text stringlengths 81 112k |
|---|
Parameterized ReLU as in the paper `Delving Deep into Rectifiers: Surpassing
Human-Level Performance on ImageNet Classification
<http://arxiv.org/abs/1502.01852>`_.
Args:
x (tf.Tensor): input
init (float): initial value for the learnable slope.
name (str): name of the output.
Variable Names:
* ``alpha``: learnable slope.
def PReLU(x, init=0.001, name='output'):
"""
Parameterized ReLU as in the paper `Delving Deep into Rectifiers: Surpassing
Human-Level Performance on ImageNet Classification
<http://arxiv.org/abs/1502.01852>`_.
Args:
x (tf.Tensor): input
init (float): initial value for the learnable slope.
name (str): name of the output.
Variable Names:
* ``alpha``: learnable slope.
"""
init = tfv1.constant_initializer(init)
alpha = tfv1.get_variable('alpha', [], initializer=init)
x = ((1 + alpha) * x + (1 - alpha) * tf.abs(x))
ret = tf.multiply(x, 0.5, name=name)
ret.variables = VariableHolder(alpha=alpha)
return ret |
A shorthand of BatchNormalization + ReLU.
def BNReLU(x, name=None):
"""
A shorthand of BatchNormalization + ReLU.
"""
x = BatchNorm('bn', x)
x = tf.nn.relu(x, name=name)
return x |
More code that reproduces the paper can be found at https://github.com/ppwwyyxx/GroupNorm-reproduce/.
def GroupNorm(x, group=32, gamma_initializer=tf.constant_initializer(1.)):
"""
More code that reproduces the paper can be found at https://github.com/ppwwyyxx/GroupNorm-reproduce/.
"""
shape = x.get_shape().as_list()
ndims = len(shape)
assert ndims == 4, shape
chan = shape[1]
assert chan % group == 0, chan
group_size = chan // group
orig_shape = tf.shape(x)
h, w = orig_shape[2], orig_shape[3]
x = tf.reshape(x, tf.stack([-1, group, group_size, h, w]))
mean, var = tf.nn.moments(x, [2, 3, 4], keep_dims=True)
new_shape = [1, group, group_size, 1, 1]
beta = tf.get_variable('beta', [chan], initializer=tf.constant_initializer())
beta = tf.reshape(beta, new_shape)
gamma = tf.get_variable('gamma', [chan], initializer=gamma_initializer)
gamma = tf.reshape(gamma, new_shape)
out = tf.nn.batch_normalization(x, mean, var, beta, gamma, 1e-5, name='output')
return tf.reshape(out, orig_shape, name='output') |
Args:
freeze (bool): whether to freeze all the variables under the scope
def backbone_scope(freeze):
"""
Args:
freeze (bool): whether to freeze all the variables under the scope
"""
def nonlin(x):
x = get_norm()(x)
return tf.nn.relu(x)
with argscope([Conv2D, MaxPooling, BatchNorm], data_format='channels_first'), \
argscope(Conv2D, use_bias=False, activation=nonlin,
kernel_initializer=tf.variance_scaling_initializer(
scale=2.0, mode='fan_out')), \
ExitStack() as stack:
if cfg.BACKBONE.NORM in ['FreezeBN', 'SyncBN']:
if freeze or cfg.BACKBONE.NORM == 'FreezeBN':
stack.enter_context(argscope(BatchNorm, training=False))
else:
stack.enter_context(argscope(
BatchNorm, sync_statistics='nccl' if cfg.TRAINER == 'replicated' else 'horovod'))
if freeze:
stack.enter_context(freeze_variables(stop_gradient=False, skip_collection=True))
else:
# the layers are not completely freezed, but we may want to only freeze the affine
if cfg.BACKBONE.FREEZE_AFFINE:
stack.enter_context(custom_getter_scope(freeze_affine_getter))
yield |
Extract the images into a 4D uint8 numpy array [index, y, x, depth].
def extract_images(filename):
"""Extract the images into a 4D uint8 numpy array [index, y, x, depth]."""
with gzip.open(filename) as bytestream:
magic = _read32(bytestream)
if magic != 2051:
raise ValueError(
'Invalid magic number %d in MNIST image file: %s' %
(magic, filename))
num_images = _read32(bytestream)
rows = _read32(bytestream)
cols = _read32(bytestream)
buf = bytestream.read(rows * cols * num_images)
data = numpy.frombuffer(buf, dtype=numpy.uint8)
data = data.reshape(num_images, rows, cols, 1)
data = data.astype('float32') / 255.0
return data |
Extract the labels into a 1D uint8 numpy array [index].
def extract_labels(filename):
"""Extract the labels into a 1D uint8 numpy array [index]."""
with gzip.open(filename) as bytestream:
magic = _read32(bytestream)
if magic != 2049:
raise ValueError(
'Invalid magic number %d in MNIST label file: %s' %
(magic, filename))
num_items = _read32(bytestream)
buf = bytestream.read(num_items)
labels = numpy.frombuffer(buf, dtype=numpy.uint8)
return labels |
When a dependency of a class is not available, create a dummy class which throws ImportError when used.
Args:
klass (str): name of the class.
dependency (str): name of the dependency.
Returns:
class: a class object
def create_dummy_class(klass, dependency):
"""
When a dependency of a class is not available, create a dummy class which throws ImportError when used.
Args:
klass (str): name of the class.
dependency (str): name of the dependency.
Returns:
class: a class object
"""
assert not building_rtfd()
class _DummyMetaClass(type):
# throw error on class attribute access
def __getattr__(_, __):
raise AttributeError("Cannot import '{}', therefore '{}' is not available".format(dependency, klass))
@six.add_metaclass(_DummyMetaClass)
class _Dummy(object):
# throw error on constructor
def __init__(self, *args, **kwargs):
raise ImportError("Cannot import '{}', therefore '{}' is not available".format(dependency, klass))
return _Dummy |
When a dependency of a function is not available, create a dummy function which throws ImportError when used.
Args:
func (str): name of the function.
dependency (str or list[str]): name(s) of the dependency.
