def spectrum_dct(self):
"""
Calculo del espectro segun transformada discreta del coseno
"""
return Transforms.dct(self)
def temp_add_transforms(self):
# TODO: accept multiple arguments for features to add
# this function is currently only for debugging
# create features/transforms
self.df['mmi'] = Transforms.mmi(self.df, 500)
self.df['sma'] = self.df.price.rolling(window=5, center=False).mean()
def validationTransform(self, rgb, depth):
first_resize = tuple(
map(
int,
list((250.0 / IMAGE_HEIGHT) *
np.array([IMAGE_HEIGHT, IMAGE_WIDTH]))))
depth_np = depth
transform = T.Compose([
T.Resize(first_resize),
T.CenterCrop((228, 304)),
T.Resize(self.output_size),
])
rgb_np = transform(rgb)
rgb_np = np.asfarray(rgb_np, dtype='float') / 255
depth_np = transform(depth_np)
return rgb_np, depth_np
def trainTransform(self, rgb, depth):
s = np.random.uniform(1.0, 1.5) # random scaling
depth_np = depth / s
angle = np.random.uniform(-5.0, 5.0) # random rotation degrees
do_flip = np.random.uniform(0.0, 1.0) < 0.5 # random horizontal flip
# perform 1st step of data augmentation
first_resize = tuple(
map(
int,
list((250.0 / IMAGE_HEIGHT) *
np.array([IMAGE_HEIGHT, IMAGE_WIDTH]))))
second_resize = tuple(
map(int, list(s * np.array([IMAGE_HEIGHT, IMAGE_WIDTH]))))
transform = T.Compose([
T.Resize(
first_resize
), # this is for computational efficiency, since rotation can be slow
T.Rotate(angle),
T.Resize(second_resize),
T.CenterCrop((228, 304)),
T.HorizontalFlip(do_flip),
T.Resize(self.output_size),
])
rgb_np = transform(rgb)
rgb_np = self.color_jitter(rgb_np) # random color jittering
rgb_np = np.asfarray(rgb_np, dtype='float') / 255
depth_np = transform(depth_np)
return rgb_np, depth_np
def new_event(self, date, event_index=None):
'''returns a new event'''
# if not any memory is loaded
if not "_self" in self.current_streamview.atoms.keys():
return None
# if event is specified, play it now
if event_index != None:
self.reset()
event_index = int(event_index)
z_, event = self.current_streamview.atoms["_self"].memorySpace[
event_index]
# using actual transformation?
transforms = [Transforms.NoTransform()]
self.waiting_to_jump = False
else:
# get global activity
global_activity = self.get_merged_activity(
date, merge_actions=self.merge_actions)
# if going to jump, erases peak in neighbour event
if self.waiting_to_jump:
zetas = global_activity.get_dates_list()
states, _ = self.current_streamview.atoms[
"_self"].memorySpace.get_events(zetas)
for i in range(0, len(states)):
if states[i].index == self.improvisation_memory[-1][
0].index + 1:
del global_activity[i]
self.waiting_to_jump = False
if len(global_activity) != 0 and len(
self.improvisation_memory) > 0:
event, transforms = self.decide(global_activity)
if event == None:
# if no event returned, choose default
event, transforms = self.decide_default()
if type(transforms) != list:
transforms = [transforms]
else:
# if activity is empty, choose default
event, transforms = self.decide_default()
for transform in transforms:
event = transform.decode(event)
# add event to improvisation memory
self.improvisation_memory.append((event, transforms))
# influences private streamview if auto-influence activated
if self.self_influence:
self.current_streamview.influence("_self", date, event.get_label())
# sends state num
self.send([
event.index,
event.get_contents().get_zeta(),
event.get_contents().get_state_length()
], "/state")
return event
def decide_default(self):
'''default decision method : selecting conjoint event'''
if len(self.improvisation_memory) != 0:
previousState = self.improvisation_memory[-1][0]
new = self.current_streamview.atoms["_self"].memorySpace[
(previousState.index + 1) %
len(self.current_streamview.atoms["_self"].memorySpace)]
trans = self.improvisation_memory[-1][1]
else:
new = self.current_streamview.atoms["_self"].memorySpace[0]
trans = [Transforms.NoTransform()]
return new[1], trans
def __init__(self, root_dir,shape_data,img_transofrm=None,label_transform=None,nameImageFile="images",nameMasksFile="masks"):
self.shape_data=shape_data
self.root_dir=root_dir
self.img_transofrm = img_transofrm
self.label_transform = label_transform
self.nameImageFile=nameImageFile
self.nameMasksFile=nameMasksFile
self.files = list()
self.files=os.listdir(self.root_dir)
####Augmentation###
self.TrChannels=trs.TransformColorChannels()
self.TrFlip=trs.TransformFlip()
self.colorJitter=trs.TransformsColorJitter(0.65,0.8,0.8,0.09)
self.TrNoisy=trs.TransformNoisyNormal((128,128,128),(128,128,128),0.98,0.02)
self.TrBlur=trs.TransformBlur(3)
###################
print("Found files: ",len(self.files))
def __init__ (self,rootData,shapeData,imgTransforms=None, labelTransforms=None):
self.rootData=rootData
self.shapeData=shapeData
self.imgTransforms=imgTransforms
self.labelTransforms=labelTransforms
self.filesDataset=list()
####Augmentation###
self.TrChannels=trs.TransformColorChannels()
self.TrFlip=trs.TransformFlip()
self.colorJitter=trs.TransformsColorJitter(0.65,0.8,0.8,0.09)
self.TrNoisy=trs.TransformNoisyNormal((128,128,128),(128,128,128),0.9,0.1)
self.TrBlur=trs.TransformBlur(5)
###################
rootFiles1=os.listdir(rootData)
for fileName in rootFiles1:
cropped_train=os.path.join(rootData,fileName,"cropped_train")
instrument_datasets=os.listdir(cropped_train)
for instrument_dataset in instrument_datasets:
imagesPath=os.path.join(cropped_train,instrument_dataset,"images")
masksPath=os.path.join(cropped_train,instrument_dataset,"binary_masks")
