Artificial Intelligence
Please fill the required field.

IMG Models


(2006 - 2010), Agency First Access Entertainment Shanina Shaik (born 11 February 1991) is an Australian model. She has worked for leading brands including Chanel, Vivienne Westwood, Tom Ford, and Victoria's Secret. She has also walked in the Victoria's Secret Fashion Show. Early life Shanina Shaik was born on 11 February 1991 in Melbourne, Australia to Hanif Shaik, a Pakistani and Lithuanian mother, and Kim Shaik, an Australian mother of Lithuanian heritage. She has a younger brother, Shay Shaik, who is also a model. Her father is a restaurant owner and her mother is a corporate events manager. Shaik started modeling at the age of 8, appearing in a series of commercials for Myer, Target, K-Mart, Jets Pizzas, and others. Career Shaik was first scouted as a model at the age of 8, but didn't start modeling career until she was 15, when she was signed by New York-based IMG Models in 2006. She was signed to the Agency First Access Entertainment when she was 17 years old. In 2008, Shaik launched her modeling career in New York City. She graced her first international magazine cover on the British local magazine ‘Culture.’ In 2009, Shaik returned to Australia to walk in Melbourne Fashion Week for model Dion Lee. In 2010, Shaik moved back to New York and landed a number of big campaigns such as Victoria's Secret, American Eagle, and Lovable. Shaik's career continued to grow, and in 2011, she walked in the Victoria's Secret Fashion Show for the first time. She has since walked in the show three times (2012, 2014, and 2018). In addition to modeling, Shaik has also appeared in music videos for artists such as Kanye West and Justin Bieber. She has also been featured in campaigns for brands such as Ralph Lauren, Free People, and Urban Outfitters. In 2015, Shaik was named the new face of Seafolly, an Australian swimwear brand. She has also appeared in the pages of top fashion magazines such as Vogue, Marie Claire, Harper's Bazaar, and Elle. Personal life Shaik met DJ Ruckus (Greg Andrews) in Ibiza in 2015. Andrews proposed on her birthday while on holiday in the Bahamas in December 2015 and they married in the Bahamas in April 2018. The couple announced their separation in June 2019. Shaik is a supporter of various charities, including the Lotus Outreach International, a non-profit organization that helps at-risk women and children in Cambodia, and the Natal Foundation, a charitable organization that works to improve the lives of those living in the favelas of Brazil. References - Bogart, Laura. Shanina Shaik: Model with an eye on the top (https://www.smh.com.au/entertainment/shanina-shaik-model-with-an-eye-on-the-top-20101117-17uz2.html). 2010-11-17. - Shanina Shaik is striking in the Sabina Hairdo in her latest Photoshoot (https://www.popsugar.co.uk/fashion/Shanina-Shaik-Striking-Sabina-Hairdo-Photoshoot-31353826). POPSUGAR UK Fashion. 2013-08-12. - Meet the ‘Tomboy’ secretly stalking the runway at Paris Couture Week (https://www.news.com.au/lifestyle/fashion/people/meet-the-tomboy-secretly-stalking-the-runway-at-paris-couture-week/news-story/c27f9640849615e4bf41d2f8733eda62). 2013-06-28. - SHANINA SHAIK (https://www.firstaccessent.com/first-shanina-shaik.html). - Aussie fashion model Shanina Shaik set for Victoria's Secret debut after securing lingerie brand's top gig (http://www.news.com.au/entertainment/aussie-fashion-model-shanina-shaik-set-for-victorias-secret-debut-after-securing-lingerie-brands-top-gig/story-fn9076o9-1226563405442). NewsComAu. - "I was here before Shanina, so step back!" Alex Perry and Shanina Shaik dish it out on Australia's Next Top Model (http://www.thevine.com.au/fashion/celeb/i-was-here-before-shanina-so-step-back%21-alex-perry-and-shanina-shaik-dish-it-out-on-australia%27s-next-top-model-20110808-241395.aspx). - Ross, Jenna. Fahion (http://www.smh.com.au/lifestyle/fashion/head-over-heels-20111124-1nxqe.html). Smh.com.au. 2011-11-26. - Shanina Shaik - Model Profile (http://nymag.com/fashion/models/ssaik/shaninashaik/). Nymag.com. - Bureau, Entertainment. Shanina Shaik to appear in Victoria's Secret catwalk show (http://www.dailytelegraph.com.au/entertainment/sydney-confidential/shanina-shaik-to-appear-in-victorias-secret-catwalk-show/story-e6frewz0-1225996725725). Daily Telegraph. news.com.au. 20 May 2011. - Betty White Gets the Last Laugh (https://www.vogue.com/article/justin-bieber-snl-naked-cameo). Vogue. - Shanina Shaik - Fashion Model (https://models.com/models/shanina-shaik). MODELS.com. - Wedding Bells Ring for Greg Andrews (DJ Rukkus) and Shanina Shaik (http://models.com/oftheminute/?p=14749). Models.com. 2012-01-04. - Shanina Shaik is the New Face of Seafolly - on (http://www.exploredailymail.com/Shanina+Shaik+is+the+New+Face+of+Seafolly+on+Lastwaterway.com/159247). Exploredailymail.com. 2010-07-14. - Shanina Shaik Personal Success (https://www.inc.com/lewis-schiff/shanina-shaik-personal-success.html). - Shanina