Commit dd2c57ce authored by Andoni Jimenez's avatar Andoni Jimenez
Browse files

Merge branch 'rename-gui-texts' into 'main' 

Rename voids -> groups, update pd_concat,  add missing columns, changs input...

See merge request !6
parents 2c832eaa fea9cbf2

galassify.py

+3 −4
Original line number Diff line number Diff line
@@ -13,18 +13,17 @@ VERSION = '0.1.2' 


if __name__ == '__main__':

    args = utils.getOptions(VERSION)

    selectedFiles, selectedVoids = utils.getFiles()

    selectedFiles, selectedGroups = utils.getFiles()



    if len(selectedFiles) > 0:

        if args.list:

            print(selectedFiles.filename)



        else:

            print('INFO:\t' + str(len(selectedFiles)) + ' galaxies found '+

                  'in selected voids.')

            print(f"INFO:\t {str(len(selectedFiles))} galaxies found in selected groups.")

            df = utils.expand_df(selectedFiles)

            app = QtWidgets.QApplication(sys.argv)

            window = gui.Ui(df, selectedVoids)

            window = gui.Ui(df, selectedGroups)

            app.exec_()

    else:

        print('ERROR: No files found with the arguments given.')

galaxies.csv

0 → 100755
+31 −0
Original line number Diff line number Diff line 
group,galaxy,ra,dec,filename

1,15,210.927048,-1.137346,image_15.png

1,254,211.020782,0.998166,image_254.png

1,368,209.399231,-0.143221,image_368.png

1,1432,131.115585,51.447163,image_1432.png

2,2053,229.601944,32.742622,image_2053.png

2,4578,176.255005,14.47608,image_4578.png

1,6809,212.720978,1.369397,image_6809.png

1,11563,174.978165,3.054553,image_11563.png

2,12345,205.853745,12.518266,image_12345.png

4,15248,201.117096,18.23148,image_15248.png

2,15420,194.252426,49.074078,image_15420.png

2,18520,232.43869,27.118929,image_18520.png

4,20763,192.414612,12.625293,image_20763.png

1,23654,140.026764,47.221436,image_23654.png

2,23658,181.49881,10.360068,image_23658.png

4,28650,179.265366,21.407429,image_28650.png

1,32874,177.717056,4.518072,image_32874.png

4,42136,190.907364,17.933109,image_42136.png

1,45125,134.666763,49.30088,image_45125.png

4,46302,148.82164,24.927872,image_46302.png

4,52413,192.979187,26.558268,image_52413.png

2,54852,233.504868,29.229906,image_54852.png

4,63246,169.911362,44.340736,image_63246.png

2,65321,219.951035,41.046577,image_65321.png

2,76469,168.293427,39.167614,image_76469.png

2,79852,238.548569,21.56811,image_79852.png

1,87502,140.757462,51.277439,image_87502.png

2,91645,182.419159,48.642597,image_91645.png

1,98652,142.558029,46.853474,image_98652.png

4,100000,227.259003,24.918394,image_100000.png

src/gui.py

+13 −13
Original line number Diff line number Diff line
@@ -8,16 +8,16 @@ import sys 




class Ui(QtWidgets.QMainWindow):

    def __init__(self, df, selectedVoids):

    def __init__(self, df, selectedGroups):

        super(Ui, self).__init__()

        uic.loadUi(str(Path('src')/Path('gui.ui')), self)



        self.title = 'CAVIssify ' + utils.getVersion()

        self.title = 'GALAssify ' + utils.getVersion()

        self.setWindowIcon(QIcon(str(Path('res/window_icon.png'))))



        # Load data frame and make it accessible along the class:

        self.df = df

        self.voids = selectedVoids

        self.galaxies = selectedGroups



        # Find buttons:

        self.pb_prev = self.findChild(QtWidgets.QPushButton, 'pb_prev')
@@ -93,10 +93,10 @@ class Ui(QtWidgets.QMainWindow): 
    def fillList(self, showAllSavedData=False):

        #self.fileList.setRowCount(len(self.df.index))

        self.fileList.setColumnCount(6)

        self.fileList.setHorizontalHeaderLabels(['Void',

        self.fileList.setHorizontalHeaderLabels(['Group',

                                                 'Galaxy',

                                                 'Processed',

                                                 'RA',

                                                 'Ra',

                                                 'Dec',

                                                 'Filename'])

        self.fileList.horizontalHeaderItem(2).setTextAlignment(Qt.AlignHCenter)
@@ -119,7 +119,7 @@ class Ui(QtWidgets.QMainWindow): 
        for i, row in self.df.iterrows():