images=os.listdir(imagesPath)
masks=os.listdir(masksPath)
lenImgs=len(images)
lenMasks=len(masks)
if lenImgs!=lenMasks:
strError = "error of the dimension of the list of images (" + str(lenIms) + ") and masks (" + str(lenMsks) + ")"
raise IOError(strError)
for i in range(lenImgs):
img_file = os.path.join(imagesPath,images[i])
msk_file = os.path.join(masksPath,masks[i])
self.filesDataset.append({
"image":img_file,
"label":msk_file
})
def open_transform_operations(self):
transform_window = tk.Tk(screenName=None,
baseName=None,
className='Tk',
useTk=1)
transform_window.title('Transform Operations')
transform_window.geometry("400x400")
swirl_button = tk.Button(transform_window,
text='Swirl Operation with custom image',
width=self.width,
command=lambda: transforms.swirled(self.img))
swirl_button.pack()
swirl_with_checker_board_button = tk.Button(
transform_window,
text='Swirl Operation with checker board',
width=self.width,
command=lambda: transforms.swirled_with_checkerboard())
swirl_with_checker_board_button.pack()
rescale_button = tk.Button(
transform_window,
text="Rescale with anti anti aliasing",
width=self.width,
command=lambda: transforms.rescale(self.img))
rescale_button.pack()
resize_button = tk.Button(transform_window,
text="Resize Operation with anti aliasing",
width=self.width,
command=lambda: transforms.resize(self.img))
resize_button.pack()
downscale_button = tk.Button(
transform_window,
text='Downscale Operation',
width=self.width,
command=lambda: transforms.downscale(self.img_gray))
downscale_button.pack()
rotation_button = tk.Button(
transform_window,
text="Rotation Operation",
width=self.width,
command=lambda: transforms.rotation(self.img))
rotation_button.pack()
def __getitem__(self, index):
rgb, depth = self.__getraw__(index)
if self.transform is not None:
rgb_np, depth_np = self.transform(rgb, depth)
else:
raise (RuntimeError("transform not defined"))
# color normalization
# rgb_tensor = normalize_rgb(rgb_tensor)
# rgb_np = normalize_np(rgb_np)
if self.modality == 'rgb':
input_np = rgb_np
to_tensor = T.ToTensor()
input_tensor = to_tensor(input_np)
while input_tensor.dim() < 3:
input_tensor = input_tensor.unsqueeze(0)
depth_tensor = to_tensor(depth_np)
depth_tensor = depth_tensor.unsqueeze(0)
return input_tensor, depth_tensor
class CustomDataLoader(data.Dataset):
modality_names = ['rgb']
def isImageFile(self, filename):
IMG_EXTENSIONS = ['.h5']
return any(
filename.endswith(extension) for extension in IMG_EXTENSIONS)
def findClasses(self, dir):
classes = [
d for d in os.listdir(dir) if os.path.isdir(os.path.join(dir, d))
]
classes.sort()
class_to_idx = {classes[i]: i for i in range(len(classes))}
return classes, class_to_idx
def makeDataset(self, dir, class_to_idx):
images = []
print(dir)
dir = os.path.expanduser(dir)
for target in sorted(os.listdir(dir)):
d = os.path.join(dir, target)
if not os.path.isdir(d):
continue
for root, _, fnames in sorted(os.walk(d)):
for fname in sorted(fnames):
if self.isImageFile(fname):
path = os.path.join(root, fname)
item = (path, class_to_idx[target])
images.append(item)
return images
color_jitter = T.ColorJitter(0.4, 0.4, 0.4)
def __init__(self, root, split, modality='rgb', loader=h5Loader):
classes, class_to_idx = self.findClasses(root)
imgs = self.makeDataset(root, class_to_idx)
assert len(imgs) > 0, "Found 0 images in subfolders of: " + root + "\n"
# print("Found {} images in {} folder.".format(len(imgs), split))
self.root = root
self.imgs = imgs
self.classes = classes
self.class_to_idx = class_to_idx
if split == 'train':
self.transform = self.trainTransform
elif split == 'holdout':
self.transform = self.validationTransform
elif split == 'val':
self.transform = self.validationTransform
else:
raise (RuntimeError("Invalid dataset split: " + split + "\n"
"Supported dataset splits are: train, val"))
self.loader = loader
assert (modality in self.modality_names), "Invalid modality split: " + modality + "\n" + \
"Supported dataset splits are: " + ''.join(self.modality_names)
self.modality = modality
# def trainTransform(self, rgb, depth):
# raise (RuntimeError("train_transform() is not implemented. "))
#
# def validationTransform(rgb, depth):
# raise (RuntimeError("val_transform() is not implemented."))
def __getraw__(self, index):
"""
Args:
index (int): Index
Returns:
tuple: (rgb, depth) the raw data.
"""
path, target = self.imgs[index]
rgb, depth = self.loader(path)
return rgb, depth
def __getitem__(self, index):
rgb, depth = self.__getraw__(index)
if self.transform is not None:
rgb_np, depth_np = self.transform(rgb, depth)
else:
raise (RuntimeError("transform not defined"))
# color normalization
# rgb_tensor = normalize_rgb(rgb_tensor)
# rgb_np = normalize_np(rgb_np)
if self.modality == 'rgb':
input_np = rgb_np
to_tensor = T.ToTensor()
input_tensor = to_tensor(input_np)
while input_tensor.dim() < 3:
input_tensor = input_tensor.unsqueeze(0)
depth_tensor = to_tensor(depth_np)
depth_tensor = depth_tensor.unsqueeze(0)
return input_tensor, depth_tensor
def __len__(self):
return len(self.imgs)
def trainValidateSegmentation(args):
print('Data file: ' + str(args.cached_data_file))
print(args)
# check if processed data file exists or not
if not os.path.isfile(args.cached_data_file):
dataLoader = ld.LoadData(args.data_dir, args.data_dir_val,
args.classes, args.cached_data_file)
data = dataLoader.processData()
if data is None:
print('Error while pickling data. Please check.')
