from fastai import *
from fastai.gen_doc.nbdoc import *

jekyll_note("""As usual, this page is generated from a notebook that you can find in the docs_srs folder of the
[fastai repo](https://github.com/fastai/fastai). We use the saved models from [this tutorial](/tutorial.data.html) to
have this notebook run fast.
""")

from fastai import *
from fastai.vision import *

mnist = untar_data(URLs.MNIST_TINY)
tfms = get_transforms(do_flip=False)

data = (ImageItemList.from_folder(mnist)
        .split_by_folder()          
        .label_from_folder()
        .transform(tfms, size=32)
        .databunch()
        .normalize(imagenet_stats)) 

data.export()

empty_data = ImageDataBunch.load_empty(mnist, tfms=tfms[1],size=32).normalize(imagenet_stats)
learn = create_cnn(empty_data, models.resnet18)
learn.load('mini_train');

img = data.train_ds[0][0]
learn.predict(img)

planet = untar_data(URLs.PLANET_TINY)
planet_tfms = get_transforms(flip_vert=True, max_lighting=0.1, max_zoom=1.05, max_warp=0.)

data = (ImageItemList.from_csv(planet, 'labels.csv', folder='train', suffix='.jpg')
        .random_split_by_pct()
        .label_from_df(sep=' ')
        .transform(planet_tfms, size=128)
        .databunch()
        .normalize(imagenet_stats))

data.export()

empty_data = ImageDataBunch.load_empty(planet, tfms=tfms[1],size=32).normalize(imagenet_stats)
learn = create_cnn(empty_data, models.resnet18)
learn.load('mini_train');

img = data.train_ds[0][0]
learn.predict(img)

learn.predict(img, thresh=0.3)

biwi = untar_data(URLs.BIWI_SAMPLE)
fn2ctr = pickle.load(open(biwi/'centers.pkl', 'rb'))

data = (ImageItemList.from_folder(biwi)
        .random_split_by_pct()
        .label_from_func(lambda o:fn2ctr[o.name], label_cls=PointsItemList)
        .transform(get_transforms(), tfm_y=True, size=(120,160))
        .databunch()
        .normalize(imagenet_stats))

data.export()

empty_data = ImageDataBunch.load_empty(biwi, tfms=get_transforms()[1], tfm_y=True, size=(120,60)).normalize(imagenet_stats)
learn = create_cnn(empty_data, models.resnet18)
learn.load('mini_train');

img = data.train_ds[0][0]
learn.predict(img)

img.show(y=learn.predict(img)[0])

camvid = untar_data(URLs.CAMVID_TINY)
path_lbl = camvid/'labels'
path_img = camvid/'images'

codes = np.loadtxt(camvid/'codes.txt', dtype=str)
get_y_fn = lambda x: path_lbl/f'{x.stem}_P{x.suffix}'

data = (SegmentationItemList.from_folder(path_img)
        .random_split_by_pct()
        .label_from_func(get_y_fn, classes=codes)
        .transform(get_transforms(), tfm_y=True, size=128)
        .databunch(bs=16, path=camvid)
        .normalize(imagenet_stats))

data.export()

empty_data = ImageDataBunch.load_empty(camvid, tfms=get_transforms()[1], tfm_y=True, size=128).normalize(imagenet_stats)
learn = Learner.create_unet(empty_data, models.resnet18)
learn.load('mini_train');

img = data.train_ds[0][0]
learn.predict(img)

img.show(y=learn.predict(img)[0])

from fastai import *
from fastai.text import *

imdb = untar_data(URLs.IMDB_SAMPLE)

vocab = Vocab(pickle.load(open(imdb/'tmp'/'itos.pkl', 'rb')))
data_lm = (TextList.from_csv(imdb, 'texts.csv', cols='text', vocab=vocab)
                   .random_split_by_pct()
                   .label_for_lm()
                   .databunch())

data_lm.export()

empty_data = TextLMDataBunch.load_empty(imdb)
learn = language_model_learner(empty_data)
learn.load('mini_train_lm');

learn.predict('This is a simple test of', n_words=20)

data_clas = (TextList.from_csv(imdb, 'texts.csv', cols='text', vocab=vocab)
                   .split_from_df(col='is_valid')
                   .label_from_df(cols='label')
                   .databunch(bs=42))

data_clas.export()

empty_data = TextClasDataBunch.load_empty(imdb)
learn = text_classifier_learner(empty_data)
learn.load('mini_train_clas');

learn.predict('I really loved that movie!')

from fastai import *
from fastai.tabular import *

adult = untar_data(URLs.ADULT_SAMPLE)
df = pd.read_csv(adult/'adult.csv')
dep_var = '>=50k'
cat_names = ['workclass', 'education', 'marital-status', 'occupation', 'relationship', 'race', 'sex', 'native-country']
cont_names = ['education-num', 'hours-per-week', 'age', 'capital-loss', 'fnlwgt', 'capital-gain']
procs = [FillMissing, Categorify, Normalize]

data = (TabularList.from_df(df, path=adult, cat_names=cat_names, cont_names=cont_names, procs=procs)
                           .split_by_idx(valid_idx=range(800,1000))
                           .label_from_df(cols=dep_var)
                           .databunch())

learn = tabular_learner(data, layers=[200,100], metrics=accuracy)
learn.fit(1, 1e-2)
learn.save('mini_train')

data.export()

data = TabularDataBunch.load_empty(adult)
learn = tabular_learner(data, layers=[200,100])
learn.load('mini_train');

learn.predict(df.iloc[0])