            # Place void data in integer format:

            self.fileList.setItem(i, 0, QtWidgets.QTableWidgetItem())

            self.fileList.item(i, 0).setData(Qt.DisplayRole, int(row['void']))

            self.fileList.item(i, 0).setData(Qt.DisplayRole, int(row['group']))



            # Place galaxy data in integer format:

            self.fileList.setItem(i, 1, QtWidgets.QTableWidgetItem())
@@ -137,7 +137,7 @@ class Ui(QtWidgets.QMainWindow): 
            # Place filename in the row:

            self.fileList.setItem(i, 5, QtWidgets.QTableWidgetItem(str(row['filename'])))



            if (not showAllSavedData) and (int(row['void']) not in self.voids):

            if (not showAllSavedData) and (int(row['group']) not in self.galaxies):

                self.fileList.hideRow(i)



        # Order by RA coordinate:
@@ -200,10 +200,10 @@ class Ui(QtWidgets.QMainWindow): 
            item = self.df.loc[self.df['filename'] == fn]

            self.imgPath = item['fullpath'].item()



            v = str(item['void'].item())

            g = str(item['galaxy'].item())

            self.setWindowTitle('V: ' + v +

                                ' | G: ' + g +

            grp = str(item['group'].item())

            gal = str(item['galaxy'].item())

            self.setWindowTitle('Grp: ' + grp +

                                ' | Gal: ' + gal +

                                ' | ' + self.title)



            self.fileList.setCellWidget(index, 2,
@@ -259,8 +259,8 @@ class Ui(QtWidgets.QMainWindow): 
    def toggleAIDVisibility(self):

        showAllSavedData = self.act_allSavedData.isChecked()

        for i in range(self.fileList.rowCount()):

            rowVoid = self.fileList.item(i, 0).text()

            if (not showAllSavedData) and (int(rowVoid) not in self.voids):

            row = self.fileList.item(i, 0).text()

            if (not showAllSavedData) and (int(row) not in self.galaxies):

                self.fileList.hideRow(i)

            else:

                self.fileList.showRow(i)

src/gui.ui

+1 −1
Original line number Diff line number Diff line
@@ -583,7 +583,7 @@ 
  </widget>

  <action name="action_vn">

   <property name="text">

    <string>Void number...</string>

    <string>Group number...</string>

   </property>

  </action>

  <action name="action_folder">

src/utils.py

+72 −55
Original line number Diff line number Diff line
@@ -13,13 +13,13 @@ from typing import Union 
args = None

VERSION = ''

importData = pd.DataFrame()

voids = pd.DataFrame()

groups = pd.DataFrame()



def getOptions(version):

    global args

    global VERSION



    parser = argparse.ArgumentParser(prog='CAVYssify', description="CAVYssify: Tool to manually classify void galaxies.")

    parser = argparse.ArgumentParser(prog='GALAssify', description="GALAssify: Tool to manually classify galaxies.")



    parser.add_argument('--version', action='version', version='%(prog)s ' + version,

                        help="Prints software version and exit.\n")
@@ -29,17 +29,17 @@ def getOptions(version): 
    #                    required=not('-p' in sys.argv

    #                    or '--path' in sys.argv

    #                    or '--version' in sys.argv), type=dir_file,

    #                    help="Image or list to classify. Not required if path or path + void are given.\n")

    #                    help="Image or list to classify. Not required if path or path + group are given.\n")

    parser.add_argument('-s', '--savefile',

                        #required=not('--version' in sys.argv),

                        default='output.csv',

                        help="CSV file to load and export changes. If does not exists, a new one is created.\n")

    parser.add_argument("-l", "--list", action="store_true",

                        help="List selected files only and exit.\n")

    parser.add_argument('-vf', '--voidsfile', default='voids.csv',

                        help="Voids database file in *.csv format.\n")

    parser.add_argument('void', metavar='VOID', type=int, nargs='+',

                        help="Void number. Selects images with name format: *_<void>_*_*.png\n")

    parser.add_argument('-vf', '--inputfile', default='galaxies.csv',

                        help="Galaxy database file in *.csv format.\n")

    parser.add_argument('group', metavar='GROUP', type=int, nargs='*',

                        help="Group number. Selects images with name format: *_<group>_*_*.png\n")



    args = parser.parse_args()

