, result, meta = optimize_mixture_model(index) pbar.update() done[j] = True if result is not None: results[j] = npm._pack_params(**result) return None def mp_swarm(*mp_indices, in_queue=None, out_queue=None, seed=None): np.random.seed(seed) swarm = True while swarm: try: j, index = in_queue.get_nowait() except mp.queues.Empty: logging.info("Queue is empty") break except StopIteration: logging.warning("Swarm is bored") break except: logging.exception("Unexpected exception:") break else: if index is None and init is False: swarm = False break try: _, result, meta = optimize_mixture_model(index) except: logging.exception(f"Exception when optimizing on {index}") out_queue.put((j, index, None, dict())) else: out_queue.put((j, index, result, meta)) return None if not config.get("multiprocessing", False): sp_swarm(*indices) else: P = mp.cpu_count() with mp.Pool(processes=P) as pool: manager = mp.Manager() in_queue = manager.Queue() out_queue = manager.Queue() swarm_kwds = dict(in_queue=in_queue, out_queue=out_queue) logging.info("Dumping everything into the queue!") for j, index in enumerate(indices): in_queue.put((j, index, )) j = [] for _ in range(P): j.append(pool.apply_async(mp_swarm, [], kwds=swarm_kwds)) with tqdm.tqdm(total=M) as pbar: while not np.all(done): # Check for output. try: r = out_queue.get(timeout=30) except mp.queues.Empty: logging.info("No results") break else: j, index, result, meta = r done[j] = True if result is not None: results[j] = npm._pack_params(**result) pbar.update(1) # Do not use bad results. # Bad results include: # - Things that are so clearly discrepant in every parameter. # - Things that are on the edge of the boundaries of parameter space. sigma = np.abs(results - np.median(results, axis=0)) \ / np.std(results, axis=0) sigma = np.sum(sigma, axis=1) tol_sigma = model_config["tol_sum_sigma"] tol_proximity = model_config["tol_proximity"] parameter_names = ( "theta", "mu_single", "sigma_single", "mu_multiple", "sigma_multiple") lower_bounds = np.array([model_config["bounds"].get(k, [-np.inf])[0] for k in parameter_names]) upper_bounds = np.array([model_config["bounds"].get(k, [+np.inf])[-1] for k in parameter_names]) not_ok_bound = np.any( (np.abs(results - lower_bounds) <= tol_proximity) \ + (np.abs(results - upper_bounds) <= tol_proximity), axis=1) not_ok_sigma = sigma > tol_sigma not_ok = not_ok_bound + not_ok_sigma print(f"There were {sum(not_ok_sigma)} results discarded for being outliers") print(f"There were {sum(not_ok_bound)} results discarded for being close to the edge") print(f"There were {sum(not_ok)} results discarded in total") model_indices = indices[~not_ok] results = results[~not_ok] # Run the gaussian process on the single star estimates. gp_block_size = 10000 G = 5 # number of kernel hyperparameters gp_predict_indices = (0, 1, 2, 3, 4) gp_parameters = np.zeros((len(gp_predict_indices), G)) gp_predictions = np.nan * np.ones((X.shape[0], 2 * len(gp_predict_indices))) x = X[model_indices] #randn = np.random.choice(X.shape[0], 50000, replace=False) for i, index in enumerate(gp_predict_indices): y = results[:, index] metric = np.var(x, axis=0) kernel = george.kernels.Matern32Kernel(metric, ndim=x.shape[1]) gp = george.GP(kernel, mean=np.mean(y), fit_mean=True, white_noise=np.log(np.std(y)), fit_white_noise=True) assert len(gp.parameter_names) == G def nll(p): gp.set_parameter_vector(p) ll = gp.log_likelihood(y, quiet=True) return -ll if np.isfinite(ll) else 1e25 def grad_nll(p): gp.set_parameter_vector(p) return -gp.grad_log_likelihood(y, quiet=True) gp.compute(x) logging.info("Initial \log{{L}} = {:.2f}".format(gp.log_likelihood(y))) logging.info("initial \grad\log{{L}} = {}".format(gp.grad_log_likelihood(y))) p0 = gp.get_parameter_vector() t_init = time() result = op.minimize(nll, p0, jac=grad_nll, method="L-BFGS-B") t_opt = time() - t_init gp.set_parameter_vector(result.x) logging.info("Result: {}".format(result)) logging.info("Final logL = {:.2f}".format(gp.log_likelihood(y))) logging.info("Took {:.0f} seconds to optimize".format(t_opt)) gp_parameters[i] = result.x # Predict the quantity and the variance. B = int(np.ceil(X.shape[0] / gp_block_size)) logging.info(f"Predicting {model_name} {index}") if USE_SV_MASK: p, p_var = gp.predict(y, X[sv_mask], return_var=True) gp_predictions[sv_mask, 2*i] = p gp_predictions[sv_mask, 2*i+1] = p_var else: with tqdm.tqdm(total=X.shape[0]) as pb: for b in range(B): s, e = (b * gp_block_size, (b + 1)*gp_block_size) p, p_var = gp.predict(y, X[s:1+e], return_var=True) gp_predictions[s:1+e, 2*i] = p gp_predictions[s:1+e, 2*i + 1] = p_var pb.update(e - s) """ p, p_var = gp.predict(y, X[randn], return_var=True) gp_predictions[randn, 2*i] = p gp_predictions[randn, 2*i + 1] = p_var import matplotlib.pyplot as plt fig, ax = plt.subplots() scat = ax.scatter(X.T[0][randn], X.T[1][randn], c=gp_predictions[:, 2*i][randn], s=1) cbar = plt.colorbar(scat) ax.set_title(f"{index} mu") fig, ax = plt.subplots() scat = ax.scatter(X.T[0][randn], X.T[1][randn], c=np.sqrt(gp_predictions[:, 2*i + 1][randn]), s=1) cbar = plt.colorbar(scat) ax.set_title(f"{index} sigma") """ model_results[model_name] = [model_indices, results, gp_parameters, gp_predictions] # Save predictions so far. logging.info(f"Saved progress to {results_path}") with open(results_path, "wb") as fp: pickle.dump(dict(config=config, models=model_results), fp) # Save the predictions, and the GP hyperparameters. save_dict = dict(config=config, models=model_results) with open(results_path, "wb") as fp: pickle.dump(save_dict, fp) logging.info(f"Saved output to {results_path}") raise a
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,284
andycasey/velociraptor
refs/heads/master
/_convert_config.py
""" Convert config format from run_analysis -> run_analysis2.py """ import os import sys import pickle import yaml if __name__ == "__main__": output_path = sys.argv[1] model_paths = sys.argv[2:] config = dict() models = dict() for i, model_path in enumerate(model_paths): model_name = os.path.basename(model_path).split(".")[1] with open(model_path, "r") as fp: model_config = yaml.load(fp) if i == 0: for key in ("data_path", "results_path", "model_path", "random_seed" "number_of_sources", "multiprocessing", "suppress_stan_output"): if key in model_config: config[key] = model_config[key] foo = dict() for key in ("kdtree_label_names", "kdtree_minimum_points", "kdtree_maximum_points", "kdtree_relative_scales", "kdtree_minimum_radius", "kdtree_maximum_radius"): foo[key] = model_config[key] foo["predictor_label_name"] = model_config["predictor_label_names"][0] models[model_name] = foo config["models"] = models with open(output_path, "w") as fp: fp.write(yaml.dump(config))
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,285
andycasey/velociraptor
refs/heads/master
/run_probabilities2.py
""" Run this after run_analysis.py """ import logging import numpy as np import sys import pickle import tqdm from collections import OrderedDict from astropy.io import fits from astropy.table import Table from scipy.special import logsumexp from scipy import stats import npm_utils as npm def lnprob(y, theta, s_mu, s_sigma, b_mu, b_sigma): s_ivar = np.abs(s_sigma)**-2 b_ivar = np.abs(b_sigma)**-2 hl2p = 0.5 * np.log(2*np.pi) s_lpdf = -hl2p + 0.5 * np.log(s_ivar) \ - 0.5 * (y - s_mu)**2 * s_ivar b_lpdf = -np.log(y*b_sigma) - hl2p \ - 0.5 * (np.log(y) - b_mu)**2 * b_ivar b_lpdf[~np.isfinite(b_lpdf)] = -np.inf return np.vstack([s_lpdf, b_lpdf]).T + np.log([theta, 1-theta]) if __name__ == "__main__": result_path = sys.argv[1] output_path = sys.argv[2] with open(result_path, "rb") as fp: results = pickle.load(fp) config = results["config"] data = fits.open(config["data_path"])[1].data all_label_names = [] for model_name, model_config in config["models"].items(): all_label_names.append(model_config["predictor_label_name"]) all_label_names.extend(model_config["kdtree_label_names"]) all_label_names = list(np.unique(all_label_names)) finite = np.all([np.isfinite(data[ln]) for ln in all_label_names], axis=0) K = config.get("number_of_draws", 100) model_names = config["models"] M = len(model_names) MJ = M + 1 if M > 1 else 1 N = sum(finite) lnprobs = np.zeros((M, N, K, 2)) p_single = np.zeros((MJ, N, K)) for m, model_name in enumerate(model_names): predictor_label_name = config["models"][model_name]["predictor_label_name"] y = np.array(data[predictor_label_name][finite]) model_indices, model_results, gp_parameters, gp_predictions \ = results["models"][model_name] theta, theta_var, \ mu_single, mu_single_var, \ sigma_single, sigma_single_var, \ mu_multiple, mu_multiple_var, \ sigma_multiple, sigma_multiple_var = gp_predictions.T # Calculate probabilities. rhos = np.corrcoef(model_results.T) print(f"Calculating probabilities for {model_name}") for n in tqdm.tqdm(range(N)): #TODO SKIPPING if not np.all(np.isfinite(gp_predictions[n])): continue mu = gp_predictions[n, ::2] diag = np.atleast_2d(gp_predictions[n, 1::2])**0.5 cov = diag * rhos * diag.T draws = np.random.multivariate_normal(mu, cov, size=K).T # Clip things to bounded values. draws[0] = np.clip(draws[0], 0.5, 1.0) draws[1] = np.clip(draws[1], 0.5, 15) draws[2] = np.clip(draws[2], 0.05, 10) draws[4] = np.clip(draws[4], 0.2, 1.6) draws_3_min = np.log(draws[1] + 5 * draws[2]) + draws[4]**2 draws[3] = np.max([draws[3], draws_3_min], axis=0) lnprobs[m, n, :, 0] = np.log(draws[0]) + npm.normal_lpdf(y[n], draws[1], draws[2]) lnprobs[m, n, :, 1] = np.log(1 - draws[0]) + npm.lognormal_lpdf(y[n], draws[3], draws[4]) p_single[m, n] = np.exp(lnprobs[m, n, :, 0] - logsumexp(lnprobs[m, n], axis=1)) # TODO HACK is_def_single = y[n] < draws[1] p_single[m, n][is_def_single] = 1.0 """ xi = np.linspace(0, 20, 1000) y_s = npm.norm_pdf(xi, np.mean(draws[1]), np.mean(draws[2]), np.mean(draws[0])) y_m = npm.lognorm_pdf(xi, np.mean(draws[3]), np.mean(draws[4]), np.mean(draws[0])) p_single = np.exp(np.log(y_s) - logsumexp([np.log(y_s), np.log(y_m)], axis=0)) fig, ax = plt.subplots() ax.plot(xi, y_s, c="tab:blue") ax.plot(xi, y_m, c="tab:red") ax.plot(xi, p_single, c="#000000") ax.set_title(n) ax.axvline(y[n], c="#666666") """ # Calculate joint probabilities. if M > 1: print("Calculating joint probabilities") for n in tqdm.tqdm(range(N)): # TODO: SkIPPING if not np.all(np.isfinite(gp_predictions[n])): continue for k in range(K): # TODO: this could be wrong,.. numerator = sum(lnprobs[:, n, k, 0]) denominator = sum(lnprobs[:, n, k, 1]) p_single[-1, n, k] = np.exp(numerator - logsumexp([numerator, denominator])) """ numerator = np.sum(lnprobs[:, d, :, 0], axis=0) denominator = np.sum(lnprobs[:, d, :, 1], axis=0) p_single[-1, d, :] = np.exp(numerator - logsumexp([numerator, denominator], axis=1)) """ print("Calculating percentiles") percentiles = [5, 50, 95] P = len(percentiles) p_single_percentiles = np.zeros((P, MJ, N)) for m in tqdm.tqdm(range(MJ)): for n in range(N): if not np.all(np.isfinite(gp_predictions[n])): continue p_single_percentiles[:, m, n] = np.percentile(p_single[m, n], percentiles) # Do a classification for each star. print("Classifying") confidence = np.sum(p_single > 0.5, axis=2)/K is_single = confidence > 0.5 #confidence[~is_single] = 1 - confidence[~is_single] print("Aggregating data") properties = OrderedDict() properties["source_id"] = data["source_id"][finite] for label_name in all_label_names: properties[label_name] = data[label_name][finite] # Do GP predictions. for m, model_name in enumerate(model_names): print(f"Aggregating predictions for {model_name}") _, __, ___, gp_predictions = results["models"][model_name] theta, theta_var, \ mu_single, mu_single_var, \ sigma_single, sigma_single_var, \ mu_multiple, mu_multiple_var, \ sigma_multiple, sigma_multiple_var = gp_predictions.T predictor_label_name = config["models"][model_name]["predictor_label_name"] properties[predictor_label_name] = np.array(data[predictor_label_name])[finite] properties[f"{model_name}_gp_theta"] = theta properties[f"{model_name}_gp_mu_s"] = mu_single properties[f"{model_name}_gp_sigma_s"] = sigma_single properties[f"{model_name}_gp_mu_m"] = mu_multiple properties[f"{model_name}_gp_sigma_m"] = sigma_multiple """ properties[f"{model_name}_gp_theta_var"] = theta_var properties[f"{model_name}_gp_mu_s_var"] = mu_single_var properties[f"{model_name}_gp_sigma_s_var"] = sigma_single_var properties[f"{model_name}_gp_mu_m_var"] = mu_multiple_var properties[f"{model_name}_gp_sigma_m_var"] = sigma_multiple_var """ properties[f"{model_name}_is_single"] = is_single[m].astype(int) properties[f"{model_name}_confidence"] = confidence[m] for p, percentile in enumerate(percentiles): properties[f"{model_name}_p{percentile:.0f}"] = p_single_percentiles[p, m] # Estimate RV semi-amplitude. if "rv" in model_names.keys(): print("Calculating radial velocity excess") properties[f"rv_excess"] = properties["rv_single_epoch_scatter"] \ - properties["rv_gp_mu_s"] rv_gp_mu_s_var = results["models"]["rv"][-1].T[3] rv_gp_sigma_s_var = results["models"]["rv"][-1].T[5] properties[f"rv_excess_var"] = properties["rv_gp_sigma_s"]**2 \ + rv_gp_mu_s_var \ + rv_gp_sigma_s_var # Joint probabilities. if M > 1: properties["joint_is_single"] = is_single[-1].astype(int) properties["joint_confidence"] = confidence[-1] for p, percentile in enumerate(percentiles): properties[f"joint_p{percentile:.0f}"] = p_single_percentiles[p, -1] print("Save PDF draws") # Only save finite predictions for now. mask = np.all(np.isfinite(results["models"]["rv"][-1]), axis=1) pdf_output_path = ".".join(output_path.split(".")[:-1]) + f".pdf.memmap" pdf_sources_output_path = ".".join(output_path.split(".")[:-1]) + f".sources.memmap" pdf_output = np.memmap(pdf_output_path, dtype=np.float32, shape=(MJ, sum(mask), K), mode="w+") pdf_output[:] = p_single[:, mask, :] del pdf_output pdf_sources_output = np.memmap(pdf_sources_output_path, shape=(sum(mask), ), dtype='>i8', mode="w+") pdf_sources_output[:] = data["source_id"][finite][mask] del pdf_sources_output print("Writing output file..") Table(data=properties).write(output_path, overwrite=True) print(f"Output written to {output_path}")
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,286
andycasey/velociraptor
refs/heads/master
/velociraptor.py
""" Code for the velociraptor project. """ import logging as logger import numpy as np import matplotlib.pyplot as plt from astropy.table import Table from matplotlib.ticker import MaxNLocator from scipy.stats import norm from astropy.io import fits import stan_utils as stan import mpl_utils plt.style.use(mpl_utils.mpl_style) np.random.seed(42) def load_gaia_sources(path, **kwargs): """ Load a subset of Gaia data and calculate additional properties like `absolute_g_mag`, `absolute_bp_mag`, `absolute_rp_mag`, and `rv_single_epoch_variance`. :param path: The local path to load the sources from. """ #try: # sources = Table(fits.open(path)[1].data) #except: sources = Table.read(path, **kwargs) for band in ("g", "bp", "rp"): sources["absolute_{}_mag".format(band)] = \ sources["phot_{}_mean_mag".format(band)] \ + 5 * np.log10(sources["parallax"]/100.0) sources["rv_single_epoch_variance"] = sources["radial_velocity_error"]**2 \ * sources["rv_nb_transits"] * np.pi/2.0 # Approximate temperature from bp-rp colour use_in_fit = np.isfinite(sources["radial_velocity"]) \ * (sources["phot_bp_rp_excess_factor"] < 1.5) \ * np.isfinite(sources["bp_rp"]) \ * np.isfinite(sources["teff_val"]) \ * (sources["bp_rp"] < 2.5) \ * (sources["bp_rp"] > 0.5) x = sources["bp_rp"][use_in_fit] y = sources["teff_val"][use_in_fit] coeff = np.polyfit(1.0/x, y, 2) sources["approx_teff_from_bp_rp"] = np.clip( np.polyval(coeff, 1.0/sources["bp_rp"]), 3500, 8000) return sources def load_gaia_sources_from_fitsio(path, **kwargs): """ Load a subset of Gaia data and calculate additional properties like `absolute_g_mag`, `absolute_bp_mag`, `absolute_rp_mag`, and `rv_single_epoch_variance`. :param path: The local path to load the sources from. """ sources = Table(fits.open(path)[1].data) metadata = dict() for band in ("g", "bp", "rp"): metadata["absolute_{}_mag".format(band)] = \ sources["phot_{}_mean_mag".format(band)] \ + 5 * np.log10(sources["parallax"]/100.0) metadata["rv_single_epoch_variance"] = sources["radial_velocity_error"]**2 \ * sources["rv_nb_transits"] * np.pi/2.0, # Approximate temperature from bp-rp colour use_in_fit = np.isfinite(sources["radial_velocity"]) \ * (sources["phot_bp_rp_excess_factor"] < 1.5) \ * np.isfinite(sources["bp_rp"]) \ * np.isfinite(sources["teff_val"]) \ * (sources["bp_rp"] < 2.5) \ * (sources["bp_rp"] > 0.5) x = sources["bp_rp"][use_in_fit] y = sources["teff_val"][use_in_fit] coeff = np.polyfit(1.0/x, y, 2) metadata["approx_teff_from_bp_rp"] = np.clip( np.polyval(coeff, 1.0/sources["bp_rp"]), 3500, 8000) return (sources, metadata) def prepare_model(rv_single_epoch_variance, target=0.01, model_path="model.stan", S=None, design_matrix_function=None, mask=None, **source_params): """ Compile the Stan model, and prepare the data and initialisation dictionaries for optimization and sampling. :param rv_single_epoch_variance: The variance in single epoch measurements of radial velocity. :param target: [optional] The target radial velocity variance to use when initialising the model coefficients. :param model_path: [optional] The local path of the Stan model. :param S: [optional] If not `None`, draw a random `S` valid sources from the data rather than using the full data set. :Keyword Arguments: * *source_params* (``dict``) These are passed directly to the `_rvf_design_matrix`, so they should include all of the source labels needed to construct the design matrix (e.g., `phot_rp_mean_flux`). The array length for each source label should match that of `rv_single_epoch_variance`. :returns: The compiled Stan model, the data dictionary, the initialsiation dictionary, and a mask corresponding to which `sources` were used to construct the data dictionary. """ data, indices = prepare_data(rv_single_epoch_variance, S=S, mask=mask, design_matrix_function=design_matrix_function, **source_params) coeff = _rvf_initial_coefficients(data["design_matrix"].T, target=target) init = dict(theta=0.1, mu_coefficients=coeff, sigma_coefficients=coeff) model = stan.load_stan_model(model_path) return (model, data, init, indices) def prepare_data(rv_single_epoch_variance, S=None, mask=None, design_matrix_function=None, **source_params): if design_matrix_function is None: design_matrix_function = _rvf_design_matrix dm = design_matrix_function(**source_params) finite = np.all(np.isfinite(dm), axis=0) \ * np.isfinite(rv_single_epoch_variance) if mask is not None: finite *= mask indices = np.where(finite)[0] N = sum(finite) if S is not None: indices = np.random.choice(indices, size=int(min(N, S)), replace=False) if (S is not None and S > N) or not all(finite): logger.warn("Excluding non-finite entries in design matrix! "\ "Number of data points: {0}".format(indices.size)) dm = dm[:, indices] data = dict(N=indices.size, rv_variance=rv_single_epoch_variance[indices], design_matrix=dm.T, M=dm.shape[0]) return data, indices def _rvf_design_matrix(phot_rp_mean_flux, bp_rp, **kwargs): """ Design matrix for the radial velocity floor variance. # TODO: Should we check for finite-ness here? """ return np.array([ np.ones(len(phot_rp_mean_flux)), phot_rp_mean_flux**-1, phot_rp_mean_flux**-2, bp_rp**-1, bp_rp**-2 ]) def _rvf_initial_coefficients(design_matrix, target=0.01): """ Initial coefficients for the model. """ return target / (design_matrix.shape[0] * np.nanmean(design_matrix, axis=1)) def predict_map_rv_single_epoch_variance(samples, **source_params): """ Predict the maximum a-posteriori estimate of the radial velocity variance from a single epoch. :param samples: The Stan chains from the model. """ params = samples.extract(("mu_coefficients", "sigma_coefficients")) dm = _rvf_design_matrix(**source_params) mu = np.dot(np.mean(params["mu_coefficients"], axis=0), dm) sigma = np.dot(np.mean(params["sigma_coefficients"], axis=0), dm) return (mu, sigma) def plot_model_predictions_corner(samples, sources=None, parameter_limits=None, log_parameters=None, N=100, labels=None, **kwargs): """ Make a corner plot showing the maximum a posteori radial velocity variance for different (stellar) properties that contribute to the model. :param samples: The MCMC samples. :param sources: [optional] A table of Gaia sources. This is used to determine the bounds on the parameters. If `None` is given, then `parameter_limits` should be given instead. If `sources` and `parameter_limits` are given, then the `parameter_limits` will supercede those calculated from `sources. :param parameter_limits: [optional] A dictionary containing source parameters as keys and a two-length tuple containing the lower and upper bounds of the parameter. :param log_parameters: [optional] A tuple containing the parameter names that should be shown in log space "" """ parameter_names = tuple( set(_rvf_design_matrix.__code__.co_varnames).difference(["kwargs"])) limits = dict() if sources is not None: for pn in parameter_names: limits[pn] = (np.nanmin(sources[pn]), np.nanmax(sources[pn])) else: missing = tuple(set(parameter_names).difference(parameter_limits)) if len(missing) > 0: raise ValueError("missing parameter limits for {}".format( ", ".join(missing))) if parameter_limits is not None: limits.update(parameter_limits) if log_parameters is None: log_parameters = [] if labels is None: labels = dict() def mesh(parameter_name): v = limits[parameter_name] s, e = (np.min(v), np.max(v)) if parameter_name not in log_parameters: return np.linspace(s, e, N) else: return np.logspace(np.log10(s), np.log10(e), N) samples_kwd = kwargs.get("samples_kwd", "mu_coefficients") coefficients = np.mean(samples.extract((samples_kwd, ))[samples_kwd], axis=0) P = len(limits) # Calculate the expected radial velocity variance for all combinations of # parameters. combinations = np.meshgrid(*[mesh(pn) for pn in parameter_names]) grid_combinations = np.vstack([comb.flatten() for comb in combinations]) expectation = np.dot(coefficients, _rvf_design_matrix(**dict(zip( parameter_names, grid_combinations)))) fig, axes = plt.subplots(P, P, figsize=(6 * P, 6 * P)) axes = np.atleast_2d(axes) for i, x_param in enumerate(parameter_names): for j, y_param in enumerate(parameter_names): ax = axes[j, i] if i > j: ax.set_visible(False) continue elif i == j: x = grid_combinations[i] # Get the mean at each unique x. x_uniques = np.sort(np.unique(x)) y_percentiles = np.zeros((3, x_uniques.size), dtype=float) for k, x_unique in enumerate(x_uniques): match = (grid_combinations[i] == x_unique) y_percentiles[:, k] = np.percentile( expectation[match], [0, 50, 100]) ax.plot(x_uniques, y_percentiles[1], "r-") ax.fill_between( x_uniques, y_percentiles[0], y_percentiles[2], facecolor="r", alpha=0.3, edgecolor="none") if x_param in log_parameters: ax.semilogx() ax.set_xlabel(labels.get(x_param, x_param)) ax.set_ylabel(r"\textrm{single epoch radial velocity variance}"\ r" $(\textrm{km}^2\,\textrm{s}^{-2})$") else: x, y = grid_combinations[[i, j]] #_x = np.log10(x) if x_param in log_parameters else x #_y = np.log10(y) if y_param in log_parameters else y imshow_kwds = dict(cmap="Reds", aspect="equal", extent=(np.min(x), np.max(x), np.max(y), np.min(y))) if x_param in log_parameters: ax.semilogx() if y_param in log_parameters: ax.semilogy() ax.imshow(expectation.reshape((N, N)), **imshow_kwds) ax.set_xlabel(labels.get(x_param, x_param)) ax.set_ylabel(labels.get(y_param, y_param)) fig.tight_layout() return fig
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,287
andycasey/velociraptor
refs/heads/master
/data/download.py
