I have a list which is 2 dimensional and has elements like ([0,1,2],[3,4,5]). Type of its elements is numpy.ndarray. I am trying to delete 2nd columns of each element. When I check its type, it returns list but it gives ValueError: cannot delete array elements error. I checked StackOverflow but haven't found a similar case. The code is below, any help is appreciated.
for row in trainSet:
del row[1]
Its a list that contains numpy arrays
import numpy as np
trainSet = np.array([[0,1,2],[3,4,5]])
#wrong way list(trainSet) # this do not convert nested lists
trainSet = list(map(lambda x: x.tolist(),trainSet)) #to make sure it is a list, do not contain nested numpy arrays
for row in trainSet:
del row[1]
print(trainSet)
[[0, 2], [3, 5]]
import numpy as np
import pandas as pd
import os
for dirname, _, filenames in os.walk('/datasets/train'):
for filename in filenames:
os.path.join(dirname, filename
import cv2
import matplotlib.pyplot as plt
import tensorflow as tf
from tqdm.notebook import tqdm
from glob import glob
from keras.layers import Flatten, Dense
from keras.applications.vgg16 import VGG16
import xml.etree.ElementTree as xet
from keras.models import load_model, Sequential
from sklearn.model_selection import train_test_split
def load_images(directory):
images = []
for filepath in tqdm(os.listdir(directory)):
img_path = os.path.join(directory,filepath)
img = cv2.imread(img_path)
img = cv2.cvtColor(img,cv2.COLOR_BGR2RGB)
img = cv2.resize(img,(224,224))
img = cv2.medianBlur(img,5)
images.append(img)
images = np.array(images).reshape(len(images),224,224,3)
return images
path = glob("./datasets/train/annotations/*.xml")
labels_dict = dict(filepath=[],xmin=[],xmax=[],ymin=[],ymax=[])
for filename in path:
data = xet.parse(filename)
root = data.getroot()
obj = root.find('object')
labels_info = obj.find('bndbox')
xmin = int(labels_info.find('xmin').text)
xmax = int(labels_info.find('xmax').text)
ymin = int(labels_info.find('ymin').text)
ymax = int(labels_info.find('ymax').text)
labels_dict['filepath'].append(filename)
labels_dict['xmin'].append(xmin)
labels_dict['xmax'].append(xmax)
labels_dict['ymin'].append(ymin)
labels_dict['ymax'].append(ymax)
df = pd.DataFrame(labels_dict)
df.to_csv('labels.csv',index=False)
df.head()
images = load_images('./datasets/train/images/')
images.shape
filename = df['filepath'][0]
def getFilename(filename):
filename_image = xet.parse(filename).getroot().find('filename').text
filepath_image = os.path.join('./datasets/train/images/',filename_image)
return filepath_image
getFilename(filename)
image_path=list(df['filepath'].apply(getFilename))
image_path[:10]
bbox_coords = df.iloc[:,1:].values
labels = []
for ind in range(len(image_path)):
image = image_path[ind]
img_arr = cv2.imread(image)
h,w,d = img_arr.shape
xmin,xmax,ymin,ymax = bbox_coords[ind]
nxmin, nxmax = int(xmin/(w/255.)), int(xmax/(w/255.))
nymin, nymax = int(ymin/(h/255.)), int(ymax/(h/255.))
labels.append((nxmin,nxmax,nymin,nymax))
labels = np.array(labels)
labels
X = np.array(images,dtype=np.float32)
y = np.array(labels,dtype=np.float32)
X = X / 255.
y = y / 255.
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=42)
X_train, X_val, y_train, y_val = train_test_split(X_train,y_train,test_size=0.1,random_state=42)
model = Sequential()
model.add(VGG16(weights="imagenet", include_top=False, input_shape=(224, 224, 3)))
model.add(Flatten())
model.add(Dense(128, activation="relu"))
model.add(Dense(128, activation="relu"))
model.add(Dense(64, activation="relu"))
model.add(Dense(4, activation="sigmoid"))
model.layers[-6].trainable = False
model.summary()
model.compile(loss='mean_squared_error', optimizer='adam', metrics=['accuracy'])
r = model.fit(X_train,
y_train,
epochs=10,
batch_size=32,
validation_data=
(X_val,y_val),
verbose=1)
model.save('license_plate_detector.h5')
