from fastai.torch_basics import *
from fastai.data.core import *
from fastai.data.load import *
from fastai.data.external import *
from sklearn.model_selection import train_test_split#hide
from nbdev.showdoc import *Helper functions for processing data and basic transforms
Functions for getting, splitting, and labeling data, as well as generic transforms
Get, split, and label
For most data source creation we need functions to get a list of items, split them in to train/valid sets, and label them. fastai provides functions to make each of these steps easy (especially when combined with fastai.data.blocks).
Get
First we’ll look at functions that get a list of items (generally file names).
We’ll use tiny MNIST (a subset of MNIST with just two classes, 7s and 3s) for our examples/tests throughout this page.
path = untar_data(URLs.MNIST_TINY)
(path/'train').ls()(#2) [Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/7')]
def _get_files(p, fs, extensions=None):
p = Path(p)
res = [p/f for f in fs if not f.startswith('.')
and ((not extensions) or f'.{f.split(".")[-1].lower()}' in extensions)]
return res
def get_files(path, extensions=None, recurse=True, folders=None, followlinks=True):
"Get all the files in `path` with optional `extensions`, optionally with `recurse`, only in `folders`, if specified."
path = Path(path)
folders=L(folders)
extensions = setify(extensions)
extensions = {e.lower() for e in extensions}
if recurse:
res = []
for i,(p,d,f) in enumerate(os.walk(path, followlinks=followlinks)): # returns (dirpath, dirnames, filenames)
if len(folders) !=0 and i==0: d[:] = [o for o in d if o in folders]
else: d[:] = [o for o in d if not o.startswith('.')]
if len(folders) !=0 and i==0 and '.' not in folders: continue
res += _get_files(p, f, extensions)
else:
f = [o.name for o in os.scandir(path) if o.is_file()]
res = _get_files(path, f, extensions)
return L(res)This is the most general way to grab a bunch of file names from disk. If you pass extensions (including the .) then returned file names are filtered by that list. Only those files directly in path are included, unless you pass recurse, in which case all child folders are also searched recursively. folders is an optional list of directories to limit the search to.
path.ls()(#5) [Path('/home/ubuntu/.fastai/data/mnist_tiny/valid'),Path('/home/ubuntu/.fastai/data/mnist_tiny/models'),Path('/home/ubuntu/.fastai/data/mnist_tiny/test'),Path('/home/ubuntu/.fastai/data/mnist_tiny/labels.csv'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train')]t3 = get_files(path/'train'/'3', extensions='.png', recurse=False)
t7 = get_files(path/'train'/'7', extensions='.png', recurse=False)
t = get_files(path/'train', extensions='.png', recurse=True)
test_eq(len(t), len(t3)+len(t7))
test_eq(len(get_files(path/'train'/'3', extensions='.jpg', recurse=False)),0)
test_eq(len(t), len(get_files(path, extensions='.png', recurse=True, folders='train')))
t(#709) [Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/7634.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/8672.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/8406.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/9386.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/7678.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/7332.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/9424.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/7189.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/9759.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/9141.png')...]#hide
test_eq(len(get_files(path/'train'/'3', recurse=False)),346)
test_eq(len(get_files(path, extensions='.png', recurse=True, folders=['train', 'test'])),729)
test_eq(len(get_files(path, extensions='.png', recurse=True, folders='train')),709)
test_eq(len(get_files(path, extensions='.png', recurse=True, folders='training')),0)It’s often useful to be able to create functions with customized behavior. fastai.data generally uses functions named as CamelCase verbs ending in er to create these functions. FileGetter is a simple example of such a function creator.
def FileGetter(suf='', extensions=None, recurse=True, folders=None):
"Create `get_files` partial function that searches path suffix `suf`, only in `folders`, if specified, and passes along args"
def _inner(o, extensions=extensions, recurse=recurse, folders=folders):
return get_files(o/suf, extensions, recurse, folders)
return _innerfpng = FileGetter(extensions='.png', recurse=False)
test_eq(len(t7), len(fpng(path/'train'/'7')))
test_eq(len(t), len(fpng(path/'train', recurse=True)))
fpng_r = FileGetter(extensions='.png', recurse=True)
test_eq(len(t), len(fpng_r(path/'train')))
image_extensions = set(k for k,v in mimetypes.types_map.items() if v.startswith('image/'))
def get_image_files(path, recurse=True, folders=None):
"Get image files in `path` recursively, only in `folders`, if specified."
return get_files(path, extensions=image_extensions, recurse=recurse, folders=folders)This is simply get_files called with a list of standard image extensions.
test_eq(len(t), len(get_image_files(path, recurse=True, folders='train')))
def ImageGetter(suf='', recurse=True, folders=None):
"Create `get_image_files` partial that searches suffix `suf` and passes along `kwargs`, only in `folders`, if specified"
def _inner(o, recurse=recurse, folders=folders): return get_image_files(o/suf, recurse, folders)
return _innerSame as FileGetter, but for image extensions.
test_eq(len(get_files(path/'train', extensions='.png', recurse=True, folders='3')),
len(ImageGetter( 'train', recurse=True, folders='3')(path)))
def get_text_files(path, recurse=True, folders=None):
"Get text files in `path` recursively, only in `folders`, if specified."
return get_files(path, extensions=['.txt'], recurse=recurse, folders=folders)
class ItemGetter(ItemTransform):
"Creates a proper transform that applies `itemgetter(i)` (even on a tuple)"
_retain = False
def __init__(self, i): self.i = i
def encodes(self, x): return x[self.i]test_eq(ItemGetter(1)((1,2,3)), 2)
test_eq(ItemGetter(1)(L(1,2,3)), 2)
test_eq(ItemGetter(1)([1,2,3]), 2)
test_eq(ItemGetter(1)(np.array([1,2,3])), 2)
class AttrGetter(ItemTransform):
"Creates a proper transform that applies `attrgetter(nm)` (even on a tuple)"
_retain = False
def __init__(self, nm, default=None): store_attr()
def encodes(self, x): return getattr(x, self.nm, self.default)test_eq(AttrGetter('shape')(torch.randn([4,5])), [4,5])
test_eq(AttrGetter('shape', [0])([4,5]), [0])Split
The next set of functions are used to split data into training and validation sets. The functions return two lists - a list of indices or masks for each of training and validation sets.