Returns:
function: a function object
def create_dummy_func(func, dependency):
"""
When a dependency of a function is not available, create a dummy function which throws ImportError when used.
Args:
func (str): name of the function.
dependency (str or list[str]): name(s) of the dependency.
Returns:
function: a function object
"""
assert not building_rtfd()
if isinstance(dependency, (list, tuple)):
dependency = ','.join(dependency)
def _dummy(*args, **kwargs):
raise ImportError("Cannot import '{}', therefore '{}' is not available".format(dependency, func))
return _dummy |
Log deprecation warning.
Args:
name (str): name of the deprecated item.
text (str, optional): information about the deprecation.
eos (str, optional): end of service date such as "YYYY-MM-DD".
def log_deprecated(name="", text="", eos=""):
"""
Log deprecation warning.
Args:
name (str): name of the deprecated item.
text (str, optional): information about the deprecation.
eos (str, optional): end of service date such as "YYYY-MM-DD".
"""
assert name or text
if eos:
eos = "after " + datetime(*map(int, eos.split("-"))).strftime("%d %b")
if name:
if eos:
warn_msg = "%s will be deprecated %s. %s" % (name, eos, text)
else:
warn_msg = "%s was deprecated. %s" % (name, text)
else:
warn_msg = text
if eos:
warn_msg += " Legacy period ends %s" % eos
logger.warn("[Deprecated] " + warn_msg) |
Args:
text, eos: same as :func:`log_deprecated`.
Returns:
a decorator which deprecates the function.
Example:
.. code-block:: python
@deprecated("Explanation of what to do instead.", "2017-11-4")
def foo(...):
pass
def deprecated(text="", eos=""):
"""
Args:
text, eos: same as :func:`log_deprecated`.
Returns:
a decorator which deprecates the function.
Example:
.. code-block:: python
@deprecated("Explanation of what to do instead.", "2017-11-4")
def foo(...):
pass
"""
def get_location():
import inspect
frame = inspect.currentframe()
if frame:
callstack = inspect.getouterframes(frame)[-1]
return '%s:%i' % (callstack[1], callstack[2])
else:
stack = inspect.stack(0)
entry = stack[2]
return '%s:%i' % (entry[1], entry[2])
def deprecated_inner(func):
@functools.wraps(func)
def new_func(*args, **kwargs):
name = "{} [{}]".format(func.__name__, get_location())
log_deprecated(name, text, eos)
return func(*args, **kwargs)
return new_func
return deprecated_inner |
Clear the queue, then call dataflow.__iter__() again and fill into the queue.
def refill_queue(self):
"""
Clear the queue, then call dataflow.__iter__() again and fill into the queue.
"""
self.thread.pause() # pause enqueue
opt = tfv1.RunOptions()
opt.timeout_in_ms = 2000 # 2s
sess = tfv1.get_default_session()
# dequeue until empty
try:
while True:
sess.run(self._dequeue_op, options=opt)
except tf.errors.DeadlineExceededError:
pass
# reset dataflow, start thread
self.thread.reinitialize_dataflow()
self.thread.resume() |
Create a hook-only callback which maintain EMA of the queue size.
Also tf.summary.scalar the EMA.
def _create_ema_callback(self):
"""
Create a hook-only callback which maintain EMA of the queue size.
Also tf.summary.scalar the EMA.
"""
with self.cached_name_scope():
# in TF there is no API to get queue capacity, so we can only summary the size
size = tf.cast(self.queue.size(), tf.float32, name='queue_size')
size_ema_op = add_moving_summary(size, collection=None, decay=0.5)[0].op
ret = RunOp(
lambda: size_ema_op,
run_before=False,
run_as_trigger=False,
run_step=True)
ret.name_scope = "InputSource/EMA"
return ret |
shapes except for the batch dimension
def _setup(self, inputs):
logger.info("Setting up the queue for CPU prefetching ...")
self.input_placehdrs = [build_or_reuse_placeholder(v) for v in inputs]
assert len(self.input_placehdrs) > 0, \
"BatchQueueInput has to be used with some input signature!"
# prepare placeholders without the first dimension
placehdrs_nobatch = []
for p in self.input_placehdrs:
placehdrs_nobatch.append(tfv1.placeholder(
dtype=p.dtype, shape=p.get_shape().as_list()[1:],
name=get_op_tensor_name(p.name)[0] + '-nobatch'))
# dequeue_many requires fully-defined shapes
shape_err = "Use of BatchQueueInput requires inputs to have fully-defined "
"shapes except for the batch dimension"
shapes = []
for p in placehdrs_nobatch:
assert p.get_shape().is_fully_defined(), shape_err
shapes.append(p.get_shape())
with self.cached_name_scope():
if self.queue is None:
self.queue = tf.FIFOQueue(
3000, [x.dtype for x in self.input_placehdrs],
shapes=shapes,
name='input_queue')
for shp in self.queue.shapes:
assert shp.is_fully_defined(), shape_err
self.thread = EnqueueThread(self.queue, self._inf_ds, placehdrs_nobatch) |
Wrap a dataflow to tf.data.Dataset.
This function will also reset the dataflow.
If the dataflow itself is finite, the returned dataset is also finite.
Therefore, if used for training, you'll need to add `.repeat()` on the returned
dataset.
Args:
df (DataFlow): a dataflow which produces lists
types([tf.DType]): list of types
Returns:
(tf.data.Dataset)
def dataflow_to_dataset(df, types):
"""
Wrap a dataflow to tf.data.Dataset.
This function will also reset the dataflow.
If the dataflow itself is finite, the returned dataset is also finite.
Therefore, if used for training, you'll need to add `.repeat()` on the returned
dataset.
Args:
df (DataFlow): a dataflow which produces lists
types([tf.DType]): list of types
Returns:
(tf.data.Dataset)
"""
# TODO theoretically it can support dict
assert isinstance(df, DataFlow), df
assert isinstance(types, (list, tuple)), types
df = MapData(df, lambda dp: tuple(dp))
df.reset_state()
ds = tf.data.Dataset.from_generator(
df.get_data, tuple(types))
return ds |
Returns:
OnlinePredictor: the nth predictor on the nth tower.
def get_predictor(self, n):
"""
Returns:
OnlinePredictor: the nth predictor on the nth tower.