exit(-1)
else:
data = pickle.load(open(args.cached_data_file, "rb"))
print('=> Loading the model')
model = net.ESPNet(classes=args.classes, channels=args.channels)
args.savedir = args.savedir + os.sep
if args.onGPU:
model = model.cuda()
# create the directory if not exist
if not os.path.exists(args.savedir):
os.mkdir(args.savedir)
if args.onGPU:
model = model.cuda()
if args.visualizeNet:
import VisualizeGraph as viz
x = Variable(
torch.randn(1, args.channels, args.inDepth, args.inWidth,
args.inHeight))
if args.onGPU:
x = x.cuda()
y = model(x, (128, 128, 128)) #, _, _
g = viz.make_dot(y)
g.render(args.savedir + os.sep + 'model', view=False)
total_paramters = 0
for parameter in model.parameters():
i = len(parameter.size())
p = 1
for j in range(i):
p *= parameter.size(j)
total_paramters += p
print('Parameters: ' + str(total_paramters))
# define optimization criteria
weight = torch.from_numpy(
data['classWeights']) # convert the numpy array to torch <- Sachin
print('Class Imbalance Weights')
print(weight)
criteria = torch.nn.CrossEntropyLoss(weight)
if args.onGPU:
criteria = criteria.cuda()
# We train at three different resolutions (144x144x144, 96x96x96 and 128x128x128)
# and validate at one resolution (128x128x128)
trainDatasetA = myTransforms.Compose([
myTransforms.MinMaxNormalize(),
myTransforms.ScaleToFixed(dimA=144, dimB=144, dimC=144),
myTransforms.RandomFlip(),
myTransforms.ToTensor(args.scaleIn),
])
trainDatasetB = myTransforms.Compose([
myTransforms.MinMaxNormalize(),
myTransforms.ScaleToFixed(dimA=96, dimB=96, dimC=96),
myTransforms.RandomFlip(),
myTransforms.ToTensor(args.scaleIn),
])
trainDatasetC = myTransforms.Compose([
myTransforms.MinMaxNormalize(),
myTransforms.ScaleToFixed(dimA=args.inWidth,
dimB=args.inHeight,
dimC=args.inDepth),
myTransforms.RandomFlip(),
myTransforms.ToTensor(args.scaleIn),
])
valDataset = myTransforms.Compose([
myTransforms.MinMaxNormalize(),
myTransforms.ScaleToFixed(dimA=args.inWidth,
dimB=args.inHeight,
dimC=args.inDepth),
myTransforms.ToTensor(args.scaleIn),
#
])
trainLoaderA = torch.utils.data.DataLoader(
myDataLoader.MyDataset(data['trainIm'],
data['trainAnnot'],
transform=trainDatasetA),
batch_size=args.batch_size,
shuffle=True,
num_workers=args.num_workers,
pin_memory=False) #disabling pin memory because swap usage is high
trainLoaderB = torch.utils.data.DataLoader(myDataLoader.MyDataset(
data['trainIm'], data['trainAnnot'], transform=trainDatasetB),
batch_size=args.batch_size,
shuffle=True,
num_workers=args.num_workers,
pin_memory=False)
trainLoaderC = torch.utils.data.DataLoader(myDataLoader.MyDataset(
data['trainIm'], data['trainAnnot'], transform=trainDatasetC),
batch_size=args.batch_size,
shuffle=True,
num_workers=args.num_workers,
pin_memory=False)
valLoader = torch.utils.data.DataLoader(myDataLoader.MyDataset(
data['valIm'], data['valAnnot'], transform=valDataset),
batch_size=1,
shuffle=False,
num_workers=args.num_workers,
pin_memory=False)
# define the optimizer
optimizer = torch.optim.Adam(filter(lambda p: p.requires_grad,
model.parameters()),
args.lr, (0.9, 0.999),
eps=1e-08,
weight_decay=2e-4)
if args.onGPU == True:
cudnn.benchmark = True
start_epoch = 0
stored_loss = 100000000.0
if args.resume:
if os.path.isfile(args.resumeLoc):
print("=> loading checkpoint '{}'".format(args.resumeLoc))
checkpoint = torch.load(args.resumeLoc)
start_epoch = checkpoint['epoch']
stored_loss = checkpoint['stored_loss']
model.load_state_dict(checkpoint['state_dict'])
optimizer.load_state_dict(checkpoint['optimizer'])
print("=> loaded checkpoint '{}' (epoch {})".format(
args.resume, checkpoint['epoch']))
else:
print("=> no checkpoint found at '{}'".format(args.resume))
logFileLoc = args.savedir + args.logFile
if os.path.isfile(logFileLoc):
logger = open(logFileLoc, 'a')
logger.write("Parameters: %s" % (str(total_paramters)))
logger.write(
"\n%s\t%s\t%s\t%s\t%s\t" %
('Epoch', 'Loss(Tr)', 'Loss(val)', 'mIOU (tr)', 'mIOU (val'))
logger.flush()
else:
logger = open(logFileLoc, 'w')
logger.write("Arguments: %s" % (str(args)))
logger.write("\n Parameters: %s" % (str(total_paramters)))
logger.write(
"\n%s\t%s\t%s\t%s\t%s\t" %
('Epoch', 'Loss(Tr)', 'Loss(val)', 'mIOU (tr)', 'mIOU (val'))
logger.flush()
# reduce the learning rate by 0.5 after every 100 epochs
scheduler = torch.optim.lr_scheduler.StepLR(optimizer,
step_size=args.step_loss,
gamma=0.5) #40
best_val_acc = 0
loader_idxs = [
0, 1, 2
] # Three loaders at different resolutions are mapped to three indexes
for epoch in range(start_epoch, args.max_epochs):
# step the learning rate
scheduler.step(epoch)
lr = 0
for param_group in optimizer.param_groups:
lr = param_group['lr']
print('Running epoch {} with learning rate {:.5f}'.format(epoch, lr))
if epoch > 0:
# shuffle the loaders
np.random.shuffle(loader_idxs)
for l_id in loader_idxs:
if l_id == 0:
train(args, trainLoaderA, model, criteria, optimizer, epoch)
elif l_id == 1:
train(args, trainLoaderB, model, criteria, optimizer, epoch)
else:
lossTr, overall_acc_tr, per_class_acc_tr, per_class_iu_tr, mIOU_tr = \
train(args, trainLoaderC, model, criteria, optimizer, epoch)
# evaluate on validation set
lossVal, overall_acc_val, per_class_acc_val, per_class_iu_val, mIOU_val = val(
args, valLoader, model, criteria)
print('saving checkpoint') ## added
save_checkpoint(
{
'epoch': epoch + 1,
'arch': str(model),
'state_dict': model.state_dict(),
'optimizer': optimizer.state_dict(),
'lossTr': lossTr,
'lossVal': lossVal,
'iouTr': mIOU_tr,
'iouVal': mIOU_val,
'stored_loss': stored_loss,
}, args.savedir + '/checkpoint.pth.tar')
# save the model also
if mIOU_val >= best_val_acc:
best_val_acc = mIOU_val
torch.save(model.state_dict(), args.savedir + '/best_model.pth')
with open(args.savedir + 'acc_' + str(epoch) + '.txt', 'w') as log:
log.write(
"\nEpoch: %d\t Overall Acc (Tr): %.4f\t Overall Acc (Val): %.4f\t mIOU (Tr): %.4f\t mIOU (Val): %.4f"
% (epoch, overall_acc_tr, overall_acc_val, mIOU_tr, mIOU_val))
log.write('\n')
log.write('Per Class Training Acc: ' + str(per_class_acc_tr))
log.write('\n')
log.write('Per Class Validation Acc: ' + str(per_class_acc_val))
log.write('\n')
log.write('Per Class Training mIOU: ' + str(per_class_iu_tr))
log.write('\n')
log.write('Per Class Validation mIOU: ' + str(per_class_iu_val))
logger.write("\n%d\t\t%.4f\t\t%.4f\t\t%.4f\t\t%.4f\t\t%.6f" %
(epoch, lossTr, lossVal, mIOU_tr, mIOU_val, lr))
logger.flush()
print("Epoch : " + str(epoch) + ' Details')
print(
"\nEpoch No.: %d\tTrain Loss = %.4f\tVal Loss = %.4f\t mIOU(tr) = %.4f\t mIOU(val) = %.4f"
% (epoch, lossTr, lossVal, mIOU_tr, mIOU_val))
logger.close()
def idct(self):
"""
Compute the Inverse Discrete Cosine Transform (IDCT)
"""
return Transforms.idct(self)
def dct(self):
"""
Compute the Discrete Cosine Transform (DCT)
"""
return Transforms.dct(self)
def show_price(self):
# plot prices of self.df
plt.plot(self.df.price)
plt.show()
def temp_add_transforms(self):
# TODO: accept multiple arguments for features to add
# this function is currently only for debugging
# create features/transforms
self.df['mmi'] = Transforms.mmi(self.df, 500)
self.df['sma'] = self.df.price.rolling(window=5, center=False).mean()
if __name__ == "__main__":
month_list = ['df201609']
dh = DataHandler()
for month in month_list:
dh.concat_month(month)
dh.temp_add_transforms()
dh.df["rsi"] = Transforms.rsi(dh.df.price,120)
dh.df["expiry"] = Transforms.expiry(dh.df, 300)
dh.temp_add_transforms()
ax1 = plt.subplot(211)
ax1.plot(dh.df.index, dh.df.price)
ax2 = plt.subplot(212, sharex=ax1)
ax2.plot(dh.df.index, dh.df.rsi)
plt.show()
def idct(self):
"""
Compute the Inverse Discrete Cosine Transform (IDCT)
"""
return Transforms.idct(self)
def trainValidateSegmentation(args):
'''
Main function for trainign and validation
:param args: global arguments
:return: None
'''
# check if processed data file exists or not
if not os.path.isfile(args.cached_data_file):
dataLoad = ld.LoadData(args.data_dir, args.classes, args.cached_data_file)
data = dataLoad.processData()
if data is None:
print('Error while pickling data. Please check.')