def RandomSplitter(valid_pct=0.2, seed=None):
"Create function that splits `items` between train/val with `valid_pct` randomly."
def _inner(o):
if seed is not None: torch.manual_seed(seed)
rand_idx = L(list(torch.randperm(len(o)).numpy()))
cut = int(valid_pct * len(o))
return rand_idx[cut:],rand_idx[:cut]
return _innersrc = list(range(30))
f = RandomSplitter(seed=42)
trn,val = f(src)
assert 0<len(trn)<len(src)
assert all(o not in val for o in trn)
test_eq(len(trn), len(src)-len(val))
# test random seed consistency
test_eq(f(src)[0], trn)Use scikit-learn train_test_split. This allow to split items in a stratified fashion (uniformely according to the ’labels‘ distribution)
def TrainTestSplitter(test_size=0.2, random_state=None, stratify=None, train_size=None, shuffle=True):
"Split `items` into random train and test subsets using sklearn train_test_split utility."
def _inner(o, **kwargs):
train,valid = train_test_split(range_of(o), test_size=test_size, random_state=random_state,
stratify=stratify, train_size=train_size, shuffle=shuffle)
return L(train), L(valid)
return _innersrc = list(range(30))
labels = [0] * 20 + [1] * 10
test_size = 0.2
src
labels
f = TrainTestSplitter(test_size=test_size, random_state=42, stratify=labels)
trn,val = f(src)
assert 0<len(trn)<len(src)
assert all(o not in val for o in trn)
test_eq(len(trn), len(src)-len(val))
# test random seed consistency
test_eq(f(src)[0], trn)
# test labels distribution consistency
# there should be test_size % of zeroes and ones respectively in the validation set
test_eq(len([t for t in val if t < 20]) / 20, test_size)
test_eq(len([t for t in val if t > 20]) / 10, test_size)[0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26,
27,
28,
29][0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1]
def IndexSplitter(valid_idx):
"Split `items` so that `val_idx` are in the validation set and the others in the training set"
def _inner(o):
train_idx = np.setdiff1d(np.array(range_of(o)), np.array(valid_idx))
return L(train_idx, use_list=True), L(valid_idx, use_list=True)
return _inneritems = list(range(10))
splitter = IndexSplitter([3,7,9])
test_eq(splitter(items),[[0,1,2,4,5,6,8],[3,7,9]])
def _grandparent_idxs(items, name):
def _inner(items, name): return mask2idxs(Path(o).parent.parent.name == name for o in items)
return [i for n in L(name) for i in _inner(items,n)]
def GrandparentSplitter(train_name='train', valid_name='valid'):
"Split `items` from the grand parent folder names (`train_name` and `valid_name`)."
def _inner(o):
return _grandparent_idxs(o, train_name),_grandparent_idxs(o, valid_name)
return _innerfnames = [path/'train/3/9932.png', path/'valid/7/7189.png',
path/'valid/7/7320.png', path/'train/7/9833.png',
path/'train/3/7666.png', path/'valid/3/925.png',
path/'train/7/724.png', path/'valid/3/93055.png']
splitter = GrandparentSplitter()
test_eq(splitter(fnames),[[0,3,4,6],[1,2,5,7]])fnames2 = fnames + [path/'test/3/4256.png', path/'test/7/2345.png', path/'valid/7/6467.png']
splitter = GrandparentSplitter(train_name=('train', 'valid'), valid_name='test')
test_eq(splitter(fnames2),[[0,3,4,6,1,2,5,7,10],[8,9]])
def FuncSplitter(func):
"Split `items` by result of `func` (`True` for validation, `False` for training set)."
def _inner(o):
val_idx = mask2idxs(func(o_) for o_ in o)
return IndexSplitter(val_idx)(o)
return _innersplitter = FuncSplitter(lambda o: Path(o).parent.parent.name == 'valid')
test_eq(splitter(fnames),[[0,3,4,6],[1,2,5,7]])
def MaskSplitter(mask):
"Split `items` depending on the value of `mask`."
def _inner(o): return IndexSplitter(mask2idxs(mask))(o)
return _inneritems = list(range(6))
splitter = MaskSplitter([True,False,False,True,False,True])
test_eq(splitter(items),[[1,2,4],[0,3,5]])
def FileSplitter(fname):
"Split `items` by providing file `fname` (contains names of valid items separated by newline)."
valid = Path(fname).read_text().split('\n')
def _func(x): return x.name in valid
def _inner(o): return FuncSplitter(_func)(o)
return _innerwith tempfile.TemporaryDirectory() as d:
fname = Path(d)/'valid.txt'
fname.write_text('\n'.join([Path(fnames[i]).name for i in [1,3,4]]))
splitter = FileSplitter(fname)
test_eq(splitter(fnames),[[0,2,5,6,7],[1,3,4]])26
def ColSplitter(col='is_valid'):
"Split `items` (supposed to be a dataframe) by value in `col`"
def _inner(o):
assert isinstance(o, pd.DataFrame), "ColSplitter only works when your items are a pandas DataFrame"
valid_idx = (o.iloc[:,col] if isinstance(col, int) else o[col]).values.astype('bool')
return IndexSplitter(mask2idxs(valid_idx))(o)
return _innerdf = pd.DataFrame({'a': [0,1,2,3,4], 'b': [True,False,True,True,False]})
splits = ColSplitter('b')(df)
test_eq(splits, [[1,4], [0,2,3]])
#Works with strings or index
splits = ColSplitter(1)(df)
test_eq(splits, [[1,4], [0,2,3]])
# does not get confused if the type of 'is_valid' is integer, but it meant to be a yes/no
df = pd.DataFrame({'a': [0,1,2,3,4], 'is_valid': [1,0,1,1,0]})
splits_by_int = ColSplitter('is_valid')(df)
test_eq(splits_by_int, [[1,4], [0,2,3]])
def RandomSubsetSplitter(train_sz, valid_sz, seed=None):
"Take randoms subsets of `splits` with `train_sz` and `valid_sz`"
assert 0 < train_sz < 1
assert 0 < valid_sz < 1
assert train_sz + valid_sz <= 1.