"""
l = len(self.predictors)
if n >= l:
logger.warn("n > #towers, will assign predictor to GPU by round-robin")
return [self.predictors[k % l] for k in range(n)] |
Compute pairwise intersection areas between boxes.
Args:
boxes1: a numpy array with shape [N, 4] holding N boxes
boxes2: a numpy array with shape [M, 4] holding M boxes
Returns:
a numpy array with shape [N*M] representing pairwise intersection area
def intersection(boxes1, boxes2):
"""Compute pairwise intersection areas between boxes.
Args:
boxes1: a numpy array with shape [N, 4] holding N boxes
boxes2: a numpy array with shape [M, 4] holding M boxes
Returns:
a numpy array with shape [N*M] representing pairwise intersection area
"""
[y_min1, x_min1, y_max1, x_max1] = np.split(boxes1, 4, axis=1)
[y_min2, x_min2, y_max2, x_max2] = np.split(boxes2, 4, axis=1)
all_pairs_min_ymax = np.minimum(y_max1, np.transpose(y_max2))
all_pairs_max_ymin = np.maximum(y_min1, np.transpose(y_min2))
intersect_heights = np.maximum(
np.zeros(all_pairs_max_ymin.shape, dtype='f4'),
all_pairs_min_ymax - all_pairs_max_ymin)
all_pairs_min_xmax = np.minimum(x_max1, np.transpose(x_max2))
all_pairs_max_xmin = np.maximum(x_min1, np.transpose(x_min2))
intersect_widths = np.maximum(
np.zeros(all_pairs_max_xmin.shape, dtype='f4'),
all_pairs_min_xmax - all_pairs_max_xmin)
return intersect_heights * intersect_widths |
Computes pairwise intersection-over-union between box collections.
Args:
boxes1: a numpy array with shape [N, 4] holding N boxes.
boxes2: a numpy array with shape [M, 4] holding M boxes.
Returns:
a numpy array with shape [N, M] representing pairwise iou scores.
def iou(boxes1, boxes2):
"""Computes pairwise intersection-over-union between box collections.
Args:
boxes1: a numpy array with shape [N, 4] holding N boxes.
boxes2: a numpy array with shape [M, 4] holding M boxes.
Returns:
a numpy array with shape [N, M] representing pairwise iou scores.
"""
intersect = intersection(boxes1, boxes2)
area1 = area(boxes1)
area2 = area(boxes2)
union = np.expand_dims(area1, axis=1) + np.expand_dims(
area2, axis=0) - intersect
return intersect / union |
Computes pairwise intersection-over-area between box collections.
Intersection-over-area (ioa) between two boxes box1 and box2 is defined as
their intersection area over box2's area. Note that ioa is not symmetric,
that is, IOA(box1, box2) != IOA(box2, box1).
Args:
boxes1: a numpy array with shape [N, 4] holding N boxes.
boxes2: a numpy array with shape [M, 4] holding N boxes.
Returns:
a numpy array with shape [N, M] representing pairwise ioa scores.
def ioa(boxes1, boxes2):
"""Computes pairwise intersection-over-area between box collections.
Intersection-over-area (ioa) between two boxes box1 and box2 is defined as
their intersection area over box2's area. Note that ioa is not symmetric,
that is, IOA(box1, box2) != IOA(box2, box1).
Args:
boxes1: a numpy array with shape [N, 4] holding N boxes.
boxes2: a numpy array with shape [M, 4] holding N boxes.
Returns:
a numpy array with shape [N, M] representing pairwise ioa scores.
"""
intersect = intersection(boxes1, boxes2)
inv_areas = np.expand_dims(1.0 / area(boxes2), axis=0)
return intersect * inv_areas |
Download the data from Marlin's website, unless it's already here.
def maybe_download(url, work_directory):
"""Download the data from Marlin's website, unless it's already here."""
filename = url.split("/")[-1]
filepath = os.path.join(work_directory, filename)
if not os.path.exists(filepath):
logger.info("Downloading to {}...".format(filepath))
download(url, work_directory)
return filepath |
Returns:
dict: {cls_number: synset_id}
def get_synset_1000(self):
"""
Returns:
dict: {cls_number: synset_id}
"""
fname = os.path.join(self.dir, 'synsets.txt')
assert os.path.isfile(fname)
lines = [x.strip() for x in open(fname).readlines()]
return dict(enumerate(lines)) |
Args:
name (str): 'train' or 'val' or 'test'
dir_structure (str): same as in :meth:`ILSVRC12.__init__()`.
Returns:
list: list of (image filename, label)
def get_image_list(self, name, dir_structure='original'):
"""
Args:
name (str): 'train' or 'val' or 'test'
dir_structure (str): same as in :meth:`ILSVRC12.__init__()`.
Returns:
list: list of (image filename, label)
"""
assert name in ['train', 'val', 'test']
assert dir_structure in ['original', 'train']
add_label_to_fname = (name != 'train' and dir_structure != 'original')
if add_label_to_fname:
synset = self.get_synset_1000()
fname = os.path.join(self.dir, name + '.txt')
assert os.path.isfile(fname), fname
with open(fname) as f:
ret = []
for line in f.readlines():
name, cls = line.strip().split()
cls = int(cls)
if add_label_to_fname:
name = os.path.join(synset[cls], name)
ret.append((name.strip(), cls))
assert len(ret), fname
return ret |
Args:
size (tuple): image size in (h, w). Defaults to (256, 256).
Returns:
np.ndarray: per-pixel mean of shape (h, w, 3 (BGR)) in range [0, 255].
def get_per_pixel_mean(self, size=None):
"""
Args:
size (tuple): image size in (h, w). Defaults to (256, 256).
Returns:
np.ndarray: per-pixel mean of shape (h, w, 3 (BGR)) in range [0, 255].