exit(-1)
else:
data = pickle.load(open(args.cached_data_file, "rb"))
q = args.q
p = args.p
# load the model
if not args.decoder:
model = net.ESPNet_Encoder(args.classes, p=p, q=q)
args.savedir = args.savedir + '_enc_' + str(p) + '_' + str(q) + '/'
else:
model = net.ESPNet(args.classes, p=p, q=q, encoderFile=args.pretrained)
args.savedir = args.savedir + '_dec_' + str(p) + '_' + str(q) + '/'
if args.onGPU:
model = model.cuda()
# create the directory if not exist
if not os.path.exists(args.savedir):
os.mkdir(args.savedir)
if args.visualizeNet:
x = Variable(torch.randn(1, 3, args.inWidth, args.inHeight))
if args.onGPU:
x = x.cuda()
y = model.forward(x)
g = viz.make_dot(y)
g.render(args.savedir + 'model.png', view=False)
total_paramters = netParams(model)
print('Total network parameters: ' + str(total_paramters))
# define optimization criteria
weight = torch.from_numpy(data['classWeights']) # convert the numpy array to torch
if args.onGPU:
weight = weight.cuda()
criteria = CrossEntropyLoss2d(weight) #weight
if args.onGPU:
criteria = criteria.cuda()
print('Data statistics')
print(data['mean'], data['std'])
print(data['classWeights'])
#compose the data with transforms
trainDataset_main = myTransforms.Compose([
myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.Scale(1024, 512),
myTransforms.RandomCropResize(32),
myTransforms.RandomFlip(),
#myTransforms.RandomCrop(64).
myTransforms.ToTensor(args.scaleIn),
#
])
trainDataset_scale1 = myTransforms.Compose([
myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.Scale(1536, 768), # 1536, 768
myTransforms.RandomCropResize(100),
myTransforms.RandomFlip(),
#myTransforms.RandomCrop(64),
myTransforms.ToTensor(args.scaleIn),
#
])
trainDataset_scale2 = myTransforms.Compose([
myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.Scale(1280, 720), # 1536, 768
myTransforms.RandomCropResize(100),
myTransforms.RandomFlip(),
#myTransforms.RandomCrop(64),
myTransforms.ToTensor(args.scaleIn),
#
])
trainDataset_scale3 = myTransforms.Compose([
myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.Scale(768, 384),
myTransforms.RandomCropResize(32),
myTransforms.RandomFlip(),
#myTransforms.RandomCrop(64),
myTransforms.ToTensor(args.scaleIn),
#
])
trainDataset_scale4 = myTransforms.Compose([
myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.Scale(512, 256),
#myTransforms.RandomCropResize(20),
myTransforms.RandomFlip(),
#myTransforms.RandomCrop(64).
myTransforms.ToTensor(args.scaleIn),
#
])
valDataset = myTransforms.Compose([
myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.Scale(1024, 512),
myTransforms.ToTensor(args.scaleIn),
#
])
# since we training from scratch, we create data loaders at different scales
# so that we can generate more augmented data and prevent the network from overfitting
trainLoader = torch.utils.data.DataLoader(
myDataLoader.MyDataset(data['trainIm'], data['trainAnnot'], transform=trainDataset_main),
batch_size=args.batch_size + 2, shuffle=True, num_workers=args.num_workers, pin_memory=True)
trainLoader_scale1 = torch.utils.data.DataLoader(
myDataLoader.MyDataset(data['trainIm'], data['trainAnnot'], transform=trainDataset_scale1),
batch_size=args.batch_size, shuffle=True, num_workers=args.num_workers, pin_memory=True)
trainLoader_scale2 = torch.utils.data.DataLoader(
myDataLoader.MyDataset(data['trainIm'], data['trainAnnot'], transform=trainDataset_scale2),
batch_size=args.batch_size, shuffle=True, num_workers=args.num_workers, pin_memory=True)
trainLoader_scale3 = torch.utils.data.DataLoader(
myDataLoader.MyDataset(data['trainIm'], data['trainAnnot'], transform=trainDataset_scale3),
batch_size=args.batch_size + 4, shuffle=True, num_workers=args.num_workers, pin_memory=True)
trainLoader_scale4 = torch.utils.data.DataLoader(
myDataLoader.MyDataset(data['trainIm'], data['trainAnnot'], transform=trainDataset_scale4),
batch_size=args.batch_size + 4, shuffle=True, num_workers=args.num_workers, pin_memory=True)
valLoader = torch.utils.data.DataLoader(
myDataLoader.MyDataset(data['valIm'], data['valAnnot'], transform=valDataset),
batch_size=args.batch_size + 4, shuffle=False, num_workers=args.num_workers, pin_memory=True)
if args.onGPU:
cudnn.benchmark = True
start_epoch = 0
if args.resume:
if os.path.isfile(args.resumeLoc):
print("=> loading checkpoint '{}'".format(args.resume))
checkpoint = torch.load(args.resumeLoc)
start_epoch = checkpoint['epoch']
#args.lr = checkpoint['lr']