def _inner(o):
if seed is not None: torch.manual_seed(seed)
train_len,valid_len = int(len(o)*train_sz),int(len(o)*valid_sz)
idxs = L(list(torch.randperm(len(o)).numpy()))
return idxs[:train_len],idxs[train_len:train_len+valid_len]
return _inneritems = list(range(100))
valid_idx = list(np.arange(70,100))
splits = RandomSubsetSplitter(0.3, 0.1)(items)
test_eq(len(splits[0]), 30)
test_eq(len(splits[1]), 10)Label
The final set of functions is used to label a single item of data.
def parent_label(o):
"Label `item` with the parent folder name."
return Path(o).parent.nameNote that parent_label doesn’t have anything customize, so it doesn’t return a function - you can just use it directly.
test_eq(parent_label(fnames[0]), '3')
test_eq(parent_label("fastai_dev/dev/data/mnist_tiny/train/3/9932.png"), '3')
[parent_label(o) for o in fnames]['3', '7', '7', '7', '3', '3', '7', '3']#hide
#test for MS Windows when os.path.sep is '\\' instead of '/'
test_eq(parent_label(os.path.join("fastai_dev","dev","data","mnist_tiny","train", "3", "9932.png") ), '3')
class RegexLabeller():
"Label `item` with regex `pat`."
def __init__(self, pat, match=False):
self.pat = re.compile(pat)
self.matcher = self.pat.match if match else self.pat.search
def __call__(self, o):
res = self.matcher(str(o))
assert res,f'Failed to find "{self.pat}" in "{o}"'
return res.group(1)RegexLabeller is a very flexible function since it handles any regex search of the stringified item. Pass match=True to use re.match (i.e. check only start of string), or re.search otherwise (default).
For instance, here’s an example the replicates the previous parent_label results.
fnames[0]Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/9932.png')f = RegexLabeller(fr'{os.path.sep}(\d){os.path.sep}')
test_eq(f(fnames[0]), '3')
[f(o) for o in fnames]['3', '7', '7', '7', '3', '3', '7', '3']f = RegexLabeller(r'(\d*)', match=True)
test_eq(f(fnames[0].name), '9932')
class ColReader(DisplayedTransform):
"Read `cols` in `row` with potential `pref` and `suff`"
def __init__(self, cols, pref='', suff='', label_delim=None):
store_attr()
self.pref = str(pref) + os.path.sep if isinstance(pref, Path) else pref
self.cols = L(cols)
def _do_one(self, r, c):
o = r[c] if isinstance(c, int) else r[c] if c=='name' or c=='cat' else getattr(r, c)
if len(self.pref)==0 and len(self.suff)==0 and self.label_delim is None: return o
if self.label_delim is None: return f'{self.pref}{o}{self.suff}'
else: return o.split(self.label_delim) if len(o)>0 else []
def __call__(self, o, **kwargs):
if len(self.cols) == 1: return self._do_one(o, self.cols[0])
return L(self._do_one(o, c) for c in self.cols)cols can be a list of column names or a list of indices (or a mix of both). If label_delim is passed, the result is split using it.
df = pd.DataFrame({'a': 'a b c d'.split(), 'b': ['1 2', '0', '', '1 2 3']})
f = ColReader('a', pref='0', suff='1')
test_eq([f(o) for o in df.itertuples()], '0a1 0b1 0c1 0d1'.split())
f = ColReader('b', label_delim=' ')
test_eq([f(o) for o in df.itertuples()], [['1', '2'], ['0'], [], ['1', '2', '3']])
df['a1'] = df['a']
f = ColReader(['a', 'a1'], pref='0', suff='1')
test_eq([f(o) for o in df.itertuples()], [L('0a1', '0a1'), L('0b1', '0b1'), L('0c1', '0c1'), L('0d1', '0d1')])
df = pd.DataFrame({'a': [L(0,1), L(2,3,4), L(5,6,7)]})
f = ColReader('a')
test_eq([f(o) for o in df.itertuples()], [L(0,1), L(2,3,4), L(5,6,7)])
df['name'] = df['a']
f = ColReader('name')
test_eq([f(df.iloc[0,:])], [L(0,1)])Categorize -
class CategoryMap(CollBase):
"Collection of categories with the reverse mapping in `o2i`"
def __init__(self, col, sort=True, add_na=False, strict=False):
if is_categorical_dtype(col):
items = L(col.cat.categories, use_list=True)
#Remove non-used categories while keeping order
if strict: items = L(o for o in items if o in col.unique())
else:
if not hasattr(col,'unique'): col = L(col, use_list=True)
# `o==o` is the generalized definition of non-NaN used by Pandas
items = L(o for o in col.unique() if o==o)
if sort: items = items.sorted()
self.items = '#na#' + items if add_na else items
self.o2i = defaultdict(int, self.items.val2idx()) if add_na else dict(self.items.val2idx())
def map_objs(self,objs):
"Map `objs` to IDs"
return L(self.o2i[o] for o in objs)
def map_ids(self,ids):
"Map `ids` to objects in vocab"
return L(self.items[o] for o in ids)
def __eq__(self,b): return all_equal(b,self)aa = CategoryMap(L(1,2,3,2,1,5,6,1))
aa[1, 2, 3, 5, 6]t = CategoryMap([4,2,3,4])
t
test_eq(t, [2,3,4])
test_eq(t.o2i, {2:0,3:1,4:2})
test_eq(t.map_objs([2,3]), [0,1])
test_eq(t.map_ids([0,1]), [2,3])
test_fail(lambda: t.o2i['unseen label'])[2, 3, 4]t = CategoryMap([4,2,3,4], add_na=True)
test_eq(t, ['#na#',2,3,4])
test_eq(t.o2i, {'#na#':0,2:1,3:2,4:3})t = CategoryMap(pd.Series([4,2,3,4]), sort=False)
test_eq(t, [4,2,3])
test_eq(t.o2i, {4:0,2:1,3:2})col = pd.Series(pd.Categorical(['M','H','L','M'], categories=['H','M','L'], ordered=True))
t = CategoryMap(col)
test_eq(t, ['H','M','L'])