"""
if self.caffepb is None:
self.caffepb = get_caffe_pb()
obj = self.caffepb.BlobProto()
mean_file = os.path.join(self.dir, 'imagenet_mean.binaryproto')
with open(mean_file, 'rb') as f:
obj.ParseFromString(f.read())
arr = np.array(obj.data).reshape((3, 256, 256)).astype('float32')
arr = np.transpose(arr, [1, 2, 0])
if size is not None:
arr = cv2.resize(arr, size[::-1])
return arr |
Return the directory structure of "dir".
Args:
dir(str): something like '/path/to/imagenet/val'
Returns:
either 'train' or 'original'
def guess_dir_structure(dir):
"""
Return the directory structure of "dir".
Args:
dir(str): something like '/path/to/imagenet/val'
Returns:
either 'train' or 'original'
"""
subdir = os.listdir(dir)[0]
# find a subdir starting with 'n'
if subdir.startswith('n') and \
os.path.isdir(os.path.join(dir, subdir)):
dir_structure = 'train'
else:
dir_structure = 'original'
logger.info(
"[ILSVRC12] Assuming directory {} has '{}' structure.".format(
dir, dir_structure))
return dir_structure |
Args:
json_file (str): path to the results json file in coco format
Returns:
dict: the evaluation metrics
def print_coco_metrics(self, json_file):
"""
Args:
json_file (str): path to the results json file in coco format
Returns:
dict: the evaluation metrics
"""
from pycocotools.cocoeval import COCOeval
ret = {}
cocoDt = self.coco.loadRes(json_file)
cocoEval = COCOeval(self.coco, cocoDt, 'bbox')
cocoEval.evaluate()
cocoEval.accumulate()
cocoEval.summarize()
fields = ['IoU=0.5:0.95', 'IoU=0.5', 'IoU=0.75', 'small', 'medium', 'large']
for k in range(6):
ret['mAP(bbox)/' + fields[k]] = cocoEval.stats[k]
json_obj = json.load(open(json_file))
if len(json_obj) > 0 and 'segmentation' in json_obj[0]:
cocoEval = COCOeval(self.coco, cocoDt, 'segm')
cocoEval.evaluate()
cocoEval.accumulate()
cocoEval.summarize()
for k in range(6):
ret['mAP(segm)/' + fields[k]] = cocoEval.stats[k]
return ret |
Args:
add_gt: whether to add ground truth bounding box annotations to the dicts
add_mask: whether to also add ground truth mask
Returns:
a list of dict, each has keys including:
'image_id', 'file_name',
and (if add_gt is True) 'boxes', 'class', 'is_crowd', and optionally
'segmentation'.
def load(self, add_gt=True, add_mask=False):
"""
Args:
add_gt: whether to add ground truth bounding box annotations to the dicts
add_mask: whether to also add ground truth mask
Returns:
a list of dict, each has keys including:
'image_id', 'file_name',
and (if add_gt is True) 'boxes', 'class', 'is_crowd', and optionally
'segmentation'.
"""
if add_mask:
assert add_gt
with timed_operation('Load Groundtruth Boxes for {}'.format(self.name)):
img_ids = self.coco.getImgIds()
img_ids.sort()
# list of dict, each has keys: height,width,id,file_name
imgs = self.coco.loadImgs(img_ids)
for img in tqdm.tqdm(imgs):
img['image_id'] = img.pop('id')
self._use_absolute_file_name(img)
if add_gt:
self._add_detection_gt(img, add_mask)
return imgs |
Change relative filename to abosolute file name.
def _use_absolute_file_name(self, img):
"""
Change relative filename to abosolute file name.
"""
img['file_name'] = os.path.join(
self._imgdir, img['file_name'])
assert os.path.isfile(img['file_name']), img['file_name'] |
Add 'boxes', 'class', 'is_crowd' of this image to the dict, used by detection.
If add_mask is True, also add 'segmentation' in coco poly format.
def _add_detection_gt(self, img, add_mask):
"""
Add 'boxes', 'class', 'is_crowd' of this image to the dict, used by detection.
If add_mask is True, also add 'segmentation' in coco poly format.
"""
# ann_ids = self.coco.getAnnIds(imgIds=img['image_id'])
# objs = self.coco.loadAnns(ann_ids)
objs = self.coco.imgToAnns[img['image_id']] # equivalent but faster than the above two lines
# clean-up boxes
valid_objs = []
width = img.pop('width')
height = img.pop('height')
for objid, obj in enumerate(objs):
if obj.get('ignore', 0) == 1:
continue
x1, y1, w, h = obj['bbox']
# bbox is originally in float
# x1/y1 means upper-left corner and w/h means true w/h. This can be verified by segmentation pixels.
# But we do make an assumption here that (0.0, 0.0) is upper-left corner of the first pixel
x1 = np.clip(float(x1), 0, width)
y1 = np.clip(float(y1), 0, height)
w = np.clip(float(x1 + w), 0, width) - x1
h = np.clip(float(y1 + h), 0, height) - y1
# Require non-zero seg area and more than 1x1 box size
if obj['area'] > 1 and w > 0 and h > 0 and w * h >= 4:
obj['bbox'] = [x1, y1, x1 + w, y1 + h]
valid_objs.append(obj)
if add_mask:
segs = obj['segmentation']
if not isinstance(segs, list):
assert obj['iscrowd'] == 1
obj['segmentation'] = None
else:
valid_segs = [np.asarray(p).reshape(-1, 2).astype('float32') for p in segs if len(p) >= 6]
if len(valid_segs) == 0:
logger.error("Object {} in image {} has no valid polygons!".format(objid, img['file_name']))
elif len(valid_segs) < len(segs):
logger.warn("Object {} in image {} has invalid polygons!".format(objid, img['file_name']))
obj['segmentation'] = valid_segs
# all geometrically-valid boxes are returned
boxes = np.asarray([obj['bbox'] for obj in valid_objs], dtype='float32') # (n, 4)
cls = np.asarray([
self.COCO_id_to_category_id[obj['category_id']]
for obj in valid_objs], dtype='int32') # (n,)
is_crowd = np.asarray([obj['iscrowd'] for obj in valid_objs], dtype='int8')
# add the keys
img['boxes'] = boxes # nx4
img['class'] = cls # n, always >0
img['is_crowd'] = is_crowd # n,
if add_mask:
# also required to be float32
img['segmentation'] = [
obj['segmentation'] for obj in valid_objs] |
Load and merges several instance files together.
