#hide
from utils import *

path = untar_data(URLs.MNIST)

#hide
Path.BASE_PATH = path

def get_dls(bs=64):
    return DataBlock(
        blocks=(ImageBlock(cls=PILImageBW), CategoryBlock), 
        get_items=get_image_files, 
        splitter=GrandparentSplitter('training','testing'),
        get_y=parent_label,
        batch_tfms=Normalize()
    ).dataloaders(path, bs=bs)

dls = get_dls()

def conv(ni, nf, ks=3, act=True):
    res = nn.Conv2d(ni, nf, stride=2, kernel_size=ks, padding=ks//2)
    if act: res = nn.Sequential(res, nn.ReLU())
    return res

def simple_cnn():
    return sequential(
        conv(1 ,8, ks=5),        #14x14
        conv(8 ,16),             #7x7
        conv(16,32),             #4x4
        conv(32,64),             #2x2
        conv(64,10, act=False),  #1x1
        Flatten(),
    )

from fastai2.callback.hook import *

def fit(epochs=1):
    learn = Learner(dls, simple_cnn(), loss_func=F.cross_entropy,
                    metrics=accuracy, cbs=ActivationStats(with_hist=True))
    learn.fit(epochs, 0.06)
    return learn

dls = get_dls(512)

def fit(epochs=1, lr=0.06):
    learn = Learner(dls, simple_cnn(), loss_func=F.cross_entropy,
                    metrics=accuracy, cbs=ActivationStats(with_hist=True))
    learn.fit_one_cycle(epochs, lr)
    return learn

def conv(ni, nf, ks=3, act=True):
    layers = [nn.Conv2d(ni, nf, stride=2, kernel_size=ks, padding=ks//2)]
    if act: layers.append(nn.ReLU())
    layers.append(nn.BatchNorm2d(nf))
    return nn.Sequential(*layers)