test_eq(t.o2i, {'H':0,'M':1,'L':2})col = pd.Series(pd.Categorical(['M','H','M'], categories=['H','M','L'], ordered=True))
t = CategoryMap(col, strict=True)
test_eq(t, ['H','M'])
test_eq(t.o2i, {'H':0,'M':1})
class Categorize(DisplayedTransform):
"Reversible transform of category string to `vocab` id"
loss_func,order=CrossEntropyLossFlat(),1
def __init__(self, vocab=None, sort=True, add_na=False):
if vocab is not None: vocab = CategoryMap(vocab, sort=sort, add_na=add_na)
store_attr()
def setups(self, dsets):
if self.vocab is None and dsets is not None: self.vocab = CategoryMap(dsets, sort=self.sort, add_na=self.add_na)
self.c = len(self.vocab)
def encodes(self, o):
try:
return TensorCategory(self.vocab.o2i[o])
except KeyError as e:
raise KeyError(f"Label '{o}' was not included in the training dataset") from e
def decodes(self, o): return Category (self.vocab [o])
class Category(str, ShowTitle): _show_args = {'label': 'category'}cat = Categorize()
tds = Datasets(['cat', 'dog', 'cat'], tfms=[cat])
test_eq(cat.vocab, ['cat', 'dog'])
test_eq(cat('cat'), 0)
test_eq(cat.decode(1), 'dog')
test_stdout(lambda: show_at(tds,2), 'cat')
test_fail(lambda: cat('bird'))cat.decode(0)'cat'cat = Categorize(add_na=True)
tds = Datasets(['cat', 'dog', 'cat'], tfms=[cat])
test_eq(cat.vocab, ['#na#', 'cat', 'dog'])
test_eq(cat('cat'), 1)
test_eq(cat.decode(2), 'dog')
test_stdout(lambda: show_at(tds,2), 'cat')cat.decode(0)'#na#'cat = Categorize(vocab=['dog', 'cat'], sort=False, add_na=True)
tds = Datasets(['cat', 'dog', 'cat'], tfms=[cat])
test_eq(cat.vocab, ['#na#', 'dog', 'cat'])
test_eq(cat('dog'), 1)
test_eq(cat.decode(2), 'cat')
test_stdout(lambda: show_at(tds,2), 'cat')Multicategorize -
class MultiCategorize(Categorize):
"Reversible transform of multi-category strings to `vocab` id"
loss_func,order=BCEWithLogitsLossFlat(),1
def __init__(self, vocab=None, add_na=False): super().__init__(vocab=vocab,add_na=add_na,sort=vocab==None)
def setups(self, dsets):
if not dsets: return
if self.vocab is None:
vals = set()
for b in dsets: vals = vals.union(set(b))
self.vocab = CategoryMap(list(vals), add_na=self.add_na)
def encodes(self, o):
if not all(elem in self.vocab.o2i.keys() for elem in o):
diff = [elem for elem in o if elem not in self.vocab.o2i.keys()]
diff_str = "', '".join(diff)
raise KeyError(f"Labels '{diff_str}' were not included in the training dataset")
return TensorMultiCategory([self.vocab.o2i[o_] for o_ in o])
def decodes(self, o): return MultiCategory ([self.vocab [o_] for o_ in o])
class MultiCategory(L):
def show(self, ctx=None, sep=';', color='black', **kwargs):
return show_title(sep.join(self.map(str)), ctx=ctx, color=color, **kwargs)cat = MultiCategorize()
tds = Datasets([['b', 'c'], ['a'], ['a', 'c'], []], tfms=[cat])
test_eq(tds[3][0], TensorMultiCategory([]))
test_eq(cat.vocab, ['a', 'b', 'c'])
test_eq(cat(['a', 'c']), tensor([0,2]))
test_eq(cat([]), tensor([]))
test_eq(cat.decode([1]), ['b'])
test_eq(cat.decode([0,2]), ['a', 'c'])
test_stdout(lambda: show_at(tds,2), 'a;c')
# if vocab supplied, ensure it maintains its order (i.e., it doesn't sort)
cat = MultiCategorize(vocab=['z', 'y', 'x'])
test_eq(cat.vocab, ['z','y','x'])
test_fail(lambda: cat('bird'))
class OneHotEncode(DisplayedTransform):
"One-hot encodes targets"
order=2
def __init__(self, c=None): store_attr()
def setups(self, dsets):
if self.c is None: self.c = len(L(getattr(dsets, 'vocab', None)))
if not self.c: warn("Couldn't infer the number of classes, please pass a value for `c` at init")
def encodes(self, o): return TensorMultiCategory(one_hot(o, self.c).float())
def decodes(self, o): return one_hot_decode(o, None)Works in conjunction with MultiCategorize or on its own if you have one-hot encoded targets (pass a vocab for decoding and do_encode=False in this case)
_tfm = OneHotEncode(c=3)
test_eq(_tfm([0,2]), tensor([1.,0,1]))
test_eq(_tfm.decode(tensor([0,1,1])), [1,2])tds = Datasets([['b', 'c'], ['a'], ['a', 'c'], []], [[MultiCategorize(), OneHotEncode()]])
test_eq(tds[1], [tensor([1.,0,0])])
test_eq(tds[3], [tensor([0.,0,0])])
test_eq(tds.decode([tensor([False, True, True])]), [['b','c']])
test_eq(type(tds[1][0]), TensorMultiCategory)
test_stdout(lambda: show_at(tds,2), 'a;c')#hide
#test with passing the vocab
tds = Datasets([['b', 'c'], ['a'], ['a', 'c'], []], [[MultiCategorize(vocab=['a', 'b', 'c']), OneHotEncode()]])
test_eq(tds[1], [tensor([1.,0,0])])
test_eq(tds[3], [tensor([0.,0,0])])
test_eq(tds.decode([tensor([False, True, True])]), [['b','c']])
test_eq(type(tds[1][0]), TensorMultiCategory)
test_stdout(lambda: show_at(tds,2), 'a;c')
class EncodedMultiCategorize(Categorize):
"Transform of one-hot encoded multi-category that decodes with `vocab`"
loss_func,order=BCEWithLogitsLossFlat(),1
def __init__(self, vocab):
super().__init__(vocab, sort=vocab==None)
self.c = len(vocab)
def encodes(self, o): return TensorMultiCategory(tensor(o).float())
def decodes(self, o): return MultiCategory (one_hot_decode(o, self.vocab))_tfm = EncodedMultiCategorize(vocab=['a', 'b', 'c'])
test_eq(_tfm([1,0,1]), tensor([1., 0., 1.]))