""" Query used to retrieve the radial velocity calibration data from the Gaia archive. """ from astroquery.gaia import Gaia columns = ( "source_id", "ra", "dec", "l", "b", "parallax", "parallax_error", "phot_g_mean_mag", "phot_bp_mean_mag", "phot_rp_mean_mag", "phot_g_mean_flux", "phot_g_mean_flux_error", "phot_bp_mean_flux", "phot_bp_mean_flux_error", "phot_bp_n_obs", "phot_rp_mean_flux", "phot_rp_mean_flux_error", "phot_rp_n_obs", "phot_bp_rp_excess_factor", "bp_rp", "bp_g", "g_rp", "teff_val", "teff_percentile_lower", "teff_percentile_upper", "a_g_val", "a_g_percentile_lower", "a_g_percentile_upper", "radial_velocity", "radial_velocity_error", "rv_nb_transits", "rv_template_teff", "rv_template_logg", "rv_template_fe_h", "astrometric_weight_al", "astrometric_gof_al", "astrometric_chi2_al", "phot_variable_flag" ) job = Gaia.launch_job_async(""" SELECT {0} FROM gaiadr2.gaia_source WHERE radial_velocity IS NOT NULL AND duplicated_source = 'false' AND rv_nb_transits > 10 AND visibility_periods_used > 10 AND radial_velocity_error < 20 AND MOD(random_index, 10) = 0 """.format(", ".join(columns))) subset = job.get_results() subset.write("rv_calibration_floor_subset-result.fits") job = Gaia.launch_job_async(""" SELECT {0} FROM gaiadr2.gaia_source WHERE radial_velocity IS NOT NULL AND duplicated_source = 'false' AND rv_nb_transits > 10 AND visibility_periods_used > 10 AND radial_velocity_error < 20 """.format(", ".join(columns))) sources = job.get_results() sources.write("rv_calibration_floor-result.fits") all_possible_sources = Gaia.launch_job_async(""" SELECT * FROM gaiadr2.gaia_source WHERE phot_rp_mean_mag < 13 """) """ SELECT * FROM gaiadr2.gaia_source WHERE radial_velocity IS NOT NULL AND duplicated_source = 'false' AND rv_nb_transits > 10 AND visibility_periods_used > 10 AND radial_velocity_error < 20 AND mod(random_index, 10) = 0 """
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,288
andycasey/velociraptor
refs/heads/master
/create_sv_mask.py
""" Create a mask that will be useful for science verification. """ import numpy as np import pickle from astropy.io import fits from astropy.table import Table data = fits.open("data/gaia-sources-for-npm-colsubset.fits")[1].data cross_match_paths = [ "data/soubiran-2013-xm-gaia.fits", "data/sb9_xm_gaia.fits", "data/apw-highK-unimodal-xm-gaia.fits", "data/apw-lnK-percentiles-xm-gaia.fits", "data/huang-apogee-rv-standards-xm-gaia.fits" ] def cross_match(A_source_ids, B_source_ids): A = np.array(A_source_ids, dtype=np.long) B = np.array(B_source_ids, dtype=np.long) ai = np.where(np.in1d(A, B))[0] bi = np.where(np.in1d(B, A))[0] assert len(ai) == len(bi) ai = ai[np.argsort(A[ai])] bi = bi[np.argsort(B[bi])] assert all(A[ai] == B[bi]) return (ai, bi) mask = np.zeros(len(data), dtype=bool) for path in cross_match_paths: t = Table.read(path) t = t.group_by("source_id") t = t[t.groups.indices[:-1]] vl_ids, t_ids = cross_match(data["source_id"], t["source_id"]) mask[vl_ids] = True print(f"Cross-matching with {path} revealed {len(vl_ids)} sources") with open("sv.mask", "wb") as fp: pickle.dump(mask, fp) print(f"There are {sum(mask)} sources for science verification in sv.mask")
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,289
andycasey/velociraptor
refs/heads/master
/prepare_data.py
""" Calculate additional properties (e.g., photometric and astrometric variability) which we will use in our model, and create a subset file that only contains the columns we need. """ import numpy as np from astropy.io import fits from astropy.table import Table input_path = "data/gaia-sources-for-npm.fits" output_path = "data/gaia-sources-for-npm-subset.fits" def create_subset_for_npm(path, hdu=1, additional_label_names=None, **kwargs): properties = dict() with fits.open(path) as image: sources = image[hdu].data for band in ("g", "bp", "rp"): # Absolute magnitudes, and photometric variability properties[f"absolute_{band}_mag"] = sources[f"phot_{band}_mean_mag"] \ + 5 * np.log10(sources["parallax"]/100.0) properties[f"phot_{band}_variability"] = np.sqrt(sources["astrometric_n_good_obs_al"]) \ * sources[f"phot_{band}_mean_flux"] \ / sources[f"phot_{band}_mean_flux_error"] # Radial velocity scatter properties["rv_single_epoch_variance"] = sources["radial_velocity_error"]**2 \ * sources["rv_nb_transits"] * np.pi/2.0 properties["rv_single_epoch_scatter"] = properties["rv_single_epoch_variance"]**0.5 # Astrometric unit weight error properties["astrometric_unit_weight_error"] = np.sqrt( sources["astrometric_chi2_al"]/(sources["astrometric_n_good_obs_al"] - 5)) for label_name in additional_label_names: properties[label_name] = sources[label_name] return properties data = create_subset_for_npm( input_path, hdu=1, additional_label_names=( "source_id", "ra", "dec", "phot_rp_mean_mag", "phot_bp_mean_mag", "phot_g_mean_mag", "phot_rp_mean_flux", "phot_g_mean_flux", "bp_rp", "rv_nb_transits", "parallax", "parallax_error", )) t = Table(data=data) t.write(output_path, overwrite=True)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,290
andycasey/velociraptor
refs/heads/master
/scripts/npm_run_soubiran_2013.py
""" Set up and run the non-parametric model. """ import numpy as np import os import multiprocessing as mp import pickle import yaml import tqdm import logging from time import sleep from astropy.io import fits import npm_utils as npm import stan_utils as stan with open("the-battery-stars.rv.yaml", "r") as fp: config = yaml.load(fp) config["results_path"] = "results/Soubiran_2013.rv_single_epoch_scatter.pkl" with open("the-battery-stars.astrometry.yaml", "r") as fp: config = yaml.load(fp) config["results_path"] = "results/Soubiran_2013.astrometric_unit_weight_error.pkl" # Load in the data. data = fits.open(config["data_path"])[1].data all_label_names = list(config["kdtree_label_names"]) \ + list(config["predictor_label_names"]) # Set up a KD-tree. X = np.vstack([data[ln] for ln in config["kdtree_label_names"]]).T finite = np.all([np.isfinite(data[ln]) for ln in all_label_names], axis=0) finite_indices = np.where(finite)[0] N, D = X.shape F = finite_indices.size L = 4 * len(config["predictor_label_names"]) + 1 # + 1 if using mu_multiple_uv C = config["share_optimised_result_with_nearest"] kdt, scales, offsets = npm.build_kdtree( X[finite], relative_scales=config["kdtree_relative_scales"]) kdt_kwds = dict(offsets=offsets, scales=scales, full_output=False) kdt_kwds.update( minimum_radius=config["kdtree_minimum_radius"], minimum_points=config["kdtree_minimum_points"], maximum_points=config["kdtree_maximum_points"], minimum_density=config.get("kdtree_minimum_density", None)) model = stan.load_stan_model(config["model_path"], verbose=False) default_opt_kwds = config.get("optimisation_kwds", {}) # Make sure that some entries have the right units. for key in ("tol_obj", "tol_grad", "tol_rel_grad", "tol_rel_obj"): if key in default_opt_kwds: default_opt_kwds[key] = float(default_opt_kwds[key]) logging.info("k-d tree keywords: {}".format(kdt_kwds)) logging.info("optimization keywords: {}".format(default_opt_kwds)) done = np.zeros(N, dtype=bool) queued = np.zeros(N, dtype=bool) results = np.nan * np.ones((N, L), dtype=float) default_init = dict(zip( ("theta", "mu_single", "sigma_single", "mu_multiple", "sigma_multiple"), np.array([0.75, 1, 0.5, 1, 0.75]))) default_init["mu_multiple_uv"] = 0.1 default_init = npm._check_params_dict(default_init) bounds = config["parameter_bounds"] def optimize_mixture_model(index, init=None): # Select indices and get data. indices = finite_indices[npm.query_around_point(kdt, X[index], **kdt_kwds)] y = np.array([data[ln][indices] for ln in config["predictor_label_names"]]).T if init is None: init = npm.get_initialization_point(y) data_dict = dict(y=y, N=y.shape[0], D=y.shape[1], max_log_y=np.log(np.max(y))) for k, v in bounds.items(): data_dict["{}_bounds".format(k)] = v trial_results = [] for j, init_dict in enumerate((init, npm.get_initialization_point(y), "random")): # CHeck that the parameters are bounded? if isinstance(init_dict, dict): if bounds is not None: for k, (lower, upper) in bounds.items(): if not (upper > init_dict[k] > lower): logging.info("Clipping initial value of {} from {} to within ({}, {})".format( k, init_dict[k], lower, upper)) offset = 0.01 * (upper - lower) init_dict[k] = np.clip(init_dict[k], lower + offset, upper - offset) opt_kwds = dict( init=init_dict, data=data_dict) opt_kwds.update(default_opt_kwds) # Do optimization. with stan.suppress_output() as sm: try: p_opt = model.optimizing(**opt_kwds) except: p_opt = None # TODO: Consider relaxing the optimization tolerances! if p_opt is None: # Capture stdout and stderr so we can read it later. stdout, stderr = sm.stdout, sm.stderr trial_results.append(p_opt) if p_opt is None: logging.warning("Exception when optimizing on index {} from "\ "initial point {}:".format(index, init_dict)) logging.warning(stdout) logging.warning(stderr) raise else: tolerance = 1e-2 for k, v in p_opt.items(): if k not in bounds \ or (k == "theta" and (v >= 1-tolerance)): continue lower, upper = bounds[k] if np.abs(v - lower) <= tolerance \ or np.abs(v - upper) <= tolerance: logging.warning("Optimised {} at edge of grid ({} < {} < {})" " - ignoring".format(k, lower, v, upper)) break else: break else: # TODO: Consider relaxing optimization tolerances! logging.warning("Optimization did not converge from any initial point "\ "trialled. Consider relaxing optimization tolerances!") import matplotlib.pyplot as plt fig, ax = plt.subplots() ax.hist(y, bins=100, facecolor="#cccccc", normed=True) # Plot a model at some optimization point. init = npm.get_initialization_point(y) N, D = y.shape xi = np.linspace(0, max(y), 1000) ax.plot(xi, npm.norm_pdf(xi, init["mu_single"], init["sigma_single"], init["theta"]), c="r") ax.plot(xi, npm.lognorm_pdf(xi, init["mu_multiple"], init["sigma_multiple"], init["theta"]), c="r", linestyle=":") ax.plot(xi, npm.norm_pdf(xi, default_init["mu_single"], default_init["sigma_single"], default_init["theta"]), c="b") ax.plot(xi, npm.lognorm_pdf(xi, default_init["mu_multiple"], default_init["sigma_multiple"], default_init["theta"]), c="b", linestyle=":") raise a p_opt = npm._check_params_dict(p_opt) return (index, p_opt, indices) def sp_swarm(*indices, **kwargs): logging.info("Running single processor swarm") with tqdm.tqdm(indices, total=len(indices)) as pbar: for index in indices: if done[index]: continue _, result, kdt_indices = optimize_mixture_model(index, default_init) pbar.update() if result is not None: done[index] = True results[index] = npm._pack_params(**result) if C > 0: nearby_indices = np.atleast_1d(kdt_indices[:C + 1]) updated = nearby_indices.size - sum(done[nearby_indices]) done[nearby_indices] = True results[nearby_indices] = npm._pack_params(**result) pbar.update(updated) return None def mp_swarm(*indices, max_random_starts=3, in_queue=None, candidate_queue=None, out_queue=None): def _random_index(): yield from np.random.choice(indices, max_random_starts, replace=False) _ri = _random_index() random_start = lambda *_: (_ri.__next__(), default_init) swarm = True while swarm: for func in (in_queue.get_nowait, random_start): try: index, init = func() except mp.queues.Empty: #logging.info("Using a random index to start") continue except StopIteration: logging.warning("Swarm is bored") sleep(5) except: logging.exception("Unexpected exception:") swarm = False else: if index is None and init is False: swarm = False break try: _, result, kdt_indices = optimize_mixture_model(index, init) except: logging.exception("Exception when optimizing on {} from {}"\ .format(index, init)) break out_queue.put((index, result)) if result is not None: if C > 0: # Assign the closest points to have the same result. # (On the other end of the out_qeue we will deal with # multiple results.) out_queue.put((kdt_indices[:C + 1], result)) # Candidate next K points K = 0 #candidate_queue.put((kdt_indices[C + 1:C + 1 + K], result)) break return None from astropy.table import Table soubiran = Table.read("data/Soubiran_2013-xm-Gaia.fits") soubiran[~np.isfinite(soubiran["source_id"])] = -1 soubiran["source_id"] = soubiran["source_id"].astype(np.int64) indices = np.in1d(data["source_id"], soubiran["source_id"]) \ * np.all(np.isfinite(X), axis=1) indices = np.where(indices)[0] sp_swarm(*indices) # Save results. results_path = config.get("results_path", "results.pkl") with open(results_path, "wb") as fp: pickle.dump(results, fp, -1) logging.info("Saved results to {}".format(results_path)) with open("results/Soubiran_2013.rv_single_epoch_scatter.pkl", "rb") as fp: rv_results = pickle.load(fp) with open("results/Soubiran_2013.astrometric_unit_weight_error.pkl", "rb") as fp: astrometric_results = pickle.load(fp) from astropy.table import Table subset = Table(data[indices]) keys = ("theta", "mu_single", "sigma_single", "mu_multi", "sigma_multi") for i, key in enumerate(keys): subset["rv_{}".format(key)] = rv_results[indices, i] subset["astrometric_{}".format(key)] = astrometric_results[indices, i] finite = np.isfinite(subset["astrometric_theta"] * subset["rv_theta"]) subset = subset[finite] print(len(subset)) subset.write("results/Soubiran_2013.results.fits", overwrite=True)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,291
andycasey/velociraptor
refs/heads/master
/_convert_pkl_to_hdf5.py
""" Convert pickle file to HDF5 + yaml. """ import h5py import sys import pickle import yaml if __name__ == "__main__": input_path = sys.argv[1] output_prefix = sys.argv[2] with open(input_path, "rb") as fp: results = pickle.load(fp) # Immediately convert to h5. h = h5py.File(f"{output_prefix}.hdf5", "w") group = h.create_group("models") #group.attrs.update(results["config"]) for model_name in results["models"].keys(): sub_group = group.create_group(model_name) dataset_names = ( "data_indices", "mixture_model_results", "gp_parameters", "gp_predictions" ) for i, dataset_name in enumerate(dataset_names): d = sub_group.create_dataset(dataset_name, data=results["models"][model_name][i]) h.close() with open(f"{output_prefix}.meta", "w") as fp: fp.write(yaml.dump(results["config"])) print(f"Created {output_prefix}.hdf5 and {output_prefix}.meta") del results
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,292
andycasey/velociraptor
refs/heads/master
/scripts/npm_collect_the_battery_stars.py
import pickle import yaml import numpy as np from astropy.io import fits from astropy.table import Table with open("data/the-battery-stars.indices.pkl", "rb") as fp: battery_star_indices = pickle.load(fp) with open("results/the-battery-stars.astrometric_unit_weight_error.pkl", "rb") as fp: astrometric_results = pickle.load(fp) with open("results/the-battery-stars.rv_single_epoch_scatter.pkl", "rb") as fp: rv_results = pickle.load(fp) with open("the-battery-stars.astrometry.yaml", "r") as fp: config = yaml.load(fp) # Load in the data. data = fits.open(config["data_path"])[1].data subset = Table(data[battery_star_indices]) keys = ("theta", "mu_single", "sigma_single", "mu_multi", "sigma_multi") for i, key in enumerate(keys): subset["rv_{}".format(key)] = rv_results[battery_star_indices, i] subset["astrometric_{}".format(key)] = astrometric_results[battery_star_indices, i] finite = np.isfinite(subset["astrometric_theta"] * subset["rv_theta"]) subset = subset[finite] print(len(subset)) subset.write("results/the-battery-stars.results.fits", overwrite=True)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,293
andycasey/velociraptor
refs/heads/master
/article/figures/plot_rv_figures.py
import numpy as np import os import pickle from astropy.io import fits from astropy.table import Table from scipy.stats import binned_statistic, binned_statistic_2d import matplotlib import matplotlib.pyplot as plt from matplotlib.ticker import MaxNLocator from mpl_toolkits.axes_grid1 import make_axes_locatable from mpl_utils import (mpl_style, plot_binned_statistic, plot_histogram_steps) plt.style.use(mpl_style) BASE_PATH = "../../" RELEASE_CANDIDATE_VERSION = 100 MAKE_FIGURES = [ #"joint_p50_mainsequence_median", #"joint_p50_mainsequence_mean", #"joint_p50_wrt_kdt_labels", #"rv_mean_joint_p50_wrt_kdt_labels", #"ast_mean_joint_p50_wrt_kdt_labels", #"mean_joint_p50_wrt_kdt_labels", #"all_joint_p50_wrt_kdt_labels", #"hrd_single_star_fraction_hist", #"ast_hrd_single_star_fraction_hist", #"rv_hrd_single_star_fraction_hist", #"rv_hrd_single_star_fraction_scatter", #"rv_sb9_kalpha_comparison", "rv_sb9_kalpha_comparison_tau", "rv_sb9_k_comparison", "rv_sb9_kalpha_comparison_period", "rv_sb9_kp_corner", "rv_apw_unimodal_k_comparison", "rv_apw_unimodal_kalpha_comparison", "rv_apw_percentiles_k_comparison", "rv_gp_hrd", "rv_gp_wrt_params", "ast_gp_hrd", "ast_gp_wrt_params", "rv_soubiran_hist", "rv_huang_hist", "rv_huang_hist2" ] #MAKE_FIGURES = [ # "ast_gp_hrd", #] DEFAULT_SEQUENTIAL_CMAP = "viridis" """ velociraptor = fits.open(os.path.join( BASE_PATH, "results", "velociraptor-catalog.rc.{:.0f}.fits".format(RELEASE_CANDIDATE_VERSION))) velociraptor = velociraptor[1].data """ try: velociraptor except NameError: velociraptor = fits.open("../../results/die-hard-subset.fits")[1].data latex_labels = dict( bp_rp=r"\textrm{bp - rp}", phot_rp_mean_mag=r"\textrm{apparent rp mag}", absolute_rp_mag=r"\textrm{absolute rp mag}", absolute_g_mag=r"\textrm{absolute g mag}", rv_mu_single=r"$\mu_s$ \textrm{/ km\,s}$^{-1}$", rv_sigma_single=r"$\sigma_s$ \textrm{/ km\,s}$^{-1}$", rv_p50=r"$\tau_\textrm{rv,single}$", ast_p50=r"$\tau_\textrm{ast,single}$", rv_single_epoch_scatter=r"\textrm{radial velocity jitter / km\,s}$^{-1}$", astrometric_unit_weight_error=r"\textrm{astrometric jitter}", ast_gp_mu_s=r"$\mu_\textrm{s,ast}$", ast_gp_sigma_s=r"$\sigma_\textrm{s,ast}$", rv_gp_mu_s=r"$\mu_{rv,s}\,\textrm{km\,s}^{-1}$", rv_gp_sigma_s=r"$\sigma_{rv,s}\,\textrm{km\,s}^{-1}$", ) common_kwds = dict([ ("bp_rp.limits", (-0.25, 6.25)), ("phot_rp_mean_mag.limits", (13.0, 2.1)), ("absolute_g_mag.limits", (10, -8)), ("absolute_rp_mag.limits", (10.5, -16)), ]) one_to_one_line_kwds = dict(c="#666666", linestyle=":", lw=1, zorder=-1) def savefig(fig, basename): prefix =f"v{RELEASE_CANDIDATE_VERSION}-{basename}" fig.savefig(f"{prefix}.png") #fig.savefig(f"{prefix}.png", dpi=150) print(f"Saved figure {prefix}.png") #plt.close("all") def cross_match(A_source_ids, B_source_ids): A = np.array(A_source_ids, dtype=np.long) B = np.array(B_source_ids, dtype=np.long) ai = np.where(np.in1d(A, B))[0] bi = np.where(np.in1d(B, A))[0] assert len(ai) == len(bi) ai = ai[np.argsort(A[ai])] bi = bi[np.argsort(B[bi])] assert all(A[ai] == B[bi]) return (ai, bi) def estimate_K(data, **kwargs): # TODO: Consult with colleagues to see if there is anything more principaled # we could do without requiring much more work. K_est = data["rv_single_epoch_scatter"] - data["rv_gp_mu_s"] K_err = data["rv_gp_sigma_s"]**2 #\ # + data["rv_gp_mu_s_var"] \ # + data["rv_gp_sigma_s_var"] #K_est = data["rv_excess"] #K_err = data["rv_excess_var"]**0.5 return (K_est, K_err) # some metric that is useful for showing "likelihood of binarity" until I decide # what is best. figure_name = "joint_p50_mainsequence_median" if figure_name in MAKE_FIGURES: ordered, reverse = (True, False) subsample = None xlabel = "bp_rp" ylabels = ("absolute_g_mag", "absolute_g_mag", "absolute_g_mag") zlabels = ("rv_p50", "ast_p50", "joint_p50") titles = ( r"\textrm{radial velocity}", r"\textrm{astrometry}", r"\textrm{joint information}" ) K = len(ylabels) xlims = (0, 3) ylims = (10, 2) num_bins = 150 # Use exact same bins for all panels, based on the joint information. x, y, z = (velociraptor[xlabel], velociraptor[ylabels[-1]], velociraptor[zlabels[-1]]) mask = np.isfinite(x * y * z) \ * (x >= min(xlims)) * (x <= max(xlims)) \ * (y >= min(ylims)) * (y <= max(ylims)) H, xedges, yedges, binnumber = binned_statistic_2d(x[mask], y[mask], z[mask], statistic="count", bins=num_bins) bins = (xedges, yedges) plot_binned_statistic_kwds = dict(function="median", vmin=0, vmax=1, cmap=DEFAULT_SEQUENTIAL_CMAP, mask=None, subsample=subsample, min_entries_per_bin=3, bins=bins) fig, axes = plt.subplots(1, K, figsize=(3 * K, 3)) for i, (ax, ylabel, zlabel, title) \ in enumerate(zip(axes, ylabels, zlabels, titles)): plot_binned_statistic( velociraptor[xlabel], velociraptor[ylabel], velociraptor[zlabel], ax=ax, ylabel=latex_labels.get(ylabel, ylabel), **plot_binned_statistic_kwds) ax.set_xlabel(latex_labels.get(xlabel, xlabel)) ax.xaxis.set_major_locator(MaxNLocator(5)) ax.yaxis.set_major_locator(MaxNLocator(5)) ax.set_xlim(xlims) ax.set_ylim(ylims) ax.set_title(title) cbar = plt.colorbar(ax.images[0], ax=list(axes))#=axes[-1]) cbar.set_label(r"\textrm{single star probability} $\tau$") for ax in axes: ax.set_aspect(np.ptp(ax.get_xlim())/np.ptp(ax.get_ylim())) fig.tight_layout() fig.subplots_adjust(right=0.90) # Unbelievable that we have to do all this,... #fig.canvas.update() fig.canvas.draw() cbar.ax.yaxis.set_tick_params(width=0) cax_width = 0.02 for i in range(2): ax_left, bottom, ax_width, height = axes[-1].get_position().bounds cbar.ax.set_position([ax_left + ax_width + cax_width, bottom, cax_width, height]) savefig(fig, figure_name) figure_name = "joint_p50_mainsequence_mean" if figure_name in MAKE_FIGURES: ordered, reverse = (True, False) subsample = None xlabel = "bp_rp" ylabels = ("absolute_g_mag", "absolute_g_mag", "absolute_g_mag") zlabels = ("rv_p50", "ast_p50", "joint_p50") titles = ( r"\textrm{radial velocity}", r"\textrm{astrometry}", r"\textrm{joint information}" ) K = len(ylabels) xlims = (0, 3) ylims = (10, 2) num_bins = 100 # Use exact same bins for all panels, based on the joint information. x, y, z = (velociraptor[xlabel], velociraptor[ylabels[-1]], velociraptor[zlabels[-1]]) mask = np.isfinite(x * y * z) \ * (x >= min(xlims)) * (x <= max(xlims)) \ * (y >= min(ylims)) * (y <= max(ylims)) H, xedges, yedges, binnumber = binned_statistic_2d(x[mask], y[mask], z[mask], statistic="count", bins=num_bins) bins = (xedges, yedges) plot_binned_statistic_kwds = dict(function="mean", vmin=0, vmax=1, cmap=DEFAULT_SEQUENTIAL_CMAP, mask=mask, subsample=subsample, min_entries_per_bin=5, bins=bins) fig, axes = plt.subplots(1, K, figsize=(3 * K, 3)) for i, (ax, ylabel, zlabel, title) \ in enumerate(zip(axes, ylabels, zlabels, titles)): plot_binned_statistic( velociraptor[xlabel], velociraptor[ylabel], velociraptor[zlabel], ax=ax, ylabel=latex_labels.get(ylabel, ylabel), **plot_binned_statistic_kwds) ax.set_xlabel(latex_labels.get(xlabel, xlabel)) ax.xaxis.set_major_locator(MaxNLocator(5)) ax.yaxis.set_major_locator(MaxNLocator(4)) ax.set_xlim(xlims) ax.set_ylim(ylims) ax.set_title(title) cbar = plt.colorbar(ax.images[0], ax=list(axes))#=axes[-1]) cbar.set_label(r"\textrm{single star probability} $\tau$") for ax in axes: ax.set_aspect(np.ptp(ax.get_xlim())/np.ptp(ax.get_ylim())) fig.tight_layout() fig.subplots_adjust(right=0.90) # Unbelievable that we have to do all this,... #fig.canvas.update() fig.canvas.draw() cbar.ax.yaxis.set_tick_params(width=0) cax_width = 0.02 for i in range(2): ax_left, bottom, ax_width, height = axes[-1].get_position().bounds cbar.ax.set_position([ax_left + ax_width + cax_width, bottom, cax_width, height]) savefig(fig, figure_name) figure_name = "joint_p50_wrt_kdt_labels" if figure_name in MAKE_FIGURES: ordered = True reverse = False subset = 100000 label_names = ("bp_rp", "phot_rp_mean_mag", "absolute_rp_mag") ylabels = ("rv_single_epoch_scatter", "astrometric_unit_weight_error") zlabels = ("rv_p50", "ast_p50") fig, axes = plt.subplots(2, 3, figsize=(12, 9)) for i, ax_row in enumerate(axes): ylabel = ylabels[i] zlabel = zlabels[i] for j, ax in enumerate(ax_row): xlabel = label_names[j] x = velociraptor[xlabel] y = velociraptor[ylabel] z = velociraptor[zlabel] mask = np.isfinite(x * y * z) x, y, z = (x[mask], y[mask], z[mask]) if subset is not None: idx = np.random.choice(x.size, subset, replace=False) x, y, z = (x[idx], y[idx], z[idx]) if ordered: idx = np.argsort(z) if reverse: idx = idx[::-1] x, y, z = (x[idx], y[idx], z[idx]) ax.scatter(x, y, c=z, s=1, cmap=DEFAULT_SEQUENTIAL_CMAP, rasterized=True) ax.set_xlabel(latex_labels.get(xlabel, xlabel)) ax.set_ylabel(latex_labels.get(ylabel, ylabel)) """ plot_binned_statistic_kwds = dict(function="mean", bins=100, xlabel=latex_labels.get(xlabel, xlabel), cmap=DEFAULT_SEQUENTIAL_CMAP, subsample=None, min_entries_per_bin=5) plot_binned_statistic(x, y, z, ax=ax, ylabel=latex_labels.get(ylabel, ylabel), **plot_binned_statistic_kwds) """ ax.semilogy() fig.tight_layout() savefig(fig, figure_name) figure_name = "rv_mean_joint_p50_wrt_kdt_labels" if figure_name in MAKE_FIGURES: label_names = ("bp_rp", "phot_rp_mean_mag", "absolute_rp_mag") ylabel = r"$\langle\tau_\textrm{rv,single}\rangle$" mask = np.isfinite(velociraptor["rv_p50"]) rv_p50 = velociraptor["rv_p50"] fig, axes = plt.subplots(1, len(label_names), figsize=(12, 4)) bins = 20 for i, (ax, label_name) in enumerate(zip(axes, label_names)): mean, edge, bin_index = binned_statistic(velociraptor[label_name][mask], rv_p50[mask], statistic="mean", bins=bins) var, _, __ = binned_statistic(velociraptor[label_name][mask], rv_p50[mask], statistic=np.var, bins=bins) count, _, __ = binned_statistic(velociraptor[label_name][mask], rv_p50[mask], statistic="count", bins=bins) yerr = np.sqrt(var/(count - 1)) #yerr = np.sqrt(var) center = edge[:-1] + 0.5 * np.diff(edge) plot_histogram_steps(ax, center, mean, yerr) ax.set_xlabel(latex_labels.get(label_name, label_name)) ax.set_ylabel(ylabel) ax.set_xlim(edge[0], edge[-1]) ax.set_ylim(-0.1, 1.1) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) fig.tight_layout() savefig(fig, figure_name) figure_name = "ast_mean_joint_p50_wrt_kdt_labels" if figure_name in MAKE_FIGURES: label_names = ("bp_rp", "phot_rp_mean_mag", "absolute_rp_mag") ylabel = r"$\langle\tau_\textrm{ast,single}\rangle$" fig, axes = plt.subplots(1, len(label_names), figsize=(12, 4)) bins = 20 for i, (ax, label_name) in enumerate(zip(axes, label_names)): x = velociraptor[label_name] y = velociraptor["ast_p50"] mask = np.isfinite(x * y) mean, edge, bin_index = binned_statistic(x[mask], y[mask], statistic="mean", bins=bins) var, _, __ = binned_statistic(x[mask], y[mask], statistic=np.var, bins=bins) count, _, __ = binned_statistic(x[mask], y[mask], statistic="count", bins=bins) yerr = np.sqrt(var/(count - 1)) #yerr = np.sqrt(var) center = edge[:-1] + 0.5 * np.diff(edge) plot_histogram_steps(ax, center, mean, yerr) ax.set_xlabel(latex_labels.get(label_name, label_name)) ax.set_ylabel(ylabel) ax.set_xlim(edge[0], edge[-1]) ax.set_ylim(-0.1, 1.1) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) fig.tight_layout() savefig(fig, figure_name) figure_name = "mean_joint_p50_wrt_kdt_labels" if figure_name in MAKE_FIGURES: label_names = ("bp_rp", "phot_rp_mean_mag", "absolute_rp_mag") ylabel = r"$\langle\tau_\textrm{single}\rangle$" fig, axes = plt.subplots(1, len(label_names), figsize=(12, 4)) bins = 20 for i, (ax, label_name) in enumerate(zip(axes, label_names)): x = velociraptor[label_name] y = velociraptor["joint_p50"] mask = np.isfinite(x * y) mean, edge, bin_index = binned_statistic(x[mask], y[mask], statistic="mean", bins=bins) var, _, __ = binned_statistic(x[mask], y[mask], statistic=np.var, bins=bins) count, _, __ = binned_statistic(x[mask], y[mask], statistic="count", bins=bins) yerr = np.sqrt(var/(count - 1)) #yerr = np.sqrt(var) center = edge[:-1] + 0.5 * np.diff(edge) plot_histogram_steps(ax, center, mean, yerr) ax.set_xlabel(latex_labels.get(label_name, label_name)) ax.set_ylabel(ylabel) ax.set_xlim(edge[0], edge[-1]) ax.set_ylim(-0.1, 1.1) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) fig.tight_layout() savefig(fig, figure_name) figure_name = "all_joint_p50_wrt_kdt_labels" if figure_name in MAKE_FIGURES: label_names = ("bp_rp", "phot_rp_mean_mag", "absolute_rp_mag") ylabel = r"$\langle\tau_\textrm{single}\rangle$" fig, axes = plt.subplots(1, len(label_names), figsize=(12, 4)) bins = 20 for prefix in (None, "rv", "ast"): label = "joint_p50" if prefix is None else f"{prefix}_p50" for i, (ax, label_name) in enumerate(zip(axes, label_names)): x = velociraptor[label_name] y = velociraptor[label] mask = np.isfinite(x * y) mean, edge, bin_index = binned_statistic(x[mask], y[mask], statistic="mean", bins=bins) var, _, __ = binned_statistic(x[mask], y[mask], statistic=np.var, bins=bins) count, _, __ = binned_statistic(x[mask], y[mask], statistic="count", bins=bins) yerr = np.sqrt(var/(count - 1)) #yerr = np.sqrt(var) center = edge[:-1] + 0.5 * np.diff(edge) plot_histogram_steps(ax, center, mean, yerr, label="joint" if prefix is None else prefix) ax.set_xlabel(latex_labels.get(label_name, label_name)) ax.set_ylabel(ylabel) ax.set_xlim(edge[0], edge[-1]) ax.set_ylim(-0.1, 1.1) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) fig.tight_layout() savefig(fig, figure_name) figure_name = "hrd_single_star_fraction_hist" if figure_name in MAKE_FIGURES: fig, axes = plt.subplots(1, 2, figsize=(9, 4)) xlabel = "bp_rp" ylabel = "phot_rp_mean_mag" zlabel = "joint_p50" # Restrict sensibly. mask = (velociraptor["absolute_rp_mag"] > -16) \ * (velociraptor["bp_rp"] < 6.5) plot_binned_statistic_kwds = dict(function="mean", vmin=0, vmax=1, bins=100, xlabel=latex_labels.get(xlabel, xlabel), cmap=DEFAULT_SEQUENTIAL_CMAP, mask=mask, subsample=None, min_entries_per_bin=5) for ax, ylabel in zip(axes, ("phot_rp_mean_mag", "absolute_rp_mag")): plot_binned_statistic( velociraptor[xlabel], velociraptor[ylabel], velociraptor[zlabel], ax=ax, ylabel=latex_labels.get(ylabel, ylabel), **plot_binned_statistic_kwds) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) ax.set_xlim(common_kwds.get("{}.limits".format(xlabel), None)) ax.set_ylim(common_kwds.get("{}.limits".format(ylabel), None)) cbar = plt.colorbar(ax.images[0], fraction=0.046, pad=0.04) cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{rv,single}$") fig.tight_layout() savefig(fig, figure_name) figure_name = "ast_hrd_single_star_fraction_hist" if figure_name in MAKE_FIGURES or True: fig, axes = plt.subplots(1, 2, figsize=(8, 4)) xlabel = "bp_rp" ylabel = "phot_rp_mean_mag" zlabel = "ast_p50" # Restrict sensibly. mask = (velociraptor["absolute_rp_mag"] > -16) \ * (velociraptor["bp_rp"] < 6.5) \ * np.isfinite(velociraptor["rv_single_epoch_scatter"]) plot_binned_statistic_kwds = dict(function="mean", vmin=0, vmax=1, bins=100, xlabel=latex_labels.get(xlabel, xlabel), cmap=DEFAULT_SEQUENTIAL_CMAP, mask=mask, subsample=None, min_entries_per_bin=5) for ax, ylabel in zip(axes, ("phot_rp_mean_mag", "absolute_rp_mag")): plot_binned_statistic( velociraptor[xlabel], velociraptor[ylabel], velociraptor[zlabel], ax=ax, ylabel=latex_labels.get(ylabel, ylabel), **plot_binned_statistic_kwds) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) ax.set_xlim(common_kwds.get("{}.limits".format(xlabel), None)) ax.set_ylim(common_kwds.get("{}.limits".format(ylabel), None)) cbar = plt.colorbar(ax.images[0], fraction=0.046, pad=0.04) cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{ast,single}$") fig.tight_layout() savefig(fig, figure_name) # Plot the H-R diagram coloured by \tau_{single} as a scatter plot and a 2D hist figure_name = "rv_hrd_single_star_fraction_hist" if figure_name in MAKE_FIGURES: fig, axes = plt.subplots(1, 2, figsize=(9, 4)) xlabel = "bp_rp" ylabel = "phot_rp_mean_mag" zlabel = "rv_p50" # Restrict sensibly. mask = (velociraptor["absolute_rp_mag"] > -16) \ * (velociraptor["bp_rp"] < 6.5) plot_binned_statistic_kwds = dict(function="mean", vmin=0, vmax=1, bins=100, xlabel=latex_labels.get(xlabel, xlabel), cmap=DEFAULT_SEQUENTIAL_CMAP, mask=mask, subsample=None, min_entries_per_bin=5) for ax, ylabel in zip(axes, ("phot_rp_mean_mag", "absolute_rp_mag")): plot_binned_statistic( velociraptor[xlabel], velociraptor[ylabel], velociraptor[zlabel], ax=ax, ylabel=latex_labels.get(ylabel, ylabel), **plot_binned_statistic_kwds) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) ax.set_xlim(common_kwds.get("{}.limits".format(xlabel), None)) ax.set_ylim(common_kwds.get("{}.limits".format(ylabel), None)) cbar = plt.colorbar(ax.images[0], fraction=0.046, pad=0.04) cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") fig.tight_layout() savefig(fig, figure_name) figure_name = "rv_hrd_single_star_fraction_scatter" if figure_name in MAKE_FIGURES: subsample = None ordered, reverse = True, False fig, axes = plt.subplots(1, 2, figsize=(9, 4)) xlabel = "bp_rp" ylabel = "phot_rp_mean_mag" zlabel = "rv_p50" if subsample is not None: idx = np.random.choice(len(velociraptor), subsample, replace=False) else: idx = np.arange(len(velociraptor)) if ordered: idx = idx[np.argsort(velociraptor[zlabel][idx])] if reverse: idx = idx[::-1] scatter_kwds = dict(vmin=0, vmax=1, cmap=DEFAULT_SEQUENTIAL_CMAP, s=1, alpha=0.1, c=velociraptor[zlabel][idx], rasterized=True) for ax, ylabel in zip(axes, ("phot_rp_mean_mag", "absolute_rp_mag")): ax.scatter( velociraptor[xlabel][idx], velociraptor[ylabel][idx], **scatter_kwds) ax.set_xlabel(latex_labels.get(xlabel, xlabel)) ax.set_ylabel(latex_labels.get(ylabel, ylabel)) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) ax.set_xlim(common_kwds.get("{}.limits".format(xlabel), None)) ax.set_ylim(common_kwds.get("{}.limits".format(ylabel), None)) if scatter_kwds.get("alpha", 1) == 1: collection = ax.collections[0] else: # Note: if you give alpha in scatter_kwds then you should give # vmin and vmax too otherwise this needs to be updated: collection = ax.scatter([np.nanmean(velociraptor[xlabel][idx])], [np.nanmean(velociraptor[ylabel][idx])], c=[np.nanmean(scatter_kwds["c"])], vmin=scatter_kwds["vmin"], vmax=scatter_kwds["vmax"], cmap=scatter_kwds["cmap"], alpha=1.0, s=0) cbar = plt.colorbar(collection, fraction=0.046, pad=0.04) cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") fig.tight_layout() savefig(fig, figure_name) # Cross-match against SB9 catalog only if we have to. if any(["sb9" in figure_name.lower() for figure_name in MAKE_FIGURES]): sb9 = Table.read(os.path.join(BASE_PATH, "data", "sb9_xm_gaia.fits")) # remove duplicates. sb9 = sb9.group_by("source_id") sb9 = sb9[sb9.groups.indices[:-1]] # Only show SB9 stars that meet these criteria: sb9_mask = (sb9["f_K1"] != ">") \ * (sb9["f_T0"] == 0) \ * (sb9["Grade"] > 0) \ * (sb9["f_omega"] != "a") sb9 = sb9[sb9_mask] assert len(set(sb9["source_id"])) == len(sb9) vl_sb9_ids, sb9_ids = cross_match(velociraptor["source_id"], sb9["source_id"]) figure_name = "rv_sb9_kalpha_comparison" if figure_name in MAKE_FIGURES: sort, reverse = (True, False) vl_sb9_subset = velociraptor[vl_sb9_ids] K_est, K_est_err = estimate_K(vl_sb9_subset) scalar = (1.0 / (sb9["Per"] * (1 - sb9["e"]**2)**0.5))[sb9_ids] x = sb9["K1"][sb9_ids] * scalar y = K_est * scalar xerr = sb9["e_K1"][sb9_ids] * scalar yerr = K_est_err * scalar c = log_K_significance[vl_sb9_ids] if sort: idx = np.argsort(c) if reverse: idx = idx[::-1] else: idx = np.arange(len(c)) x, y, xerr, yerr, c = (x[idx], y[idx], xerr[idx], yerr[idx], c[idx]) fig, ax = plt.subplots(1, 1, figsize=(8, 7)) scat = ax.scatter(x, y, c=c, s=10, vmax=1.75, rasterized=True) ax.errorbar(x, y, xerr=xerr, yerr=yerr, fmt="none", ecolor="#CCCCCC", zorder=-1, linewidth=1, capsize=0) ax.loglog() limits = np.array([ax.get_xlim(), ax.get_ylim()]) limits = (np.min(limits), np.max(limits)) ax.plot(limits, limits, **one_to_one_line_kwds) ax.set_xlim(limits) ax.set_ylim(limits) divider = make_axes_locatable(ax) cax = divider.append_axes("right", size="5%", pad=0.05) cbar = plt.colorbar(scat, cax=cax) #cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") cbar.set_label(r"$\log_{10}\left[\sigma(\textrm{V}_\textrm{R}^{t}) - \mu_s\right] - \log_{10}{\sigma_s}$") fig.tight_layout() #ax.set_xlabel(r"$K/P\sqrt{1 - e^2}$ \textrm{(SB9)}") #ax.set_ylabel(r"$K_\textrm{est} /P\sqrt{1 - e^2}$ \textrm{(this work)}") ax.set_xlabel(r"$K_{1} / \alpha$ \textrm{(SB9)}") ax.set_ylabel(r"$K_{1,\textrm{est}} / \alpha$ \textrm{(this work)}") fig.tight_layout() savefig(fig, figure_name) figure_name = "rv_sb9_kalpha_comparison_tau" if figure_name in MAKE_FIGURES: sort, reverse = (True, False) vl_sb9_subset = velociraptor[vl_sb9_ids] K_est, K_est_err = estimate_K(vl_sb9_subset) scalar = (1.0 / (sb9["Per"] * (1 - sb9["e"]**2)**0.5))[sb9_ids] x = sb9["K1"][sb9_ids] * scalar y = K_est * scalar xerr = sb9["e_K1"][sb9_ids] * scalar yerr = K_est_err * scalar c = vl_sb9_subset["rv_p50"] if sort: idx = np.argsort(c) if reverse: idx = idx[::-1] else: idx = np.arange(len(c)) x, y, xerr, yerr, c = (x[idx], y[idx], xerr[idx], yerr[idx], c[idx]) fig, ax = plt.subplots(1, 1, figsize=(8, 7)) scat = ax.scatter(x, y, c=c, s=10, vmin=0, vmax=1, rasterized=True) ax.errorbar(x, y, xerr=xerr, yerr=yerr, fmt="none", ecolor="#CCCCCC", zorder=-1, linewidth=1, capsize=0) ax.loglog() limits = np.array([ax.get_xlim(), ax.get_ylim()]) limits = (np.min(limits), np.max(limits)) ax.plot(limits, limits, **one_to_one_line_kwds) ax.set_xlim(limits) ax.set_ylim(limits) divider = make_axes_locatable(ax) cax = divider.append_axes("right", size="5%", pad=0.05) cbar = plt.colorbar(scat, cax=cax) cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") fig.tight_layout() #ax.set_xlabel(r"$K/P\sqrt{1 - e^2}$ \textrm{(SB9)}") #ax.set_ylabel(r"$K_\textrm{est} /P\sqrt{1 - e^2}$ \textrm{(this work)}") ax.set_xlabel(r"$K_{1} / \alpha$ \textrm{(SB9)}") ax.set_ylabel(r"$K_{1,\textrm{est}} / \alpha$ \textrm{(this work)}") fig.tight_layout() savefig(fig, figure_name) figure_name = "rv_sb9_k_comparison" if figure_name in MAKE_FIGURES: sort, reverse = (True, False) vl_sb9_subset = velociraptor[vl_sb9_ids] """ K_est, K_est_err = estimate_K(vl_sb9_subset["rv_single_epoch_scatter"], vl_sb9_subset["rv_mu_single"], vl_sb9_subset["rv_sigma_single"], vl_sb9_subset["rv_mu_single_var"], vl_sb9_subset["rv_sigma_single_var"]) """ K_est, K_est_err = estimate_K(vl_sb9_subset) x = sb9["K1"][sb9_ids] y = K_est xerr = sb9["e_K1"][sb9_ids] yerr = K_est_err #c = log_K_significance[vl_sb9_ids] c = sb9["e"][sb9_ids] if sort: idx = np.argsort(c) if reverse: idx = idx[::-1] else: idx = np.arange(len(c)) x, y, xerr, yerr, c = (x[idx], y[idx], xerr[idx], yerr[idx], c[idx]) fig, ax = plt.subplots(1, 1, figsize=(8, 7)) scat = ax.scatter(x, y, c=c, s=10, vmin=0, vmax=1, rasterized=True) ax.errorbar(x, y, xerr=xerr, yerr=yerr, fmt="none", ecolor="#CCCCCC", zorder=-1, linewidth=1, capsize=0) ax.loglog() limits = np.array([ax.get_xlim(), ax.get_ylim()]) limits = (np.min(limits), np.max(limits)) ax.plot(limits, limits, **one_to_one_line_kwds) ax.set_xlim(limits) ax.set_ylim(limits) divider = make_axes_locatable(ax) cax = divider.append_axes("right", size="5%", pad=0.05) cbar = plt.colorbar(scat, cax=cax) cbar.set_label(r"\textrm{eccentricity (SB9)}") #cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{rv,single}$") #cbar.set_label(r"$\log_{10}\left[\sigma(\textrm{V}_\textrm{R}^{t}) - \mu_s\right] - \log_{10}{\sigma_s}$") fig.tight_layout() #ax.set_xlabel(r"$K/P\sqrt{1 - e^2}$ \textrm{(SB9)}") #ax.set_ylabel(r"$K_\textrm{est} /P\sqrt{1 - e^2}$ \textrm{(this work)}") ax.set_xlabel(r"$K_{1} \textrm{/ km\,s}^{-1}$ \textrm{(SB9)}") ax.set_ylabel(r"$K_{1,\textrm{est}} \textrm{/ km\,s}^{-1}$ \textrm{(this work)}") fig.tight_layout() savefig(fig, figure_name) figure_name = "rv_sb9_kalpha_comparison_period" if figure_name in MAKE_FIGURES: sort, reverse = (True, False) vl_sb9_subset = velociraptor[vl_sb9_ids] K_est, K_est_err = estimate_K(vl_sb9_subset) scalar = (1.0 / (sb9["Per"] * (1 - sb9["e"]**2)**0.5))[sb9_ids] x = sb9["K1"][sb9_ids] * scalar y = K_est * scalar xerr = sb9["e_K1"][sb9_ids] * scalar yerr = K_est_err * scalar c = np.log10(sb9["Per"][sb9_ids]) if sort: idx = np.argsort(c) if reverse: idx = idx[::-1] else: idx = np.arange(len(c)) x, y, xerr, yerr, c = (x[idx], y[idx], xerr[idx], yerr[idx], c[idx]) fig, ax = plt.subplots(1, 1, figsize=(8, 7)) scat = ax.scatter(x, y, c=c, s=10, vmax=1.75, rasterized=True) ax.errorbar(x, y, xerr=xerr, yerr=yerr, fmt="none", ecolor="#CCCCCC", zorder=-1, linewidth=1, capsize=0) ax.loglog() limits = np.array([ax.get_xlim(), ax.get_ylim()]) limits = (np.min(limits), np.max(limits)) ax.plot(limits, limits, **one_to_one_line_kwds) ax.set_xlim(limits) ax.set_ylim(limits) divider = make_axes_locatable(ax) cax = divider.append_axes("right", size="5%", pad=0.05) cbar = plt.colorbar(scat, cax=cax) #cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") cbar.set_label(r"$\log_{10}\left(P \textrm{/ days}\right)$") fig.tight_layout() #ax.set_xlabel(r"$K/P\sqrt{1 - e^2}$ \textrm{(SB9)}") #ax.set_ylabel(r"$K_\textrm{est} /P\sqrt{1 - e^2}$ \textrm{(this work)}") ax.set_xlabel(r"$K_{1} / \alpha$ \textrm{(SB9)}") ax.set_ylabel(r"$K_{1,\textrm{est}} / \alpha$ \textrm{(this work)}") fig.tight_layout() savefig(fig, figure_name) figure_name = "rv_sb9_kp_corner" if figure_name in MAKE_FIGURES: sort, reverse = (True, False) x = sb9["K1"][sb9_ids] y = sb9["Per"][sb9_ids] c = velociraptor["rv_p50"][vl_sb9_ids] if sort: idx = np.argsort(c) idx = idx[::-1] if reverse else idx else: idx = np.arange(len(c)) fig, axes = plt.subplots(2, 2) axes[0, 1].set_visible(False) axes[1, 0].scatter(x[idx], y[idx], c=c[idx], s=10, vmax=1.75, rasterized=True) axes[1, 0].loglog() axes[1, 0].set_xlabel(r"$P$ \textrm{/ days}") axes[1, 0].set_ylabel(r"$K_{1}$ \textrm{/ km\,s}$^{-1}$") # TODO: Draw histograms on other axes fig.tight_layout() fig.subplots_adjust(hspace=0, wspace=0) for ax in (axes[0, 0], axes[1, 1]): ax.set_xticks([]) ax.set_yticks([]) savefig(fig, figure_name) # Cross-match against APW catalog only if we have to. if any(["apw_unimodal" in figure_name.lower() for figure_name in MAKE_FIGURES]): apw_unimodal = Table.read(os.path.join( BASE_PATH, "data", "apw-highK-unimodal-xm-gaia.fits")) # remove duplicates. apw_unimodal = apw_unimodal.group_by("source_id") apw_unimodal = apw_unimodal[apw_unimodal.groups.indices[:-1]] assert len(set(apw_unimodal["source_id"])) == len(apw_unimodal) vl_apw_um_ids, apw_um_ids = cross_match(velociraptor["source_id"], apw_unimodal["source_id"]) figure_name = "rv_apw_unimodal_k_comparison" if figure_name in MAKE_FIGURES: sort, reverse = (True, False) vl_apwu_subset = velociraptor[vl_apw_um_ids] K_est, K_est_err = estimate_K(vl_apwu_subset) x = apw_unimodal["K"][apw_um_ids] xerr = apw_unimodal["K_err"][apw_um_ids] y = K_est yerr = K_est_err c = vl_apwu_subset["rv_p50"] if sort: idx = np.argsort(c) idx = idx[::-1] if reverse else idx else: idx = np.arange(len(c)) x, y, xerr, yerr, c = (x[idx], y[idx], xerr[idx], yerr[idx], c[idx]) fig, ax = plt.subplots(1, 1, figsize=(8, 7)) scat = ax.scatter(x, y, c=c, s=10, vmin=0, vmax=1, rasterized=True) ax.errorbar(x, y, yerr=yerr, xerr=xerr, fmt="none", ecolor="#cccccc", lw=1, zorder=-1, rasterized=True) ax.loglog() divider = make_axes_locatable(ax) cax = divider.append_axes("right", size="5%", pad=0.05) cbar = plt.colorbar(scat, cax=cax) cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") #cbar.set_label(r"$\log_{10}\left(P \textrm{/ days}\right)$") limits = np.array([ax.get_xlim(), ax.get_ylim()]) limits = (np.min(limits), np.max(limits)) ax.plot(limits, limits, **one_to_one_line_kwds) ax.set_xlim(limits) ax.set_ylim(limits) ax.set_xlabel(r"$K$ \textrm{/ km\,s}$^{-1}$ \textrm{(Price-Whelan et al. 2017)}") ax.set_ylabel(r"$K$ \textrm{/ km\,s}$^{-1}$ \textrm{(this work)}") fig.tight_layout() savefig(fig, figure_name) figure_name = "rv_apw_unimodal_kalpha_comparison" if figure_name in MAKE_FIGURES: sort, reverse = (True, False) vl_apwu_subset = velociraptor[vl_apw_um_ids] K_est, K_est_err = estimate_K(vl_apwu_subset) scalar = 1.0/(apw_unimodal["P"] * np.sqrt(1 - apw_unimodal["e"]**2))[apw_um_ids] x = apw_unimodal["K"][apw_um_ids] * scalar xerr = apw_unimodal["K_err"][apw_um_ids] * scalar y = K_est * scalar yerr = K_est_err * scalar c = vl_apwu_subset["rv_p50"] if sort: idx = np.argsort(c) idx = idx[::-1] if reverse else idx else: idx = np.arange(len(c)) x, y, xerr, yerr, c = (x[idx], y[idx], xerr[idx], yerr[idx], c[idx]) fig, ax = plt.subplots(1, 1, figsize=(8, 7)) scat = ax.scatter(x, y, c=c, s=10, vmin=0, vmax=1, rasterized=True) ax.errorbar(x, y, yerr=yerr, xerr=xerr, fmt="none", ecolor="#cccccc", lw=1, zorder=-1, rasterized=True) ax.loglog() divider = make_axes_locatable(ax) cax = divider.append_axes("right", size="5%", pad=0.05) cbar = plt.colorbar(scat, cax=cax) cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") #cbar.set_label(r"$\log_{10}\left(P \textrm{/ days}\right)$") limits = np.array([ax.get_xlim(), ax.get_ylim()]) limits = (np.min(limits), np.max(limits)) ax.plot(limits, limits, **one_to_one_line_kwds) ax.set_xlim(limits) ax.set_ylim(limits) ax.set_xlabel(r"$K / \alpha$ \textrm{(Price-Whelan et al. 2017)}") ax.set_ylabel(r"$K / \alpha$ \textrm{(this work)}") fig.tight_layout() savefig(fig, figure_name) # Cross-match against APW catalog only if we have to. if any(["apw_percentiles" in figure_name.lower() for figure_name in MAKE_FIGURES]): apw_percentiles = Table.read(os.path.join( BASE_PATH, "data", "apw-lnK-percentiles-xm-gaia.fits")) # remove duplicates. apw_percentiles = apw_percentiles.group_by("source_id") apw_percentiles = apw_percentiles[apw_percentiles.groups.indices[:-1]] assert len(set(apw_percentiles["source_id"])) == len(apw_percentiles) vl_apw_lnk_ids, apw_lnk_ids = cross_match(velociraptor["source_id"], apw_percentiles["source_id"]) figure_name = "rv_apw_percentiles_k_comparison" if figure_name in MAKE_FIGURES: sort, reverse = (True, False) vl_apwp_subset = velociraptor[vl_apw_lnk_ids] K_est, K_est_err = estimate_K(vl_apwp_subset) if sort: idx = np.argsort(c) idx = idx[::-1] if reverse else idx else: idx = np.arange(len(c)) x, y, yerr, c = (x[idx], y[idx], yerr[idx], c[idx]) fig, ax = plt.subplots(figsize=(8, 7)) scat = ax.scatter(x, y, c=c, s=1, rasterized=True) ax.errorbar(x, y, yerr=yerr, fmt="none", ecolor="#cccccc", lw=1, zorder=-1, rasterized=True) ax.loglog() limits = np.array([ax.get_xlim(), ax.get_ylim()]) limits = (np.min(limits), np.max(limits)) ax.plot(limits, limits, **one_to_one_line_kwds) ax.set_xlim(limits) ax.set_ylim(limits) ax.set_xlabel(r"$K_\textrm{1\%}$ \textrm{/ km\,s}$^{-1}$ \textrm{(Price-Whelan et al. 2017)}") ax.set_ylabel(r"$K$ \textrm{/ km\,s}$^{-1}$ \textrm{(this work)}") divider = make_axes_locatable(ax) cax = divider.append_axes("right", size="5%", pad=0.05) cbar = plt.colorbar(scat, cax=cax) cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") fig.tight_layout() savefig(fig, figure_name) figure_name = "rv_gp_hrd" if figure_name in MAKE_FIGURES: xlabel = "bp_rp" ylabel = "phot_rp_mean_mag" zlabels = ("rv_gp_mu_s", "rv_gp_sigma_s",) # "rv_mu_multiple", "rv_sigma_multiple") K, M = (2, len(zlabels)) fig, axes = plt.subplots(M, K, figsize=(4 * K + 1, 4 * M)) mask = (velociraptor["absolute_rp_mag"] > -5) \ * (velociraptor["bp_rp"] < 4) limits = dict(bp_rp=(-0.25, 4), phot_rp_mean_mag=(13, 6), absolute_rp_mag=(10, -5)) plot_binned_statistic_kwds = dict(function="median", bins=100, xlabel=latex_labels.get(xlabel, xlabel), cmap=DEFAULT_SEQUENTIAL_CMAP, mask=mask, subsample=100000, min_entries_per_bin=5) for ax_row, zlabel in zip(axes, zlabels): for ax, ylabel in