model.load_state_dict(checkpoint['state_dict'])
print("=> loaded checkpoint '{}' (epoch {})"
.format(args.resume, checkpoint['epoch']))
else:
print("=> no checkpoint found at '{}'".format(args.resume))
logFileLoc = args.savedir + args.logFile
if os.path.isfile(logFileLoc):
logger = open(logFileLoc, 'a')
else:
logger = open(logFileLoc, 'w')
logger.write("Parameters: %s" % (str(total_paramters)))
logger.write("\n%s\t%s\t%s\t%s\t%s\t" % ('Epoch', 'Loss(Tr)', 'Loss(val)', 'mIOU (tr)', 'mIOU (val'))
logger.flush()
optimizer = torch.optim.Adam(model.parameters(), args.lr, (0.9, 0.999), eps=1e-08, weight_decay=5e-4)
# we step the loss by 2 after step size is reached
scheduler = torch.optim.lr_scheduler.StepLR(optimizer, step_size=args.step_loss, gamma=0.5)
for epoch in range(start_epoch, args.max_epochs):
scheduler.step(epoch)
lr = 0
for param_group in optimizer.param_groups:
lr = param_group['lr']
print("Learning rate: " + str(lr))
# train for one epoch
# We consider 1 epoch with all the training data (at different scales)
train(args, trainLoader_scale1, model, criteria, optimizer, epoch)
train(args, trainLoader_scale2, model, criteria, optimizer, epoch)
train(args, trainLoader_scale4, model, criteria, optimizer, epoch)
train(args, trainLoader_scale3, model, criteria, optimizer, epoch)
lossTr, overall_acc_tr, per_class_acc_tr, per_class_iu_tr, mIOU_tr = train(args, trainLoader, model, criteria, optimizer, epoch)
# evaluate on validation set
lossVal, overall_acc_val, per_class_acc_val, per_class_iu_val, mIOU_val = val(args, valLoader, model, criteria)
save_checkpoint({
'epoch': epoch + 1,
'arch': str(model),
'state_dict': model.state_dict(),
'optimizer': optimizer.state_dict(),
'lossTr': lossTr,
'lossVal': lossVal,
'iouTr': mIOU_tr,
'iouVal': mIOU_val,
'lr': lr
}, args.savedir + 'checkpoint.pth.tar')
#save the model also
model_file_name = args.savedir + '/model_' + str(epoch + 1) + '.pth'
torch.save(model.state_dict(), model_file_name)
with open(args.savedir + 'acc_' + str(epoch) + '.txt', 'w') as log:
log.write("\nEpoch: %d\t Overall Acc (Tr): %.4f\t Overall Acc (Val): %.4f\t mIOU (Tr): %.4f\t mIOU (Val): %.4f" % (epoch, overall_acc_tr, overall_acc_val, mIOU_tr, mIOU_val))
log.write('\n')
log.write('Per Class Training Acc: ' + str(per_class_acc_tr))
log.write('\n')
log.write('Per Class Validation Acc: ' + str(per_class_acc_val))
log.write('\n')
log.write('Per Class Training mIOU: ' + str(per_class_iu_tr))
log.write('\n')
log.write('Per Class Validation mIOU: ' + str(per_class_iu_val))
logger.write("\n%d\t\t%.4f\t\t%.4f\t\t%.4f\t\t%.4f\t\t%.7f" % (epoch, lossTr, lossVal, mIOU_tr, mIOU_val, lr))
logger.flush()
print("Epoch : " + str(epoch) + ' Details')
print("\nEpoch No.: %d\tTrain Loss = %.4f\tVal Loss = %.4f\t mIOU(tr) = %.4f\t mIOU(val) = %.4f" % (epoch, lossTr, lossVal, mIOU_tr, mIOU_val))
logger.close()
"F_beta": [],
"MAE": []
}
}
bests = {"F_beta_tr": 0., "F_beta_val": 0., "MAE_tr": 1., "MAE_val": 1.}
logger.info('Data statistics:')
logger.info("mean: [%.5f, %.5f, %.5f], std: [%.5f, %.5f, %.5f]" %
(*data['mean'], *data['std']))
# compose the data with transforms
trainTransform = myTransforms.Compose([
# myTransforms.ColorJitter(brightness=0.2, contrast=0.2, saturation=0.2, hue=0.1),
myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.Scale(args.inWidth, args.inHeight),
# myTransforms.RandomCropResize(int(7./224.*args.inWidth)),
# myTransforms.RandomFlip(),
myTransforms.ToTensor()
])
valTransform = myTransforms.Compose([
myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.Scale(args.inWidth, args.inHeight),
myTransforms.ToTensor()
])
train_set = myDataLoader.Dataset(data['trainIm'],
data['trainDepth'],
data['trainAnnot'],
transform=trainTransform)
val_set = myDataLoader.Dataset(data['valIm'],
data['valDepth'],
def ifft(self):
"""
Compute the Inverse FFT
"""
return Transforms.ifft(self)
def trainValidateSegmentation(args):
# check if processed data file exists or not
if not os.path.isfile(args.cached_data_file):
dataLoader = ld.LoadData(args.data_dir, args.classes,
args.cached_data_file)
if dataLoader is None:
print('Error while processing the data. Please check')
exit(-1)
data = dataLoader.processData()
else:
data = pickle.load(open(args.cached_data_file, "rb"))
if args.modelType == 'C1':
model = net.ResNetC1(args.classes)
elif args.modelType == 'D1':
model = net.ResNetD1(args.classes)
else:
print('Please select the correct model. Exiting!!')