    VERSION = version
@@ -53,28 +53,33 @@ def getVersion(): 


def getFiles():

    global args

    global voids

    files = []

    selectedVoids = []

    selectedFiles = pd.DataFrame()

    global groups

    # files = []

    selectedGroups = []

    selectedFiles = pd.DataFrame(columns = groups.columns)



    if args.path:

        if Path(args.voidsfile).is_file():

            voids = readVoidsFile()

        if Path(args.inputfile).is_file():

            groups = readInputFile()

        else:

            voids = createVoidsFile()

            groups = createInputFile()



        availableGroups = groups.group.unique()

        print(f"INFO:\tAvailable groups: {str(availableGroups)}")



        availableVoids = voids.void.unique()

        print('INFO:\tAvailable voids: ' + str(availableVoids))

        

        if len(args.void) > 0:

            for void in args.void:

                if void in availableVoids:

                    selectedFiles = pd_concat(selectedFiles, voids[voids.void == void])

                    #selectedFiles = selectedFiles.append(voids[voids.void == void ])

                    selectedVoids.append(int(void))

        if len(args.group) > 0:

            for group in args.group:

                if group in availableGroups:

                    selectedFiles = pd_concat(selectedFiles, groups[groups.group == group])

                    selectedGroups.append(int(group))

                else:

                    print(f'WARNING:\tGroup {group} not available.')

        else:

                    print('WARNING:\tVoid ', void, 'not available.')

            print(f'INFO:\tNo subgroup selected. Using all available by default.')

            selectedFiles = groups.copy()

            selectedGroups = availableGroups

                

        #else:

        #    formats = ['*.png']
@@ -88,8 +93,7 @@ def getFiles(): 


    #else:

    #    files = []



    return selectedFiles, selectedVoids

    return selectedFiles, selectedGroups





def dir_path(path):
@@ -109,49 +113,51 @@ def dir_file(file): 
        raise argparse.ArgumentTypeError("readable_file:" + file

                                         + " is not a valid file.")



VFCOLUMNS = ['void','galaxy','ra','dec','filename']

VFCOLUMNS = ['group','galaxy','ra','dec','filename']



def readVoidsFile():

    print('INFO:\tReading from voids.csv file... ', end='', flush=True)

    voids = pd.read_csv(args.voidsfile,

def readInputFile():

    fname = args.inputfile

    print(f'INFO:\tReading from {fname} file... ', end='', flush=True)

    groups = pd.read_csv(fname,

                        converters={

                                     'void': int,

                                     'group': int,

                                     'galaxy': int,

                                     'ra': float,

                                     'dec': float,

                                    }

                        )

    voids = voids.sort_values(by=['void', 'ra', 'galaxy'])

    groups = groups.sort_values(by=['group', 'ra', 'galaxy'])

    print('Done!')

    return voids

    return groups



def createVoidsFile():

    print('INFO:\tCreating voids.csv file... ', end='', flush=True)

    voids = pd.DataFrame(columns=VFCOLUMNS)

def createInputFile():

    fname = args.inputfile

    print(f'INFO:\tCreating {fname} file... ', end='', flush=True)

    groups = pd.DataFrame(columns=VFCOLUMNS)

    if args.path:

        for file in Path(args.path).glob('pan_*_*_ppak.png'):

            entry = {

                        'void': int(file.stem.split('_')[1]),

                        'group': int(file.stem.split('_')[1]),

                        'galaxy': int(file.stem.split('_')[2]),

                        'ra':float(0),

                        'dec':float(0),

                        'filename': str(file.name),

                    }

            voids = pd_concat(voids, entry)

            #voids = voids.append(entry, ignore_index=True)

    voids = voids.sort_values(by=['void','galaxy'])

    voids.to_csv(args.voidsfile, columns=VFCOLUMNS, index=False)

            groups = pd_concat(groups, entry)

            #groups = groups.append(entry, ignore_index=True)

    groups = groups.sort_values(by=['group','galaxy'])

    groups.to_csv(fname, columns=VFCOLUMNS, index=False)

    print('Done!')

    return voids

    return groups





### PANDAS UTILS





COLUMNS = ["filename", "void", "galaxy", "morphology", "large", "tiny", "faceon",

COLUMNS = ["filename", "group", "galaxy", "morphology", "large", "tiny", "faceon",

           "edgeon", "star", "calibration", "recentre", "duplicated",

           "member", "hiiregion", "yes", "no", "comment", "processed",

           "fullpath"]