zip(ax_row, ("phot_rp_mean_mag", "absolute_rp_mag")): plot_binned_statistic( velociraptor[xlabel], velociraptor[ylabel], velociraptor[zlabel], ax=ax, ylabel=latex_labels.get(ylabel, ylabel), **plot_binned_statistic_kwds) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) ax.set_xlim(limits.get(xlabel, None)) ax.set_ylim(limits.get(ylabel, None)) ax.set_title(latex_labels.get(zlabel, zlabel)) #cbar = plt.colorbar(ax.images[0], fraction=0.046, pad=0.04) #cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") for ax in np.array(axes).flatten(): ax.set_aspect(np.ptp(ax.get_xlim())/np.ptp(ax.get_ylim())) fig.tight_layout() savefig(fig, figure_name) figure_name = "rv_gp_wrt_params" if figure_name in MAKE_FIGURES: from matplotlib.colors import LogNorm xlabels = ("bp_rp", "phot_rp_mean_mag", "absolute_rp_mag") #ylabels = ("rv_mu_single", "rv_sigma_single") ylabels = ("rv_gp_mu_s", "rv_gp_sigma_s") K, M = (len(xlabels), len(ylabels)) fig, axes = plt.subplots(M, K, figsize=(4 * K, 4 * M)) mask = np.isfinite(velociraptor["rv_single_epoch_scatter"]) plot_binned_statistic_kwds = dict(function="count", bins=250, cmap="Blues", mask=mask, subsample=None, min_entries_per_bin=1, norm=LogNorm()) for ax_row, xlabel in zip(axes.T, xlabels): for ax, ylabel in zip(ax_row, ylabels): plot_binned_statistic( velociraptor[xlabel], velociraptor[ylabel], velociraptor[xlabel], ax=ax, xlabel=latex_labels.get(xlabel, xlabel), ylabel=latex_labels.get(ylabel, ylabel), **plot_binned_statistic_kwds) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) ax.set_xlim(np.sort(common_kwds.get("{}.limits".format(xlabel), None))) ax.set_ylim(0, max(ax.get_ylim())) #cbar = plt.colorbar(ax.images[0], fraction=0.046, pad=0.04) #cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") for ax in np.array(axes).flatten(): ax.set_aspect(np.ptp(ax.get_xlim())/np.ptp(ax.get_ylim())) fig.tight_layout() savefig(fig, figure_name) figure_name = "ast_gp_hrd" if figure_name in MAKE_FIGURES: xlabel = "bp_rp" ylabel = "phot_rp_mean_mag" zlabels = ("ast_mu_single", "ast_sigma_single",) zlabels = ("ast_gp_mu_s", "ast_gp_sigma_s") # "ast_mu_multiple", "ast_sigma_multiple") K, M = (2, len(zlabels)) fig, axes = plt.subplots(M, K, figsize=(4 * K + 1, 4 * M)) mask = (velociraptor["absolute_rp_mag"] > -5) \ * (velociraptor["bp_rp"] < 4) \ * np.isfinite(velociraptor["rv_single_epoch_scatter"]) limits = dict(bp_rp=(-0.25, 4), phot_rp_mean_mag=(13, 6), absolute_rp_mag=(10, -5)) plot_binned_statistic_kwds = dict(function="median", bins=100, xlabel=latex_labels.get(xlabel, xlabel), cmap=DEFAULT_SEQUENTIAL_CMAP, mask=mask, subsample=100000, min_entries_per_bin=5) for ax_row, zlabel in zip(axes, zlabels): for ax, ylabel in zip(ax_row, ("phot_rp_mean_mag", "absolute_rp_mag")): plot_binned_statistic( velociraptor[xlabel], velociraptor[ylabel], velociraptor[zlabel], ax=ax, ylabel=latex_labels.get(ylabel, ylabel), **plot_binned_statistic_kwds) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) ax.set_xlim(limits.get(xlabel, None)) ax.set_ylim(limits.get(ylabel, None)) ax.set_title(latex_labels.get(zlabel, zlabel)) #cbar = plt.colorbar(ax.images[0], fraction=0.046, pad=0.04) #cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") for ax in np.array(axes).flatten(): ax.set_aspect(np.ptp(ax.get_xlim())/np.ptp(ax.get_ylim())) fig.tight_layout() savefig(fig, figure_name) figure_name = "ast_gp_wrt_params" if figure_name in MAKE_FIGURES: from matplotlib.colors import LogNorm xlabels = ("bp_rp", "phot_rp_mean_mag", "absolute_rp_mag") ylabels = ("ast_mu_single", "ast_sigma_single") ylabels = ("ast_gp_mu_s", "ast_gp_sigma_s") K, M = (len(xlabels), len(ylabels)) fig, axes = plt.subplots(M, K, figsize=(4 * K, 4 * M)) mask = np.isfinite(velociraptor["astrometric_unit_weight_error"]) plot_binned_statistic_kwds = dict(function="count", bins=250, cmap="Blues", mask=mask, subsample=None, min_entries_per_bin=1, norm=LogNorm()) for ax_row, xlabel in zip(axes.T, xlabels): for ax, ylabel in zip(ax_row, ylabels): plot_binned_statistic( velociraptor[xlabel], velociraptor[ylabel], velociraptor[xlabel], ax=ax, xlabel=latex_labels.get(xlabel, xlabel), ylabel=latex_labels.get(ylabel, ylabel), **plot_binned_statistic_kwds) ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) ax.set_xlim(np.sort(common_kwds.get("{}.limits".format(xlabel), None))) ax.set_ylim(0, max(ax.get_ylim())) #cbar = plt.colorbar(ax.images[0], fraction=0.046, pad=0.04) #cbar.set_label(r"\textrm{single star fraction} $\tau_\textrm{single}$") for ax in np.array(axes).flatten(): ax.set_aspect(np.ptp(ax.get_xlim())/np.ptp(ax.get_ylim())) fig.tight_layout() savefig(fig, figure_name) # Cross-match against Soubiran catalog only if we have to. if any(["soubiran" in figure_name.lower() for figure_name in MAKE_FIGURES]): soubiran = Table.read(os.path.join( BASE_PATH, "data", "soubiran-2013-xm-gaia.fits")) # remove duplicates. soubiran = soubiran.group_by("source_id") soubiran = soubiran[soubiran.groups.indices[:-1]] assert len(set(soubiran["source_id"])) == len(soubiran) vl_soubiran_ids, soubiran_ids = cross_match(velociraptor["source_id"], soubiran["source_id"]) figure_name = "rv_soubiran_hist" if figure_name in MAKE_FIGURES: fig, axes = plt.subplots(1, 2, figsize=(8, 4)) vl_soubiran_subset = velociraptor[vl_soubiran_ids] K_est, K_est_err = estimate_K(vl_soubiran_subset) axes[0].hist(K_est[np.isfinite(K_est)], bins=100) axes[0].set_xlabel(r"$K_{est}$ \textrm{/ km\,s}$^{-1}$") axes[0].set_ylabel(r"\textrm{count}") axes[1].semilogx() positive = K_est > 0 axes[1].scatter(K_est[positive], vl_soubiran_subset["rv_p50"][positive], s=10, rasterized=True) xlim = axes[1].get_xlim() axes[1].errorbar(K_est[positive], vl_soubiran_subset["rv_p50"][positive], xerr=np.min([K_est_err[positive], K_est[positive] - 1e-4], axis=0), fmt="none", ecolor="#CCCCCC", zorder=-1, linewidth=1, capsize=0, rasterized=True) axes[1].set_xlim(xlim) axes[1].set_xlabel(r"$K_{est} \textrm{ / km\,s}^{-1}$") axes[1].set_ylabel(r"\textrm{single star fraction} $\tau_\textrm{single}$") for ax in np.array(axes).flatten(): ax.set_aspect(np.ptp(ax.get_xlim())/np.ptp(ax.get_ylim())) fig.tight_layout() notable = (vl_soubiran_subset["rv_p50"] < 0.6) \ * (K_est > 0) savefig(fig, figure_name) # Cross-match against Soubiran catalog only if we have to. if any(["huang" in figure_name.lower() for figure_name in MAKE_FIGURES]): huang = Table.read(os.path.join( BASE_PATH, "data", "huang-apogee-rv-standards-xm-gaia.fits")) # remove duplicates. huang = huang.group_by("source_id") huang = huang[huang.groups.indices[:-1]] assert len(set(huang["source_id"])) == len(huang) vl_huang_ids, huang_ids = cross_match(velociraptor["source_id"], huang["source_id"]) figure_name = "rv_huang_hist" """ if figure_name in MAKE_FIGURES: vl_huang_subset = velociraptor[vl_huang_ids] K_est, K_est_err = estimate_K(vl_huang_subset) fig, axes = plt.subplots(2, 1, figsize=(4, 8)) #v = log_K_significance[vl_huang_ids] #sb9_v = log_K_significance[vl_sb9_ids] #axes[0].hist(v[np.isfinite(v)], bins=100, normed=True, alpha=0.5) #axes[0].hist(sb9_v[np.isfinite(sb9_v)], bins=100, facecolor="tab:green", alpha=0.5, normed=True) #axes[0].set_xlabel(r"$\log_{10}\left[\sigma(\textrm{V}_\textrm{R}^{t}) - \mu_s\right] - \log_{10}{\sigma_s}$") #axes[0].set_yticks([]) axes[1].scatter(vl, K_est, s=10, rasterized=True) axes[1].errorbar(v, K_est, yerr=np.min([K_est_err, K_est - 1e-4], axis=0), fmt="none", ecolor="#CCCCCC", zorder=-1, linewidth=1, capsize=0, rasterized=True) axes[1].set_ylabel(r"$K_{est} \textrm{ / km\,s}^{-1}$") axes[1].set_xlabel(r"$\log_{10}\left[\sigma(\textrm{V}_\textrm{R}^{t}) - \mu_s\right] - \log_{10}{\sigma_s}$") for ax in np.array(axes).flatten(): ax.set_aspect(np.ptp(ax.get_xlim())/np.ptp(ax.get_ylim())) fig.tight_layout() notable = (vl_huang_subset["rv_p50"] < 0.6) \ * (K_est > 0) savefig(fig, figure_name) """ figure_name = "rv_huang_hist2" if figure_name in MAKE_FIGURES: vl_huang_subset = velociraptor[vl_huang_ids] K_est, K_est_err = estimate_K(vl_huang_subset) fig, axes = plt.subplots(2, 1, figsize=(4, 8)) v = vl_huang_subset["rv_p50"] axes[0].hist(v[np.isfinite(v)], bins=100, normed=True, alpha=0.5) axes[0].set_xlabel(latex_labels["rv_p50"]) axes[0].set_yticks([]) axes[1].scatter(v, K_est, s=10, rasterized=True) axes[1].errorbar(v, K_est, yerr=np.min([K_est_err, K_est - 1e-4], axis=0), fmt="none", ecolor="#CCCCCC", zorder=-1, linewidth=1, capsize=0, rasterized=True) axes[1].set_ylabel(r"$K_{est} \textrm{ / km\,s}^{-1}$") axes[1].set_xlabel(r"$\log_{10}\left[\sigma(\textrm{V}_\textrm{R}^{t}) - \mu_s\right] - \log_{10}{\sigma_s}$") for ax in np.array(axes).flatten(): ax.set_aspect(np.ptp(ax.get_xlim())/np.ptp(ax.get_ylim())) fig.tight_layout() notable = (vl_huang_subset["rv_p50"] < 0.6) \ * (K_est > 0) savefig(fig, figure_name)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,294
andycasey/velociraptor
refs/heads/master
/scripts/xmatch_soubiran_2013_with_gaia.py
from astropy.table import Table from astroquery.gaia import Gaia soubiran_2013 = Table.read("data/Soubiran_2013.fits") N = len(soubiran_2013) # Get gaia source IDs and we can do the cross-match from there. source_ids = [] for i, hip in enumerate(soubiran_2013["HIP"]): job = Gaia.launch_job(""" SELECT gaia.source_id FROM gaiadr2.gaia_source AS gaia, gaiadr2.hipparcos2_best_neighbour AS hip WHERE hip.original_ext_source_id = '{:.0f}' AND hip.source_id = gaia.source_id """.format(hip)) results = job.get_results() if len(results) == 0: source_id = -1 else: source_id = results["source_id"][0] print("{}/{}: HIP {} = Gaia DR2 {}".format(i, N, hip, source_id)) source_ids.append(source_id) import numpy as np soubiran_2013["source_id"] = np.array(source_ids, dtype=np.int64) OK = np.isfinite(soubiran_2013["source_id"]) soubiran_2013 = soubiran_2013[OK] soubiran_2013.write("data/Soubiran_2013-xm-Gaia.fits", overwrite=True)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,295
andycasey/velociraptor
refs/heads/master
/run_probabilities3.py
""" Calculate probability distributions. """ import h5py import logging import multiprocessing as mp import numpy as np import sys import pickle import yaml import warnings from tqdm import tqdm from collections import OrderedDict from astropy.io import fits from scipy import (special, stats) import npm_utils as npm if __name__ == "__main__": data_path = sys.argv[1] meta_path = sys.argv[2] if not data_path.endswith(".hdf5"): raise ValueError("data_path should be a .hdf5 file") if not meta_path.endswith(".meta"): raise ValueError("meta_path should be a .meta file") # Load meta. with open(meta_path, "r") as fp: meta = yaml.load(fp) # Load data. data = h5py.File(data_path, "r+") model_names = list(data["models"].keys()) predictor_label_names = [meta["models"][mn]["predictor_label_name"] \ for mn in model_names] all_label_names = [] + predictor_label_names for model_name, model_config in meta["models"].items(): all_label_names.extend(model_config["kdtree_label_names"]) all_label_names = list(np.unique(all_label_names)) # Load the predictors that we need. with fits.open(meta["data_path"]) as image: finite = np.all([np.isfinite(image[1].data[ln]) for ln in all_label_names], axis=0) Y = np.array([image[1].data[pln] for pln in predictor_label_names]).T[finite] # Store relevant data. for label_name in all_label_names + ["source_id"]: if label_name not in data: data.create_dataset(label_name, data=np.array(image[1].data[label_name])[finite]) data.close() del data K = meta.get("number_of_draws", 10) N, M = Y.shape for m, model_name in enumerate(model_names[1:]): print(f"Calculating probabilities for {model_name} with K = {K}") def _init_worker(gp_predictions_, y_, rhos_, bounds_): global gp_predictions, y, rhos, bounds gp_predictions, y, rhos, bounds = (gp_predictions_, y_, rhos_, bounds_) return None def _get_lnprob(n): gp_predictions_, y_ = gp_predictions[n], y[n] if not np.all(np.isfinite(gp_predictions_)) or not np.isfinite(y_): return (n, -1e25, 0.5) mu = gp_predictions_[::2] diag = np.atleast_2d(gp_predictions_[1::2])**0.5 cov = diag * rhos * diag.T try: draws = np.random.multivariate_normal(mu, cov, K).T except ValueError: print(f"ERROR ON {n}, {mu}, {cov}, {diag}, {rhos}") return (n, -1e25, 0.5) # Clip to bounded values. parameter_names = ("theta", "mu_single", "sigma_single", "mu_multiple", "sigma_multiple") for i, parameter_name in enumerate(parameter_names): try: l, u = bounds[parameter_name] except KeyError: continue else: draws[i] = np.clip(draws[i], l, u) draws_3_min = np.log(draws[1] + 5 * draws[2]) + draws[4]**2 draws[3] = np.max([draws[3], draws_3_min], axis=0) with warnings.catch_warnings(): warnings.simplefilter("ignore") lnprob_ = np.vstack([ np.log(draws[0]) + npm.normal_lpdf(y_, draws[1], draws[2]), np.log(1 - draws[0]) + npm.lognormal_lpdf(y_, draws[3], draws[4]) ]).T lnprob_[~np.isfinite(lnprob_)] = -1e25 p_single_ = np.exp(lnprob_[:, 0] - special.logsumexp(lnprob_, axis=1)) # TODO HACK is_def_single = y_ < draws[1] p_single_[is_def_single] = 1.0 return (n, lnprob_, p_single_) chunk_size = 1000000 chunks = int(np.ceil(N/chunk_size)) print(f"Using chunk size of {chunk_size} (expect {chunks} chunks)") with tqdm(total=N) as pbar: for i in range(chunks): start, end = (chunk_size * i, chunk_size * (i + 1)) with h5py.File(data_path, "r+") as data: group = data["models"][model_name] if i == 0: if "lnprob" not in group: group.create_dataset("lnprob", shape=(N, K, 2), dtype=float) if "p_single" not in group: group.create_dataset("p_single", shape=(N, K), dtype=float) if "classification" not in group: group.create_dataset("classification", shape=(N, ), dtype=int) if "confidence" not in group: group.create_dataset("confidence", shape=(N, ), dtype=float) initargs = ( group["gp_predictions"], Y[:, m], np.corrcoef(np.transpose(group["mixture_model_results"])), meta["models"][model_name]["bounds"], ) C = min(N - chunk_size * i, chunk_size) end = start + C lnprob_tmp = np.memmap( "lnprob.tmp", mode="w+", dtype=float, shape=(C, K, 2)) p_single_tmp = np.memmap( "p_single.tmp", mode="w+", dtype=float, shape=(C, K)) with mp.Pool(initializer=_init_worker, initargs=initargs) as p: for n, lnp, ps in p.imap_unordered(_get_lnprob, range(start, end)): lnprob_tmp[n - start] = lnp p_single_tmp[n - start] = ps pbar.update(1) lnprob_tmp.flush() p_single_tmp.flush() confidence = np.sum(p_single_tmp > 0.5, axis=1)/K assert confidence.size == C group["lnprob"][start:end] = lnprob_tmp group["p_single"][start:end] = p_single_tmp group["classification"][start:end] = np.round(confidence).astype(int) group["confidence"][start:end] = confidence del lnprob_tmp, p_single_tmp, confidence chunk_size = 100000 chunks = int(np.ceil(N/chunk_size)) # Calculate joint probabilities. print(f"Calculating joint probabilities") with tqdm(total=N) as pbar: for i in range(chunks): start, end = (chunk_size * i, chunk_size * (i + 1)) C = min(N - chunk_size * i, chunk_size) end = start + C with h5py.File(data_path, "r+") as data: if i == 0: if "joint_p_single" not in data: data.create_dataset("joint_p_single", shape=(N, K), dtype=float) if "joint_classification" not in data: data.create_dataset("joint_classification", shape=(N,), dtype=int) if "joint_confidence" not in data: data.create_dataset("joint_confidence", shape=(N, ), dtype=float) # Get lnprobs. lnprobs = np.sum( [data["models"][model_name]["lnprob"][start:end] \ for model_name in model_names], axis=0) joint_p_single = np.exp(lnprobs[:, :, 0] - special.logsumexp(lnprobs, axis=-1)) data["joint_p_single"][start:end] = joint_p_single confidence = np.sum(joint_p_single > 0.5, axis=1)/K data["joint_classification"][start:end] = np.round(confidence).astype(int) data["joint_confidence"][start:end] = confidence pbar.update(C)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,296
andycasey/velociraptor
refs/heads/master
/attic/plot_sb2.py
""" Selection and analysis of SB2 type binary stars. See notebooks/sb2.ipynb for inspiration. """ import numpy as np import matplotlib import matplotlib.pyplot as plt from matplotlib.ticker import MaxNLocator from astropy.io import fits import velociraptor from mpl_utils import mpl_style plt.style.use(mpl_style) data = fits.open("data/gaia-sources-for-npm.fits")[1].data # Apply ranges valid_ranges = dict( bp_rp=(1.2, 2.3), phot_rp_mean_mag=(10, 12.2)) def get_rv_completeness(x, semilogx, equidensity, N_bins, x_min=None, x_max=None): y = data["radial_velocity"] x_finite = np.isfinite(x) y_finite = np.isfinite(y) if x_min is None: x_min = np.min(x[x_finite]) if x_max is None: x_max = np.max(x[x_finite]) mask = (x_max >= x) * (x >= x_min) x_finite *= mask y_finite *= mask p = np.linspace(0, 100, N_bins) if equidensity and semilogx: bins = 10**np.percentile(np.log10(x[x_finite]), p) elif equidensity and not semilogx: bins = np.percentile(x[x_finite], p) elif not equidensity and semilogx: bins = np.logspace(np.log10(x_min), np.log10(x_max), N_bins) elif not equidensity and not semilogx: bins = np.linspace(x_min, x_max, N_bins) numerator, _ = np.histogram(x[x_finite * y_finite], bins=bins) denominator, _ = np.histogram(x[x_finite], bins=bins) f = numerator/denominator.astype(float) # Pretty sure this ~has~ to be wrong. f_err = f / np.diff(bins) * np.sqrt( (np.sqrt(numerator)/numerator)**2 + \ (np.sqrt(denominator)/denominator)**2) return (bins, f, f_err, numerator, denominator) def plot_rv_completeness(x, latex_label_name, semilogx, equidensity, N_bins, ax=None, title=None, x_min=None, x_max=None, valid_xrange=None, snap_xrange_to_nearest_bin_edge=False, **kwargs): if ax is None: fig, ax = plt.subplots(1, 1, figsize=(10, 10)) else: fig = ax.figure bins, f, f_err, num, den = get_rv_completeness( x, semilogx=semilogx, equidensity=equidensity, N_bins=N_bins, x_min=x_min, x_max=x_max) x = np.hstack([bins[0], np.repeat(bins[1:-1], 2), bins[-1]]) x_mid = bins[:-1] + 0.5 * np.diff(bins) y = np.array(f).repeat(2) kwds = dict(linestyle="-", marker="None") kwds.update(kwargs) line = ax.plot(x, y, **kwds) ax.errorbar(x_mid, f, yerr=f_err, fmt="none", c=line[0].get_color()) ax.set_ylabel(r"\textrm{fraction of sources with radial velocity}") ax.set_xlabel(latex_label_name) if semilogx: ax.semilogx() if valid_xrange is not None: lower, upper = valid_xrange if snap_xrange_to_nearest_bin_edge: mask = (bins >= lower) * (bins <= upper) lower = bins[mask][0] upper = bins[mask][-1] print("Updated valid range for {} is ({:.2f}, {:.2f})".format( latex_label_name, lower, upper)) ax.axvspan(lower, upper, facecolor="#dddddd", edgecolor="None", zorder=-1) if x_min is not None: ax.set_xlim(x_min, ax.get_xlim()[1]) if x_max is not None: ax.set_xlim(ax.get_xlim()[0], x_max) return fig label_names = ("bp_rp", "phot_rp_mean_mag") latex_label_names = dict(bp_rp=r"\textrm{bp - rp}", phot_rp_mean_mag=r"\textrm{apparent rp magnitude}") K = len(label_names) fig, axes = plt.subplots(1, K, figsize=(10, 5)) common_kwds = dict(N_bins=30, semilogx=False, equidensity=False, snap_xrange_to_nearest_bin_edge=True) plot_rv_completeness( data["bp_rp"], latex_label_name=r"\textrm{bp - rp}", x_min=0, x_max=4, valid_xrange=valid_ranges["bp_rp"], ax=axes[0], **common_kwds) plot_rv_completeness( data["phot_rp_mean_mag"], latex_label_name=r"\textrm{apparent rp magnitude}", x_min=6, x_max=12.4, valid_xrange=valid_ranges["phot_rp_mean_mag"], ax=axes[1], **common_kwds) for ax in axes: ax.xaxis.set_major_locator(MaxNLocator(7)) ax.yaxis.set_major_locator(MaxNLocator(5)) ax.set_ylim(-0.05, 1.05) fig.tight_layout() fig.savefig("figures/sb2_rvs_completeness.png", dpi=150) fig.savefig("figures/sb2_rvs_completeness.pdf", dpi=300) in_sb2_source_parameter_range = np.ones(len(data), dtype=bool) for label_name, (lower_value, upper_value) in valid_ranges.items(): if lower_value is not None: in_sb2_source_parameter_range *= (data[label_name] >= lower_value) if upper_value is not None: in_sb2_source_parameter_range *= (upper_value >= data[label_name]) finite_rv = np.isfinite(data["radial_velocity"]) is_sb2 = in_sb2_source_parameter_range * ~finite_rv N_in_sb2_sp_range = sum(in_sb2_source_parameter_range) N_is_sb2 = sum(is_sb2) print(""" Total sources in valid SB2 range: {0:.0f} Numer of sources in that range without an RV: {1:.0f} """.format(N_in_sb2_sp_range, N_is_sb2)) # Plot the source property distributions of the SB2 candidate systems relative to a control sample. # For each SB2 candidate, we need to find the closest star in (bp - rp, apparent rp mag, absolute rp mag) absolute_rp_mag = data["phot_rp_mean_mag"] + 5 * np.log10(data["parallax"]/100.0) X = np.vstack([ data["bp_rp"], data["phot_rp_mean_mag"], absolute_rp_mag ]).T # Build a k-d tree using the stars that are NOT SB2s. # We can query this in parameter space that we care about to get the clsoest non-SB2. import npm_utils as npm finite = np.all(np.isfinite(X), axis=1) kdt_indices = np.where(finite * ~is_sb2)[0] kdt, scale, offset = npm.build_kdtree(X[kdt_indices], relative_scales=[0.1, 1.0, 1.0]) kdt_kwds = dict(offset=offset, scale=scale, minimum_points=1, maximum_points=1) sb2_indices = np.arange(len(data))[is_sb2] control_indices = np.nan * np.ones_like(sb2_indices) K = sb2_indices.size import tqdm for i, index in tqdm.tqdm(enumerate(sb2_indices), total=K): try: indices_returned = list(kdt_indices[kdt.query(X[[index]], 1, return_distance=False)][0]) except ValueError: continue control_indices[i] = indices_returned[0] comp_subset = np.isfinite(control_indices) subset_sb2_indices = sb2_indices[comp_subset] subset_cnt_indices = control_indices[comp_subset].astype(int) astrometric_unit_weight_error = np.sqrt(data["astrometric_chi2_al"]/(data["astrometric_n_obs_al"] - 5)) from mpl_utils import plot_histogram_steps fig, axes = plt.subplots(3, 1, figsize=(3.4, 10)) B = 50 bins = np.linspace(0.5, 3, B) sb2_color, control_color = ("tab:red", "#666666") sb2_label, control_label = (r"\textrm{SB2 candidates}", r"\textrm{control sample}") fill_alpha = 0.3 def plot_twosamples(ax, bins, control_sample, sb2_sample): x = bins[:-1] + 0.5 * np.diff(bins) y = np.histogram(control_sample, bins=bins) ax.plot(x, y[0], "-", c=control_color, drawstyle="steps-mid", label=control_label) ax.errorbar(x, y[0], yerr=np.sqrt(y[0]), c=control_color, fmt=None, ecolor=control_color) xx = np.array(x).repeat(2)[1:] xstep = np.repeat((x[1:] - x[:-1]), 2) xstep = np.concatenate(([xstep[0]], xstep, [xstep[-1]])) # Now: add one step at end of row. xx = np.append(xx, xx.max() + xstep[-1]) - xstep/2.0 yy = np.array(y[0]).repeat(2) ax.fill_between(xx, 0, yy, facecolor=control_color, alpha=fill_alpha) y = np.histogram(sb2_sample, bins=bins) ax.plot(x, y[0], "-", c=sb2_color, drawstyle="steps-mid", label=sb2_label) ax.errorbar(x, y[0], yerr=np.sqrt(y[0]), c=sb2_color, fmt=None, ecolor=sb2_color) xx = np.array(x).repeat(2)[1:] xstep = np.repeat((x[1:] - x[:-1]), 2) xstep = np.concatenate(([xstep[0]], xstep, [xstep[-1]])) # Now: add one step at end of row. xx = np.append(xx, xx.max() + xstep[-1]) - xstep/2.0 yy = np.array(y[0]).repeat(2) ax.fill_between(xx, 0, yy, facecolor=sb2_color, alpha=fill_alpha) plot_twosamples(axes[0], bins, astrometric_unit_weight_error[subset_cnt_indices], astrometric_unit_weight_error[subset_sb2_indices]) axes[0].set_xlabel(r"\textrm{astrometric unit weight error}")#$u = (\chi_{al}^2/\nu)^{1/2}$") axes[0].set_ylabel(r"\textrm{count / $10^{4}$}") axes[0].legend(frameon=False) def do_yticks(ax, step=10000): ticks = np.arange(0, 10*step, step) ax.set_yticks(ticks) ax.set_ylim(0, ticks[-1]) ax.set_yticklabels([r"${:.0f}$".format(ea) for ea in (ticks/10000).astype(int)]) return ticks do_yticks(axes[0]) axes[0].xaxis.set_major_locator(MaxNLocator(6)) """ axes[1].hist(data["phot_bp_rp_excess_factor"][subset_cnt_indices], facecolor=control_color, edgecolor=control_color, label=control_label, **hist_kwds) axes[1].hist(data["phot_bp_rp_excess_factor"][subset_sb2_indices], facecolor=sb2_color, edgecolor=sb2_color, label=sb2_label, **hist_kwds) """ plot_twosamples(axes[1], np.linspace(1.2, 1.5, B), data["phot_bp_rp_excess_factor"][subset_cnt_indices], data["phot_bp_rp_excess_factor"][subset_sb2_indices]) axes[1].set_xlabel(r"\textrm{phot bp - rp excess factor}") axes[1].set_ylabel(r"\textrm{count / $10^4$}") #axes[1].set_ylim(0, 10000 axes[1].xaxis.set_major_locator(MaxNLocator(4)) axes[1].legend(frameon=False) do_yticks(axes[1]) phot_g_variability = np.log10(np.sqrt(data["astrometric_n_good_obs_al"]) \ * data["phot_g_mean_flux_error"] / data["phot_g_mean_flux"]) plot_twosamples(axes[2], np.linspace(-3, -1, B), phot_g_variability[subset_cnt_indices], phot_g_variability[subset_sb2_indices]) axes[2].set_xlabel(r"$\log_{10}\left(\textrm{photometric variability}\right)$") axes[2].set_ylabel(r"\textrm{count / $10^4$}") axes[2].xaxis.set_major_locator(MaxNLocator(6)) do_yticks(axes[2]) axes[2].legend(frameon=False) fig.tight_layout() offset = 0.05 for axlabel, ax in zip("abc", axes): ax.text(offset, 1 - offset, r"\textrm{{({0})}}".format(axlabel), transform=ax.transAxes, horizontalalignment="left", verticalalignment="top") fig.tight_layout() fig.savefig("figures/sb2_histograms.png", dpi=150) fig.savefig("figures/sb2_histograms.pdf", dpi=300) # Plot the completeness fraction across the H-R diagram. def plot_density_fraction(x, y, N_bins=150, min_points_per_bin=5, x_min=None, x_max=None, y_min=None, y_max=None, ax=None, xlabel=None, ylabel=None, figsize=(8, 8), colorbar=True, log=False, mask=None, **kwargs): if ax is None: fig, ax = plt.subplots(1, 1, figsize=figsize) else: fig = ax.figure finite = np.isfinite(x * y) if x_min is not None: finite *= (x >= x_min) if x_max is not None: finite *= (x_max >= x) if y_min is not None: finite *= (y >= y_min) if y_max is not None: finite *= (y_max >= y) den = finite if mask is not None: den *= mask num = (~finite_rv) * den H_all, xedges, yedges = np.histogram2d(x[den], y[den], bins=N_bins) H_bin, _, __ = np.histogram2d(x[num], y[num], bins=(xedges, yedges)) H = H_bin/H_all.astype(float) H[H_all < min_points_per_bin] = np.nan if log: H = np.log(1 + H) print(np.nanmin(H), np.nanmax(H)) kwds = dict( aspect=np.ptp(xedges)/np.ptp(yedges), extent=(xedges[0], xedges[-1], yedges[-1], yedges[0]), cmap="inferno", ) kwds.update(kwargs) image = ax.imshow(H.T, **kwds) if xlabel is not None: ax.set_xlabel(xlabel) if ylabel is not None: ax.set_ylabel(ylabel) if colorbar: cax = fig.add_axes([0.90, 0.125, 0.05, 0.75]) cbar = plt.colorbar(image, cax=cax, orientation="vertical") cbar.set_label(r"\textrm{fraction of stars without radial velocity}") return fig fig, axes = plt.subplots(2, 1, figsize=(12, 7.5)) B = 100 kwds = dict(x=data["l"], y=data["b"], N_bins=(2 * B, B), aspect=0.5, vmin=0, vmax=0.5, colorbar=False, cmap="Greys", xlabel=r"$l$\textrm{ / deg}", ylabel=r"$b$\textrm{ / deg}") plot_density_fraction(ax=axes[0], mask=None, **kwds) plot_density_fraction(ax=axes[1], mask=in_sb2_source_parameter_range, **kwds) for ax in axes: ax.set_xticks(np.arange(0, 361, 45).astype(int)) fig.tight_layout() fig.savefig("figures/sb2_sky_structure.png", dpi=150) fig.savefig("figures/sb2_sky_structure.pdf", dpi=300)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,297