test_eq(type(_tfm([1,0,1])), TensorMultiCategory)
test_eq(_tfm.decode(tensor([False, True, True])), ['b','c'])
_tfm2 = EncodedMultiCategorize(vocab=['c', 'b', 'a'])
test_eq(_tfm2.vocab, ['c', 'b', 'a'])
class RegressionSetup(DisplayedTransform):
"Transform that floatifies targets"
loss_func=MSELossFlat()
def __init__(self, c=None): store_attr()
def encodes(self, o): return tensor(o).float()
def decodes(self, o): return TitledFloat(o) if o.ndim==0 else TitledTuple(o_.item() for o_ in o)
def setups(self, dsets):
if self.c is not None: return
try: self.c = len(dsets[0]) if hasattr(dsets[0], '__len__') else 1
except: self.c = 0_tfm = RegressionSetup()
dsets = Datasets([0, 1, 2], RegressionSetup)
test_eq(dsets.c, 1)
test_eq_type(dsets[0], (tensor(0.),))
dsets = Datasets([[0, 1, 2], [3,4,5]], RegressionSetup)
test_eq(dsets.c, 3)
test_eq_type(dsets[0], (tensor([0.,1.,2.]),))
def get_c(dls):
if getattr(dls, 'c', False): return dls.c
if getattr(getattr(dls.train, 'after_item', None), 'c', False): return dls.train.after_item.c
if getattr(getattr(dls.train, 'after_batch', None), 'c', False): return dls.train.after_batch.c
vocab = getattr(dls, 'vocab', [])
if len(vocab) > 0 and is_listy(vocab[-1]): vocab = vocab[-1]
return len(vocab)End-to-end dataset example with MNIST
Let’s show how to use those functions to grab the mnist dataset in a Datasets. First we grab all the images.
path = untar_data(URLs.MNIST_TINY)
items = get_image_files(path)Then we split between train and validation depending on the folder.
splitter = GrandparentSplitter()
splits = splitter(items)
train,valid = (items[i] for i in splits)
train[:3],valid[:3]((#3) [Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/7634.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/8672.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/train/3/8406.png')],
(#3) [Path('/home/ubuntu/.fastai/data/mnist_tiny/valid/3/7215.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/valid/3/7312.png'),Path('/home/ubuntu/.fastai/data/mnist_tiny/valid/3/8050.png')])Our inputs are images that we open and convert to tensors, our targets are labeled depending on the parent directory and are categories.
from PIL import Imagedef open_img(fn:Path): return Image.open(fn).copy()
def img2tensor(im:Image.Image): return TensorImage(array(im)[None])
tfms = [[open_img, img2tensor],
[parent_label, Categorize()]]
train_ds = Datasets(train, tfms)train_ds(#709) [(TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 26, 138, 233, 253,
255, 253, 169, 97, 24, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 120, 252, 252, 252,
253, 252, 252, 252, 252, 43, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 17, 228, 252, 195, 79,
69, 141, 234, 252, 252, 237, 25, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 24, 252, 252, 45, 0,
0, 0, 33, 173, 252, 253, 92, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 117, 252, 45, 0,
0, 0, 0, 93, 252, 253, 92, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 32, 126, 0,
0, 0, 0, 208, 253, 252, 84, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 30, 130, 236, 252, 135, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
49, 228, 252, 227, 130, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 76, 144,
233, 252, 185, 50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 64, 136, 230, 248, 252,
243, 64, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 179, 255, 253, 253, 253, 253,
244, 138, 138, 44, 24, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 230, 253, 252, 252, 252, 252,
253, 252, 252, 252, 252, 151, 17, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 115, 184, 183, 183, 183, 215,
253, 252, 252, 252, 252, 253, 92, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 21,
46, 45, 45, 215, 252, 253, 92, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 116, 248, 252, 253, 92, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 93, 201, 0, 0, 0, 0, 0,
24, 159, 253, 253, 253, 158, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 166, 252, 0, 0, 0, 0, 22,
212, 252, 252, 252, 231, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 207, 252, 123, 70, 153, 184, 215,
253, 252, 252, 153, 37, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 113, 252, 253, 252, 252, 252, 252,
253, 193, 77, 9, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 9, 128, 201, 252, 252, 252, 200,
75, 8, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8), TensorCategory(0)),(TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 13, 41, 141, 104, 29, 141, 141,
229, 203, 141, 47, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 19, 144, 253, 252, 252, 252, 253, 252,
252, 252, 253, 159, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 57, 243, 244, 168, 168, 118, 106, 56,
209, 252, 253, 109, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 25, 25, 0, 0, 0, 0, 0,
197, 252, 228, 9, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 128,
253, 253, 214, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 104, 252,
252, 227, 38, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 89, 229, 252,
252, 103, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 26, 150, 237, 253, 252,
214, 116, 13, 38, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 104, 253, 253, 253, 255, 253,
253, 253, 255, 253, 128, 4, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 228, 252, 252, 227, 197, 196,
196, 221, 253, 252, 252, 128, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 44, 168, 130, 31, 0, 0,
0, 25, 119, 224, 252, 252, 89, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 44, 228, 252, 213, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 198, 253, 226, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 197, 252, 225, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 57, 215, 252, 187, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 26, 113, 63, 0, 0,
0, 0, 26, 243, 252, 252, 63, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 170, 253, 153, 0, 0,
0, 26, 204, 253, 253, 128, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 144, 252, 252, 169, 169,
169, 243, 253, 252, 170, 9, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 13, 209, 252, 253, 252,
252, 252, 244, 93, 13, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 59, 190, 253, 252,
252, 202, 125, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8), TensorCategory(0)),(TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 37, 187, 254, 254, 255,
233, 149, 41, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 8, 202, 252, 234, 234, 234,
239, 254, 242, 147, 13, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 99, 254, 135, 0, 0, 0,
18, 86, 230, 254, 155, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 8, 57, 3, 0, 0, 0,
0, 0, 13, 236, 208, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 31, 242, 254, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 5, 195, 254, 133, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 20,
119, 203, 254, 182, 32, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 40, 254,
254, 254, 169, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 34, 214,
254, 238, 18, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