Shaik Helps Children in Brazil (https://www.youtube.com/watch?v=W6K8_QHmYZ8). External links * * * Shanina Shaik (https://models.com/models/shanina-shaik) on Models.com Category:1991 births Category:Living people Category:Australian female models Category:Australian people of Pakistani descent Category:Australian people of Lithuanian descent Category:IMG Models models Category:Models from Melbourne Update (2024-04-21): . Trying to make better images import numpy as np import tensorflow as tf import matplotlib.pyplot as plt import cv2 from tensorflow.keras.preprocessing.image import ImageDataGenerator from tensorflow.keras.callbacks import ReduceLROnPlateau, ModelCheckpoint from tensorflow.keras.layers import Conv2D, MaxPooling2D, Dense, Flatten, BatchNormalization, Activation from tensorflow.keras.models import Sequential from tensorflow.keras.optimizers import Adam data_dir = 'data/' num_of_classes = len(list(data_dir + 'train/')) def build_sequence(datadir, img_size): seq = ImageDataGenerator(rescale=1./255, zoom_range=(0.9, 1.1), height_shift_range=3, width_shift_range=3, rotation_range=30, horizontal_flip=True, vertical_flip=True, fill_mode='constant', cval=0, validation_split=0.155) val_seq = ImageDataGenerator(rescale=1./255, validation_split=0.155) train_generator = seq.flow_from_directory(datadir, target_size=(img_size, img_size), batch_size=96, classes=['BadImg', 'GoodImg'], class_mode='binary', shuffle=True, subset='training') val_generator = val_seq.flow_from_directory(datadir, target_size=(img_size, img_size), batch_size=96, classes=['BadImg', 'GoodImg'], class_mode='binary', shuffle=True, subset='validation') # more sizes = better model (if done right) return train_generator, val_generator def build_model(cnn=True): """Use cnn = False for testing the MLP of the model vs the CNN""" if cnn: model = Sequential() model.add(Conv2D(64, (3,3), padding='same', input_shape=(144, 144, 3))) model.add(BatchNormalization()) model.add(Activation('relu')) model.add(MaxPooling2D((2,2))) model.add(Conv2D(64, (3,3), padding='same')) model.add(BatchNormalization()) model.add(Activation('relu')) model.add(MaxPooling2D((2,2))) model.add(Conv2D(128, (3,3), padding='same')) model.add(BatchNormalization()) model.add(Activation('relu')) model.add(MaxPooling2D((2,2))) model.add(Conv2D(256, (3,3), padding='same')) model.add(BatchNormalization()) model.add(Activation('relu')) model.add(MaxPooling2D((2,2))) model.add(Conv2D(512, (5,5), padding='same')) model.add(BatchNormalization()) model.add(Activation('relu')) model.add(MaxPooling2D((2,2))) model.add(Conv2D(512, (5,5), padding='same')) model.add(BatchNormalization()) model.add(Activation('relu')) model.add(MaxPooling2D((2,2))) model.add(Flatten()) model.add(Dense(512)) model.add(BatchNormalization()) model.add(Activation('relu')) model.add(Dense(128)) model.add(BatchNormalization()) model.add(Activation('relu')) # Use sigmoid for binary classification model.add(Dense(1, activation='sigmoid')) model.summary() return model # The MLP version of the model model = Sequential() model.add(Flatten(input_shape=(144, 144, 3))) model.add(Dense(512)) model.add(BatchNormalization()) model.add(Activation('relu')) model.add(Dense(128)) model.add(BatchNormalization()) model.add(Activation('relu')) model.add(Dense(1, activation='sigmoid')) model.summary() return model def train_model(model, train_generator, val_generator): """For testing: history = train_single_layer_model(model, x_train, y_train) """ opt = Adam(lr=0.0001) cb = [ModelCheckpoint('models/weights.h5', save_best_only=True, monitor='val_loss', mode='min'), ReduceLROnPlateau(patience=3, verbose=1)] model.compile(optimizer=opt, loss='binary_crossentropy', metrics=['accuracy']) history = model.fit_generator(train_generator, steps_per_epoch=len(train_generator), epochs=40, validation_data=val_generator, validation_steps=len(val_generator), callbacks=cb) return history # Main img_size = 144 train_generator, val_generator = build_sequence(data_dir, img_size) model = build_model(cnn=True) # Train the model hist = train_model(model, train_generator, val_generator) # access data from history print(hist.history.keys()) # Visualize the accuracy and the loss plt.plot(hist.history['accuracy']) plt.plot(hist.history['loss']) plt.title('Accuracy and Loss over time') plt.ylabel('value') plt.xlabel('epoch') plt.legend(['accuracy', 'loss'], loc='upper left') plt.show() # Example of a bad Image img = cv2.imread('/Users/ria/desktop/pyvenv/cnn/data/train/BadImg/IMG_0333.JPG'.format(np.random.randint(200))) print(img.shape) plt.imshow(img) plt.show() # Because it's an ndarray, we can manipulate it and create copies of it to add to the dataset. # Let's try flipping it for example example_img_debug = np.fliplr(img) print(example_img_debug.shape) plt.imshow(example_img_debug) plt.show() def predict(img): return model.predict(np.array([img]))[0][0] print('Predicted value: ', predict(img)) aug = ImageDataGenerator(rescale=1./255, zoom_range=(0.95, 1.05),