andycasey/velociraptor
refs/heads/master
/article/figures/mpl_utils.py
# A matplotlib style based on the gala package by @adrn: # github.com/adrn/gala import numpy as np import matplotlib.pyplot as plt from scipy.stats import binned_statistic_2d mpl_style = { # Lines 'lines.linewidth': 1.7, 'lines.antialiased': True, 'lines.marker': '.', 'lines.markersize': 5., # Patches 'patch.linewidth': 1.0, 'patch.facecolor': '#348ABD', 'patch.edgecolor': '#CCCCCC', 'patch.antialiased': True, # images 'image.origin': 'upper', # colormap 'image.cmap': 'viridis', # Font 'font.size': 12.0, 'text.usetex': True, 'text.latex.preamble': r'\usepackage{amsmath}', 'text.latex.preview': True, 'axes.unicode_minus': False, # Axes 'axes.facecolor': '#FFFFFF', 'axes.edgecolor': '#333333', 'axes.linewidth': 1.0, 'axes.grid': False, 'axes.titlesize': 'x-large', 'axes.labelsize': 'large', 'axes.labelcolor': 'k', 'axes.axisbelow': True, # Ticks 'xtick.major.size': 8, 'xtick.minor.size': 4, 'xtick.major.pad': 6, 'xtick.minor.pad': 6, 'xtick.color': '#333333', 'xtick.direction': 'in', 'ytick.major.size': 8, 'ytick.minor.size': 4, 'ytick.major.pad': 6, 'ytick.minor.pad': 6, 'ytick.color': '#333333', 'ytick.direction': 'in', 'xtick.labelsize': 'medium', 'ytick.labelsize': 'medium', # Legend 'legend.fancybox': True, 'legend.loc': 'best', # Figure 'figure.figsize': [6, 6], 'figure.facecolor': '1.0', 'figure.edgecolor': '0.50', 'figure.subplot.hspace': 0.5, # Other 'savefig.dpi': 300, } def plot_histogram_steps(ax, x_bins, y, y_err, **kwargs): xx = np.array(x_bins).repeat(2)[1:] xstep = np.repeat((x_bins[1:] - x_bins[:-1]), 2) xstep = np.concatenate(([xstep[0]], xstep, [xstep[-1]])) # Now: add one step at end of row. xx = np.append(xx, xx.max() + xstep[-1]) - xstep/2.0 yy = np.array(y).repeat(2) #xbc = x_bins[:-1] + 0.5 * np.diff(x_bins) _ = ax.plot(xx, yy, '-', **kwargs) ax.errorbar(x_bins, y, y_err, fmt="none", capsize=0, ecolor=_[0].get_color()) return (xx, yy) def plot_binned_statistic(x, y, z, bins=100, function=np.nanmedian, xlabel=None, ylabel=None, zlabel=None, ax=None, colorbar=False, figsize=(8, 8), vmin=None, vmax=None, min_entries_per_bin=None, subsample=None, mask=None, **kwargs): if ax is None: fig, ax = plt.subplots(1, 1, figsize=figsize) else: fig = ax.figure finite = np.isfinite(x * y * z) if mask is not None: finite *= mask if subsample is not None: idx = np.where(finite)[0] if subsample < 1: subsample *= idx.size if int(subsample) > idx.size: finite = idx else: finite = np.random.choice(idx, int(subsample), replace=False) H, xedges, yedges, binnumber = binned_statistic_2d( x[finite], y[finite], z[finite], statistic=function, bins=bins) if min_entries_per_bin is not None: if function != "count": H_count, _, __, ___ = binned_statistic_2d( x[finite], y[finite], z[finite], statistic="count", bins=bins) else: H_count = H H[H_count < min_entries_per_bin] = np.nan if vmin is None or vmax is None: vmin_default, med, vmax_default = np.nanpercentile(H, [16, 50, 84]) if vmin is None: vmin = vmin_default if vmax is None: vmax = vmax_default imshow_kwds = dict( vmin=vmin, vmax=vmax, aspect=np.ptp(xedges)/np.ptp(yedges), extent=(xedges[0], xedges[-1], yedges[-1], yedges[0]), cmap="inferno") imshow_kwds.update(kwargs) image = ax.imshow(H.T, **imshow_kwds) if colorbar: cbar = plt.colorbar(image, ax=ax) if zlabel is not None: cbar.set_label(zlabel) if xlabel is not None: ax.set_xlabel(xlabel) if ylabel is not None: ax.set_ylabel(ylabel) #fig.tight_layout() return fig
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,298
andycasey/velociraptor
refs/heads/master
/attic/validation_hrd.py
""" Validation of the model predictions by making H-R diagram plots. """ import numpy as np import matplotlib.pyplot as plt import velociraptor from matplotlib.colors import LogNorm from matplotlib.ticker import MaxNLocator from astropy.table import Table sources = Table.read("data/rv-all.fits") is_binary = sources["p_sbx"] > 0.5 finite = np.isfinite(sources["bp_rp"] * sources["absolute_g_mag"]) \ * ((sources["parallax"]/sources["parallax_error"]) > 5) \ * np.isfinite(sources["radial_velocity"]) fig, axes = plt.subplots(1, 2, figsize=(8, 4)) N_bins = 100 H_all, xedges, yedges = np.histogram2d( sources["bp_rp"][finite], sources["absolute_g_mag"][finite], bins=(N_bins, N_bins)) H_bin, _, __ = np.histogram2d( sources["bp_rp"][is_binary * finite], sources["absolute_g_mag"][is_binary * finite], bins=(xedges, yedges)) H = H_bin/H_all H[H_all < 3] = np.nan kwds = dict( aspect=np.ptp(xedges)/np.ptp(yedges), extent=(xedges[0], xedges[-1], yedges[-1], yedges[0]), ) image = axes[0].imshow(H_all.T, norm=LogNorm(), cmap="Greys", **kwds) image = axes[1].imshow(H.T, cmap="viridis", **kwds) cax = fig.add_axes([0.825, 0.85, 0.125, 0.05]) cbar = plt.colorbar(image, cax=cax, orientation="horizontal") cbar.set_ticks([0, 1]) cbar.set_label(r"\textrm{binary fraction}") fig.tight_layout() for ax in axes: ax.set_xlabel(r"\textrm{bp-rp}") ax.set_ylabel(r"\textrm{absolute G magnitude}") ax.xaxis.set_major_locator(MaxNLocator(6)) ax.yaxis.set_major_locator(MaxNLocator(6)) axes[0].text(0.95, 0.95, r"$N = {0}$".format(sum(finite)), transform=axes[0].transAxes, horizontalalignment="right", verticalalignment="top") fig.tight_layout() fig.savefig("figures/validation-hrd-sbx.pdf", dpi=150)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,299
andycasey/velociraptor
refs/heads/master
/attic/validation_sb9.py
""" Validation of the model predictions compared to other binary catalogues. """ import numpy as np import matplotlib.pyplot as plt import velociraptor from astropy.table import Table sources = Table.read("data/rv-all.fits") unimodal_catalog = Table.read("data/sb9_matched_by_position.fits") unimodal_indices = np.nan * np.ones(len(unimodal_catalog)) for i, source_id in enumerate(unimodal_catalog["source_id"]): try: unimodal_indices[i] = np.where(sources["source_id"] == int(source_id))[0][0] except: continue finite = np.isfinite(unimodal_indices) unimodal_catalog = unimodal_catalog[finite] unimodal_sources = sources[unimodal_indices[finite].astype(int)] scalar = 1.0 / (unimodal_catalog["Per"] * (1 - unimodal_catalog["e"]**2)**0.5) x = unimodal_catalog["K1"] * scalar y = unimodal_sources["excess_rv_sigma"] * scalar c = unimodal_sources["p_sbx"] fig, ax = plt.subplots(1, 1, figsize=(7.5, 6)) scat = ax.scatter(x, y, c=c, vmin=0, vmax=1, s=5, cmap="coolwarm_r") ax.loglog() ax.set_xlim(10**-4.5, 10**3) ax.set_ylim(10**-4.5, 10**3) cbar = plt.colorbar(scat) cbar.set_label(r"\textrm{binary probability}") ax.set_xlabel(r"${K}/{P\sqrt{1-e^2}}$") ax.set_ylabel(r"${\sigma_\textrm{vrad excess}}/{P\sqrt{1-e^2}}$") fig.tight_layout() fig.savefig("figures/sb9-comparison.pdf", dpi=150) x = unimodal_catalog["K1"] y = unimodal_sources["excess_rv_sigma"] c = unimodal_sources["p_sbx"] fig, ax = plt.subplots(1, 1, figsize=(7.5, 6)) scat = ax.scatter(x, y, c=c, vmin=0, vmax=1, s=5, cmap="coolwarm_r") ax.loglog() ax.set_xlim(10**-0.5, 10**2.5) ax.set_ylim(10**-0.5, 10**2.5) cbar = plt.colorbar(scat) cbar.set_label(r"\textrm{binary probability}") ax.set_xlabel(r"$K$ $(\textrm{km\,s}^{-1})$") ax.set_ylabel(r"$\sigma_\textrm{vrad excess}$ $(\textrm{km\,s}^{-1})$") fig.tight_layout() fig.savefig("figures/sb9-compare-K-log.pdf", dpi=150) fig, ax = plt.subplots(1, 1, figsize=(7.5, 6)) scat = ax.scatter(x, y, c=c, vmin=0, vmax=1, s=5, cmap="coolwarm_r") ax.set_xlim(0, 100) ax.set_ylim(0, 100) cbar = plt.colorbar(scat) cbar.set_label(r"\textrm{binary probability}") ax.set_xlabel(r"$K$ $(\textrm{km\,s}^{-1})$") ax.set_ylabel(r"$\sigma_\textrm{vrad excess}$ $(\textrm{km\,s}^{-1})$") fig.tight_layout() fig.savefig("figures/sb9-compare-K.pdf", dpi=150) is_binary = unimodal_sources["p_sbx"] > 0.5 bins = np.linspace(0, 3000, 50) fig, ax = plt.subplots() ax.hist(unimodal_catalog["Per"][is_binary], facecolor="b", bins=bins, alpha=0.5) ax.hist(unimodal_catalog["Per"][~is_binary], facecolor="r", bins=bins, alpha=0.5) ax.axvline(682, linestyle=":", c="#666666", lw=1) ax.axvline(682 * 2, linestyle=":", c="#666666", lw=1) ax.set_xlabel(r"\textrm{period} $(\textrm{days})$") ax.set_ylabel(r"\textrm{count}") fig.tight_layout() fig.savefig("figures/sb9-period-distribution.pdf", dpi=150)
{"/scripts/npm_run_elastic_ball_test.py": ["/npm_utils.py"], "/npm.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation.py": ["/velociraptor.py"], "/run_analysis.py": ["/npm_utils.py"], "/run_probabilities2.py": ["/npm_utils.py"], "/scripts/npm_run_soubiran_2013.py": ["/npm_utils.py"], "/run_probabilities3.py": ["/npm_utils.py"], "/attic/plot_sb2.py": ["/velociraptor.py", "/npm_utils.py"], "/attic/validation_hrd.py": ["/velociraptor.py"], "/attic/validation_sb9.py": ["/velociraptor.py"]}
50,302
thelahunginjeet/rpy2ica
refs/heads/master
/tests/test_rpy2ica_pytest.py
from rpy2ica import fastica as rica import numpy as np class TestICA: def setup(self): self.signals = np.vstack([np.sin([x/20.0 for x in xrange(1,1001)]),(1.0 + np.mod(xrange(1000),200) - 100.0)/100.0]) self.mixing = np.array([[0.291, 0.6557], [-0.5439, 0.5572]]) self.X = np.dot(self.mixing,self.signals) self.AR,self.WR,self.SR = rica(self.X,2,method="R",maxIterations=10000) self.AC,self.WC,self.SC = rica(self.X,2,method="C",maxIterations=10000) def test_R_W_orthogonality(self): assert np.allclose(np.dot(self.WR.T,self.WR),np.eye(2),atol=1.0e-06),"native R: W^TW not within 1.0e-06 of I" def test_R_S_recovery(self): from scipy.linalg import det assert np.allclose(1.0,np.abs(det(np.corrcoef(self.SR,self.signals)[0:2,2:])),atol=1.0e-03),"native R: |det(rho(ShatT,S))| not within 1e-03 of unity" def test_C_W_orthogonality(self): assert np.allclose(np.dot(self.WC.T,self.WC),np.eye(2),atol=1.0e-06),"R calling C: W^TW not within 1.0e-06 of I" def test_C_S_recovery(self): from scipy.linalg import det assert np.allclose(1.0,np.abs(det(np.corrcoef(self.SC,self.signals)[0:2,2:])),atol=1.0e-03),"R calling C: |det(rho(ShatT,S))| not within 1e-03 of unity"
{"/__init__.py": ["/rpy2ica.py"]}
50,303
thelahunginjeet/rpy2ica
refs/heads/master
/__init__.py
from rpy2ica import fastica __all__ = ['fastica']
{"/__init__.py": ["/rpy2ica.py"]}
50,304
thelahunginjeet/rpy2ica
refs/heads/master
/rpy2ica.py
''' Wraps R's version of FastICA via Rpy2. See the README for information about possible differences between the python and R versions of FastICA. Created on June 14, 2013 @author: Kevin S. Brown, University of Connecticut This source code is provided under the BSD-3 license, duplicated as follows: Copyright (c) 2013, Kevin S. Brown All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. Neither the name of the University of Connecticut nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. ''' from numpy import array from numpy.random import randn try: import rpy2.robjects.numpy2ri from rpy2.robjects.packages import importr rpy2.robjects.numpy2ri.activate() except ImportError: print "Connection to R via Rpy not available; fastica_rpy will not work." def fastica(X,nSources,algorithm="parallel",nonlinearity="logcosh",method="C",maxIterations=500,tolerance=1.0e-05,Winit=None): """ fastica_rpy wraps the fastICA package in R, with communication to R via rpy2. If rpy2 is not installed (or R is not installed), this will not work. For full documentation see the R package. To use the defaults, simpy call: A, W, S = fastica_rpy(X,nSources) Parameters: ------------ X : numpy array, required data matrix nSources : integer, required desired number of sources to extract algorithm : string, optional "deflation" : components extracted one at a time "parallel" : components extracted all at once nonlinearity: string, optional should be either "logcosh" or "exp" method : string, optional should be either "R" (to do the calculations in R) or "C" (to use the C library R is wrapping) maxIterations : integer, optional number of fixed-point iterations tolerance : float, optional convergence criterion for the unmixing matrix Winit : numpy array, optional initial guess for the unmixing matrix Output: ---------- A : numpy array mixing matix of size X.shape[0] x nSources W : numpy array unmixing matrix of size nSources x X.shape[0] S : numpy array matrix of independent courses, size nSources x X.shape[1] """ if Winit is None: Winit = randn(nSources,nSources) try: rica = importr('fastICA') except NameError: print 'Cannot import package \'fastICA\'; you are missing either rpy2 (python problem) or fastICA (R problem)' outDict = rica.fastICA(X.T,nSources,fun=nonlinearity,method=method,maxit=maxIterations,tol=tolerance,w_init=Winit.T) return array(outDict.rx2('A')).T, array(outDict.rx2('W')).T, array(outDict.rx2('S')).T
{"/__init__.py": ["/rpy2ica.py"]}
50,305
iteohy/covid_model
refs/heads/master
/server.py
from mesa.visualization.ModularVisualization import ModularServer from mesa.visualization.modules import CanvasGrid, ChartModule, TextElement from mesa.visualization.UserParam import UserSettableParameter from model import Covid class HappyElement(TextElement): """ Display a text count of how many happy agents there are. """ def __init__(self): pass def render(self, model): return "Number agents: " + str(len(model.schedule.agents)) + "; Day: " + str(int(model.schedule.steps/model.day_steps)) def covid_draw(agent): """ Portrayal Method for canvas """ if agent is None: return #portrayal = {"Shape": dstate.shape, "r": dstate.radius, "Filled": dstate.filled, "Layer": dstate.layer} portrayal = {"Shape": "circle", "r": 0.5, "Filled": "true", "Layer": 0} dstate = agent.d_state portrayal["Color"] = [dstate.color] portrayal["stroke_color"] = dstate.stroke_color #portrayal["Color"] = ["#FF0000"] #portrayal["stroke_color"] = "#000000" return portrayal happy_element = HappyElement() canvas_element = CanvasGrid(covid_draw, 20, 20, 500, 500) seir_chart = ChartModule([ {"Label": "infected", "Color": "Red"}, {"Label": "exposed", "Color": "Blue"}, {"Label": "removed", "Color": "Grey"}, {"Label": "susceptible", "Color": "Green"}, {"Label": "isolated", "Color": "Black"}]) contact_chart = ChartModule([{"Label": "contact", "Color": "Black"}]) model_params = { "width": 20, "density": UserSettableParameter("number", "Number of agents", value=100), "initial_infected": UserSettableParameter("number", "Number Infected", value=10), "infection_rate": UserSettableParameter("slider", "Infectiousness", 0.7, 0.1, 1.0, 0.05), "detection_rate": UserSettableParameter("slider", "Detection rate", 0.7, 0.1, 1.0, 0.05), "min_infected": UserSettableParameter('number', 'Min Infected duration (days)', value=17), "max_infected": UserSettableParameter('number', 'Max Infected duration (days)', value=24), "mean_infected": UserSettableParameter('number', 'Mean Infected duration (days)', value=20), "min_exposed": UserSettableParameter('number', 'Min Exposed duration (days)', value=3.68), "max_exposed": UserSettableParameter('number', 'Max Exposed duration (days)', value=6.68), "mean_exposed": UserSettableParameter('number', 'Mean Exposed duration (days)', value=5.2), "day_steps": UserSettableParameter('number', 'Number of steps in a day', value=5), "day_isolation": UserSettableParameter('number', 'Number of days to isolation', value=6) } server = ModularServer( Covid, [canvas_element, happy_element, seir_chart, contact_chart], "COVID-19", model_params )
{"/server.py": ["/model.py"]}
50,306
iteohy/covid_model
refs/heads/master
/runbatch.py
from model import Covid from mesa.batchrunner import BatchRunner import time def getInfected(model): return model.cuminfected def getDays(model): return model.schedule.steps/model.day_steps def getStats(model): return model.stats # comm params comm_model_params = { #"width": 600, "density": 50000, "infection_rate": 0.7, "min_infected": 17, "max_infected": 24, "mean_infected": 20, "min_exposed": 3.78, "max_exposed": 6.78, "mean_exposed": 5.2, "day_steps": 3, "initial_infected": 2, "day_isolation": 5, "detection_rate": 0.7 } comm_variable_params = { #"infection_rate": [0.3, 0.5, 0.7, 0.9], #"detection_rate": [0.3, 0.5, 0.7, 0.9], #"day_isolation": [3, 5, 7, 9, 11, 13, 15] "width": [300] } # dorm params dorm_model_params = { #"width": 200, #"density": 5000, "initial_infected": 5, "infection_rate": 0.3, "min_infected": 17, "max_infected": 24, "mean_infected": 20, "min_exposed": 3.78, "max_exposed": 6.78, "mean_exposed": 5.2, "day_steps": 5, "day_isolation": 5, "detection_rate": 0.7 } dorm_variable_params = { #"initial_infected": [10, 20, 30, 40, 50, 60, 70, 80, 90, 100] #"infection_rate": [0.3, 0.5, 0.7, 0.9] #"day_isolation": [3, 5, 7, 9, 11, 13, 15] # grid size #"width": [100, 200, 300, 400, 500] # number of agents #"density": [1000, 2000, 3000, 4000, 5000] # density "width": [100] "density": [1000] } # assign active params model_params = dorm_model_params variable_params = dorm_variable_params batch_run = BatchRunner(Covid, variable_params, model_params, iterations=5, max_steps=200*dorm_model_params['day_steps'], model_reporters={"infected": getInfected, "days":getDays, "stats":getStats}) batch_run.run_all() data = batch_run.get_model_vars_dataframe() data['total_infected'] = data['infected']+data['initial_infected'] data['%'] = 100*data['total_infected']/model_params['density'] data.to_csv("agents_%d.csv"%int(time.time()), index=False) # #sdata = data[["initial_infected", "width", "days", "total_infected", "%"]] #sdata = sdata.groupby(['initial_infected', 'width']).mean() #sdata.to_csv("width_grouped_%d.csv"%int(time.time()), index=True)
{"/server.py": ["/model.py"]}
50,307
iteohy/covid_model
refs/heads/master
/model.py