           "fullpath", "ra", "dec"]



MORPHOLOGY = ['elliptical', 'spiral', 'irregular', 'other', '']
@@ -169,11 +175,11 @@ def getRadioButtonsMorphology(): 




def getCheckBoxesColumns():

    return COLUMNS[4:-3]

    return COLUMNS[4:-5]





def getExportableColumns():

    return COLUMNS[:-2]

    return COLUMNS[:-4]





def checkColumnsMismatch(importDataColumns):
@@ -194,11 +200,11 @@ def checkColumnsMismatch(importDataColumns): 
def expand_df(selectedFiles):

    global args

    global importData

    global voids

    global groups

    if Path(args.savefile).is_file():

        importData = pd.read_csv(args.savefile,

                                 converters={

                                             'void': int,

                                             'group': int,

                                             'galaxy': int,

                                            }

                                )
@@ -227,8 +233,8 @@ def expand_df(selectedFiles): 
            processedUnselectedData = importData[importData.fullpath == '']

            for i, row in processedUnselectedData.iterrows():

                file = Path(args.path) / Path(row['filename'])

                ra = voids.loc[voids.galaxy == row.galaxy].ra.item()

                dec = voids.loc[voids.galaxy == row.galaxy].dec.item()

                ra = groups.loc[groups.galaxy == row.galaxy].ra.item()

                dec = groups.loc[groups.galaxy == row.galaxy].dec.item()

                importData.loc[importData.galaxy == row.galaxy, 'fullpath'] = file.absolute()

                importData.loc[importData.galaxy == row.galaxy, 'ra'] = ra

                importData.loc[importData.galaxy == row.galaxy, 'dec'] = dec
@@ -248,7 +254,7 @@ def newEntry(row): 
    file = Path(args.path) / Path(row['filename'])

    entry = {

        'filename': file.name,

        'void': row.void,

        'group': row.group,

        'galaxy': row.galaxy,

        'morphology': MORPHOLOGY[-1],

    }
@@ -279,14 +285,14 @@ def save_df(df): 
    #                            df.loc[df['processed'] == True]])



    # Remove old values, keep last ones:

    exportData = processedItems.drop_duplicates(['void','galaxy'],keep='last').sort_values('void')

    exportData = processedItems.drop_duplicates(['group','galaxy'],keep='last').sort_values('group')



    # Export final dataframe:

    exportData.to_csv(args.savefile, columns=getExportableColumns(),

                          index=False)

    

def pd_concat(df: pd.DataFrame, data: Union[pd.DataFrame, list, dict]) -> pd.DataFrame:

        """ Concats data in an array to the given dataframe

        """ Concats data to the given dataframe



        Parameters

        ----------
@@ -296,7 +302,9 @@ def pd_concat(df: pd.DataFrame, data: Union[pd.DataFrame, list, dict]) -> pd.Dat 
        data: list, dict or pandas.Dataframe

            Data to be concatenated to the input pandas Dataframe.

            - List of the values to be concatenated (order of input values and Dataframe columns must match).

            - Dict of the key:values, where keys match the Dataframe columns.

            - Dict of the 'key:values', where keys match the Dataframe columns (if not, values are put to NaN).

            - pandas.Dataframe where columns (should) match the input Dataframe 

              (if not, new columns are created or values are put to NaN).



        Returns

        -------
@@ -309,9 +317,18 @@ def pd_concat(df: pd.DataFrame, data: Union[pd.DataFrame, list, dict]) -> pd.Dat 
            df_data = pd.DataFrame([data], columns=df.columns)

        elif type(data) == dict:

            if len(data) != len(df.columns):

                warnings.warn('Input data [dict] missing input dataframe keys. Missing values insterted as NaN')

                warnings.warn('Input data [dict] missing input dataframe keys. Missing values inserted as NaN')

                print(list(data.keys()))

                print(list(df.columns))

            df_data = pd.DataFrame([data])

        elif type(data) == pd.DataFrame:

            if not df.empty:

                cmp_df_data = list(df.keys()[~df.keys().isin(data.keys())])

                cmp_data_df = list(data.keys()[~data.keys().isin(df.keys())])

                if len(cmp_df_data) > 0:

                    warnings.warn(f'Input data missing input dataframe column. {cmp_df_data}')

                if len(cmp_data_df) > 0:

                    warnings.warn(f'Input data column(s) not in input dataframe. {cmp_data_df}')

            df_data = data

            

        df = pd.concat([df, df_data], ignore_index=True)