def main(args):
data_transform = transforms.Compose(
[transforms.Resize((224, 224)),
RGB_to_LAB()])
#Loading the Training Set
train_dataset = datasets.ImageFolder(root=args.trainset_path,
transform=data_transform)
train_loader = torch.utils.data.DataLoader(train_dataset,
batch_size=args.batch_size,
shuffle=True,
pin_memory=True,
num_workers=args.num_workers)
#Loading the Validation Set
val_dataset = datasets.ImageFolder(root=args.valset_path,
transform=data_transform)
val_loader = torch.utils.data.DataLoader(val_dataset,
batch_size=args.batch_size,
shuffle=False,
pin_memory=True,
num_workers=args.num_workers)
print('Total images in training set: ', len(train_dataset))
print('Total images in validation set: ', len(val_dataset))
train = training(args)
if args.infer_iter:
train.test(val_loader, args.infer_iter, args.infer_iter)
else:
#Send to train
train.train(train_loader, val_loader)
def thick_img_cls(dir_, IMG_DIMS, INPUT_SHAPE): train_imgs, train_labels, test_imgs, test_labels, val_imgs, val_labels = data_creation(dir_, IMG_DIMS, w=40, h=40, num_neg=20) model = CNN_model(INPUT_SHAPE) model_trained = training(model, train_imgs, train_labels, val_imgs, val_labels, BATCH_SIZE, EPOCHS, model_name='CNN_thick_imgs') return model_trained, test_imgs, test_labels
def ind_cells_cls(base_dir, IMG_DIMS, INPUT_SHAPE ): train_imgs, train_labels, val_imgs, val_labels, test_imgs, test_labels = data_preparation(base_dir, IMG_DIMS) model = CNN_model(INPUT_SHAPE) model_trained = training(model, train_imgs, train_labels, val_imgs, val_labels, BATCH_SIZE, EPOCHS, model_name='CNN_ind_cells') return model_trained, test_imgs, test_labels
from __future__ import division
import numpy as np
from scipy.io.wavfile import read
from LBG import EUDistance
from mel_coefficients import mfcc
from LPC import lpc
from Train import training
import os
nSpeaker = 8
nfiltbank = 12
orderLPC = 15
(codebooks_mfcc, codebooks_lpc) = training(nfiltbank, orderLPC)
directory = os.getcwd() + '/test'
fname = str()
nCorrect_MFCC = 0
nCorrect_LPC = 0
def minDistance(features, codebooks):
speaker = 0
distmin = np.inf
for k in range(np.shape(codebooks)[0]):
D = EUDistance(features, codebooks[k, :, :])
dist = np.sum(np.min(D, axis=1)) / (np.shape(D)[0])
if dist < distmin:
distmin = dist
print(distmin)
speaker = k
return speaker
def main():
training(pretrained_weights=None, )
import numpy as np
from scipy.io.wavfile import read
from LBG import EucledianDistance
from MFCC import MFCC_Coeff
from LinearPredictionCoefficients import lpc
from Train import training
import os
nCorrect_MFCC = 0
nCorrect_LPC = 0
trainingSet = 4
q1 = 12
q2 = 15
(cbMfcc, cbLpc) = training(q1, q2)
directory = os.getcwd() + '/test'
fname = str()
def minDistance(f, c):
person = 0
minDist = np.inf
for k in range(np.shape(c)[0]):
D = EucledianDistance(f, c[k, :, :])
dist = np.sum(np.min(D, axis=1)) / (np.shape(D)[0])
if dist < minDist:
minDist = dist
person = k
return person
for i in range(nPerson):