179, 254, 183, 46, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 193, 254, 177, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 18, 188, 248, 95, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 176, 254, 155, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 176, 254, 140, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 176, 238, 36, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 35, 187, 192, 0,
14, 149, 254, 95, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 28, 218, 179, 18, 8,
225, 254, 141, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 41, 242, 169, 94, 206,
254, 141, 24, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 162, 254, 254, 254,
132, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 73, 186, 238, 81,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8), TensorCategory(0)),(TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 32, 43, 32, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 8, 85, 173, 191, 237, 252, 237, 190,
191, 111, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 13, 136, 234, 252, 235, 170, 126, 126, 126, 170,
236, 252, 172, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 174, 252, 252, 183, 14, 0, 0, 0, 0, 0,
14, 120, 247, 196, 28, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 253, 252, 190, 14, 0, 0, 0, 0, 0, 0,
0, 0, 144, 252, 86, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 237, 239, 17, 0, 0, 0, 0, 0, 0, 0,
0, 0, 22, 253, 253, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 55, 112, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 22, 252, 252, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 16, 85, 252, 252, 9, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 59, 91, 169,
169, 232, 252, 252, 252, 204, 133, 32, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 18, 141, 249, 252, 164,
209, 252, 252, 252, 252, 253, 252, 223, 35, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 36, 210, 253, 170, 11, 87,
255, 253, 250, 185, 62, 175, 253, 253, 253, 60, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 211, 253, 217, 125, 206, 252,
253, 252, 196, 0, 0, 11, 127, 252, 252, 226, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 211, 253, 252, 252, 252, 252,
241, 117, 11, 0, 0, 0, 7, 136, 252, 252, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 211, 253, 205, 189, 189, 101,
35, 0, 0, 0, 0, 0, 0, 127, 252, 252, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 35, 86, 11, 0, 0, 0,
0, 0, 0, 0, 0, 0, 116, 232, 252, 252, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 4, 43, 43, 210, 253, 253, 253, 218, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 109, 7, 0, 0, 0, 0, 0, 0, 71,
86, 164, 195, 252, 252, 253, 252, 252, 196, 28, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 181, 225, 127, 127, 189, 232, 232, 232, 249,
253, 252, 252, 252, 252, 241, 196, 73, 7, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 64, 194, 210, 242, 216, 189, 189, 189, 189,
128, 184, 185, 84, 84, 35, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 4, 14, 35, 18, 0, 0, 0, 0,
0, 14, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8), TensorCategory(0)),(TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 29, 154,
253, 253, 140, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 13, 172, 252, 247,
208, 252, 65, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 172, 252, 177, 50,
144, 186, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 16, 215, 127, 3, 0,
169, 118, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 13, 204,
209, 25, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 57, 206, 247,
65, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 57, 191, 214, 50,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 63, 206, 252, 240, 126,
225, 225, 225, 226, 225, 187, 13, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 141, 253, 253, 253, 255,
234, 225, 238, 255, 253, 206, 13, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 60, 196, 158, 84, 84,
28, 0, 38, 103, 252, 168, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 66, 252, 168, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 191, 252, 118, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 63, 255, 134, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 19, 194, 247, 65, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10,
28, 123, 252, 150, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 51, 154,
110, 246, 252, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 179, 79,
216, 206, 13, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 110, 252, 216,
240, 81, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 19, 215, 252, 244,
81, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 32, 228, 151, 25,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8), TensorCategory(0)),(TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 19, 149, 231, 254, 254,
231, 145, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 49, 238, 243, 243, 243,
246, 251, 182, 13, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
45, 244, 253, 120, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 128, 253, 230, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 115, 253, 246, 53, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 127, 253, 228, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
21, 228, 253, 228, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 24, 40,
91, 253, 253, 158, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 149, 229, 253,
254, 253, 253, 30, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 194, 253, 253,
253, 253, 226, 18, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 38, 49, 49,
73, 196, 253, 151, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 53, 253, 254, 74, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 15, 253, 253, 158, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 15, 253, 253, 173, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 16, 253, 254, 173, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 124, 153, 6, 0, 0, 0,
0, 15, 253, 253, 104, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 70, 252, 168, 0, 0, 0, 0,
0, 122, 253, 250, 64, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 149, 253, 232, 82, 0, 0, 0,
81, 244, 253, 193, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 66, 232, 254, 252, 245, 245, 245,
250, 251, 181, 16, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 28, 54, 164, 253, 253, 253,
230, 144, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8), TensorCategory(0)),(TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 99, 219,
253, 255, 253, 253, 182, 12, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 18, 78, 208, 249, 252,
252, 253, 252, 252, 252, 64, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 171, 252, 252, 252, 249,
237, 176, 223, 252, 252, 177, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 153, 252, 188, 132, 110,
0, 50, 229, 252, 252, 107, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 9, 29, 14, 0, 0,
0, 105, 252, 252, 216, 20, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 6,
30, 210, 252, 223, 39, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 45,
252, 253, 252, 88, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 32, 206,
252, 253, 142, 5, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 204, 252,
252, 173, 10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 80, 246, 252,
252, 149, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 31, 253, 253, 253,
