add script

get_time_detection.py

+import pandas as pd
+import os
+import ipdb
+from tqdm import tqdm
+import argparse
+from datetime import date
+
+def arg_directory(path):
+    if os.path.isdir(path):
+        return path
+    else:
+        raise argparse.ArgumentTypeError(f'`{path}` is not a valid path')
+
+parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter, description='TODO')
+parser.add_argument('-p','--path_to_data', type=arg_directory, help = 'Path of the folder that contain the .txt files',required=True)
+parser.add_argument('-d','--direction', type=arg_directory, help = 'Directory to wich the dataframe will be stored',required=True)
+args = parser.parse_args()
+
+annots = args.path_to_data
+
+today = date.today()
+out_file = str('YOLO_detection_'+str('_'+today.day+'_'+today.month))
+
+outdir = args.direction
+
+df = pd.DataFrame(columns = ['file','idx','espece','x','y','w','h','conf'])
+for i in tqdm(os.listdir(annots)):
+	if i == 'list_results_YOLO.csv':
+		continue
+	else:
+		table = pd.read_csv(str(annots+i), sep=' ')
+		l=[]
+		tab= pd.DataFrame(columns = ['file','idx','espece','x','y','w','h','conf'])
+		tab2 = pd.DataFrame(columns = ['file','idx','espece','x','y','w','h','conf'])
+
+		if len (table) == 0:
+			for j in table.columns:
+				l.append(j)
+
+			name = i.split('.')[0]
+
+			if len(name.split('_')[-1]) == 2:
+				name = name.split('.')[0][0:-3]
+			elif len(name.split('_')[-1]) == 3:
+				name = name.split('.')[0][0:-4]
+			else :
+				name= name.split('.')[0][0:-2]
+
+			name = str(name+'.wav')
+			idx = i.split('_')[-1]
+			idx = idx.split('.')[0]
+
+			table = pd.DataFrame([[name, idx, l[0],l[1],l[2],l[3],l[4],l[5]]], columns = ['file', 'idx','espece','x','y','w','h','conf'])
+			df = pd.concat([df,table])
+		else:
+			nb = len(table)
+
+			name = i.split('.')[0]
+
+			if len(name.split('_')[-1]) == 2:
+				name = name.split('.')[0][0:-3]
+			elif len(name.split('_')[-1]) == 3:
+				name = name.split('.')[0][0:-4]
+			else :
+				name= name.split('.')[0][0:-2]
+
+			name = str(name+'.wav' )
+			idx = i.split('_')[-1]
+			idx = idx.split('.')[0]
+
+			for j in table.columns:
+				l.append(j)
+			try :
+				tab2 = pd.DataFrame([[name, idx, l[0],l[1],l[2],l[3],l[4],l[5]]],columns = ['file', 'idx','espece','x','y','w','h','conf'])
+			except:
+				ipdb.set_trace()
+			for v in range(nb):
+				new = pd.DataFrame([[name, idx, table.iloc[v][0],table.iloc[v][1], table.iloc[v][2],table.iloc[v][3],table.iloc[v][4],table.iloc[v][5]]], columns = ['file', 'idx','espece','x','y','w','h','conf'])
+				tab = pd.concat([tab , new])
+
+			table = pd.concat([tab2, tab])
+			df = pd.concat([df, table])
+
+#put the classes here
+names = []
+
+df['annot'] = 'None'
+for j in range (len(df)):
+	df['annot'].iloc[j] = names[int(df.espece.iloc[j])]
+
+print('Calculating the positions','\n')
+
+corres = pd.DataFrame(columns=['start','stop'])
+new = pd.DataFrame([['0','8']],columns = ['start','stop'])
+corres = pd.concat([corres,new])
+
+new = pd.DataFrame([['6','14']],columns = ['start','stop'])
+corres = pd.concat([corres,new])
+
+DUREE_SPECTRO = 8
+OVERLAP = 2
+
+for k in range(50):
+	val = k+1
+	new = pd.DataFrame([[int(corres.start.iloc[val])+(DUREE_SPECTRO - OVERLAP),int(corres.stop.iloc[val])+(DUREE_SPECTRO - OVERLAP)]],columns = ['start','stop'])
+	corres = pd.concat([corres,new])
+
+df['midl'] = 'None'
+
+for w in range(len(df.midl)):
+	if int(df.idx.iloc[w]) == 0:
+		try:
+			df.midl.iloc[w] = float(df.x.iloc[w])*DUREE_SPECTRO+float(corres.start.iloc[int(df.idx.iloc[w])])
+		except:
+			ipdb.set_trace()
+	else:
+		df.midl.iloc[w] = float(df.x.iloc[w])*DUREE_SPECTRO+float(corres.start.iloc[int(df.idx.iloc[w])])
\ No newline at end of file