Returns the same format as :meth:`COCODetection.load`.
def load_many(basedir, names, add_gt=True, add_mask=False):
"""
Load and merges several instance files together.
Returns the same format as :meth:`COCODetection.load`.
"""
if not isinstance(names, (list, tuple)):
names = [names]
ret = []
for n in names:
coco = COCODetection(basedir, n)
ret.extend(coco.load(add_gt, add_mask=add_mask))
return ret |
Args:
names (list[str]): name of the training datasets, e.g. ['train2014', 'valminusminival2014']
Returns:
roidbs (list[dict]):
Produce "roidbs" as a list of dict, each dict corresponds to one image with k>=0 instances.
and the following keys are expected for training:
file_name: str, full path to the image
boxes: numpy array of kx4 floats, each row is [x1, y1, x2, y2]
class: numpy array of k integers, in the range of [1, #categories], NOT [0, #categories)
is_crowd: k booleans. Use k False if you don't know what it means.
segmentation: k lists of numpy arrays (one for each instance).
Each list of numpy arrays corresponds to the mask for one instance.
Each numpy array in the list is a polygon of shape Nx2,
because one mask can be represented by N polygons.
If your segmentation annotations are originally masks rather than polygons,
either convert it, or the augmentation will need to be changed or skipped accordingly.
Include this field only if training Mask R-CNN.
def load_training_roidbs(self, names):
"""
Args:
names (list[str]): name of the training datasets, e.g. ['train2014', 'valminusminival2014']
Returns:
roidbs (list[dict]):
Produce "roidbs" as a list of dict, each dict corresponds to one image with k>=0 instances.
and the following keys are expected for training:
file_name: str, full path to the image
boxes: numpy array of kx4 floats, each row is [x1, y1, x2, y2]
class: numpy array of k integers, in the range of [1, #categories], NOT [0, #categories)
is_crowd: k booleans. Use k False if you don't know what it means.
segmentation: k lists of numpy arrays (one for each instance).
Each list of numpy arrays corresponds to the mask for one instance.
Each numpy array in the list is a polygon of shape Nx2,
because one mask can be represented by N polygons.
If your segmentation annotations are originally masks rather than polygons,
either convert it, or the augmentation will need to be changed or skipped accordingly.
Include this field only if training Mask R-CNN.
"""
return COCODetection.load_many(
cfg.DATA.BASEDIR, names, add_gt=True, add_mask=cfg.MODE_MASK) |
Args:
name (str): name of one inference dataset, e.g. 'minival2014'
Returns:
roidbs (list[dict]):
Each dict corresponds to one image to run inference on. The
following keys in the dict are expected:
file_name (str): full path to the image
image_id (str): an id for the image. The inference results will be stored with this id.
def load_inference_roidbs(self, name):
"""
Args:
name (str): name of one inference dataset, e.g. 'minival2014'
Returns:
roidbs (list[dict]):
Each dict corresponds to one image to run inference on. The
following keys in the dict are expected:
file_name (str): full path to the image
image_id (str): an id for the image. The inference results will be stored with this id.
"""
return COCODetection.load_many(cfg.DATA.BASEDIR, name, add_gt=False) |
Args:
results (list[dict]): the inference results as dicts.
Each dict corresponds to one __instance__. It contains the following keys:
image_id (str): the id that matches `load_inference_roidbs`.
category_id (int): the category prediction, in range [1, #category]
bbox (list[float]): x1, y1, x2, y2
score (float):
segmentation: the segmentation mask in COCO's rle format.
dataset (str): the name of the dataset to evaluate.
output (str): the output file to optionally save the results to.
Returns:
dict: the evaluation results.
def eval_or_save_inference_results(self, results, dataset, output=None):
"""
Args:
results (list[dict]): the inference results as dicts.
Each dict corresponds to one __instance__. It contains the following keys:
image_id (str): the id that matches `load_inference_roidbs`.
category_id (int): the category prediction, in range [1, #category]
bbox (list[float]): x1, y1, x2, y2
score (float):
segmentation: the segmentation mask in COCO's rle format.
dataset (str): the name of the dataset to evaluate.
output (str): the output file to optionally save the results to.
Returns:
dict: the evaluation results.
"""
continuous_id_to_COCO_id = {v: k for k, v in COCODetection.COCO_id_to_category_id.items()}
for res in results:
# convert to COCO's incontinuous category id
res['category_id'] = continuous_id_to_COCO_id[res['category_id']]
# COCO expects results in xywh format
box = res['bbox']
box[2] -= box[0]
box[3] -= box[1]
res['bbox'] = [round(float(x), 3) for x in box]
assert output is not None, "COCO evaluation requires an output file!"
with open(output, 'w') as f:
json.dump(results, f)
if len(results):
# sometimes may crash if the results are empty?
return COCODetection(cfg.DATA.BASEDIR, dataset).print_coco_metrics(output)
else:
return {} |
Surround a context with a timer.
Args:
msg(str): the log to print.
log_start(bool): whether to print also at the beginning.
Example:
.. code-block:: python
with timed_operation('Good Stuff'):
time.sleep(1)
Will print:
.. code-block:: python
Good stuff finished, time:1sec.
def timed_operation(msg, log_start=False):
"""
Surround a context with a timer.
Args:
msg(str): the log to print.
log_start(bool): whether to print also at the beginning.
Example:
.. code-block:: python
with timed_operation('Good Stuff'):
time.sleep(1)
Will print:
.. code-block:: python
Good stuff finished, time:1sec.