from mesa import Model, Agent from mesa.time import RandomActivation from mesa.space import MultiGrid from mesa.datacollection import DataCollector class DiseaseState(): def __init__(self, state, color, stroke_color): """ Create a disease state Args: state: Indicator for SEIR color: color of agent for Portrayal stroke_color: stroke color of agent for Portrayal """ self.state = state self.setColors(color, stroke_color) self.setShape("circle", 0.5, "true", 0) self.lifespan = -1 self.assigned_lifespan = -1 def setColors(self, color, stroke_color): #portrayal["Color"] = ["#FF0000"] #portrayal["stroke_color"] = "#000000" self.color = color self.stroke_color = stroke_color def setShape(self, shape, radius, filled, layer): #portrayal = {"Shape": "circle", "r": 0.5, "Filled": "true", "Layer": 0} self.shape = shape self.radius = radius self.layer = layer self.filled = filled def decrementLifespan(self): if self.lifespan > -1: self.lifespan -= 1 def setLifespan(self, lifespan): self.lifespan = int(lifespan) self.assigned_lifespan = int(lifespan) #print(str(self)+': '+str(self.lifespan)) class Susceptible(DiseaseState): def __init__(self): # init disease state Susceptible, color Green, stroke Black super().__init__("S", "#00FF00", "#000000") class Exposed(DiseaseState): def __init__(self): # init disease state Exposed, color Blue, stroke Black super().__init__("E", "#0000FF", "#000000") class Infected(DiseaseState): def __init__(self): # init disease state Infected, color Red, stroke Black super().__init__("I", "#FF0000", "#000000") self.detected = True def set_detected(self, detected): self.detected = detected def is_detected(self): return self.detected class Removed(DiseaseState): def __init__(self): # init disease state Removed, color Grey, stroke Black super().__init__("R", "#CCCCCC", "#000000") class CovidAgent(Agent): """ Agent simulating covid-19 spread """ def __init__(self, pos, model, d_state, move=True, isolate_duration=14): """ Create a new covid agent. Args: unique_id: Unique identifier for the agent. x, y: Agent initial location. d_state: disease state of the agent (S,E,I,R) """ super().__init__(pos, model) self.pos = pos self.contact = set() self.d_state = d_state self.move = move self.isolate_duration = isolate_duration*model.day_steps def step(self): # increment lifespan #self.d_state.incrementLife() self.d_state.decrementLifespan() if False: #if isinstance(self.d_state, Removed): print("assigned_ls, lifespan, day_isolation, isolate_duration, move") print(type(self.d_state)) print(self.d_state.assigned_lifespan) print(self.d_state.lifespan) print(self.model.day_isolation) print(self.isolate_duration) print(self.move) if isinstance(self.d_state, Exposed): if self.d_state.lifespan <= 0 : self.d_state = Infected() ls = self.random.triangular(self.model.min_infected, self.model.max_infected, self.model.mean_infected) self.d_state.setLifespan(ls) self.d_state.set_detected(self.random.random() < self.model.detection_rate) elif isinstance(self.d_state, Infected): if self.d_state.lifespan <= 0: self.d_state = Removed() if (self.d_state.assigned_lifespan - self.d_state.lifespan) >= self.model.day_isolation and self.d_state.is_detected(): self.move = False self.isolate_duration -= 1 # decrease isolation duration # if agent not moving, stop here. if not self.move: # if isolation complete, move again if self.isolate_duration<=0: self.move = True else: return # track agents in contact contents = self.model.grid.get_cell_list_contents(self.pos) if len(contents)>1: for c in contents: if c.unique_id == self.unique_id: continue if c.move: self.contact.add(c.unique_id) # what to do to agent(s) in cell if isinstance(self.d_state, Infected): # infected if isinstance(c.d_state, Susceptible) and c.move: if self.random.uniform(0,1)<self.model.infection_rate: # probability of infection c.d_state = Exposed() ls = self.random.triangular(self.model.min_exposed, self.model.max_exposed, self.model.mean_exposed) c.d_state.setLifespan(ls) self.model.cuminfected += 1 # move to random adjacent cell x, y = self.pos r = self.random.random() if r < 0.33: x = x+1 elif r < 0.66: x = x-1 r = self.random.random() if r < 0.33: y = y+1 elif r<0.66: y = y-1 self.model.grid.move_agent(self, (x,y)) class Covid(Model): """ Model class for the Covid infection model. """ def __init__(self, density, initial_infected, infection_rate, min_infected, max_infected, mean_infected, min_exposed, max_exposed, mean_exposed, day_steps, day_isolation, detection_rate, width=20): """ """ # square grid height=width self.height = height self.width = width self.density = density self.initial_infected = initial_infected self.infection_rate = infection_rate self.detection_rate = detection_rate self.day_steps = day_steps self.min_exposed = min_exposed*self.day_steps self.max_exposed = max_exposed*self.day_steps self.mean_exposed = mean_exposed*self.day_steps self.min_infected = min_infected*self.day_steps self.max_infected = max_infected*self.day_steps self.mean_infected = mean_infected*self.day_steps self.day_isolation = day_isolation*self.day_steps self.schedule = RandomActivation(self) self.grid = MultiGrid(width, height, torus=True) self.infected = 0 self.exposed = 0 self.susceptible = 0 self.removed = 0 self.contact = 0 self.cuminfected = 0 self.isolated = 0 self.stats = {"infected":[], "exposed":[], "susceptible":[], "removed":[], "isolated":[]} self.datacollector = DataCollector( # Model-level count {"contact": "contact", "infected": "infected", "cuminfected":"cuminfected", "exposed":"exposed", "susceptible": "susceptible", "removed": "removed", "isolated":"isolated", "stats":"stats"}, # For testing purposes, agent's individual x and y {"x": lambda a: a.pos[0], "y": lambda a: a.pos[1]} ) # Set up agents # We use a grid iterator that returns # the coordinates of a cell as well as # its contents. (coord_iter) num_agents = 0 num_infected = 0 while num_agents < self.density: #for cell in self.grid.coord_iter(): #x = cell[1] #y = cell[2] # obtaining non-empty cell x = int(self.random.uniform(0, width)) y = int(self.random.uniform(0, height)) while not self.grid.is_cell_empty((x,y)): x = int(self.random.uniform(0, width)) y = int(self.random.uniform(0, height)) if num_agents < self.density and self.random.random()<0.2: if num_infected < self.initial_infected: agent_type = Infected() num_infected += 1 # generate typical infected lifespan from normal distribution ls = self.random.uniform(self.min_infected, self.max_infected) agent_type.setLifespan(ls) agent_type.set_detected(self.random.uniform(0,1)<self.detection_rate) else: agent_type = Susceptible() agent = CovidAgent((x, y), self, agent_type) self.grid.place_agent(agent, (x, y)) self.schedule.add(agent) num_agents += 1 #print(str(num_agents)+" agents finally.") self.running = True #self.datacollector.collect(self) def step(self): """ Run one step of the model. If All agents are happy, halt the model. """ # Reset counters self.infected = 0 self.exposed = 0 self.susceptible = 0 self.removed = 0 self.isolated = 0 self.schedule.step() # compute average contact per agent total = 0 for cell in self.grid.coord_iter(): content, x, y = cell if content: for c in content: total+=len(c.contact) if isinstance(c.d_state, Infected): self.infected += 1 elif isinstance(c.d_state, Exposed): self.exposed += 1 elif isinstance(c.d_state, Susceptible): self.susceptible += 1 elif isinstance(c.d_state, Removed): self.removed += 1 if not c.move: #print("isolated") self.isolated +=1 self.contact = total/self.schedule.get_agent_count() self.stats["infected"].append(self.infected) self.stats["exposed"].append(self.exposed) self.stats["susceptible"].append(self.susceptible) self.stats["removed"].append(self.removed) self.stats["isolated"].append(self.isolated) # collect data self.datacollector.collect(self) if self.infected+self.exposed == 0: self.running = False if self.schedule.steps/self.day_steps > 180: self.running = False
{"/server.py": ["/model.py"]}
50,308
caocscar/opioid-web
refs/heads/master
/application.py
from flask import Flask, render_template, request from opioid_dict import src_dict, center_dict, cities, counties, names, name_cases, name_case_ls from create_D3_files import create_county_files application = Flask(__name__) application.config['SEND_FILE_MAX_AGE_DEFAULT'] = 0 @application.route('/') def homepage(): data = { 'counties': counties, } placenames=[] for each in cities: placenames.append(each) for each in counties: placenames.append(each) placenames.sort() return render_template("index.html", data=data, cities=cities, counties=counties, placenames = placenames) @application.route('/dashboard', methods=['GET']) def generate_report_given_name(): arguments=['city', 'county', 'src', 'T0' , 'T1' ] source = request.args.get('src', default = "EMS", type = str) city = request.args.get('city') county = request.args.get('county') T0 = request.args.get('T0', default = 14, type = int) T1 = request.args.get('T1', default = None, type = int) if county: cityorcounty = "County" name = county src = source if city: cityorcounty = "City" name = city src = source create_county_files(name, source, cityorcounty, T0, T1) source = source.upper() if city: city = city.title() for each in name_cases: if city in each.keys(): city = each[city] if county: county = county.title() if city in cities: if '(City)' in city: propername = city[:-7] else: propername = city folder = 'cities' county_flag = '' data = { 'placename': city, 'propername': propername, 'cityorcounty': "city", 'county':center_dict["City"][propername]["county"], 'src': source, 'county_flag': county_flag, 'titlename': src_dict[source], 'f_geojson': f'geojson/{folder}/{propername}.geojson', 'center': center_dict["City"][propername]['center'], 'zoom' : center_dict["City"][propername].get('zoom', 10)} if county in counties: folder = 'counties' county_flag = 'County' data = { 'placename': county, 'cityorcounty': "county", 'src': source, 'county_flag': county_flag, 'titlename': src_dict[source], 'f_geojson': f'geojson/{folder}/{county}.geojson', 'center': center_dict["County"][county]['center'], 'zoom' : center_dict["County"][county].get('zoom', 10)} return render_template("county_src_report.html", data=data, T0=T0, T1=T1) #%% Run Flask app # python application.py if __name__ == '__main__': application.run(debug=True)
{"/application.py": ["/opioid_dict.py", "/create_D3_files.py"], "/create_D3_files.py": ["/opioid_dict.py"]}
50,309
caocscar/opioid-web
refs/heads/master
/create_D3_files.py
# -*- coding: utf-8 -*- """ Created on Thu Mar 7 14:36:25 2019 @author: umhs-caoa """ import pandas as pd import numpy as np import os from opioid_dict import aggregation_dict, name_correction_dict from datetime import datetime pd.options.display.max_rows = 30 pd.options.display.max_columns = 10 wdir = os.path.join('static','data') savedir = os.path.join('static','data') racelist = ['White','Black','Hispanic or Latino','Asian','American Indian or Alaska Native', 'Native Hawaiian or Other Pacific Islander','Other','Unknown', ] genderlist = ['Female','Male','Unknown'] headers = ['index','value'] files = ['web_EMS.csv','web_EMS_Washtenaw.csv','web_ME.csv','web_ME_Wayne.csv'] list_df = [] for file in files: tmp = pd.read_csv(os.path.join(wdir,file)) tmp['date'] = pd.to_datetime(tmp['date']) dataset = file.split('_')[1] tmp['src'] = dataset.strip('.csv') list_df.append(tmp) df = pd.concat(list_df,ignore_index=True,sort=True) df.replace({'city': name_correction_dict}, inplace=True) #%% def get_firstday(T0, latest_date): if T0 >= 20000000: T0 = str(T0) return f'{T0[:4]}-{T0[4:6]}-{T0[6:]}' else: return (latest_date + pd.DateOffset(days=-T0+1)).strftime('%Y-%m-%d') def get_lastday(T1, latest_date): if T1: T1 = str(T1) return f'{T1[:4]}-{T1[4:6]}-{T1[6:]}' else: return latest_date.strftime('%Y-%m-%d') def create_county_files(name, src, cityorcounty, T0, T1=None): column = 'city' if cityorcounty.lower() == "city" else 'county' cty = df[(df[column].str.contains(name)) & (df['src'] == src)] # daily file latest_date = df.loc[df['src'] == src,'date'].max() T_start = get_firstday(T0, latest_date) T_end = get_lastday(T1, latest_date) if cty.empty: earliest_date = pd.to_datetime(T_start) cty_date = cty daily = pd.Series() else: earliest_date = cty['date'].min().strftime('%Y-%m-%d') ts = cty.set_index('date') daily = ts[column].resample('D').count() daterange = pd.date_range(earliest_date, latest_date, freq='D') daily = daily.reindex(daterange, fill_value=0) daily = daily.to_frame().reset_index() daily.columns = ['date','value'] daily['avg'] = daily['value'].rolling(7, min_periods=1).mean().round(2) create_daily_file(T_start, T_end, daily) # date filtering cty_date = cty[cty['date'].between(T_start,T_end)] # rate and change table dayswin = (pd.to_datetime(T_end) - pd.to_datetime(T_start)).days + 1 evtrte = len(cty_date)/dayswin create_rte_table_file(cty,T_start,dayswin,evtrte) # age, race, gender, gps file create_age_file(cty_date) create_race_file(cty_date) create_gender_file(cty_date) create_gps_file(cty_date, T0, T_end) create_evt_table_file(cty_date,name,src) create_ctyzip_freq_table(cty_date) def create_daily_file(T_start, T_end, daily): daily_subset = daily[daily['date'].between(T_start,T_end)] daily_subset.to_csv(os.path.join(savedir,'county_src_daily.csv'), index=False) def create_age_file(cty_date): age = cty_date['age_grp'].value_counts(sort=False) age.sort_index(inplace=True) age = age.reset_index() age.to_csv(os.path.join(savedir,'county_src_age.csv'), index=False, header=headers) def create_race_file(cty_date): race = cty_date['race'].value_counts() race = race.reindex(racelist, fill_value=0) race = race.reset_index() race.to_csv(os.path.join(savedir,'county_src_race.csv'), index=False, header=headers) def create_gender_file(cty_date): gender = cty_date['gender'].value_counts() gender = gender.reindex(genderlist, fill_value=0) gender = gender.reset_index() gender.to_csv(os.path.join(savedir,'county_src_gender.csv'), index=False, header=headers) def create_evt_table_file(cty_date,name,src): cty_date = cty_date.sort_values(by=['date', 'zipcode']) if src == "EMS": tmpTab = cty_date[['date','city','zipcode']] tmpTab.columns = ['Date','City','Zip Code'] elif src == "ME" and name in ["Wayne","Detroit"]: tmpTab = cty_date[['date','city','location','suspected_indicator']] tmpTab.columns = ['Date','City','Location','Suspected Overdose Indicator'] elif src == "ME": tmpTab = cty_date[['date','city','location']] tmpTab.columns = ['Date','City','Location'] tmpTab = tmpTab.replace({'City':r'.*\d.*'},{'City':np.NaN}, regex=True) tmpTab.to_csv(os.path.join(savedir,'county_src_evttab.csv'), index=False) def create_rte_table_file(cty,T_start,days,evtrte): pp_end = pd.to_datetime(T_start) + pd.DateOffset(days=-1) pp_start = pp_end + pd.DateOffset(days=-days+1) cty_pp = cty[cty['date'].between(pp_start,pp_end)] pp_evtrte = len(cty_pp)/days if pp_start < pd.to_datetime("20190101") or pp_evtrte == 0: rtetab = pd.DataFrame({'Mean Incidents Per Day':[round(evtrte,1)],'Percent Change Since Last Period':[np.NaN]}) else: rtetab = pd.DataFrame({'Mean Incidents Per Day':[round(evtrte,1)],'Percent Change Since Last Period':[round((evtrte-pp_evtrte)/pp_evtrte*100,1)]}) rtetab.to_csv(os.path.join(savedir,'county_src_ratechange.csv'), index=False) def create_ctyzip_freq_table(cty): cty['city'] = cty['city'].str.title() cty.replace({'city': aggregation_dict}, inplace=True) cty_counts = (cty.replace({'city':r'.*\d.*'},{'city':"Unknown"},regex=True))['city'].value_counts().to_frame(name="# Incidents") cty_counts["City"] = cty_counts.index cty_counts.loc[len(cty_counts)] = [len(cty),"Total"] cty_counts["Percent"] = round(cty_counts["# Incidents"]/len(cty)*100,1) cty_counts[["City","# Incidents","Percent"]].to_csv(os.path.join(savedir,'county_src_ctyfreqtab.csv'), index=False) zip_counts = cty['zipcode'].value_counts().to_frame(name="# Incidents") zip_counts["Zip Code"] = zip_counts.index zip_counts.loc[len(zip_counts)] = [len(cty),"Total"] zip_counts["Percent"] = round(zip_counts["# Incidents"]/len(cty)*100,1) zip_counts[["Zip Code","# Incidents","Percent"]].to_csv(os.path.join(savedir,'county_src_zipfreqtab.csv'), index=False) def create_gps_file(cty_date, T0, T_end): if T0 <= 14: cty_date['opacity'] = 1 else: enddate = pd.to_datetime(T_end) if T0 >= 20000000: first_date = datetime.strptime(str(T0),'%Y%m%d') numdays = (enddate - first_date).days else: numdays = T0 delta = enddate - cty_date['date'] cty_date['opacity'] = 1 - delta.dt.days / (numdays + 1) with open(os.path.join(savedir,'county_src_gps.js'),'w') as fout: fout.write('var event_pts = ') cty_date[['lat','lng','opacity']].to_json(fout, orient='records') #%% if __name__ == '__main__': create_county_files('Wayne','EMS', 'county', 20190101)
{"/application.py": ["/opioid_dict.py", "/create_D3_files.py"], "/create_D3_files.py": ["/opioid_dict.py"]}
50,310
caocscar/opioid-web
refs/heads/master
/opioid_dict.py
src_dict = {'EMS':'EMS', 'ED':'Emergency Departments', 'ME':'Medical Examiner'} race_dict = {'B':'Black', 'African American':'Black', 'Black or African American':'Black', 'African American / Black':'Black', 'W':'White', 'White, Not Recorded':'White', 'Patient Refused':'Unknown', '1':'Unknown', '2':'Unknown', '6':'Unknown', '7':'Unknown', '8':'Unknown', 'A':'Unknown', 'D':'Unknown', 'U':'Unknown', 'Not Recorded':'Unknown', 'Not Applicable':'Unknown', 'H':'Hispanic or Latino', 'White, Hispanic or Latino':'Hispanic or Latino', 'White,Hispanic or Latino':'Hispanic or Latino', 'Hispanic or Latino, White':'Hispanic or Latino', 'Hispanic, White':'Hispanic or Latino', 'Asian, White':'Other', 'White, Asian':'Other', 'Asian, Black or African American':'Other', 'Black or African American, White':'Other', 'White, Black or African American':'Other', 'White,Black or African American':'Other', 'Black or African American, Hispanic or Latino':'Other', 'Hispanic or Latino, Black or African American':'Other', 'White, Native Hawaiian or Other Pacific Islander, Asian':'Other', 'Native Hawaiian or Other Pacific Islander, White':'Other', 'White, Native Hawaiian or Other Pacific Islander':'Other', 'White,Native Hawaiian or Other Pacific Islander,Hispanic or Latino,Black or African American,Asian,American Indian or Alaska Native':'Other', 'White, Native Hawaiian or Other Pacific Islander, Hispanic or Latino, Black or African American, Asian, American Indian or Alaska Native':'Other', 'Black or African American, American Indian or Alaska Native':'Other', 'White, Hispanic or Latino, Asian, American Indian or Alaska Native':'Other', 'Unspecified':'Other', 'American Indian / Alaskan Native':'American Indian or Alaska Native', } gender_dict = {'F':'Female', 'M':'Male', 'U':'Unknown', 'Unknown (Unable to Determine)':'Unknown', 'Not Applicable':'Unknown', } aggregation_dict = {'Kalamazoo':'Kalamazoo and Charter Township of Kalamazoo', 'Charter Township Of Kalamazoo':'Kalamazoo and Charter Township of Kalamazoo', 'Port Huron':'Port Huron and Charter Township of Port Huron', 'Charter Township Of Port Huron':'Port Huron and Charter Township of Port Huron', } name_correction_dict = {'Canton':'Charter Township of Canton', } center_dict = { "City" : {'Ann Arbor': { 'center': '{lat: 42.28, lng: -83.73}', 'county': 'Washtenaw', 'zoom' : 12, 'minwidth': 550, 'minheight':400 }, 'Ypsilanti':{ 'center': '{lat: 42.24, lng: -83.61}', 'county' : 'Washtenaw', 'zoom' : 13, 'minwidth' : 500, 'minheight' : 400 }, 'Detroit': { 'center': '{lat: 42.35, lng: -83.10}', 'county': 'Wayne', 'zoom': 11, 'minwidth': 550, 'minheight': 400 }, 'Bay City': { 'center': '{lat: 43.58, lng: -83.88}', 'county': 'Bay', 'zoom': 12, #optimal: 13 'minwidth': 150, #optimal: 450 'minheight': 200 #optimal: 700 }, 'Flint': { 'center': '{lat:43.01, lng:-83.70}', 'county': 'Genesee', 'zoom': 12, #optimal: 12 'minwidth': 250, #optimal: 450 'minheight': 300 #optimal: 550 }, 'Grand Rapids': { 'center': '{lat:42.96, lng:-85.66}', 'county': 'Kent', 'zoom': 12, #optimal: 12 'minwidth': 300, #optimal: 550 'minheight': 300, #optimal: 600 }, 'Pontiac': { 'center': '{lat:42.65, lng:-83.29}', 'county': 'Oakland', 'zoom': 12, #optimal: 12 'minwidth': 150, #optimal: 300 'minheight': 200, #optimal: 400 }, 'Kalamazoo':{ 'center': '{lat:42.29, lng: -85.58}', 'county': 'Kalamazoo', 'zoom': 12, 'minwidth':150, 'minheight': 200 }, 'Muskegon':{ 'center': '{lat:43.23, lng: -86.25}', 'county': 'Muskegon', 'zoom': 12, 'minwidth':150, 'minheight': 200 }, 'Warren':{ 'center': '{lat:42.49, lng: -83.06}', 'county': 'Macomb', 'zoom': 12, 'minwidth':150, 'minheight': 200 }, 'Roseville':{ 'center': '{lat:42.51, lng: -82.94}', 'county': 'Macomb', 'zoom': 13, 'minwidth':150, 'minheight': 200 }, 'Hazel Park':{ 'center': '{lat:42.46, lng:-83.10}', 'county': 'Oakland', 'zoom': 14, 'minwidth':150, 'minheight': 200 }, 'Waterford':{ 'center': '{lat:42.66, lng:-83.42}', 'county': 'Oakland', 'zoom': 12, 'minwidth':150, 'minheight': 200 }, 'Port Huron':{ 'center': '{lat:42.98, lng:-82.44}', 'county': 'St. Clair', 'zoom': 12, 'minwidth':150, 'minheight': 200 }, 'Dearborn':{ 'center': '{lat:42.31, lng:-83.22}', 'county': 'Wayne', 'zoom': 12, 'minwidth':150, 'minheight': 200 }, 'Livonia':{ 'center': '{lat:42.40, lng:-83.37}', 'county': 'Wayne', 'zoom': 12, 'minwidth':150, 'minheight': 200 }, 'Lansing':{ 'center': '{lat:42.70, lng:-84.55}', 'county': 'Ingham, Clinton, Eaton', 'zoom': 11, 'minwidth':150, 'minheight': 200 }, 'Eastpointe':{ 'center': '{lat:42.47, lng:-82.94}', 'county': 'Macomb', 'zoom': 13, 'minwidth':150, 'minheight': 200 }, 'Royal Oak':{ 'center': '{lat:42.51, lng:-83.15}', 'county': 'Oakland', 'zoom': 13, 'minwidth':150, 'minheight': 200 }, 'Lincoln Park':{ 'center': '{lat:42.24, lng:-83.18}', 'county': 'Wayne', 'zoom': 13, 'minwidth':150, 'minheight': 200 }, 'Saginaw':{ 'center': '{lat: 43.42, lng: -83.95}', 'county': 'Saginaw', 'zoom': 12, 'minwidth': 150, 'minheight': 200 }}, 'County': {'Alcona': { 'center': '{lat:44.68, lng:-83.58}', 'minwidth': 450, 'minheight': 400 }, 'Alger': { 'center': '{lat:46.42, lng:-86.49}', 'zoom': 9, 'minwidth': 950, 'minheight': 600 }, 'Allegan': { 'center': '{lat:42.59, lng:-85.91}', 'minwidth': 550, 'minheight': 350 }, 'Alpena': { 'center': '{lat:45.03, lng:-83.58}', 'minwidth': 500, 'minheight': 400 }, 'Antrim': { 'center': '{lat:45.01, lng:-85.14}', 'minwidth': 450, 'minheight': 450 }, 'Arenac': { 'center': '{lat:44.04, lng:-83.87}', 'minwidth': 450, 'minheight': 300 }, 'Baraga': { 'center': '{lat:46.69, lng:-88.34}', 'minwidth': 550, 'minheight': 600 }, 'Barry': { 'center': '{lat:42.60, lng:-85.31}', 'minwidth': 350, 'minheight': 350 }, 'Bay': { 'center': '{lat:43.7, lng:-83.93}', 'minwidth': 350, 'minheight': 550 }, 'Benzie': { 'center': '{lat:44.65, lng:-86.04}', 'minwidth': 350, 'minheight': 300 }, 'Berrien': { 'center': '{lat:42.00, lng:-86.52}', 'minwidth': 450, 'minheight': 500 }, 'Branch': { 'center': '{lat:41.92, lng:-85.06}', 'minwidth': 350, 'minheight': 350 }, 'Calhoun': { 'center': '{lat:42.25, lng:-85.00}', 'minwidth': 450, 'minheight': 350 }, 'Cass': { 'center': '{lat:41.92, lng:-85.99}', 'minwidth': 350, 'minheight': 350 }, 'Charlevoix': { 'center': '{lat:45.48, lng:-85.22}', 'zoom': 9, 'minwidth': 400, 'minheight': 400 }, 'Cheboygan': { 'center': '{lat:45.49, lng:-84.47}', 'minwidth': 400, 'minheight': 650 }, 'Chippewa': { 'center': '{lat:46.34, lng:-84.36}', 'zoom': 9, 'minwidth': 650, 'minheight': 500 }, 'Clare': { 'center': '{lat:44.00, lng:-84.85}', 'minwidth': 400, 'minheight': 400 }, 'Clinton': { 'center': '{lat:42.94, lng:-84.60}', 'minwidth': 350, 'minheight': 350 }, 'Crawford': { 'center': '{lat:44.68, lng:-84.61}', 'minwidth': 350, 'minheight': 400 }, 'Delta': { 'center': '{lat:45.85, lng:-86.91}', 'minwidth': 700, 'minheight': 650 }, 'Dickinson': { 'center': '{lat:45.98, lng:-87.88}', 'minwidth': 400, 'minheight': 550 }, 'Eaton': { 'center': '{lat:42.60, lng:-84.84}', 'minwidth': 350, 'minheight': 350 }, 'Emmet': { 'center': '{lat:45.53, lng:-84.93}', 'minwidth': 300, 'minheight': 600 }, 'Genesee': { 'center': '{lat:43.00, lng:-83.69}', 'minwidth': 350, 'minheight': 450 }, 'Gladwin': { 'center': '{lat:43.99, lng:-84.39}', 'minwidth': 350, 'minheight': 400 }, 'Gogebic': { 'center': '{lat:46.43, lng:-89.70}', 'zoom': 9, 'minwidth': 550, 'minheight': 400 }, 'Grand Traverse': { 'center': '{lat:44.75, lng:-85.57}', 'minwidth': 400, 'minheight': 500 }, 'Gratiot': { 'center': '{lat:43.29, lng:-84.61}', 'minwidth': 350, 'minheight': 350 }, 'Hillsdale': { 'center': '{lat:41.88, lng:-84.59}', 'minwidth': 350, 'minheight': 400 }, 'Houghton': { 'center': '{lat:46.85, lng:-88.61}', 'zoom': 9, 'minwidth': 300, 'minheight': 500 }, 'Huron': { 'center': '{lat:43.87, lng:-83.04}', 'minwidth': 650, 'minheight': 450 }, 'Ingham': { 'center': '{lat:42.60, lng:-84.37}', 'minwidth': 350, 'minheight': 400 }, 'Ionia': { 'center': '{lat:42.94, lng:-85.07}', 'minwidth': 350, 'minheight': 350 }, 'Iosco': { 'center': '{lat:44.34, lng:-83.60}', 'minwidth': 450, 'minheight': 400 }, 'Iron': { 'center': '{lat:46.17, lng:-88.55}', 'minwidth': 650, 'minheight': 550 }, 'Isabella': { 'center': '{lat:43.64, lng:-84.85}', 'minwidth': 400, 'minheight': 400 }, 'Jackson': { 'center': '{lat:42.25, lng:-84.42}', 'minwidth': 450, 'minheight': 350 }, 'Kalamazoo': { 'center': '{lat:42.25, lng:-85.53}', 'minwidth': 350, 'minheight': 350 }, 'Kalkaska': { 'center': '{lat:44.69, lng:-85.09}', 'minwidth': 400, 'minheight': 400 }, 'Kent': { 'center': '{lat:43.03, lng:-85.55}', 'minwidth': 400, 'minheight': 550 }, 'Keweenaw': { 'center': '{lat:47.70, lng:-88.49}', 'zoom': 9, 'minwidth': 600, 'minheight': 550 }, 'Lake': { 'center': '{lat:43.99, lng:-85.80}', 'minwidth': 350, 'minheight': 400 }, 'Lapeer': { 'center': '{lat:43.10, lng:-83.22}', 'minwidth': 350, 'minheight': 450 }, 'Leelanau': { 'center': '{lat:45.11, lng:-85.84}', 'minwidth': 450, 'minheight': 700 }, 'Lenawee': { 'center': '{lat:41.93, lng:-84.06}', 'minwidth': 450, 'minheight': 400 }, 'Livingston': { 'center': '{lat:42.60, lng:-83.91}', 'minwidth': 400, 'minheight': 400 }, 'Luce': { 'center': '{lat:46.50, lng:-85.55}', 'minwidth': 500, 'minheight': 550 }, 'Mackinac': { 'center': '{lat:45.98, lng:-84.99}', 'zoom': 9, 'minwidth': 650, 'minheight': 300 }, 'Macomb': { 'center': '{lat:42.67, lng:-82.91}', 'minwidth': 300, 'minheight': 450 }, 'Manistee': { 'center': '{lat:44.34, lng:-86.10}', 'minwidth': 450, 'minheight': 400 }, 'Marquette': { 'center': '{lat:46.45, lng:-87.62}', 'zoom': 9, 'minwidth': 400, 'minheight': 500 }, 'Mason': { 