def predict(original_image): model.load_weights("models/_mini_XCEPTION.87-0.65.hdf5") highlighted_face_image, faces = find_face(original_image) nfaces = len(faces) prediction = None predicted_emotion = None if len(faces) > 0: for [x, y, w, h] in faces: image = original_image[y:y+h,x:x+w] image = preprocess_live_image(image) image = [[image]] prediction = model.predict(image) maximum = max(prediction) predicted_emotion = emotion_list[prediction.argmax()] cv2.putText(highlighted_face_image, predicted_emotion, (x,y-20), cv2.FONT_HERSHEY_SIMPLEX, 2, (0,0,255), 10) return image_extraction(highlighted_face_image, faces, nfaces, prediction, predicted_emotion,original_image)
def test(image_path, snapshot_path, transform=False): img = Image.open( image_path) # WARNING : this image is well centered and square img = img.resize(model.inputs[0].shape) imarr = np.array(img).astype(np.float32) imarr = imarr.transpose((2, 0, 1)) imarr = np.expand_dims(imarr, axis=0) model.load_weights(snapshot_path) out = model.predict(imarr) best_index = np.argmax(out, axis=1)[0] print best_index
def predict(word): model.load_weights('data/model.h5') windows = dataset.process_word(word.lower(), training=False) hyphenated = word[:2] for offset, window in enumerate(windows): result = model.predict(np.array([window])) #print('>>> {}{}{} => {: >5.2f} %'.format( # word[:2 + offset], # dataset.HYPHENATION_INDICATOR, # word[2 + offset:], # result[0][0] * 100 #)) if result[0][0] > 0.5: hyphenated += dataset.HYPHENATION_INDICATOR hyphenated += word[offset + 2] hyphenated += word[-1:] return hyphenated
# -*- coding: utf-8 -*- # @File : eval.py # @Author : AaronJny # @Time : 2019/12/30 # @Desc : from dataset import tokenizer from model import model import settings import utils # 加载训练好的模型 model.load_weights(settings.BEST_MODEL_PATH) # 随机生成一首诗 print(utils.generate_random_poetry(tokenizer, model)) # 给出部分信息的情况下,随机生成剩余部分 print(utils.generate_random_poetry(tokenizer, model, s='床前明月光,')) # 生成藏头诗 print(utils.generate_acrostic(tokenizer, model, head='海阔天空')) print(utils.generate_acrostic(tokenizer, model, head='天道酬勤'))
import cv2 import numpy as np import os import uuid os.environ["CUDA_VISIBLE_DEVICES"] = "" import tensorflow as tf from model import model, preprocess import time import pickle model_name = 'model_train_v2.h5' model.load_weights(model_name) cap = cv2.VideoCapture(0) last_warning = 0 def save_image(image, pos=False): if pos: fname = os.path.join("./pos_frames", str(uuid.uuid4()) + ".png") else: fname = os.path.join("./raw_data", str(uuid.uuid4()) + ".png") cv2.imwrite(fname, image) bad_count = 0 avg_prediction = [] frame_array = [] while (True):
import numpy as np from model import model import matplotlib.pyplot as plt # Read subset of data all_data = np.load('simple_data.npz') imgs_color = all_data['imgs'] speedx = np.concatenate((all_data['spds'], all_data['accel'])) speedx = speedx.reshape((-1, 2)) steer = all_data['steer'] #make predictions start = 45000 stop = 65000 model.load_weights('steer_comma_0_0.00057615.h5') preds = model.predict([speedx[start:stop], imgs_color[start:stop]]) steer_preds = preds.reshape([-1]) # Video of prediction import matplotlib.animation as animation from PIL import Image, ImageDraw figure = plt.figure() imageplot = plt.imshow(np.zeros((64, 64, 3), dtype=np.uint8)) val_idx = start def get_point(s, start=0, end=63, height=16): X = int(s * (end - start)) if X < start: X = start if X > end:
from tensorflow.keras.datasets import mnist from model import model from skimage import io import matplotlib.pyplot as plt import numpy as np SAVED_MODEL = 'ToReport/Checkpoints/digits-cnn_best_weights.h5' _, (test_images, test_labels) = mnist.load_data() N = len(test_images) k = np.random.randint(N) test_img = test_images[k].reshape(1, *test_images[k].shape, 1) test_img = test_img / 255.0 model.load_weights(SAVED_MODEL) pred = model.predict(test_img) io.imshow(test_images[k]) plt.title( f'Истинное значение: {test_labels[k]}, Предсказание модели: {np.argmax(pred)}' ) plt.show()
tb = TensorBoard(log_dir='./logs') update_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.8, patience=5, min_lr=1e-6, verbose=1) checkpoint = ModelCheckpoint(weights_path, monitor='val_loss', verbose=1, save_best_only=True, save_weights_only=True) # MODEL AND TRAIN model = model(img_height, img_width, num_classes, lr) model.fit(X_train, Y_train, epochs=epochs, batch_size=batch_size, verbose=1, validation_data=(X_val, Y_val), shuffle=True, callbacks=[checkpoint, update_lr, tb]) # MODEL TEST model.load_weights(weights_path) preds = model.evaluate(X_test, Y_test) print("Loss = " + str(preds[0])) print("Test Accuracy = " + str(preds[1]))
