ptrblck

"""
Script to update parameters and the optimizer on the fly.
It is not recommended to use this approach.
This script just shows the disadvantages using this approach.

@author: ptrblck
"""

import torch
import torch.nn as nn

ptrblck

import numpy as np
import pandas as pd

import torch
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F
from torch.utils.data import Dataset
from torch.utils.data.dataloader import DataLoader

ptrblck

from __future__ import print_function
import argparse
import os
import random
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.backends.cudnn as cudnn
import torch.optim as optim
import torch.utils.data

ptrblck

from copy import deepcopy
import torch
import matplotlib
matplotlib.use("agg")
from torch.backends import cudnn
from apex import amp
import argparse
from torch import cuda
from torch import nn
from urllib import request

ptrblck

import torch
import torch.nn as nn
import torch.optim as optim

from apex import amp

class SiameseTower(nn.Module):
    def __init__(self, inplanes=3, planes=32,blocks=3):
        super(SiameseTower, self).__init__()
        

ptrblck

import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torch.utils.data import DataLoader, TensorDataset

from apex import amp

from torchcontrib.optim import SWA

ptrblck

import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

from apex import amp


class MyModel(nn.Module):

ptrblck

import torch
import torch.nn as nn

import pandas as pd

import time

torch.backends.cudnn.deterministic = False
torch.backends.cudnn.benchmark = True

ptrblck

import torch
import torch.nn as nn



def print_types(input, bn, output):
    print('input.type(): {}'.format(input.type()))
    if bn.weight is not None:
        print('bn.weight.type(): {}'.format(bn.weight.type()))
    else:

ptrblck

import torch
import time

torch.backends.cudnn.benchmark = True


# 1a)
I, J, K = 64, 1024, 1024
A = torch.randn(I, J, device='cuda', dtype=torch.half)
B = torch.randn(J, K, device='cuda', dtype=torch.half)