def __init__(self):
'''
Constructor
'''
ml_alg_base.__init__(self)
self.dsr = DatasetReader()
self.learning_model = naive_bayes.GaussianNB()
def __init__(self):
'''
Constructor
'''
self.dsr = DatasetReader()
self.fenc = FreemanEncoder()
self.training_data = []
def build_binary_classifiers(path_g1_sg2m, path_g1s_g2m):
"""
Build the stacked neural network with single output neuron for binary classification to G1 vs. S+G2M and
G1+S vs. G2M phases, and evaluate its performance
:param path_g1_sg2m: Path to the labeled dataset in two labels : G1 and SG2M
:param path_g1s_g2m: Path to the labeled dataset in two labels : G1S and G2M
:return: Accuracy of classification of each model.
"""
############### Ordinal Classifier #################
dr1 = DatasetReader(path_g1_sg2m)
dr2 = DatasetReader(path_g1s_g2m)
binary_train1 = dr1.load_data()
binary_train2 = dr2.load_data()
oc1 = OrdinalClassifier(binary_train1[0], binary_train1[1])
oc2 = OrdinalClassifier(binary_train2[0], binary_train2[1])
r1 = oc1.classify()
r2 = oc2.classify()
return r1, r2
def __init__(self, n_neighbors=1):
'''
Constructor
'''
self.dsr = DatasetReader()
self.fenc = FreemanEncoder()
self.data = []
self.knn = KNeighborsClassifier(n_neighbors=n_neighbors,
algorithm='auto',
metric=self.lev_metric)
def __init__(self):
'''
Constructor
'''
self.dsr = DatasetReader()
self.fenc = FreemanEncoder()
states = ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']
symbols = ['0', '1', '2', '3', '4', '5', '6', '7']
self.learning_model = HiddenMarkovModelTrainer(states=states,
symbols=symbols)
self.model = None
def build_stacked_ae(path):
"""
Build the stacked auto-encoder neural network, and evaluate its performance
:param path: Path to the genetic dataset
:return: Accuracy of classification of cell cycle phase.
"""
############### Stacked Auto-Encoders ##############
dr = DatasetReader(path)
train = dr.load_data()
ae = StackedAutoencoder(train[0], train[1], train[2], 3)
ae.create_autoencoder()
result = ae.evaluate_autoencoder()
return result[1] * 100
print("Accuracy: %.2f%%" % (result[1] * 100))
def __init__(self, dataset_path, args):
self._dataset_path = dataset_path
self._documents = DatasetReader(dataset_path, args).read_dataset()
m20Connector.initDB()
# Init to read
moviesDS = DatasetReader.initWithFraction('datasets/data/movies.csv',
1.0,
',',
init=True)
gtagsDS = DatasetReader.initWithFraction(
'datasets/data/genome-tags.csv', 1.0, ',', init=True)
linksDS = DatasetReader.initWithFraction('datasets/data/links.csv',
1.0,
',',
init=True)
#Just init
ratingsDS = DatasetReader("datasets/data/ratings.csv", init=True)
tagsDS = DatasetReader("datasets/data/tags.csv", init=True)
gscoresDS = DatasetReader("datasets/data/genome-scores.csv", init=True)
if (init_clear == False):
for movie in moviesDS.readPercentage():
# print str(movie)
m20Connector.insert(
M20Movie(movie['movieId'], movie['title'],
movie['genres']))
for tag in gtagsDS.readPercentage():
# print str(tag)
m20Connector.insert(M20GenomeTag(tag['tagId'], tag['tag']))
for link in linksDS.readPercentage():
from datasets.caltechpedestrian import CaltechPedestrian
from datasets.bdd100k import BDD100K
from datasets.citypersons import CityPersons
from DatasetReader import DatasetReader
import logging
import os
logging.basicConfig(filename='example.log',level=logging.DEBUG)
base_dir_bdd100k = '/data/stars/share/STARSDATASETS/bdd100k'
for subset in ['train', 'val']:
db = BDD100K(name='bdd100k-{}'.format(subset), base_dir=base_dir_bdd100k, save_dir='./bdd100k-{}'.format(subset), subset=subset)
db.writedataframe()
reader = DatasetReader('./bdd100k-{}'.format(subset))
df = reader.get_annotations(query='category == "person"')
reader.plot_annotations(df=df, plot_cols=['xmin', 'ymin', 'xmax', 'ymax', 'category'])
base_dir_caltech = '/data/stars/user/uujjwal/datasets/pedestrian/caltech/caltechall-train'
db = CaltechPedestrian(name='caltechall-train', base_dir=base_dir_caltech, save_dir='./caltechall-train')
db.writedataframe()
reader = DatasetReader('./caltechall-train')
df = reader.get_annotations(query='object == "person"')
reader.plot_annotations(df=df, plot_cols=['xmin_full', 'ymin_full', 'xmax_full', 'ymax_full', 'object'])
base_dir_caltech = '/data/stars/user/uujjwal/datasets/pedestrian/caltech/caltechall-test'
db = CaltechPedestrian(name='caltechall-test', base_dir=base_dir_caltech, save_dir='./caltechall-test')
db.writedataframe()
reader = DatasetReader('./caltechall-test')
def __init__(self):
self.reader = DatasetReader()
import numpy as np
from matplotlib import pyplot as plt
from Args import DIM, ROOT, EPOCHS, BATCH_SIZE, NUM_WORKERS, LEARNING_RATE
from DatasetReader import DatasetReader
from model import UNet
import torch.optim as optim
from copy import deepcopy
from Evaluation import MeanDiceCoefficient
if (__name__ == "__main__"):
model = UNet().cuda()
loss_fn = nn.BCELoss()
optimiser = optim.Adam(model.parameters(), lr=LEARNING_RATE)
trainset = DatasetReader(ROOT + "train/")
testset = deepcopy(trainset)
testset.setTrainMode(False)
trainloader = torch.utils.data.DataLoader(trainset,
batch_size=BATCH_SIZE,
shuffle=True,
num_workers=NUM_WORKERS)
testloader = torch.utils.data.DataLoader(testset,
batch_size=BATCH_SIZE,
shuffle=True,
num_workers=NUM_WORKERS)
for epoch in range(EPOCHS):
# Training phase
moviesDS = DatasetReader.initWithFraction(dataset_path + '/movies.csv',
1.0,
',',
init=True)
gtagsDS = DatasetReader.initWithFraction(dataset_path +
'/genome-tags.csv',
1.0,
',',
init=True)
linksDS = DatasetReader.initWithFraction(dataset_path + '/links.csv',
1.0,
',',
init=True)
#Just init
ratingsDS = DatasetReader(dataset_path + "/ratings.csv", init=True)
tagsDS = DatasetReader(dataset_path + "/tags.csv", init=True)
gscoresDS = DatasetReader(dataset_path + "/genome-scores.csv",
init=True)
if (init_clear == False):
for movie in moviesDS.readPercentage():
# print str(movie)
m20Connector.insert(
M20Movie(movie['movieId'], movie['title'],
movie['genres']))
for tag in gtagsDS.readPercentage():
# print str(tag)
m20Connector.insert(M20GenomeTag(tag['tagId'], tag['tag']))