253, 193, 105, 105, 105, 12, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 153, 252, 252, 252,
252, 253, 252, 252, 252, 169, 18, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 82, 252, 252, 252,
252, 253, 252, 252, 252, 252, 170, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 16, 132, 132, 132,
132, 133, 132, 210, 252, 252, 151, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 98, 242, 252, 252, 74, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 85, 47, 9, 0, 0, 0,
18, 92, 239, 252, 252, 243, 65, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 120, 252, 152, 0, 128, 134,
203, 253, 252, 252, 252, 99, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 120, 252, 250, 238, 251, 252,
252, 253, 252, 184, 14, 2, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 99, 220, 252, 252, 252, 244,
207, 208, 84, 21, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 31, 121, 252, 111, 85,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8), TensorCategory(0)),(TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 71, 228, 196,
143, 73, 73, 73, 53, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 141, 253, 254,
253, 253, 253, 253, 249, 216, 21, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 16, 101, 127,
127, 127, 127, 127, 163, 253, 114, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 119, 253, 88, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 33, 254, 253, 36, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 38, 228, 255, 115, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 73, 243, 253, 115, 2, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 50,
146, 248, 218, 94, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 15, 116, 234,
229, 99, 12, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 59, 238, 253, 254,
218, 13, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 46, 91, 91, 137,
254, 185, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 5,
133, 224, 12, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
55, 253, 67, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
55, 253, 106, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
120, 253, 54, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 47,
247, 217, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 20, 209,
241, 106, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 19, 76, 204, 243,
137, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 68, 199, 134, 110, 128, 226, 253, 207, 104,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 49, 168, 253, 254, 253, 207, 124, 21, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8), TensorCategory(0)),(TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 44,
105, 202, 254, 254, 254, 254, 255, 228, 43, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 73, 209, 229,
253, 253, 253, 253, 253, 253, 253, 253, 139, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 11, 95, 164, 233, 253, 254,
253, 243, 238, 238, 240, 253, 253, 253, 155, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 179, 253, 253, 253, 253, 205,
133, 47, 0, 0, 102, 253, 253, 253, 104, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 38, 178, 178, 178, 90, 17,
0, 0, 36, 120, 245, 253, 253, 240, 73, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 118, 214, 253, 253, 253, 189, 61, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 80, 135,
134, 212, 253, 253, 253, 181, 77, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 86, 239, 247, 254,
253, 253, 253, 253, 235, 49, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 75, 216, 253, 254,
216, 208, 229, 253, 253, 222, 38, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 19, 104, 105,
18, 0, 120, 253, 253, 253, 163, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 226, 254, 254, 164, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 136, 253, 253, 216, 35, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 224, 253, 253, 163, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 224, 253, 253, 163, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 58, 224, 250, 253, 253, 31, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 75, 144, 30, 16, 0, 0, 0, 0, 0, 0, 0,
0, 164, 253, 253, 235, 88, 4, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 184, 253, 253, 133, 0, 0, 0, 0, 0, 0, 135,
134, 212, 253, 253, 157, 56, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 255, 253, 253, 247, 239, 239, 239, 239, 239, 239, 255,
253, 253, 253, 129, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 121, 245, 253, 253, 253, 253, 253, 253, 253, 253, 254,
242, 208, 137, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 86, 113, 253, 253, 253, 253, 253, 253, 253, 105,
80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8), TensorCategory(0)),(TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 113, 254,
254, 255, 254, 133, 8, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 112, 253, 253,
253, 253, 253, 253, 106, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 166, 253, 253, 253,
176, 82, 213, 253, 185, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 80, 251, 253, 242, 100,
15, 0, 113, 253, 246, 66, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 70, 250, 192, 39, 0,
0, 0, 113, 253, 229, 16, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 28, 13, 0, 0,
0, 40, 227, 253, 119, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 98, 253, 253, 104, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
8, 217, 253, 164, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10,
161, 253, 215, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 21, 164, 177, 112, 79, 195,
253, 187, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 89, 253, 253, 253, 253, 253,
253, 104, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 15, 145, 158, 158, 139, 69,
158, 179, 9, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 213, 76, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 213, 199, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 213, 253, 5, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
73, 239, 232, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 11, 93,
237, 253, 111, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 58, 210, 253,
253, 139, 5, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 36, 150, 219, 238, 237, 152,
62, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 199, 253, 140, 69, 14, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8), TensorCategory(0))...]x,y = train_ds[3]
x
y
xd,yd = decode_at(train_ds,3)
xd
yd
test_eq(parent_label(train[3]),yd)
test_eq(array(Image.open(train[3])),xd[0].numpy())TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 32, 43, 32, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 8, 85, 173, 191, 237, 252, 237, 190,
191, 111, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 13, 136, 234, 252, 235, 170, 126, 126, 126, 170,
236, 252, 172, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 174, 252, 252, 183, 14, 0, 0, 0, 0, 0,
14, 120, 247, 196, 28, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 253, 252, 190, 14, 0, 0, 0, 0, 0, 0,
0, 0, 144, 252, 86, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 237, 239, 17, 0, 0, 0, 0, 0, 0, 0,
0, 0, 22, 253, 253, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 55, 112, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 22, 252, 252, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 16, 85, 252, 252, 9, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 59, 91, 169,
