def snapshot_pre(self, submit, imagePath, type_image):
scores_predict = []
for name in tqdm(submit['Id']):
path = os.path.join(imagePath, name)
image = data_generator.load_image(path,
self.input_shape + [self.div],
type_image)
score_predict = self.model.predict(image[np.newaxis])[0]
scores_predict += [score_predict]
return scores_predict
def snapshot_val(self, dataset_info, type_image):
scores_predict = []
for idx in range(len(dataset_info)):
image = data_generator.load_image(dataset_info[idx]['path'],
self.input_shape + [self.div],
type_image)
score_predict = self.model.predict(image[np.newaxis])[0]
scores_predict += [score_predict]
return scores_predict
def predict(self, submit, imagePath):
predicted = []
for name in tqdm(submit['Id']):
path = os.path.join(imagePath, name)
image = data_generator.load_image(path, self.input_shape)
score_predict = self.model.predict(image[np.newaxis])[0]
label_predict = np.arange(28)[score_predict >= 0.2]
str_predict_label = ' '.join(str(l) for l in label_predict)
predicted.append(str_predict_label)
submit['Predicted'] = predicted
submit.to_csv('submission.csv', index=False)
def eval(self, dataset_info, type_image):
labels_true = []
scores_predict = []
for idx in range(len(dataset_info)):
image = data_generator.load_image(dataset_info[idx]['path'],
self.input_shape + [self.div],
type_image)
labels = np.zeros(28)
labels[dataset_info[idx]['labels']] = 1
labels_true += [labels]
score_predict = self.model.predict(image[np.newaxis])[0]
#label_predict = np.arange(28)[score_predict>=label_threshold]
scores_predict += [score_predict]
return labels_true, scores_predict
from tensorflow.keras.callbacks import LambdaCallback
from tensorflow.keras.callbacks import Callback
from tensorflow.keras import metrics
from tensorflow.keras.optimizers import Adam
from tensorflow.keras import backend as K
from tensorflow import keras
import tensorflow as tf
from kaggle_data import data_generator
from uitls import f1_loss, f1
model = load_model('working/model_train2.h5',
custom_objects={
'f1_loss': f1_loss,
'f1': f1
})
submit = pd.read_csv('data/sample_submission.csv')
predicted = []
for name in tqdm(submit['Id']):
path = os.path.join('data/test/', name)
image = data_generator.load_image(path, INPUT_SHAPE)
score_predict = model.predict(image[np.newaxis])[0]
label_predict = np.arange(28)[score_predict >= 0.2]
str_predict_label = ' '.join(str(l) for l in label_predict)
predicted.append(str_predict_label)
submit['Predicted'] = predicted
submit.to_csv('submission.csv', index=False)