ayulockin

import cv2
from scipy.spatial import distance as dist
import numpy as np

def resize(image, width=None, height=None, inter=cv2.INTER_AREA):
    dim = None
    (h, w) = image.shape[:2]

    # check to see if the width is None
    if width is None:

ayulockin

ayulockin

## DIVIDE -> mergesort
## CONQUER -> mergesort
## COMBINE -> merge

def merge(S1, S2, S):
    '''
    S1-> First sorted seq
    S2-> Second sorted seq
    S -> Original seq
    '''

ayulockin

def quicksort(S):

    if len(S)<2:
        print("[RETURNED]length of list less than 2")
        return
    
    pivot = S[-1]
    print("[INFO] pivot: ", pivot)
    L = []
    G = []

ayulockin

class LinkedStack:
    def __init__(self):
        self._head = None
        self._size = 0

    class _node:
        __slots__ = '_element', '_nextP'

        def __init__(self, element, nextP):
            self._element = element

ayulockin

net = Net().to(device)

optimizer = optim.Adam(net.parameters())
wandb.init(project='pytorchw_b')
wandb.watch(net, log='all')

for epoch in range(10):
  train(net, device, trainloader, optimizer, epoch)
  test(net, device, testloader, classes)
  

ayulockin

if logwandb:
  wandb.log({'lr': lr_schedule.get_lr()[0], 'loss': loss})

ayulockin

lr_finder = LRFinder(net, optimizer, device)

lr_finder.range_test(trainloader, end_lr=10, num_iter=100, logwandb=True)

ayulockin

class NetforExplode(nn.Module):
  def __init__(self):
    super(NetforExplode, self).__init__()
    
    self.conv1 = nn.Conv2d(1, 32, 3, 1)
    self.conv1.weight.data.fill_(100)
    self.conv1.bias.data.fill_(-100)
    
    self.conv2 = nn.Conv2d(32, 64, 3, 1)
    self.conv2.weight.data.fill_(100)

ayulockin

class Net(nn.Module):
	def __init__(self):
		super(Net, self).__init__()
		
		self.conv1 = nn.Conv2d(3, 32, 3, 1)
	  torch.nn.init.kaiming_uniform_(self.conv1.weight, mode='fan_in', nonlinearity='relu')

	  self.conv2 = nn.Conv2d(32, 32, 3, 1)
		torch.nn.init.kaiming_uniform_(self.conv2.weight, mode='fan_in', nonlinearity='relu')