"""
assert len(msg)
if log_start:
logger.info('Start {} ...'.format(msg))
start = timer()
yield
msg = msg[0].upper() + msg[1:]
logger.info('{} finished, time:{:.4f} sec.'.format(
msg, timer() - start)) |
A context which add the time spent inside to TotalTimer.
def total_timer(msg):
""" A context which add the time spent inside to TotalTimer. """
start = timer()
yield
t = timer() - start
_TOTAL_TIMER_DATA[msg].feed(t) |
Print the content of the TotalTimer, if it's not empty. This function will automatically get
called when program exits.
def print_total_timer():
"""
Print the content of the TotalTimer, if it's not empty. This function will automatically get
called when program exits.
"""
if len(_TOTAL_TIMER_DATA) == 0:
return
for k, v in six.iteritems(_TOTAL_TIMER_DATA):
logger.info("Total Time: {} -> {:.2f} sec, {} times, {:.3g} sec/time".format(
k, v.sum, v.count, v.average)) |
Will reset state of each augmentor
def reset_state(self):
""" Will reset state of each augmentor """
super(AugmentorList, self).reset_state()
for a in self.augmentors:
a.reset_state() |
Make sure processes terminate when main process exit.
Args:
proc (multiprocessing.Process or list)
def ensure_proc_terminate(proc):
"""
Make sure processes terminate when main process exit.
Args:
proc (multiprocessing.Process or list)
"""
if isinstance(proc, list):
for p in proc:
ensure_proc_terminate(p)
return
def stop_proc_by_weak_ref(ref):
proc = ref()
if proc is None:
return
if not proc.is_alive():
return
proc.terminate()
proc.join()
assert isinstance(proc, mp.Process)
atexit.register(stop_proc_by_weak_ref, weakref.ref(proc)) |
Set the "death signal" of the current process, so that
the current process will be cleaned with guarantee
in case the parent dies accidentally.
def enable_death_signal(_warn=True):
"""
Set the "death signal" of the current process, so that
the current process will be cleaned with guarantee
in case the parent dies accidentally.
"""
if platform.system() != 'Linux':
return
try:
import prctl # pip install python-prctl
except ImportError:
if _warn:
log_once('"import prctl" failed! Install python-prctl so that processes can be cleaned with guarantee.',
'warn')
return
else:
assert hasattr(prctl, 'set_pdeathsig'), \
"prctl.set_pdeathsig does not exist! Note that you need to install 'python-prctl' instead of 'prctl'."
# is SIGHUP a good choice?
prctl.set_pdeathsig(signal.SIGHUP) |
Returns:
If called in main thread, returns a context where ``SIGINT`` is ignored, and yield True.
Otherwise yield False.
def mask_sigint():
"""
Returns:
If called in main thread, returns a context where ``SIGINT`` is ignored, and yield True.
Otherwise yield False.
"""
if is_main_thread():
sigint_handler = signal.signal(signal.SIGINT, signal.SIG_IGN)
yield True
signal.signal(signal.SIGINT, sigint_handler)
else:
yield False |
Start process(es) with SIGINT ignored.
Args:
proc: (mp.Process or list)
Note:
The signal mask is only applied when called from main thread.
def start_proc_mask_signal(proc):
"""
Start process(es) with SIGINT ignored.
Args:
proc: (mp.Process or list)
Note:
The signal mask is only applied when called from main thread.
"""
if not isinstance(proc, list):
proc = [proc]
with mask_sigint():
for p in proc:
if isinstance(p, mp.Process):
if sys.version_info < (3, 4) or mp.get_start_method() == 'fork':
log_once(
"Starting a process with 'fork' method is not safe and may consume unnecessary extra memory."
" Use 'forkserver' method (available after Py3.4) instead if you run into any issues. "
"See https://docs.python.org/3/library/multiprocessing.html#contexts-and-start-methods",
'warn') # noqa
p.start() |
Execute a command with timeout, and return STDOUT and STDERR
Args:
cmd(str): the command to execute.
timeout(float): timeout in seconds.
Returns:
output(bytes), retcode(int). If timeout, retcode is -1.
def subproc_call(cmd, timeout=None):
"""
Execute a command with timeout, and return STDOUT and STDERR
Args:
cmd(str): the command to execute.
timeout(float): timeout in seconds.
Returns:
output(bytes), retcode(int). If timeout, retcode is -1.
"""
try:
output = subprocess.check_output(
cmd, stderr=subprocess.STDOUT,
shell=True, timeout=timeout)
return output, 0
except subprocess.TimeoutExpired as e:
logger.warn("Command '{}' timeout!".format(cmd))
logger.warn(e.output.decode('utf-8'))
return e.output, -1
except subprocess.CalledProcessError as e:
logger.warn("Command '{}' failed, return code={}".format(cmd, e.returncode))
logger.warn(e.output.decode('utf-8'))
return e.output, e.returncode
except Exception:
logger.warn("Command '{}' failed to run.".format(cmd))
return "", -2 |
Put obj to queue, but will give up when the thread is stopped
def queue_put_stoppable(self, q, obj):
""" Put obj to queue, but will give up when the thread is stopped"""
while not self.stopped():
try:
q.put(obj, timeout=5)
break
except queue.Full:
pass |
Take obj from queue, but will give up when the thread is stopped
def queue_get_stoppable(self, q):
""" Take obj from queue, but will give up when the thread is stopped"""
while not self.stopped():
try:
return q.get(timeout=5)
except queue.Empty:
pass |
Args:
rank(int): rank of th element. All elements must have different ranks.
val: an object
def put(self, rank, val):
"""
Args:
rank(int): rank of th element. All elements must have different ranks.
val: an object
"""
idx = bisect.bisect(self.ranks, rank)
self.ranks.insert(idx, rank)
self.data.insert(idx, val) |
Visualize use weights in convolution filters.
Args:
filters: tensor containing the weights [H,W,Cin,Cout]
name: label for tensorboard
Returns:
image of all weight
def visualize_conv_weights(filters, name):
"""Visualize use weights in convolution filters.