exit(-1)
args.savedir = args.savedir + args.modelType + '/'
if args.onGPU == True:
model = model.cuda()
# create the directory if not exist
if not os.path.exists(args.savedir):
os.mkdir(args.savedir)
if args.onGPU == True:
model = model.cuda()
if args.visualizeNet == True:
x = Variable(torch.randn(1, 3, args.inWidth, args.inHeight))
if args.onGPU == True:
x = x.cuda()
y = model.forward(x)
g = viz.make_dot(y)
g.render(args.savedir + '/model.png', view=False)
n_param = sum([np.prod(param.size()) for param in model.parameters()])
print('Network parameters: ' + str(n_param))
# define optimization criteria
print('Weights to handle class-imbalance')
weight = torch.from_numpy(
data['classWeights']) # convert the numpy array to torch
print(weight)
if args.onGPU == True:
weight = weight.cuda()
criteria = CrossEntropyLoss2d(weight) # weight
if args.onGPU == True:
criteria = criteria.cuda()
trainDatasetNoZoom = myTransforms.Compose([
# myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.RandomCropResize(20),
myTransforms.RandomHorizontalFlip(),
myTransforms.ToTensor(args.scaleIn)
])
trainDatasetWithZoom = myTransforms.Compose([
# myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.Zoom(512, 512),
myTransforms.RandomCropResize(20),
myTransforms.RandomHorizontalFlip(),
myTransforms.ToTensor(args.scaleIn)
])
valDataset = myTransforms.Compose([
# myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.ToTensor(args.scaleIn)
])
trainLoaderNoZoom = torch.utils.data.DataLoader(
myDataLoader.MyDataset(data['trainIm'],
data['trainAnnot'],
transform=trainDatasetNoZoom),
batch_size=args.batch_size,
shuffle=True,
num_workers=args.num_workers,
pin_memory=True)
trainLoaderWithZoom = torch.utils.data.DataLoader(
myDataLoader.MyDataset(data['trainIm'],
data['trainAnnot'],
transform=trainDatasetWithZoom),
batch_size=args.batch_size,
shuffle=True,
num_workers=args.num_workers,
pin_memory=True)
valLoader = torch.utils.data.DataLoader(myDataLoader.MyDataset(
data['valIm'], data['valAnnot'], transform=valDataset),
batch_size=args.batch_size,
shuffle=False,
num_workers=args.num_workers,
pin_memory=True)
# define the optimizer
# optimizer = torch.optim.Adam(model.parameters(), args.lr, (0.9, 0.999), eps=1e-08, weight_decay=2e-4)
optimizer = torch.optim.SGD(model.parameters(),
lr=args.lr,
momentum=0.9,
weight_decay=5e-4)
if args.onGPU == True:
cudnn.benchmark = True
start_epoch = 0
if args.resume:
if os.path.isfile(args.resumeLoc):
print("=> loading checkpoint '{}'".format(args.resumeLoc))
checkpoint = torch.load(args.resumeLoc)
start_epoch = checkpoint['epoch']
model.load_state_dict(checkpoint['state_dict'])
print("=> loaded checkpoint '{}' (epoch {})".format(
args.resume, checkpoint['epoch']))
else:
print("=> no checkpoint found at '{}'".format(args.resume))
logFileLoc = args.savedir + os.sep + args.logFile
if os.path.isfile(logFileLoc):
logger = open(logFileLoc, 'a')
logger.write("Parameters: %s" % (str(total_paramters)))
logger.write(
"\n%s\t%s\t%s\t%s\t%s\t" %
('Epoch', 'Loss(Tr)', 'Loss(val)', 'mIOU (tr)', 'mIOU (val'))
logger.flush()
else:
logger = open(logFileLoc, 'w')
logger.write("Parameters: %s" % (str(total_paramters)))
logger.write(
"\n%s\t%s\t%s\t%s\t%s\t" %
('Epoch', 'Loss(Tr)', 'Loss(val)', 'mIOU (tr)', 'mIOU (val'))
logger.flush()
#lr scheduler
scheduler = torch.optim.lr_scheduler.StepLR(optimizer,
step_size=args.step_loss,
gamma=0.1)
for epoch in range(start_epoch, args.max_epochs):
scheduler.step(epoch)
lr = 0
for param_group in optimizer.param_groups:
lr = param_group['lr']
# run at zoomed images first
train(args, trainLoaderWithZoom, model, criteria, optimizer, epoch)
lossTr, overall_acc_tr, per_class_acc_tr, per_class_iu_tr, mIOU_tr = train(
args, trainLoaderNoZoom, model, criteria, optimizer, epoch)
# evaluate on validation set
lossVal, overall_acc_val, per_class_acc_val, per_class_iu_val, mIOU_val = val(
args, valLoader, model, criteria)
save_checkpoint(
{
'epoch': epoch + 1,
'arch': str(model),
'state_dict': model.state_dict(),
'optimizer': optimizer.state_dict(),
'lossTr': lossTr,
'lossVal': lossVal,
'iouTr': mIOU_tr,
'iouVal': mIOU_val,
}, args.savedir + '/checkpoint.pth.tar')
# save the model also
model_file_name = args.savedir + '/model_' + str(epoch + 1) + '.pth'
torch.save(model.state_dict(), model_file_name)
with open(args.savedir + 'acc_' + str(epoch) + '.txt', 'w') as log:
log.write(
"\nEpoch: %d\t Overall Acc (Tr): %.4f\t Overall Acc (Val): %.4f\t mIOU (Tr): %.4f\t mIOU (Val): %.4f"
% (epoch, overall_acc_tr, overall_acc_val, mIOU_tr, mIOU_val))
log.write('\n')
log.write('Per Class Training Acc: ' + str(per_class_acc_tr))
log.write('\n')
log.write('Per Class Validation Acc: ' + str(per_class_acc_val))
log.write('\n')
log.write('Per Class Training mIOU: ' + str(per_class_iu_tr))
log.write('\n')
log.write('Per Class Validation mIOU: ' + str(per_class_iu_val))
logger.write("\n%d\t\t%.4f\t\t%.4f\t\t%.4f\t\t%.4f\t\t%.4f" %
(epoch, lossTr, lossVal, mIOU_tr, mIOU_val, lr))
logger.flush()
print("Epoch : " + str(epoch) + ' Details')
print(
"\nEpoch No.: %d\tTrain Loss = %.4f\tVal Loss = %.4f\t mIOU(tr) = %.4f\t mIOU(val) = %.4f"
% (epoch, lossTr, lossVal, mIOU_tr, mIOU_val))
logger.close()
dtype=np.float32)
# load the model
model = BiSalNet()
model.eval()
if args.onGPU and torch.cuda.device_count() > 1:
# model = torch.nn.DataParallel(model)
model = DataParallelModel(model)
if args.onGPU:
model = model.cuda()
# compose the data with transforms
valDataset = myTransforms.Compose([
myTransforms.Normalize(mean=data['mean'], std=data['std']),
myTransforms.Scale(args.inWidth, args.inHeight),
myTransforms.ToTensor()
])
# since we training from scratch, we create data loaders at different scales
# so that we can generate more augmented data and prevent the network from overfitting
valLoader = torch.utils.data.DataLoader(myDataLoader.Dataset(
data['valIm'], data['valAnnot'], transform=valDataset),
batch_size=args.batch_size,
shuffle=False,
num_workers=args.num_workers,
pin_memory=args.onGPU)
if os.path.isfile(args.resume):
print("=> loading checkpoint '{}'".format(args.resume))
model.load_state_dict(torch.load(args.resume)["state_dict"])
else:
def trainValSegmentation(args):
if not os.path.isfile(args.cached_data_file):
dataLoader = ld.LoadData(args.data_dir, args.classes, args.attrClasses,
args.cached_data_file)
if dataLoader is None:
print("Error while cacheing the data.")
exit(-1)
data = dataLoader.processData()
else:
print("load cacheing data.")
data = pickle.load(open(args.cached_data_file, 'rb'))
# only unet for segmentation now.
# model= unet.UNet(args.classes)
# model = r18unet.ResNetUNet(args.classes)
model = mobileunet.MobileUNet(args.classes)
print("UNet done...")