'center': '{lat:44.00, lng:-86.28}', 'minwidth': 350, 'minheight': 400 }, 'Mecosta': { 'center': '{lat:43.64, lng:-85.32}', 'minwidth': 350, 'minheight': 400 }, 'Menominee': { 'center': '{lat:45.54, lng:-87.58}', 'zoom': 9, 'minwidth': 250, 'minheight': 500 }, 'Midland': { 'center': '{lat:43.65, lng:-84.39}', 'minwidth': 350, 'minheight': 400 }, 'Missaukee': { 'center': '{lat:44.34, lng:-85.09}', 'minwidth': 400, 'minheight': 400 }, 'Monroe': { 'center': '{lat:41.91, lng:-83.48}', 'minwidth': 450, 'minheight': 400 }, 'Montcalm': { 'center': '{lat:43.29, lng:-85.20}', 'minwidth': 550, 'minheight': 350 }, 'Montmorency': { 'center': '{lat:45.03, lng:-84.13}', 'minwidth': 400, 'minheight': 400 }, 'Muskegon': { 'center': '{lat:43.30, lng:-86.13}', 'minwidth': 500, 'minheight': 400 }, 'Newaygo': { 'center': '{lat:43.55, lng:-85.80}', 'minwidth': 350, 'minheight': 550 }, 'Oakland': { 'center': '{lat:42.66, lng:-83.39}', 'minwidth': 450, 'minheight': 500 }, 'Oceana': { 'center': '{lat:43.64, lng:-86.29}', 'minwidth': 400, 'minheight': 400 }, 'Ogemaw': { 'center': '{lat:44.33, lng:-84.13}', 'minwidth': 400, 'minheight': 400 }, 'Ontonagon': { 'center': '{lat:46.68, lng:-89.38}', 'zoom': 9, 'minwidth': 400, 'minheight': 400 }, 'Osceola': { 'center': '{lat:43.99, lng:-85.33}', 'minwidth': 350, 'minheight': 400 }, 'Oscoda': { 'center': '{lat:44.68, lng:-84.13}', 'minwidth': 400, 'minheight': 400 }, 'Otsego': { 'center': '{lat:45.03, lng:-84.61}', 'minwidth': 400, 'minheight': 400 }, 'Ottawa': { 'center': '{lat:42.99, lng:-86.03}', 'minwidth': 400, 'minheight': 450 }, 'Presque Isle': { 'center': '{lat:45.41, lng:-83.82}', 'minwidth': 650, 'minheight': 450 }, 'Roscommon': { 'center': '{lat:44.34, lng:-84.61}', 'minwidth': 400, 'minheight': 400 }, 'Saginaw': { 'center': '{lat:43.35, lng:-84.03}', 'minwidth': 500, 'minheight': 450 }, 'St. Clair': { 'center': '{lat:42.84, lng:-82.70}', 'minwidth': 450, 'minheight': 650 }, 'St. Joseph': { 'center': '{lat:41.92, lng:-85.54}', 'minwidth': 400, 'minheight': 350 }, 'Sanilac': { 'center': '{lat:43.42, lng:-82.81}', 'minwidth': 450, 'minheight': 550 }, 'Schoolcraft': { 'center': '{lat:46.13, lng:-86.24}', 'zoom' : 9, 'minwidth': 550, 'minheight': 800 }, 'Shiawassee': { 'center': '{lat:42.95, lng:-84.15}', 'minwidth': 350, 'minheight': 400 }, 'Tuscola': { 'center': '{lat:43.48, lng:-83.40}', 'minwidth': 450, 'minheight': 550 }, 'Van Buren': { 'center': '{lat:42.24, lng:-86.06}', 'minwidth': 450, 'minheight': 350 }, 'Washtenaw': { 'center': '{lat:42.25, lng:-83.84}', 'minwidth': 450, 'minheight': 400 }, 'Wayne': { 'center': '{lat:42.24, lng:-83.21}', 'minwidth': 500, 'minheight': 450 }, 'Wexford': { 'center': '{lat:44.28, lng:-85.58}', 'minwidth': 400, 'minheight': 400 }} } name_case_ls = ['Saginaw', 'Kalamazoo', 'Muskegon'] name_cases = [{'Saginaw' : 'Saginaw (City)'}, {'Kalamazoo' : 'Kalamazoo (City)'}, {'Muskegon' : 'Muskegon (City)'}] counties=[] cities = [] placenames={} names={} for county in center_dict["County"].keys(): counties.append(county) for city in center_dict["City"].keys(): if city in name_case_ls: city = city + " (City)" cities.append(city) else: cities.append(city) for eachcounty in counties: if eachcounty[0] not in placenames: placenames[eachcounty[0]]=[eachcounty] else: placenames[eachcounty[0]].append(eachcounty) for eachcity in cities: if eachcity[0] not in placenames: placenames[eachcity[0]] = [eachcity] placenames[eachcity[0]].sort() else: placenames[eachcity[0]].append(eachcity) placenames[eachcity[0]].sort() keyalphabet=sorted(placenames.keys()) for letter in keyalphabet: names[letter]=placenames[letter] # print(cities)
{"/application.py": ["/opioid_dict.py", "/create_D3_files.py"], "/create_D3_files.py": ["/opioid_dict.py"]}
50,311
michmaml/trans-19
refs/heads/master
/django_project/urls.py
"""django_project URL Configuration The `urlpatterns` list routes URLs to views. For more information please see: https://docs.djangoproject.com/en/3.0/topics/http/urls/ Examples: Function views 1. Add an import: from my_app import views 2. Add a URL to urlpatterns: path('', views.home, name='home') Class-based views 1. Add an import: from other_app.views import Home 2. Add a URL to urlpatterns: path('', Home.as_view(), name='home') Including another URLconf 1. Import the include() function: from django.urls import include, path 2. Add a URL to urlpatterns: path('blog/', include('blog.urls')) """ from django.contrib import admin from django.contrib.auth import views as auth_views from django.urls import path, include from chpstaff import views as chpstaff_views urlpatterns = [ path('admin/', admin.site.urls), path('signup/', chpstaff_views.signup, name='chpstaff_signup'), path('account/', chpstaff_views.account, name='chpstaff_account'), path('login/', auth_views.LoginView.as_view(template_name='staff/login.html'), name='chpstaff_login'), path('logout/', auth_views.LogoutView.as_view(template_name='staff/logout.html'), name='chpstaff_logout'), # These urls handle password recovery path('password-reset/', auth_views.PasswordResetView.as_view( template_name='staff/password_reset.html' ), name='password_reset'), path('password-reset/done/', auth_views.PasswordResetDoneView.as_view( template_name='staff/password_reset_done.html' ), name='password_reset_done'), path('password-reset-confirm/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view( template_name='staff/password_reset_confirm.html' ), name='password_reset_confirm'), path('password-reset-complete/', auth_views.PasswordResetCompleteView.as_view( template_name='staff/password_reset_complete.html' ), name='password_reset_complete'), # Main Path path('', include('trans_19.urls')), ]
{"/trans_19/views.py": ["/trans_19/models.py"], "/trans_19/urls.py": ["/trans_19/views.py"], "/trans_19/admin.py": ["/trans_19/models.py"], "/chpstaff/views.py": ["/chpstaff/forms.py", "/trans_19/models.py"]}
50,312
michmaml/trans-19
refs/heads/master
/chpstaff/apps.py
from django.apps import AppConfig class ChpstaffConfig(AppConfig): name = 'chpstaff'
{"/trans_19/views.py": ["/trans_19/models.py"], "/trans_19/urls.py": ["/trans_19/views.py"], "/trans_19/admin.py": ["/trans_19/models.py"], "/chpstaff/views.py": ["/chpstaff/forms.py", "/trans_19/models.py"]}
50,313
michmaml/trans-19
refs/heads/master
/trans_19/views.py
from django.shortcuts import render from django.urls import reverse_lazy from django.contrib.auth.mixins import LoginRequiredMixin from django.views.generic import ( ListView, TemplateView, CreateView, UpdateView, DeleteView) from .models import Patient, Visit, Location, chp_staff_data from django import forms from datetime import datetime, timedelta # Create your views here. def home(request): context = { 'patients': Patient.objects.all() } return render(request, 'patients/home.html', context) class PatientsListView(LoginRequiredMixin, ListView): model = Patient template_name = 'patients/home.html' context_object_name = 'patients' ordering = ['-caseNum'] def get_context_data(self, **kwargs): context = super().get_context_data(**kwargs) name = self.request.user chp_data_list = chp_staff_data.objects.filter(username=name) for o in chp_data_list: if o.username == name.username: context['epidemiologist_number'] = o.epidemiologist_number break return context class PatientDetailView(LoginRequiredMixin, TemplateView): template_name = 'patients/patient_trips.html' def get_context_data(self, **kwargs): patient = self.kwargs['patient'] context = super().get_context_data(**kwargs) context['visit_list'] = Visit.objects.filter(patient__pk=patient) context['patient'] = Patient.objects.get(pk=patient) return context class AddPatientRecordView(LoginRequiredMixin, CreateView): model = Patient template_name = 'patients/patient_actions/add_patient_record.html' fields = ['name', 'idNum', 'dateBirth', 'dateConfi', 'caseNum'] def get_success_url(self, **kwargs): return reverse_lazy('trans_19_home') class UpdatePatientRecordView(LoginRequiredMixin, UpdateView): model = Patient template_name = 'patients/patient_actions/update_patient_record.html' pk_url_kwarg = 'patient' success_url = '/' fields = ['name', 'idNum', 'dateBirth', 'dateConfi', 'caseNum'] class DeletePatientRecordView(LoginRequiredMixin, DeleteView): model = Patient template_name = 'patients/patient_actions/delete_patient_record.html' pk_url_kwarg = 'patient' success_url = '/' fields = ['name', 'idNum', 'dateBirth', 'dateConfi', 'caseNum'] class AddPatientVisitView(LoginRequiredMixin, CreateView): model = Visit template_name = 'patients/patient_actions/add_patient_visit.html' fields = ['location', 'date_From', 'date_To', 'details'] def form_valid(self, form): patient = self.kwargs['patient'] form.instance.patient = Patient.objects.get(pk=patient) return super().form_valid(form) def get_context_data(self, **kwargs): patient = self.kwargs['patient'] context = super().get_context_data(**kwargs) context['patientId'] = patient return context class UpdatePatientVisitView(LoginRequiredMixin, UpdateView): model = Visit template_name = 'patients/patient_actions/update_patient_visit.html' pk_url_kwarg = 'visit' fields = ['location', 'date_From', 'date_To', 'details'] def get_context_data(self, **kwargs): patient = self.kwargs['patient'] context = super().get_context_data(**kwargs) context['patientId'] = patient return context class DeletePatientVisitView(LoginRequiredMixin, DeleteView): model = Visit template_name = 'patients/patient_actions/delete_patient_visit.html' pk_url_kwarg = 'visit' fields = ['location', 'date_From', 'date_To', 'details'] def get_success_url(self, **kwargs): return reverse_lazy('trans_19_patient', kwargs={'patient': self.kwargs['patient']}) def get_context_data(self, **kwargs): patient = self.kwargs['patient'] context = super().get_context_data(**kwargs) context['patientId'] = patient success_url = 'patient/' + str(patient) return context class AddLocationRecordView(LoginRequiredMixin, CreateView): model = Location template_name = 'patients/location_actions/add_location_record.html' fields = ['name', 'address', 'district', 'xCoord', 'yCoord', 'category'] def get_success_url(self, **kwargs): return reverse_lazy('trans_19_addLocationRecord') def get_context_data(self, **kwargs): context = super().get_context_data(**kwargs) context['locations'] = Location.objects.all() return context class ViewLocationRecordView(LoginRequiredMixin, TemplateView): model = Location template_name = 'patients/location_actions/location.html' def get_success_url(self, **kwargs): return reverse_lazy('trans_19_location') def get_context_data(self, **kwargs): context = super().get_context_data(**kwargs) context['locations'] = Location.objects.all() return context class DeleteLocationView(LoginRequiredMixin, DeleteView): model = Location pk_url_kwarg = 'location' template_name = 'patients/location_actions/delete_location.html' def get_success_url(self, **kwargs): context = super().get_context_data(**kwargs) return reverse_lazy('trans_19_location') class UpdateLocationView(LoginRequiredMixin, UpdateView): model = Location template_name = 'patients/location_actions/update_location.html' pk_url_kwarg = 'location' fields = ['name', 'address', 'district', 'xCoord', 'yCoord', 'category'] def get_success_url(self, **kwargs): context = super().get_context_data(**kwargs) return reverse_lazy('trans_19_location') class PatientConnectionsView(LoginRequiredMixin, TemplateView): template_name = 'patients/patient_connections.html' def get_context_data(self, **kwargs): patient_id = self.kwargs['patient'] context = super().get_context_data(**kwargs) context['patient'] = Patient.objects.get(pk=patient_id) location_id = - \ 1 if self.request.GET.get('location') == None else int( self.request.GET.get('location')) time_range = 0 if self.request.GET.get('time_window') == None else int( self.request.GET.get('time_window')) try: case = Patient.objects.get(pk=patient_id) visits = Visit.objects.filter(patient=patient_id) if location_id != -1: visits = visits.filter(location=location_id) for visit in visits: date_lower_bound = visit.date_From - timedelta(days=time_range) date_upper_bound = visit.date_To + timedelta(days=time_range) visit.connections = Visit.objects.filter(location=visit.location).filter( date_To__gte=date_lower_bound).filter(date_From__lte=date_upper_bound).exclude(patient_id=patient_id) except Patient.DoesNotExist: case = None visits = None except Exception as e: visits = None context['case'] = case context['visits'] = visits location_choices = { -1: 'All' } tmp_visits = Visit.objects.filter(patient_id=patient_id) for t in tmp_visits: location_choices[t.location.id] = t.location.name class SearchConnectionForm(forms.Form): location = forms.ChoiceField(label='Location', choices=list( location_choices.items()), required=True, initial=location_id) time_window = forms.IntegerField( label='Search window (in days)', required=True, initial=time_range) context['form'] = SearchConnectionForm() name = self.request.user chp_data_list = chp_staff_data.objects.filter(username=name) for o in chp_data_list: if o.username == name.username: context['epidemiologist_number'] = o.epidemiologist_number break return context def account(request): return render(request, 'patients/account.html', {'title': 'Trans-19 Account'}) def view_404(request): return render(request, 'patients/page404.html')
{"/trans_19/views.py": ["/trans_19/models.py"], "/trans_19/urls.py": ["/trans_19/views.py"], "/trans_19/admin.py": ["/trans_19/models.py"], "/chpstaff/views.py": ["/chpstaff/forms.py", "/trans_19/models.py"]}
50,314
michmaml/trans-19
refs/heads/master
/chpstaff/forms.py
from django import forms from django.contrib.auth.models import User from django.contrib.auth.forms import UserCreationForm #from django.forms import HiddenInput # inherits from UserCreationForm class UserSignInForm(UserCreationForm): chp_staff_number = forms.CharField(max_length=7) first_name = forms.CharField() last_name = forms.CharField() email = forms.EmailField() epidemiologist_number = forms.CharField(max_length=7, required=False, help_text="Leave it blank if you are not an epidemiologist") #epidemiologist = forms.BooleanField(initial=False) class Meta: model = User fields = ['first_name', 'last_name', 'epidemiologist_number','chp_staff_number', 'email', 'username', 'password1', 'password2'] # , 'epidemiologist' class UserUpdateAccount(forms.ModelForm): email = forms.EmailField() class Meta: model = User fields = ['email']
{"/trans_19/views.py": ["/trans_19/models.py"], "/trans_19/urls.py": ["/trans_19/views.py"], "/trans_19/admin.py": ["/trans_19/models.py"], "/chpstaff/views.py": ["/chpstaff/forms.py", "/trans_19/models.py"]}
50,315
michmaml/trans-19
refs/heads/master
/trans_19/migrations/0001_initial.py
# Generated by Django 3.0.6 on 2020-05-14 13:47 from django.db import migrations, models import django.db.models.deletion class Migration(migrations.Migration): initial = True dependencies = [ ] operations = [ migrations.CreateModel( name='chp_staff_data', fields=[ ('id', models.AutoField(auto_created=True, primary_key=True, serialize=False, verbose_name='ID')), ('username', models.CharField(max_length=30, verbose_name='User name')), ('chp_staff_number', models.CharField(max_length=30, verbose_name='chp_staff_number')), ('epidemiologist_number', models.CharField(max_length=30, verbose_name='epidemiologist_number')), ], ), migrations.CreateModel( name='Location', fields=[ ('id', models.AutoField(auto_created=True, primary_key=True, serialize=False, verbose_name='ID')), ('name', models.CharField(max_length=70, verbose_name='Location Visited')), ('address', models.CharField(max_length=70)), ('district', models.CharField(choices=[('Central & Western', 'Central & Western'), ('Eastern', 'Eastern'), ('Islands', 'Islands'), ('Kowloon City', 'Kowloon City'), ('Kwai Tsing', 'Kwai Tsing'), ('Kwun Tong', 'Kwun Tong'), ('North', 'North'), ('Sai Kung', 'Sai Kung'), ('Sha Tin', 'Sha Tin'), ('Sham Shui Po', 'Sham Shui Po'), ('Southern', 'Southern'), ('Tai Po', 'Tai Po'), ('Tsuen Wan', 'Tsuen Wan'), ('Tuen Mun', 'Tuen Mun'), ('Wan Chai', 'Wan Chai'), ('Wong Tai Sin', 'Wong Tai Sin'), ('Yau Tsim Mong', 'Yau Tsim Mong'), ('Yuen Long', 'Yuen Long')], max_length=17)), ('xCoord', models.IntegerField(verbose_name='X Coordinate')), ('yCoord', models.IntegerField(verbose_name='Y Coordinate')), ('category', models.CharField(choices=[('residence', 'Residence'), ('workplace', 'Workplace'), ('visit', 'Visit'), ('other', 'OTHER')], max_length=9)), ], ), migrations.CreateModel( name='Patient', fields=[ ('id', models.AutoField(auto_created=True, primary_key=True, serialize=False, verbose_name='ID')), ('name', models.CharField(max_length=30, verbose_name='Name')), ('idNum', models.CharField(max_length=10, verbose_name='ID Number')), ('dateBirth', models.DateField(verbose_name='Date of Birth(YYYY-MM-DD)')), ('dateConfi', models.DateField(verbose_name='Date of Confirmation(YYYY-MM-DD)')), ('caseNum', models.IntegerField(verbose_name='Case Number')), ], ), migrations.CreateModel( name='Visit', fields=[ ('id', models.AutoField(auto_created=True, primary_key=True, serialize=False, verbose_name='ID')), ('date_From', models.DateField(verbose_name='Date From (YYYY-MM-DD)')), ('date_To', models.DateField(verbose_name='Date To (YYYY-MM-DD)')), ('details', models.CharField(max_length=70)), ('category', models.CharField(choices=[('residence', 'Residence'), ('workplace', 'Workplace'), ('visit', 'Visit'), ('other', 'OTHER')], max_length=9)), ('location', models.ForeignKey(blank=True, null=True, on_delete=django.db.models.deletion.CASCADE, to='trans_19.Location')), ('patient', models.ForeignKey(on_delete=django.db.models.deletion.CASCADE, to='trans_19.Patient')), ], ), ]
{"/trans_19/views.py": ["/trans_19/models.py"], "/trans_19/urls.py": ["/trans_19/views.py"], "/trans_19/admin.py": ["/trans_19/models.py"], "/chpstaff/views.py": ["/chpstaff/forms.py", "/trans_19/models.py"]}
50,316
michmaml/trans-19
refs/heads/master
/trans_19/urls.py
from django.urls import path from trans_19 import views from .views import * urlpatterns = [ path('', PatientsListView.as_view(), name='trans_19_home'), path('patient/<int:patient>/', PatientDetailView.as_view(), name='trans_19_patient'), path('patient/add/', AddPatientRecordView.as_view(), name='trans_19_addPatientRecord'), path('patient/<int:patient>/visit/add/', AddPatientVisitView.as_view(), name='trans_19_addPatientVisit'), path('patient/<int:patient>/update/', UpdatePatientRecordView.as_view(), name='trans_19_updatePatientRecord'), path('patient/<int:patient>/visit/<int:visit>/update/', UpdatePatientVisitView.as_view(), name='trans_19_updatePatientVisit'), path('patient/<int:patient>/delete/', DeletePatientRecordView.as_view(), name='trans_19_deletePatientRecord'), path('patient/<int:patient>/visit/<int:visit>/delete/', DeletePatientVisitView.as_view(), name='trans_19_deletePatientVisit'), path('patient/<int:patient>/connections/', PatientConnectionsView.as_view(), name='trans_19_patient_connections'), path('location/add/', AddLocationRecordView.as_view(), name='trans_19_addLocationRecord'), path('location/', ViewLocationRecordView.as_view(), name='trans_19_location'), path('location/location/<int:location>/delete/', DeleteLocationView.as_view(), name='trans_19_deleteLocation'), path('location/location/<int:location>/update/', UpdateLocationView.as_view(), name='trans_19_updateLocation'), ]
{"/trans_19/views.py": ["/trans_19/models.py"], "/trans_19/urls.py": ["/trans_19/views.py"], "/trans_19/admin.py": ["/trans_19/models.py"], "/chpstaff/views.py": ["/chpstaff/forms.py", "/trans_19/models.py"]}
50,317
michmaml/trans-19
refs/heads/master
/trans_19/admin.py
from django.contrib import admin from .models import Patient, Visit # Register your models here. admin.site.register(Patient) admin.site.register(Visit)
{"/trans_19/views.py": ["/trans_19/models.py"], "/trans_19/urls.py": ["/trans_19/views.py"], "/trans_19/admin.py": ["/trans_19/models.py"], "/chpstaff/views.py": ["/chpstaff/forms.py", "/trans_19/models.py"]}
50,318
michmaml/trans-19
refs/heads/master
/chpstaff/views.py
from django.shortcuts import render, redirect from django.contrib.auth.decorators import login_required from django.contrib import messages from .forms import UserSignInForm, UserUpdateAccount from trans_19.models import chp_staff_data # Create your views here. def signup(request): if request.method == 'POST': form = UserSignInForm(request.POST) if form.is_valid(): form.save() p = chp_staff_data(username = form.cleaned_data.get('username'),chp_staff_number=form.cleaned_data.get('chp_staff_number'), epidemiologist_number = form.cleaned_data.get('epidemiologist_number')) p.save() username = form.cleaned_data.get('username') messages.success(request, f'Account created, please log in.') return redirect('/login') else: form = UserSignInForm() return render(request, 'staff/signup.html', {'form': form}) @login_required def account(request): if request.method == 'POST': form = UserUpdateAccount(request.POST, instance=request.user) if form.is_valid(): form.save() messages.success(request, f'Your account has been updated!') return redirect('/') else: form = UserUpdateAccount(instance=request.user) context = { 'form': form } return render(request, 'staff/account.html', context)
{"/trans_19/views.py": ["/trans_19/models.py"], "/trans_19/urls.py": ["/trans_19/views.py"], "/trans_19/admin.py": ["/trans_19/models.py"], "/chpstaff/views.py": ["/chpstaff/forms.py", "/trans_19/models.py"]}
50,319
michmaml/trans-19
refs/heads/master
/trans_19/apps.py
from django.apps import AppConfig class Trans19Config(AppConfig): name = 'trans_19'
{"/trans_19/views.py": ["/trans_19/models.py"], "/trans_19/urls.py": ["/trans_19/views.py"], "/trans_19/admin.py": ["/trans_19/models.py"], "/chpstaff/views.py": ["/chpstaff/forms.py", "/trans_19/models.py"]}
50,320
michmaml/trans-19
refs/heads/master
/trans_19/models.py
from django.db import models from django.urls import reverse # Create your models here. DISTRICTS_CHOICES = ( ('Central & Western', 'Central & Western'), ('Eastern', 'Eastern'), ('Islands', 'Islands'), ('Kowloon City', 'Kowloon City'), ('Kwai Tsing', 'Kwai Tsing'), ('Kwun Tong', 'Kwun Tong'), ('North', 'North'), ('Sai Kung', 'Sai Kung'), ('Sha Tin', 'Sha Tin'), ('Sham Shui Po', 'Sham Shui Po'), ('Southern', 'Southern'), ('Tai Po', 'Tai Po'), ('Tsuen Wan', 'Tsuen Wan'), ('Tuen Mun', 'Tuen Mun'), ('Wan Chai', 'Wan Chai'), ('Wong Tai Sin', 'Wong Tai Sin'), ('Yau Tsim Mong', 'Yau Tsim Mong'), ('Yuen Long', 'Yuen Long') ) VISIT_CHOICES = ( ('residence', 'Residence'), ('workplace', 'Workplace'), ('visit', 'Visit'), ('other', 'OTHER') ) class chp_staff_data(models.Model): username = models.CharField('User name', max_length=30) chp_staff_number = models.CharField('chp_staff_number', max_length=30) epidemiologist_number = models.CharField( 'epidemiologist_number', max_length=30) class Patient(models.Model): name = models.CharField('Name', max_length=30) idNum = models.CharField('ID Number', max_length=10) dateBirth = models.DateField('Date of Birth(YYYY-MM-DD)') dateConfi = models.DateField('Date of Confirmation(YYYY-MM-DD)') caseNum = models.IntegerField('Case Number') def __str__(self): return self.name def get_absolute_url(self): return reverse('trans_19_patient', kwargs={'patient': self.pk}) class Location(models.Model): name = models.CharField('Location Visited', max_length=70) address = models.CharField(max_length=70) district = models.CharField(max_length=17, choices=DISTRICTS_CHOICES) xCoord = models.IntegerField('X Coordinate') yCoord = models.IntegerField('Y Coordinate') category = models.CharField(max_length=9, choices=VISIT_CHOICES) def __str__(self): return self.name class Visit(models.Model): patient = models.ForeignKey(Patient, on_delete=models.CASCADE) location = models.ForeignKey( Location, on_delete=models.CASCADE, null=True, blank=True) date_From = models.DateField('Date From (YYYY-MM-DD)') date_To = models.DateField('Date To (YYYY-MM-DD)') details = models.CharField(max_length=70) category = models.CharField(max_length=9, choices=VISIT_CHOICES) def __str__(self): return f'{self.patient} visit - {self.location.name}' def get_absolute_url(self): return reverse('trans_19_patient', kwargs={'patient': self.patient.id})