Args:
filters: tensor containing the weights [H,W,Cin,Cout]
name: label for tensorboard
Returns:
image of all weight
"""
with tf.name_scope('visualize_w_' + name):
filters = tf.transpose(filters, (3, 2, 0, 1)) # [h, w, cin, cout] -> [cout, cin, h, w]
filters = tf.unstack(filters) # --> cout * [cin, h, w]
filters = tf.concat(filters, 1) # --> [cin, cout * h, w]
filters = tf.unstack(filters) # --> cin * [cout * h, w]
filters = tf.concat(filters, 1) # --> [cout * h, cin * w]
filters = tf.expand_dims(filters, 0)
filters = tf.expand_dims(filters, -1)
tf.summary.image('visualize_w_' + name, filters) |
Visualize activations for convolution layers.
Remarks:
This tries to place all activations into a square.
Args:
activation: tensor with the activation [B,H,W,C]
name: label for tensorboard
Returns:
image of almost all activations
def visualize_conv_activations(activation, name):
"""Visualize activations for convolution layers.
Remarks:
This tries to place all activations into a square.
Args:
activation: tensor with the activation [B,H,W,C]
name: label for tensorboard
Returns:
image of almost all activations
"""
import math
with tf.name_scope('visualize_act_' + name):
_, h, w, c = activation.get_shape().as_list()
rows = []
c_per_row = int(math.sqrt(c))
for y in range(0, c - c_per_row, c_per_row):
row = activation[:, :, :, y:y + c_per_row] # [?, H, W, 32] --> [?, H, W, 5]
cols = tf.unstack(row, axis=3) # [?, H, W, 5] --> 5 * [?, H, W]
row = tf.concat(cols, 1)
rows.append(row)
viz = tf.concat(rows, 2)
tf.summary.image('visualize_act_' + name, tf.expand_dims(viz, -1)) |
Make the static shape of a tensor less specific.
If you want to feed to a tensor, the shape of the feed value must match
the tensor's static shape. This function creates a placeholder which
defaults to x if not fed, but has a less specific static shape than x.
See also `tensorflow#5680 <https://github.com/tensorflow/tensorflow/issues/5680>`_.
Args:
x: a tensor
axis(int or list of ints): these axes of ``x.get_shape()`` will become
None in the output.
name(str): name of the output tensor
Returns:
a tensor equal to x, but shape information is partially cleared.
def shapeless_placeholder(x, axis, name):
"""
Make the static shape of a tensor less specific.
If you want to feed to a tensor, the shape of the feed value must match
the tensor's static shape. This function creates a placeholder which
defaults to x if not fed, but has a less specific static shape than x.
See also `tensorflow#5680 <https://github.com/tensorflow/tensorflow/issues/5680>`_.
Args:
x: a tensor
axis(int or list of ints): these axes of ``x.get_shape()`` will become
None in the output.
name(str): name of the output tensor
Returns:
a tensor equal to x, but shape information is partially cleared.
"""
shp = x.get_shape().as_list()
if not isinstance(axis, list):
axis = [axis]
for a in axis:
if shp[a] is None:
raise ValueError("Axis {} of shape {} is already unknown!".format(a, shp))
shp[a] = None
x = tf.placeholder_with_default(x, shape=shp, name=name)
return x |
Estimate H(x|s) ~= -E_{x \sim P(x|s)}[\log Q(x|s)], where x are samples, and Q is parameterized by vec.
def entropy_from_samples(samples, vec):
"""
Estimate H(x|s) ~= -E_{x \sim P(x|s)}[\log Q(x|s)], where x are samples, and Q is parameterized by vec.
"""
samples_cat = tf.argmax(samples[:, :NUM_CLASS], axis=1, output_type=tf.int32)
samples_uniform = samples[:, NUM_CLASS:]
cat, uniform = get_distributions(vec[:, :NUM_CLASS], vec[:, NUM_CLASS:])
def neg_logprob(dist, sample, name):
nll = -dist.log_prob(sample)
# average over batch
return tf.reduce_sum(tf.reduce_mean(nll, axis=0), name=name)
entropies = [neg_logprob(cat, samples_cat, 'nll_cat'),
neg_logprob(uniform, samples_uniform, 'nll_uniform')]
return entropies |
OpenAI official code actually models the "uniform" latent code as
a Gaussian distribution, but obtain the samples from a uniform distribution.
def sample_prior(batch_size):
cat, _ = get_distributions(DIST_PRIOR_PARAM[:NUM_CLASS], DIST_PRIOR_PARAM[NUM_CLASS:])
sample_cat = tf.one_hot(cat.sample(batch_size), NUM_CLASS)
"""
OpenAI official code actually models the "uniform" latent code as
a Gaussian distribution, but obtain the samples from a uniform distribution.
"""
sample_uni = tf.random_uniform([batch_size, NUM_UNIFORM], -1, 1)
samples = tf.concat([sample_cat, sample_uni], axis=1)
return samples |
Mutual information between x (i.e. zc in this case) and some
information s (the generated samples in this case):
I(x;s) = H(x) - H(x|s)
= H(x) + E[\log P(x|s)]
The distribution from which zc is sampled, in this case, is set to a fixed prior already.
So the first term is a constant.
For the second term, we can maximize its variational lower bound:
E_{x \sim P(x|s)}[\log Q(x|s)]
where Q(x|s) is a proposal distribution to approximate P(x|s).
Here, Q(x|s) is assumed to be a distribution which shares the form
of P, and whose parameters are predicted by the discriminator network.
def build_graph(self, real_sample):
real_sample = tf.expand_dims(real_sample, -1)
# sample the latent code:
zc = shapeless_placeholder(sample_prior(BATCH), 0, name='z_code')
z_noise = shapeless_placeholder(
tf.random_uniform([BATCH, NOISE_DIM], -1, 1), 0, name='z_noise')
z = tf.concat([zc, z_noise], 1, name='z')
with argscope([Conv2D, Conv2DTranspose, FullyConnected],
kernel_initializer=tf.truncated_normal_initializer(stddev=0.02)):
with tf.variable_scope('gen'):
fake_sample = self.generator(z)
fake_sample_viz = tf.cast((fake_sample) * 255.0, tf.uint8, name='viz')
tf.summary.image('gen', fake_sample_viz, max_outputs=30)
# may need to investigate how bn stats should be updated across two discrim
with tf.variable_scope('discrim'):
real_pred, _ = self.discriminator(real_sample)
fake_pred, dist_param = self.discriminator(fake_sample)
"""
Mutual information between x (i.e. zc in this case) and some
information s (the generated samples in this case):
I(x;s) = H(x) - H(x|s)
= H(x) + E[\log P(x|s)]
The distribution from which zc is sampled, in this case, is set to a fixed prior already.