# if args.onGPU == True:
model = model.cuda()
# devices_ids=[2,3], device_ids=range(2)
# device = torch.device('cuda:' + str(devices_ids[0]))
# model = model.to(device)
if args.visNet == True:
x = Variable(torch.randn(1, 3, args.inwidth, args.inheight))
if args.onGPU == True:
x = x.cuda()
print("before forward...")
y = model.forward(x)
print("after forward...")
g = viz.make_dot(y)
# g1 = viz.make_dot(y1)
g.render(args.save_dir + '/model', view=False)
model = torch.nn.DataParallel(model)
n_param = sum([np.prod(param.size()) for param in model.parameters()])
print('network parameters: ' + str(n_param))
#define optimization criteria
weight = torch.from_numpy(data['classWeights'])
print(weight)
if args.onGPU == True:
weight = weight.cuda()
criteria = CrossEntropyLoss2d(weight)
# if args.onGPU == True:
# criteria = criteria.cuda()
trainDatasetNoZoom = myTransforms.Compose([
myTransforms.RandomCropResize(args.inwidth, args.inheight),
# myTransforms.RandomHorizontalFlip(),
myTransforms.ToTensor(args.scaleIn)
])
trainDatasetWithZoom = myTransforms.Compose([
# myTransforms.Zoom(512,512),
myTransforms.RandomCropResize(args.inwidth, args.inheight),
myTransforms.RandomHorizontalFlip(),
myTransforms.ToTensor(args.scaleIn)
])
valDataset = myTransforms.Compose([
myTransforms.RandomCropResize(args.inwidth, args.inheight),
myTransforms.ToTensor(args.scaleIn)
])
trainLoaderNoZoom = torch.utils.data.DataLoader(
ld.MyDataset(data['trainIm'],
data['trainAnnot'],
transform=trainDatasetNoZoom),
batch_size=args.batch_size,
shuffle=True,
num_workers=args.num_workers,
pin_memory=True)
trainLoaderWithZoom = torch.utils.data.DataLoader(
ld.MyDataset(data['trainIm'],
data['trainAnnot'],
transform=trainDatasetWithZoom),
batch_size=args.batch_size,
shuffle=True,
num_workers=args.num_workers,
pin_memory=True)
valLoader = torch.utils.data.DataLoader(ld.MyDataset(data['valIm'],
data['valAnnot'],
transform=valDataset),
batch_size=args.batch_size_val,
shuffle=True,
num_workers=args.num_workers,
pin_memory=True)
#define the optimizer
optimizer = torch.optim.Adam(model.parameters(),
args.lr, (0.9, 0.999),
eps=1e-08,
weight_decay=2e-4)
# optimizer = torch.optim.SGD(model.parameters(), lr=args.lr, momentum=0.99, weight_decay=5e-4)
# optimizer = torch.optim.SGD([
# {'params': [param for name, param in model.named_parameters() if name[-4:] == 'bias'],
# 'lr': 2 * args.lr},
# {'params': [param for name, param in model.named_parameters() if name[-4:] != 'bias'],
# 'lr': args.lr, 'weight_decay': 5e-4}
# ], momentum=0.99)
if args.onGPU == True:
cudnn.benchmark = True
start_epoch = 0
if args.resume:
if os.path.isfile(args.resumeLoc):
print("=> loading checkpoint '{}'".format(args.resumeLoc))
checkpoint = torch.load(args.resumeLoc)
start_epoch = checkpoint['epoch']
model.load_state_dict(checkpoint['state_dict'])
print("=> loaded checkpoint '{}' (epoch{})".format(
args.resume, checkpoint['epoch']))
else:
print("=> no checkpoint found at '{}'".format(args.resumeLoc))
logfileLoc = args.save_dir + os.sep + args.logFile
print(logfileLoc)
if os.path.isfile(logfileLoc):
logger = open(logfileLoc, 'a')
logger.write("parameters: %s" % (str(n_param)))
logger.write("\n%s\t%s\t%s\t%s\t%s\t%s\t%s\t" %
('Epoch', 'Loss(Tr)', 'Loss(val)', 'Overall acc(Tr)',
'Overall acc(val)', 'mIOU (tr)', 'mIOU (val'))
logger.flush()
else:
logger = open(logfileLoc, 'w')
logger.write("Parameters: %s" % (str(n_param)))
logger.write("\n%s\t%s\t%s\t%s\t%s\t%s\t%s\t" %
('Epoch', 'Loss(Tr)', 'Loss(val)', 'Overall acc(Tr)',
'Overall acc(val)', 'mIOU (tr)', 'mIOU (val'))
logger.flush()
#lr scheduler
scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer,
milestones=[30, 60, 90],
gamma=0.1)
best_model_acc = 0
for epoch in range(start_epoch, args.max_epochs):
scheduler.step(epoch)
lr = 0
for param_group in optimizer.param_groups:
lr = param_group['lr']
# train(args,trainLoaderWithZoom,model,criteria,optimizer,epoch)
lossTr, overall_acc_tr, per_class_acc_tr, per_class_iu_tr, mIOU_tr = train(
args, trainLoaderNoZoom, model, criteria, optimizer, epoch)
# print(per_class_acc_tr,per_class_iu_tr)
lossVal, overall_acc_val, per_class_acc_val, per_class_iu_val, mIOU_val = val(
args, valLoader, model, criteria)
#save_checkpoint
torch.save(
{
'epoch': epoch + 1,
'arch': str(model),
'state_dict': model.state_dict(),
'optimizer': optimizer.state_dict(),
'lossTr': lossTr,
'lossVal': lossVal,
'iouTr': mIOU_tr,
'iouVal': mIOU_val,
}, args.save_dir + '/checkpoint.pth.tar')
#save model also
# if overall_acc_val > best_model_acc:
# best_model_acc = overall_acc_val
model_file_name = args.save_dir + '/model_' + str(epoch + 1) + '.pth'
torch.save(model.state_dict(), model_file_name)
with open('../acc/acc_' + str(epoch) + '.txt', 'w') as log:
log.write(
"\nEpoch: %d\t Overall Acc (Tr): %.4f\t Overall Acc (Val): %.4f\t mIOU (Tr): %.4f\t mIOU (Val): %.4f"
% (epoch, overall_acc_tr, overall_acc_val, mIOU_tr, mIOU_val))
log.write('\n')
log.write('Per Class Training Acc: ' + str(per_class_acc_tr))
log.write('\n')
log.write('Per Class Validation Acc: ' + str(per_class_acc_val))
log.write('\n')
log.write('Per Class Training mIOU: ' + str(per_class_iu_tr))
log.write('\n')
log.write('Per Class Validation mIOU: ' + str(per_class_iu_val))
logger.write(
"\n%d\t\t%.4f\t\t%.4f\t\t%.4f\t\t%.4f\t\t%.4f\t\t%.4f\t\t%.6f" %
(epoch, lossTr, lossVal, overall_acc_tr, overall_acc_val, mIOU_tr,
mIOU_val, lr))
logger.flush()
print("Epoch : " + str(epoch) + ' Details')
print(
"\nEpoch No.: %d\tTrain Loss = %.4f\tVal Loss = %.4f\t Train acc = %.4f\t Val acc = %.4f\t mIOU(tr) = %.4f\t mIOU(val) = %.4f"
% (epoch, lossTr, lossVal, overall_acc_tr, overall_acc_val,
mIOU_tr, mIOU_val))
logger.close()