import cv2 import numpy as np import tensorflow as tf from model import model import matplotlib.pyplot as plt classes_name = {0: 'Airplane', 1: 'Automobile', 2: 'Bird', 3: 'Cat', 4: 'Deer', 5: 'Dog', 6: 'Frog', 7: 'Horse', 8: 'Ship', 9: 'Truck'} model = model() model.load_weights('../MODEL DATA/tf-model.h5') def classify(img): img = cv2.resize(img, (32, 32)) / 255.0 img = tf.expand_dims(tf.cast(img, tf.float32), axis=0) prediction = model.predict(img)[0] index = np.argmax(prediction) class_name = classes_name[int(index)] return class_name image = plt.imread('../IMAGES/car1.jpg') cls_name = classify(image) plt.imshow(image) plt.title(cls_name, fontsize=10) plt.show()
from properties import * from model import model from data_generator import midi_input_generator, generate_song_array, save_array_to_midi print "Testing saved model" model.load_weights('net_dump.nw') save_array_to_midi([generate_song_array(model)], 'Generated.mid')
log_dir + "\\weights_epoch_{epoch:02d}-loss_{val_loss:.2f}.hdf5", monitor='val_loss', verbose=0, save_best_only=False, save_weights_only=False, mode='auto', period=1) # lr_sched = keras.callbacks.LearningRateScheduler(lambda epoch: 0.002* 0.75 ** (epoch-1) ) cfm_callback = confusion_matrix_callback(train_generator, validation_generator) callbacks = [tensorboard_callback, reduce_lr, modelCP, csv_logger] # , lr_sched] #, cfm_callback ] # ------------------------------ # saved_weights = r"C:\Users\User\PycharmProjects\PlantPathology\logs\fit\20200331-120037\weights_epoch_17-loss_1.00.hdf5" # saved_weights = r"C:\Users\User\PycharmProjects\PlantPathology\logs\fit\saved_weights\weights_epoch_34-loss_1.06.hdf5" if C.MODEL == 'VGG' and C.PRETRAINED_VGG: model.load_weights(C.PRETRAINED_VGG) model.fit_generator( train_generator, validation_data=validation_generator, epochs=500, callbacks=callbacks, verbose=True, # class_weight=get_class_weights() class_weight=1 / pd.Series(train_generator.classes).value_counts(), # initial_epoch = 20, # class_weight= 'auto', )
# Written by Markus Siemens (MIT; https://github.com/msiemens/HypheNN-de) from time import time import util from dataset import data_validation from model import model start_time = time() model.load_weights('models/model_saved.h5') data = data_validation() print('Validating model...') result = model.evaluate(data[0], data[1]) print() print('Done') print('Result:', result) print('Time:', util.time_delta(time() - start_time))
def warn(*args, **kwargs): pass import warnings warnings.warn = warn from dataset import tokenizer from model import model import config import utils import argparse # 加载训练好的模型 model.load_weights(config.BEST_MODEL_PATH) if __name__ == '__main__': parser = argparse.ArgumentParser( description='诗歌生成器 功能:1.随机生成一首诗 2.根据开头生成后面的诗句 3.生成藏头诗') parser.add_argument('--start', '-s', help='生成的开头') parser.add_argument('--acrostic', '-a', help='藏头诗') parser.add_argument('--number', '-n', help='五言、七言') parser.add_argument('--count', '-c', help='律诗、绝句') args = parser.parse_args() # 给出部分信息的情况下,随机生成剩余部分 if args.start: print( utils.generate_random_poetry(tokenizer, model,
help="Input text file") parser.add_argument("-w", "--weights", action="store", required=False, dest="weights", help="Model weights path") parser.add_argument("-i", "--input", action="store", required=False, dest="input", help="Input string for complete") parser.add_argument("-o", "--out_len", action="store", required=False, dest="out_len", help="Out length") args = parser.parse_args() _, _, vectorizer = load_data.load_data(args.text, False, False, False, 1) model = model.make_text_generator_model(1, vectorizer.vocab_size) model.load_weights(args.weights) res = generate(model, vectorizer, seed=args.input, length=int(args.out_len)) print(res)
from model import model from qlearning4k import Agent from flappy_bird import FlappyBird game = FlappyBird(frame_rate=30, sounds=True) model.load_weights('weights.dat') agent = Agent(model) agent.play(game, nb_epoch=100, epsilon=0.01, visualize=False)