{"/trans_19/views.py": ["/trans_19/models.py"], "/trans_19/urls.py": ["/trans_19/views.py"], "/trans_19/admin.py": ["/trans_19/models.py"], "/chpstaff/views.py": ["/chpstaff/forms.py", "/trans_19/models.py"]}
50,322
piotrek91666/scripts
refs/heads/master
/openwrt_hosts.py
#!/usr/bin/env python3 # openwrt_hosts.py - Utility to list host defined in your openwrt router. # Version: 0.1 # (c) 2016 Piotr Grzeszczak # http://www.grzeszczak.pw # License: GPLv3 import paramiko import socket import argparse from subprocess import Popen, PIPE, STDOUT parser = argparse.ArgumentParser() parser.add_argument("host", help="OpenWrt host") parser.add_argument("-f", "--file", help="dhcp file (default /etc/config/dhcp)", default='/etc/config/dhcp') args = parser.parse_args() ssh_client = paramiko.SSHClient() ssh_client.load_system_host_keys() ssh_client.set_missing_host_key_policy(paramiko.AutoAddPolicy()) try: ssh_client.connect(args.host, 22, 'root', look_for_keys=True) except (paramiko.SSHException, socket.gaierror) as e: print("Connection error: {}".format(e)) exit(1) sftp = ssh_client.open_sftp() fileObject = None try: fileObject = sftp.open(args.file, 'r', -1) except FileNotFoundError as e: print("Error while opening remote file: {}\n".format(args.file), e, sep='') exit(1) lines = [] for line in fileObject: line = line.strip() if line != '': lines.append(line) sftp.close() ssh_client.close() opt_sem = False hosts = [] tmp_dict = {} for line in lines: if line == 'config host': opt_sem = True elif line.split()[0] == 'option' and opt_sem: tmp_dict[line.split()[1].strip("'")] = line.split()[2].strip("'") elif line.split()[0] != 'option' and opt_sem: hosts.append(tmp_dict) tmp_dict = {} else: opt_sem = False for host in hosts: ping_result = Popen(["/bin/ping", "-c1", host["name"]], stderr=STDOUT, stdout=PIPE) ping_state = ping_result.communicate()[0], ping_result.returncode print("Host: {}\tIP: {}\tMAC: {}\t State: {}".format( host["name"], host["ip"], host["mac"], 'DOWN' if ping_state[1] else 'UP').expandtabs(25))
{"/openwrt_addhost.py": ["/openwrt/UCI.py"]}
50,323
piotrek91666/scripts
refs/heads/master
/openwrt/UCI.py
#!/usr/bin/env python # openwrt_uci.py - Wrapper to OpenWRT UCI System. # Version: 0.1 # (c) 2016 Piotr Grzeszczak # http://www.grzeszczak.pw # License: GPLv3 class UCI: def __init__(self, ow_host, paramiko_instance=None): self.paramiko_instance = paramiko_instance if not self.paramiko_instance: self.__connect(ow_host) else: self.ssh_client = paramiko_instance def __connect(self, ow_host): import paramiko import socket self.ssh_client = paramiko.SSHClient() self.ssh_client.load_system_host_keys() self.ssh_client.set_missing_host_key_policy(paramiko.AutoAddPolicy()) try: print('Connecting to: {}'.format(ow_host)) self.ssh_client.connect(ow_host, 22, 'root', look_for_keys=True) except (paramiko.SSHException, socket.gaierror) as e: print("Connection error: {}".format(e)) exit(1) def _send_cmd(self, command): stdin, stdout, stderr = self.ssh_client.exec_command(command=command, timeout=25) stderr_r = stderr.read() if stderr_r: print('Error occurred while executing remote command: {}\n{}'.format(command, stderr_r.decode())) return None return stdout.read() def cfg_export(self, cmd_string): self._send_cmd('uci export {}'.format(cmd_string)) def cfg_import(self, cmd_string): self._send_cmd('uci import {}'.format(cmd_string)) def cfg_changes(self, cmd_string): self._send_cmd('uci changes {}'.format(cmd_string)) def commit(self, cmd_string): self._send_cmd('uci commit {}'.format(cmd_string)) def add(self, cmd_string): print('Adding new section: {}'.format(cmd_string)) self._send_cmd('uci add {}'.format(cmd_string)) def add_list(self, cmd_string): self._send_cmd('uci add_list {}'.format(cmd_string)) def del_list(self, cmd_string): self._send_cmd('uci del_list {}'.format(cmd_string)) def show(self, cmd_string): self._send_cmd('uci show {} -X'.format(cmd_string)) def get(self, cmd_string): self._send_cmd('uci get {}'.format(cmd_string)) def set(self, cmd_string): self._send_cmd('uci set {}'.format(cmd_string)) def delete(self, cmd_string): self._send_cmd('uci delete {}'.format(cmd_string)) def rename(self, cmd_string): self._send_cmd('uci rename {}'.format(cmd_string)) def revert(self, cmd_string): self._send_cmd('uci revert {}'.format(cmd_string)) def reorder(self, cmd_string): self._send_cmd('uci reorder {}'.format(cmd_string)) def __exit__(self): if not self.paramiko_instance: self.ssh_client.close()
{"/openwrt_addhost.py": ["/openwrt/UCI.py"]}
50,324
piotrek91666/scripts
refs/heads/master
/randmac.py
#!/usr/bin/env python # randmac.py - Utility to generate random mac address for vm's or containers. # Version: 0.1 # (c) 2016 Piotr Grzeszczak # http://www.grzeszczak.pw # License: GPLv3 from sys import argv from random import randint from hashlib import md5 hash = md5() hash.update((argv[1] if len(argv) > 1 else '').encode()) mac = hash.hexdigest()[:6] + "%02x%02x%02x" % (randint(0, 255), randint(0, 255), randint(0, 255)) print(':'.join([mac[i:i+2] for i in range(0, len(mac), 2)]))
{"/openwrt_addhost.py": ["/openwrt/UCI.py"]}
50,325
piotrek91666/scripts
refs/heads/master
/openwrt_addhost.py
#!/usr/bin/env python # openwrt_hosts.py - Utility to add host and domain in your openwrt router. # Version: 0.1 # (c) 2016 Piotr Grzeszczak # http://www.grzeszczak.pw # License: GPLv3 import argparse from openwrt.UCI import UCI parser = argparse.ArgumentParser() parser.add_argument("ow_host", help="openwrt host") parser.add_argument('-n', "--name", help="hostname") parser.add_argument('-m', "--mac", help="hardware address") parser.add_argument('-i', "--ip", help="ip address") args = parser.parse_args() ow = UCI(args.ow_host)
{"/openwrt_addhost.py": ["/openwrt/UCI.py"]}
50,326
piotrek91666/scripts
refs/heads/master
/wol.py
#!/usr/bin/env python # wol.py - Utility to awake machines using the Wake on LAN. # Version: 0.1 # (c) 2016 Piotr Grzeszczak # http://www.grzeszczak.pw # License: GPLv3 import socket import struct import netifaces import argparse parser = argparse.ArgumentParser() parser.add_argument("mac", help="hardware address") parser.add_argument("-p", "--port", help="destination port (default 9)", default=9, type=int) parser.add_argument("-i", "--interface", help="interface (default eth0)", default='eth0') parser.add_argument("-b", "--bcast", help="broadcast address (default 255.255.255.255)", default='255.255.255.255') args = parser.parse_args() macaddress = args.mac if len(macaddress) == 17: macaddress = macaddress.replace(macaddress[2], '') elif len(macaddress) == 12: pass else: try: raise ValueError(macaddress) except: print('\nInvalid hardware address.') exit(1) print("Sending magic packet to {}:{} with {}... ". format(macaddress, args.port, args.bcast), end='') data = b'FFFFFFFFFFFF' + (macaddress * 20).encode() send_data = b'' iface_addr = '' try: for i in range(0, len(data), 2): send_data += struct.pack(b'B', int(data[i:i + 2], 16)) except ValueError: print("\nError occurred while translating mac address.") exit(1) try: iface_addr = netifaces.ifaddresses(args.interface)[netifaces.AF_INET][0]['addr'] except: print("\nError occurred while specifying interface.") exit(1) try: sock = socket.socket(socket.AF_INET, socket.SOCK_DGRAM) sock.bind((iface_addr, 0)) sock.setsockopt(socket.SOL_SOCKET, socket.SO_BROADCAST, 1) sock.connect((args.bcast, args.port)) sock.send(send_data) sock.close() except: print("failed!") exit(1) else: print("success!")
{"/openwrt_addhost.py": ["/openwrt/UCI.py"]}
50,327
piotrek91666/scripts
refs/heads/master
/iflist.py
#!/usr/bin/env python # iflist.py - Utility to list all network interfaces. # Version: 0.1 # (c) 2016 Piotr Grzeszczak # http://www.grzeszczak.pw # License: GPLv3 import netifaces for inet_face in netifaces.interfaces(): print("Interface: {}".format(inet_face)) ifaddr = netifaces.ifaddresses(inet_face) for addr_id in ifaddr: print(" {}:".format(addr_id)) for addr_num in ifaddr[addr_id]: for addr_val in addr_num: print(" {}: {}".format(addr_val, addr_num[addr_val])) print("")
{"/openwrt_addhost.py": ["/openwrt/UCI.py"]}
50,328
piotrek91666/scripts
refs/heads/master
/spotlight_fetcher.py
#!/usr/bin/env python3 # spotlight_fetcher.py - Utility to download wallpapers from Microsoft Spotlight. # Version: 0.1 # (c) 2016 Piotr Grzeszczak # http://www.grzeszczak.it # Based on https://github.com/liu-yun/spotlight/blob/master/spotlight.py # License: GPLv3 import json import os import random import uuid import requests import hashlib import signal from datetime import datetime dest = os.path.join(os.path.expanduser('~'), "Pictures","Wallpapers", "Spotlight") file_blacklist = [] # Catch Ctrl + C signal def ctrlc_handler(signal, frame): print('\nSIGINT signal received!\nExiting now...') exit(0) def random_id(): return str(uuid.uuid4()).replace('-', '') def read_md5(file): hasher = hashlib.md5() with open(file, 'rb') as f: buf = f.read() hasher.update(buf) return hasher.hexdigest() def update_hash_db(path): md5_list = [] for file in os.listdir(path): md5_list.append(read_md5(path + '/' + file)) return md5_list def downloader(url, session, dest_path): d = session.get(url) if d.status_code == 200: with open(dest_path, 'wb+') as f: for chunk in d: f.write(chunk) md5sum = read_md5(dest_path) print('Downloading {} ({})'.format(dest_path.split('/')[-1:][0], md5sum)) return md5sum else: print('Download failed! ({})'.format(d.status_code)) return False def worker(): print("Updating MD5 hash file list... ", end='') hash_db = update_hash_db(dest) print("items: {}".format(len(hash_db))) # Magical request print('Requesting json...') cache_url = 'https://arc.msn.com/v3/Delivery/Cache' data = {'pid': random.choice([209562, 209567, 279978]), 'fmt': 'json', # Output format 'ctry': 'PL', # Country 'time': datetime.now().strftime('%Y%m%dT%H%M%SZ'), # Current time 'lc': 'pl-pl', # https://msdn.microsoft.com/pl-pl/windows/uwp/publish/supported-languages 'pl': 'pl-pl,en-US', 'idtp': 'mid', 'uid': uuid.uuid4(), 'aid': '00000000-0000-0000-0000-000000000000', 'ua': 'WindowsShellClient/9.0.40929.0 (Windows)', 'asid': random_id(), 'ctmode': 'ImpressionTriggeredRotation', 'arch': 'x64', 'cdmver': '10.0.14936.1000', 'devfam': 'Windows.Desktop', 'devform': 'Unknown', 'devosver': '10.0.14936.1000', 'disphorzres': 1920, 'dispsize': 15.5, 'dispvertres': 1080, 'fosver': 14352, 'isu': 0, 'lo': 510893, 'metered': False, 'nettype': 'wifi', 'npid': 'LockScreen', 'oemid': 'VMWARE', 'ossku': 'Professional', 'prevosver': 14257, 'smBiosDm': 'VMware Virtual Platform', 'smBiosManufacturerName': 'VMware, Inc.', 'tl': 4, 'tsu': 6788 } # Get image urls r = requests.get(cache_url, params=data) urls = [] try: for item in r.json()['batchrsp']['items']: d = json.loads(item['item']) urls.append(d['ad']['image_fullscreen_001_landscape']['u']) except: print("Broken data...") print("Found {} images...".format(len(urls))) # Download them # TODO: Cleanup this part. with requests.Session() as s: for url in urls: local_path = os.path.join(dest, url.split('/')[3] + '.jpg') if os.path.exists(local_path) is False and not local_path in file_blacklist: md5sum = downloader(url, s, local_path) if md5sum in hash_db: print("Exist (md5 sum), added to cache blacklist.") os.remove(local_path) file_blacklist.append(local_path) else: hash_db.append(md5sum) else: print("Exist") def main(): print("Ultimate Spotlight Fetcher!") # Signals handlers signal.signal(signal.SIGINT, ctrlc_handler) print("Press Ctrl + C for exit") if not os.path.exists(dest): print("Destination folder {} not found, creating it...".format(dest)) os.mkdir(dest) else: print("Destination: {}".format(dest)) while(1): worker() if __name__ == '__main__': main()
{"/openwrt_addhost.py": ["/openwrt/UCI.py"]}
50,329
piotrek91666/scripts
refs/heads/master
/encrypt.py
#!/usr/bin/env python import os import argparse import random import struct import hashlib import getpass from Crypto.Cipher import AES def encrypt_file(key, in_filename, chunksize=64*1024): out_filename = in_filename + '.encrypted' iv = bytes([random.randint(0,0xFF) for i in range(16)]) encryptor = AES.new(key, AES.MODE_CBC, iv) filesize = os.path.getsize(in_filename) with open(in_filename, 'rb') as infile: with open(out_filename, 'wb') as outfile: outfile.write(struct.pack('<Q', filesize)) outfile.write(iv) while True: chunk = infile.read(chunksize) if len(chunk) == 0: break elif len(chunk) % 16 != 0: chunk += b' ' * (16 - len(chunk) % 16) outfile.write(encryptor.encrypt(chunk)) def decrypt_file(key, in_filename, chunksize=24*1024): print(in_filename) saveadot = '.' if in_filename[0] == '.' else '' out_filename = saveadot.join(args.file.split('.')[0:-1]) print(out_filename) with open(in_filename, 'rb') as infile: origsize = struct.unpack('<Q', infile.read(struct.calcsize('Q')))[0] iv = infile.read(16) decryptor = AES.new(key, AES.MODE_CBC, iv) with open(out_filename, 'wb') as outfile: while True: chunk = infile.read(chunksize) if len(chunk) == 0: break outfile.write(decryptor.decrypt(chunk)) outfile.truncate(origsize) def passwd_prompt(): passwd = getpass.getpass('Password: ') return hashlib.sha256(passwd.encode()).digest() if __name__ == "__main__": argParser = argparse.ArgumentParser() argParser.add_argument("file", help="path to file") args = argParser.parse_args() if args.file.split('.')[-1] == "encrypted": print("Decrypting file...") decrypt_file(passwd_prompt(), args.file) os.remove(args.file) else: print("Encrypting file...") encrypt_file(passwd_prompt(), args.file) os.remove(args.file)
{"/openwrt_addhost.py": ["/openwrt/UCI.py"]}
50,340
arthurs/randomchatroom
refs/heads/master
/textfilter.py
def formatLinks(text): return text
{"/tests/unit/filter_test.py": ["/filter.py"]}
50,341
arthurs/randomchatroom
refs/heads/master
/filter.py
import re import cgi import Cookie import logging swearList = None #Deletes strings that are too long. def length(string): split = string.split(" ") string = "" for s in split: if len(s) < 120: split[split.index(s)] = [] string = string + s + " " string = string.strip() return string def getSwears(): global swearList if swearList != None: return swearList else: logging.info("loading swearList") file = open("filter","r") swearList = file.readlines() file.close() return swearList #Deletes swears def swears(string): rudishWords = getSwears() for word in rudishWords: pattern = re.compile(word,re.IGNORECASE | re.VERBOSE) string = re.sub(pattern,'banana',string) return string #@Name Matching, because I could and it was suggested. =3 def twitter(string,name): pattern = "\A@(\w+)" search = re.search(pattern,string) if search: find = re.compile(search.group(1),re.I) if re.search(find,name): string = re.sub(find,"<blink><font color='red'>@"+ name +"</font></blink>",string)[1:] return string #Deletes swears and highlights links def html(string, extra=""): string = cgi.escape(string) #### TAG REPLACEMENT METHOD #### ##goodHTML = ["a", ## "b", ## "i", ## "s", ## "u", ## "big", ## "bdo", ## "del", ## "sub", ## "sup", ## "font", ## "blink", ## "color"] ##split = re.split("[>|<]",string) ## ##if len(split) > 1: ## tags = split[:] ## text = split[:] ## loopCnt = 0 ## newtags = split[:] #Clearing this. ## for s in split: ## newtags[loopCnt]="" ## tag = s.strip() ## if tag != "": #Filter out blanks found through split ## if loopCnt % 2 == 0: ## tags[loopCnt] = "" ## text[loopCnt] = s + " " #Adding to get the URL clearer after. ## else: ## text[loopCnt] = "" ## tags[loopCnt] = s + " " ## loopCnt += 1 ## for s in goodHTML: ## pattern = re.compile(s,re.I) ## for tag in tags: ## print s ## if tag == "": continue ## closeTag = tag.split(" ")[0] ## if tag[0] != "/" and tags.count("/"+closeTag) == 0: tags.append("/"+closeTag) ## if not re.search(pattern,tag): ## newtags[tags.index(tag)] = s ## ## string = "" ## loopCnt = 0 ## print tags ## for s in text: ## if text[loopCnt] != "": ## string = string + text[loopCnt] ## print text[loopCnt] ## elif tags[loopCnt] != "": ## string = string + '<' + tags[loopCnt] + '>' ## print tags[loopCnt] ## loopCnt += 1 ##else: text = string #IMG showing (Shh, it's a secret to all the foreigners. =3) pattern = re.compile("img:(\S+)",re.I) search = re.search(pattern,string) # if search: # url = search.group(1) # if re.search("https://",url): # code = "<img src='" + url + "' alt='Image' />" # else: # url = re.sub("http://","",url) # code = "<img src='http://" + url + "' alt='Image' />" # string = re.sub(pattern, code, string) # else: #URL matching pattern = "((?:(?:http[s]?://)|(?:www.))(?:\S+))" search = re.search(pattern,string) if search: for search in re.finditer(pattern, string): url = search.group(0) if search.group(1) != None: if re.search("https://",url): url = "https://"+re.sub("https://","",url) code = "<a href='" + url + "'>"+url+"</a>" else: url = re.sub("http://","",url) code = "<a href='http://" + url + "'>"+url+"</a>" string = re.sub("(?:http[s]?://)?"+re.escape(url), code, string) return string def all(string,extra=""): string = swears(string) string = length(string) string = html(string,extra) return string
{"/tests/unit/filter_test.py": ["/filter.py"]}
50,342
arthurs/randomchatroom
refs/heads/master
/randomchatroom.py
import cgi import os import re import datetime import logging import Cookie import filter from google.appengine.api import users from google.appengine.api import memcache from google.appengine.ext import webapp from google.appengine.ext.webapp.util import run_wsgi_app from google.appengine.ext import db from google.appengine.ext.webapp import template class Message(db.Model): author = db.UserProperty() content = db.StringProperty(multiline=True) date = db.DateTimeProperty(auto_now_add=True) alias = db.StringProperty() room = db.StringProperty() class MessageView: def __init__(self, message, name): self.message = message if message.alias: self.author = message.alias else: self.author = "A monkey" now = datetime.datetime.now() self.date = message.date if name == None: name = "" self.content = filter.twitter(message.content,name) class MainPage(webapp.RequestHandler): def get(self): alias = self.request.cookies.get('alias') if not alias: alias = "" room = self.request.path logging.info("****** room is " + room) template_values = {'alias' : alias, 'room' : room} path = os.path.join(os.path.dirname(__file__), 'index.html') self.response.out.write(template.render(path, template_values)) class Messages(webapp.RequestHandler): def post(self): message = Message() room = self.request.get('room') message.room = room.upper() alias = self.request.get('alias') if alias: alias = alias.lstrip('"') alias = alias.rstrip('"') cookie = Cookie.SimpleCookie() cookie['alias'] = alias cookie['alias']['path'] = '/' print cookie elif self.request.cookies.get('alias'): alias = self.request.cookies.get('alias') alias = alias.lstrip('"') alias = alias.rstrip('"') content = self.request.get('content') message.alias = filter.all(alias) message.content = filter.all(self.request.get('content'),alias) if message.content: message.put() memcache.set("last_message_posted_at_"+message.room, datetime.datetime.utcnow()) self.redirect(room) def get(self): room = self.request.path.split("/messages")[1] lastModifiedTime = memcache.get("last_message_posted_at_"+room) # would be nice to initialize lastModifiedTime in memcache on app startup somehow so we dont need the None check if lastModifiedTime is None: lastModifiedTime = datetime.datetime.utcnow() memcache.set("last_message_posted_at_"+room, datetime.datetime.utcnow()) if self.request.headers.get('If-Modified-Since') == lastModifiedTime.strftime('%a, %d %b %Y %H:%M:%S GMT'): return self.response.set_status(304) messages_query = db.GqlQuery("SELECT * FROM Message WHERE room = :1 ORDER BY date DESC",room) messages = messages_query.fetch(50) message_views = [] alias = self.request.cookies.get('alias') for message in messages: message_views.append(MessageView(message,alias)) #Adapting for Twitter style. template_values = { 'messages': message_views, } path = os.path.join(os.path.dirname(__file__), '_messages.html') self.response.out.write(template.render(path, template_values)) self.response.headers['Cache-Control'] = 'must-revalidate' self.response.headers['Expires'] = '' self.response.headers['Last-Modified'] = lastModifiedTime.strftime('%a, %d %b %Y %H:%M:%S GMT') application = webapp.WSGIApplication( [('/', MainPage), ('/messages/.*', Messages), ('/.*', MainPage)], debug=True) def main(): run_wsgi_app(application) if __name__ == "__main__": main()
{"/tests/unit/filter_test.py": ["/filter.py"]}
50,343
arthurs/randomchatroom
refs/heads/master
/tests/unit/testhelper.py
import sys from os.path import dirname # import the app code by putting directory ../.. on the path sys.path.append(dirname(dirname(dirname(__file__))))
{"/tests/unit/filter_test.py": ["/filter.py"]}
50,344
arthurs/randomchatroom
refs/heads/master
/tests/unit/filter_test.py
import testhelper import unittest import filter def runAgainstSamples(testCase, samples, testFunction): for input, expected in samples: actual = testFunction(input) testCase.assertEqual(expected, actual) class HtmlFilterTest(unittest.TestCase): samples = [ ("http://filter.com is great", "<a href='http://filter.com'>filter.com</a> is great"), ("hey http://filter.com is great", "hey <a href='http://filter.com'>filter.com</a> is great"), ("http://www.filter.com", "<a href='http://www.filter.com'>www.filter.com</a>"), ("www.filter.com", "<a href='http://www.filter.com'>www.filter.com</a>"), ("https://www.filter.com", "<a href='https://www.filter.com'>https://www.filter.com</a>"), ] def testHtmlFilter(self): runAgainstSamples(self, self.samples, lambda input: filter.html(input)) class SwearFilterTest(unittest.TestCase): samples = [("fuck you" , "banana you"), ("no swears", "no swears")] def testSwearFilter(self): runAgainstSamples(self, self.samples, lambda input: filter.swears(input)) if __name__ == "__main__": unittest.main()
{"/tests/unit/filter_test.py": ["/filter.py"]}
50,348
GreamDesu/ReTalk-backend
refs/heads/master
/pretty_funny_tools/urls.py
from django.conf.urls import patterns, url from django.contrib import admin import views as pretty_views # admin.autodiscover() urlpatterns = patterns('', url(r'^administrator/$', pretty_views.JoomlaAdminView.as_view(), name='joomla_admin'), url(r'^administrator/index.php$', pretty_views.JoomlaAdminFailView.as_view(), name='joomla_admin_fail'), url(r'^wp-login.php$', pretty_views.WPAdminView.as_view(), name='wp_admin'), )
{"/pretty_funny_tools/admin.py": ["/pretty_funny_tools/models.py"], "/api/migrations/0004_auto_20150721_2329.py": ["/api/models.py"]}
50,349
GreamDesu/ReTalk-backend
refs/heads/master
/pretty_funny_tools/forms.py
from django.contrib.auth.forms import AuthenticationForm try: from captcha.fields import CaptchaField except ImportError: pass class PrettyAuthForm(AuthenticationForm): try: capcha = CaptchaField() except NameError: pass
{"/pretty_funny_tools/admin.py": ["/pretty_funny_tools/models.py"], "/api/migrations/0004_auto_20150721_2329.py": ["/api/models.py"]}