169, 232, 252, 252, 252, 204, 133, 32, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 18, 141, 249, 252, 164,
209, 252, 252, 252, 252, 253, 252, 223, 35, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 36, 210, 253, 170, 11, 87,
255, 253, 250, 185, 62, 175, 253, 253, 253, 60, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 211, 253, 217, 125, 206, 252,
253, 252, 196, 0, 0, 11, 127, 252, 252, 226, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 211, 253, 252, 252, 252, 252,
241, 117, 11, 0, 0, 0, 7, 136, 252, 252, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 211, 253, 205, 189, 189, 101,
35, 0, 0, 0, 0, 0, 0, 127, 252, 252, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 35, 86, 11, 0, 0, 0,
0, 0, 0, 0, 0, 0, 116, 232, 252, 252, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 4, 43, 43, 210, 253, 253, 253, 218, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 109, 7, 0, 0, 0, 0, 0, 0, 71,
86, 164, 195, 252, 252, 253, 252, 252, 196, 28, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 181, 225, 127, 127, 189, 232, 232, 232, 249,
253, 252, 252, 252, 252, 241, 196, 73, 7, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 64, 194, 210, 242, 216, 189, 189, 189, 189,
128, 184, 185, 84, 84, 35, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 4, 14, 35, 18, 0, 0, 0, 0,
0, 14, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8)TensorCategory(0)TensorImage([[[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 32, 43, 32, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 8, 85, 173, 191, 237, 252, 237, 190,
191, 111, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 13, 136, 234, 252, 235, 170, 126, 126, 126, 170,
236, 252, 172, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 174, 252, 252, 183, 14, 0, 0, 0, 0, 0,
14, 120, 247, 196, 28, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 253, 252, 190, 14, 0, 0, 0, 0, 0, 0,
0, 0, 144, 252, 86, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 237, 239, 17, 0, 0, 0, 0, 0, 0, 0,
0, 0, 22, 253, 253, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 55, 112, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 22, 252, 252, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 16, 85, 252, 252, 9, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 59, 91, 169,
169, 232, 252, 252, 252, 204, 133, 32, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 18, 141, 249, 252, 164,
209, 252, 252, 252, 252, 253, 252, 223, 35, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 36, 210, 253, 170, 11, 87,
255, 253, 250, 185, 62, 175, 253, 253, 253, 60, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 211, 253, 217, 125, 206, 252,
253, 252, 196, 0, 0, 11, 127, 252, 252, 226, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 211, 253, 252, 252, 252, 252,
241, 117, 11, 0, 0, 0, 7, 136, 252, 252, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 211, 253, 205, 189, 189, 101,
35, 0, 0, 0, 0, 0, 0, 127, 252, 252, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 35, 86, 11, 0, 0, 0,
0, 0, 0, 0, 0, 0, 116, 232, 252, 252, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 4, 43, 43, 210, 253, 253, 253, 218, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 109, 7, 0, 0, 0, 0, 0, 0, 71,
86, 164, 195, 252, 252, 253, 252, 252, 196, 28, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 181, 225, 127, 127, 189, 232, 232, 232, 249,
253, 252, 252, 252, 252, 241, 196, 73, 7, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 64, 194, 210, 242, 216, 189, 189, 189, 189,
128, 184, 185, 84, 84, 35, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 4, 14, 35, 18, 0, 0, 0, 0,
0, 14, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]],
dtype=torch.uint8)'3'ax = show_at(train_ds, 3, cmap="Greys", figsize=(1,1))
assert ax.title.get_text() in ('3','7')
test_fig_exists(ax)ToTensor -
class ToTensor(Transform):
"Convert item to appropriate tensor class"
order = 5IntToFloatTensor -
class IntToFloatTensor(DisplayedTransform):
"Transform image to float tensor, optionally dividing by 255 (e.g. for images)."
order = 10 #Need to run after PIL transforms on the GPU
def __init__(self, div=255., div_mask=1): store_attr()
def encodes(self, o:TensorImage): return o.float().div_(self.div)
def encodes(self, o:TensorMask ): return o.long() // self.div_mask
def decodes(self, o:TensorImage): return ((o.clamp(0., 1.) * self.div).long()) if self.div else ot = (TensorImage(tensor(1)),tensor(2).long(),TensorMask(tensor(3)))
tfm = IntToFloatTensor()
ft = tfm(t)
test_eq(ft, [1./255, 2, 3])
test_eq(type(ft[0]), TensorImage)
test_eq(type(ft[2]), TensorMask)
test_eq(ft[0].type(),'torch.FloatTensor')
test_eq(ft[1].type(),'torch.LongTensor')
test_eq(ft[2].type(),'torch.LongTensor')Normalization -
def broadcast_vec(dim, ndim, *t, cuda=True):
"Make a vector broadcastable over `dim` (out of `ndim` total) by prepending and appending unit axes"
v = [1]*ndim
v[dim] = -1
f = to_device if cuda else noop
return [f(tensor(o).view(*v)) for o in t]
@docs
class Normalize(DisplayedTransform):
"Normalize/denorm batch of `TensorImage`"
parameters,order = L('mean', 'std'),99
def __init__(self, mean=None, std=None, axes=(0,2,3)): store_attr()
@classmethod
def from_stats(cls, mean, std, dim=1, ndim=4, cuda=True): return cls(*broadcast_vec(dim, ndim, mean, std, cuda=cuda))
def setups(self, dl:DataLoader):
if self.mean is None or self.std is None:
x,*_ = dl.one_batch()
self.mean,self.std = x.mean(self.axes, keepdim=True),x.std(self.axes, keepdim=True)+1e-7
def encodes(self, x:TensorImage): return (x-self.mean) / self.std
def decodes(self, x:TensorImage):
f = to_cpu if x.device.type=='cpu' else noop
return (x*f(self.std) + f(self.mean))
_docs=dict(encodes="Normalize batch", decodes="Denormalize batch")[1]*5[1, 1, 1, 1, 1]aa = np.random.randint(1,3,(2,2))
aa
bb = broadcast_vec(4,6,aa)
bb
aa[1] = -1
bb[0].shapearray([[1, 2],
[2, 2]])[tensor([[[[[[1],
[2],
[2],
[2]]]]]], device='cuda:0')]torch.Size([1, 1, 1, 1, 4, 1])mean,std = [0.5]*3,[0.5]*3
mean,std
mean,std = broadcast_vec(1, 4, mean, std)
mean,std
mean.shape
std.shape
batch_tfms = [IntToFloatTensor(), Normalize.from_stats(mean,std)]
tdl = TfmdDL(train_ds, after_batch=batch_tfms, bs=4, device=default_device())([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])(tensor([[[[0.5000]],
[[0.5000]],
[[0.5000]]]], device='cuda:0'),
tensor([[[[0.5000]],
[[0.5000]],
[[0.5000]]]], device='cuda:0'))torch.Size([1, 3, 1, 1])torch.Size([1, 3, 1, 1])x,y = tdl.one_batch()
xd,yd = tdl.decode((x,y))
test_eq(x.type(), 'torch.cuda.FloatTensor' if default_device().type=='cuda' else 'torch.FloatTensor')
test_eq(xd.type(), 'torch.LongTensor')
test_eq(type(x), TensorImage)
test_eq(type(y), TensorCategory)
assert x.mean()<0.0
assert x.std()>0.5
assert 0<xd.float().mean()/255.<1
assert 0<xd.float().std()/255.<0.5#hide
nrm = Normalize()
batch_tfms = [IntToFloatTensor(), nrm]
tdl = TfmdDL(train_ds, after_batch=batch_tfms, bs=4)
x,y = tdl.one_batch()
test_close(x.mean(), 0.0, 1e-4)
assert x.std()>0.9, x.std()#Just for visuals
from fastai.vision.core import *tdl.show_batch((x,y))
#hide
x,y = cast(x,Tensor),cast(y,Tensor) #Lose type of tensors (to emulate predictions)
test_ne(type(x), TensorImage)
tdl.show_batch((x,y), figsize=(1,1)) #Check that types are put back by dl.