def spectrum(self):
"""
Compute the spectrum using an FFT
Returns an instance of Spectrum
"""
return Transforms.fft(self)
def cqt(self):
"""
Compute the Constant Q Transform (CQT)
"""
return Transforms.cqt(self)
#Create Camera (very basic for now)
cam = Sh.Camera()
#Create Renderer Object: Renderer(Camera,Resolution)
rend = Rd.Renderer(cam, RESOLUTION)
if False:
#Create 4 points: Point(x,y,z,(R,G,B))
p1 = Sh.Point(-2, -2, 4, (0, 255, 255))
p2 = Sh.Point(4, 0.1, 4, (255, 255, 0))
p3 = Sh.Point(0.1, 4, 4, (255, 0, 0))
p4 = Sh.Point(-0.1, -0.1, 0, (100, 100, 100))
#Create a Tetrahedron
TH1 = Sh.Tetra(p1, p2, p3, p4)
#Create a Transformation Set
TS1 = Tf.Transform()
#Add several transformations to the set
TS1.AddTransf(Tf.Rotation(130, 3))
TS1.AddTransf(Tf.Translation(2, 2))
TS1.AddTransf(Tf.Translation(1, 3))
TS1.AddTransf(Tf.Translation(1, 1))
#Create a new Tetrahedron by applying the
#transformation set to the first tetrahedron
TH2 = TH1.ApplyTransf(TS1)
#Add the Tetrahedra to the Renderer
rend.AddObject(TH1)
rend.AddObject(TH2)
else:
#Load a mesh
TH1 = Sh.Mesh('boltmesh', (200, 20, 50))
#Add the Tetrahedra to the Renderer
def dct(self):
"""
Compute the Discrete Cosine Transform (DCT)
"""
return Transforms.dct(self)
# The data shows a normal distribution, so we're going to use a # logistic mapping function. # This time, load the entire dataset into memory. word_vecs = h5f['word_vecs'][:] # Calculate the mean using a sample of the vectors. center = np.mean(word_vecs[0:sample_size].flatten()) # Verify that the mean is roughly zero. assert(abs(center - 0.0) < 0.01) # Learn the parameters for the logistic mapping function. # The mapping function will saturate at 8 standard deviations, and will convert # the values to 8-bit integers. tf = Transforms.Transforms(np.uint8) standard_deviations = 8 tf.learn_logistic(word_vecs, num_std=standard_deviations) # Apply logistic mapping to the dataset vectors. word_vecs_int = tf.apply_logistic(word_vecs) # Load the query vectors into memory and map the values. query_vecs = h5f['query_vecs'][:] # Apply logistic mapping to the query vectors. query_vecs_int = tf.apply_logistic(query_vecs) # Take a small subset of the dataset and plot a histogram of the values.
def main():
# Parse command line arguments
parser = argparse.ArgumentParser(
description='Read an mp3 file and plot out its pitch')
parser.add_argument('-i',
dest='input',
type=str,
help='Input file path',
default='')
parser.add_argument('-n',
dest='normalize',
action='store_true',
help='Normalize input values',
default='')
parser.add_argument(
'-t',
dest='transform',
type=str,
help='Use different transforms on the input audio signal',
default='stft')
parser.add_argument(
'-s',
dest='sample',
type=int,
help=
'Sampling rate in Hz to use on the input audio signal while transforming',
default='100')
parser.add_argument(
'-w',
dest='window',
type=float,
help='Sampling window in s to use on the input audio signal for stft',
default='0.05')
options = parser.parse_args()
# Error check
if options.input == '':
print("No input given. BYE!\n")
return 1
elif not os.path.isfile(options.input):
print(f"Given input path {options.input} does not exist!")
return 2
# Read input file into frame rate and data
try:
inSignal = MP3.read(options.input, options.normalize)
except:
print("Reading MP3 failed")
return 3
figures = []
# Plot the data for quick visualization
if options.transform == 'none':
for i in range(0, inSignal.channels):
if i == 0:
figures.append(
bkfigure(plot_width=1200,
plot_height=600,
x_axis_label='Time',
y_axis_label='Amp'))
else:
figures.append(
bkfigure(plot_width=1200,
plot_height=600,
x_axis_label='Time',
y_axis_label='Amp',
x_range=figures[0].x_range,
y_range=figures[0].y_range))
figures[i].line(inSignal.time, inSignal.audioData[:, i])
elif options.transform == 'stft':
# STFT over the signal
fSignal = Transforms.STFT(inSignal,
windowEvery=1 / options.sample,
windowLength=options.window)
for i in range(0, inSignal.channels):
if i == 0:
figures.append(
bkfigure(plot_width=1200,
plot_height=400,
x_axis_label='Time',
y_axis_label='Frequency'))
else:
figures.append(
bkfigure(plot_width=1200,
plot_height=400,
x_axis_label='Time',
y_axis_label='Frequency',
x_range=figures[0].x_range,
y_range=figures[0].y_range))
channelAmp = np.max(fSignal.audioData[:, :, i])
figures[i].image(image=[fSignal.audioData[:, :, i]],
x=0,
y=0,
dw=fSignal.time[-1],
dh=fSignal.dimensionAxes[0][-1],
color_mapper=LinearColorMapper(high=channelAmp,
low=0,
palette=Inferno11))
else:
print("Unrecognized transform given!")
return 4
bkshow(bkcolumn(*figures))
return 0
def ifft(self):
"""
Compute the Inverse FFT
"""
return Transforms.ifft(self)
if my_args.output != None:
plotpath = my_args.output + "/"
# DATA_PATH = "./data/pythia/"
# filename = "reco_simu_Zdd.csv"
#
# BCG_PATH = './data/'
# Filename_background = "reco_0.csv"
h = Hits.Hits(filename)
ev = Hits.Event(h) #, 11)
# h.drawAllEvents()
# Combine events (background, or several events)
# ev.combineEvents([Hits.Event(h_background)])
# ev.combineEvents([Hits.Event(h, 12), Hits.Event(h, 11)])
ev.drawEvent3D(plotName=plotpath + "3D_Zdd.pdf")
# ev.drawEventXY()#plotName=plotpath+"3tracks_XY.pdf")
# ev.drawEventXZ()
# ev.drawEventYZ()
d = ev.data
tr = Transforms.Transforms(ev)
H, xedges, yedges = tr.HoughTransform_phi(numpoints=200,
binx=200,
biny=50,
plotName=plotpath +
"HT_Zdd_maxima.pdf")
tr.plotConformalTransform(plotpath + "CT_Zdd.pdf")
def spectrum(self):
"""
Compute the spectrum using an FFT
Returns an instance of Spectrum
"""
return Transforms.fft(self)
def cqt(self):
"""
Compute the Constant Q Transform (CQT)
"""
return Transforms.cqt(self)