from model import model from nmt_utils import * m = 10000 Tx = 30 Ty = 10 n_a = 32 n_s = 64 learning_rate = 0.005 batch_size = 100 dataset, human_vocab, machine_vocab, inv_vocab = load_dataset(m) X, Y, Xoh, Yoh = preprocess_data(dataset, human_vocab, machine_vocab, Tx, Ty) model = model(Tx, Ty, n_a, n_s, len(human_vocab), len(machine_vocab)) model.load_weights('models/model_50.h5') s0 = np.zeros((m, n_s)) c0 = np.zeros((m, n_s)) EXAMPLES = [ '3 May 1979', '5 April 09', '21th of August 2016', 'Tue 10 Jul 2007', 'Saturday May 9 2018', 'March 3 2001', 'March 3rd 2001', '1 March 2001' ] total = len(EXAMPLES) count = 1 TARGETS = [ '1979-05-03', '2009-04-05', '2016-08-21', '2007-07-10', '2018-05-09', '2001-03-03', '2001-03-03', '2001-03-01'
from model import model, preprocess_input, smodel from keras.optimizers import SGD from keras.callbacks import ReduceLROnPlateau from generator import Generator import json if __name__ == '__main__': model = smodel() opt = SGD(lr=0.01, momentum=0.9) model.compile(optimizer=opt, loss='categorical_crossentropy', metrics=['acc']) print(model.summary()) model.load_weights('weights/classifier.h5') listsss = json.load(open('list_withbndbx.json', 'r')) train_gen = Generator( listsss[:7211], '/home/palm/PycharmProjects/DATA/Tanisorn/imgCarResize/', preprocess_function=preprocess_input) test_gen = Generator( listsss[7211:], '/home/palm/PycharmProjects/DATA/Tanisorn/imgCarResize/', preprocess_function=preprocess_input) reduce_lr_01 = ReduceLROnPlateau(monitor='val_1st_acc', factor=0.2, patience=5, min_lr=0, mode='max') reduce_lr_02 = ReduceLROnPlateau(monitor='val_2nd_acc', factor=0.2,
classes_name = { 0: 'Airplane', 1: 'Automobile', 2: 'Bird', 3: 'Cat', 4: 'Deer', 5: 'Dog', 6: 'Frog', 7: 'Horse', 8: 'Ship', 9: 'Truck' } model = model() model.summary() model.load_weights('MODEL DATA/model.h5') def classify(image): img = image.resize((32, 32)) img = np.asarray(img) img = np.expand_dims(img, axis=0) img = img / 255.0 prediction = model.predict(img) prediction = prediction[0] max_index = np.argmax(prediction) class_name = classes_name.get(max_index) return class_name image_names = os.listdir('IMAGES/')
INIT_LR = 5e-3 BATCH_SIZE = 32 EPOCHS = 14 data_loader = DataLoader() x_train, y_train, x_test, y_test, classes = data_loader.load_data() x_train2 = (x_train / 255) - 0.5 x_test2 = (x_test / 255) - 0.5 y_train2 = keras.utils.to_categorical(y_train, NUM_CLASSES) y_test2 = keras.utils.to_categorical(y_test, NUM_CLASSES) model = model() model.load_weights("weights/weights.h5") # make test predictions y_pred_test = model.predict_proba(x_test) y_pred_test_classes = np.argmax(y_pred_test, axis=1) y_pred_test_max_probas = np.max(y_pred_test, axis=1) plt.figure(figsize=(7, 6)) plt.title('Confusion matrix', fontsize=16) plt.imshow(confusion_matrix(y_test, y_pred_test_classes)) plt.xticks(np.arange(10), classes, rotation=45, fontsize=12) plt.yticks(np.arange(10), classes, fontsize=12) plt.colorbar()
classes_name = { 0: 'Airplane', 1: 'Automobile', 2: 'Bird', 3: 'Cat', 4: 'Deer', 5: 'Dog', 6: 'Frog', 7: 'Horse', 8: 'Ship', 9: 'Truck' } model = model() model.summary() model.load_weights('MODEL DATA/cifar-10.h5') def classify(image): img = image.resize((32, 32)) img = np.asarray(img) img = np.expand_dims(img, axis=0) img = img / 255.0 prediction = model.predict(img) prediction = prediction[0] max_index = np.argmax(prediction) class_name = classes_name.get(max_index) return class_name image_names = os.listdir('IMAGES/')
import cv2 import numpy as np from model import model from darkflow.net.build import TFNet options = { 'model': 'cfg/tiny-yolo-voc-1c.cfg', 'load': 3250, 'threshold': 0.1, 'gpu': 1.0 } tfnet = TFNet(options) model = model() model.summary() model.load_weights('weights/fingertip_weights/Fingertip.h5') def detect_hand(image): """ Hand detection """ output = tfnet.return_predict(image) print(output) tl, br = None, None for prediction in output: # label = prediction['label'] # confidence = prediction['confidence'] tl = (prediction['topleft']['x'], prediction['topleft']['y']) br = (prediction['bottomright']['x'], prediction['bottomright']['y']) return tl, br
from time import time import util from dataset import data_validation from model import model start_time = time() model.load_weights('data/model.h5') data = data_validation() print('Validating model...') result = model.evaluate(data[0], data[1]) print() print('Done') print('Result:', result) print('Time:', util.time_delta(time() - start_time))