So the first term is a constant.
For the second term, we can maximize its variational lower bound:
E_{x \sim P(x|s)}[\log Q(x|s)]
where Q(x|s) is a proposal distribution to approximate P(x|s).
Here, Q(x|s) is assumed to be a distribution which shares the form
of P, and whose parameters are predicted by the discriminator network.
"""
with tf.name_scope("mutual_information"):
with tf.name_scope('prior_entropy'):
cat, uni = get_distributions(DIST_PRIOR_PARAM[:NUM_CLASS], DIST_PRIOR_PARAM[NUM_CLASS:])
ents = [cat.entropy(name='cat_entropy'), tf.reduce_sum(uni.entropy(), name='uni_entropy')]
entropy = tf.add_n(ents, name='total_entropy')
# Note that the entropy of prior is a constant. The paper mentioned it but didn't use it.
with tf.name_scope('conditional_entropy'):
cond_ents = entropy_from_samples(zc, dist_param)
cond_entropy = tf.add_n(cond_ents, name="total_entropy")
MI = tf.subtract(entropy, cond_entropy, name='mutual_information')
summary.add_moving_summary(entropy, cond_entropy, MI, *cond_ents)
# default GAN objective
self.build_losses(real_pred, fake_pred)
# subtract mutual information for latent factors (we want to maximize them)
self.g_loss = tf.subtract(self.g_loss, MI, name='total_g_loss')
self.d_loss = tf.subtract(self.d_loss, MI, name='total_d_loss')
summary.add_moving_summary(self.g_loss, self.d_loss)
# distinguish between variables of generator and discriminator updates
self.collect_variables() |
see "Dynamic Filter Networks" (NIPS 2016)
by Bert De Brabandere*, Xu Jia*, Tinne Tuytelaars and Luc Van Gool
Remarks:
This is the convolution version of a dynamic filter.
Args:
inputs : unfiltered input [b, h, w, 1] only grayscale images.
filters : learned filters of [b, k, k, 1] (dynamically generated by the network).
out_channel (int): number of output channel.
kernel_shape: (h, w) tuple or a int.
stride: (h, w) tuple or a int.
padding (str): 'valid' or 'same'. Case insensitive.
Returns
tf.Tensor named ``output``.
def DynamicConvFilter(inputs, filters, out_channel,
kernel_shape,
stride=1,
padding='SAME'):
""" see "Dynamic Filter Networks" (NIPS 2016)
by Bert De Brabandere*, Xu Jia*, Tinne Tuytelaars and Luc Van Gool
Remarks:
This is the convolution version of a dynamic filter.
Args:
inputs : unfiltered input [b, h, w, 1] only grayscale images.
filters : learned filters of [b, k, k, 1] (dynamically generated by the network).
out_channel (int): number of output channel.
kernel_shape: (h, w) tuple or a int.
stride: (h, w) tuple or a int.
padding (str): 'valid' or 'same'. Case insensitive.
Returns
tf.Tensor named ``output``.
"""
# tf.unstack only works with known batch_size :-(
batch_size, h, w, in_channel = inputs.get_shape().as_list()
stride = shape4d(stride)
inputs = tf.unstack(inputs)
filters = tf.reshape(filters, [batch_size] + shape2d(kernel_shape) + [in_channel, out_channel])
filters = tf.unstack(filters)
# this is ok as TF uses the cuda stream context
rsl = [tf.nn.conv2d(tf.reshape(d, [1, h, w, in_channel]),
tf.reshape(k, [kernel_shape, kernel_shape, in_channel, out_channel]),
stride, padding="SAME") for d, k in zip(inputs, filters)]
rsl = tf.concat(rsl, axis=0, name='output')
return rsl |
Estimate filters for convolution layers
Args:
theta: angle of filter
kernel_shape: size of each filter
Returns:
learned filter as [B, k, k, 1]
def _parameter_net(self, theta, kernel_shape=9):
"""Estimate filters for convolution layers
Args:
theta: angle of filter
kernel_shape: size of each filter
Returns:
learned filter as [B, k, k, 1]
"""
with argscope(FullyConnected, nl=tf.nn.leaky_relu):
net = FullyConnected('fc1', theta, 64)
net = FullyConnected('fc2', net, 128)
pred_filter = FullyConnected('fc3', net, kernel_shape ** 2, nl=tf.identity)
pred_filter = tf.reshape(pred_filter, [BATCH, kernel_shape, kernel_shape, 1], name="pred_filter")
logger.info('Parameter net output: {}'.format(pred_filter.get_shape().as_list()))
return pred_filter |
Implements a steerable Gaussian filter.
This function can be used to evaluate the first
directional derivative of an image, using the
method outlined in
W. T. Freeman and E. H. Adelson, "The Design
and Use of Steerable Filters", IEEE PAMI, 1991.
It evaluates the directional derivative of the input
image I, oriented at THETA degrees with respect to the
image rows. The standard deviation of the Gaussian kernel
is given by SIGMA (assumed to be equal to unity by default).
Args:
image: any input image (only one channel)
theta: orientation of filter [0, 2 * pi]
sigma (float, optional): standard derivation of Gaussian
filter_size (int, optional): filter support
Returns:
filtered image and the filter
def filter_with_theta(image, theta, sigma=1., filter_size=9):
"""Implements a steerable Gaussian filter.
This function can be used to evaluate the first
directional derivative of an image, using the
method outlined in
W. T. Freeman and E. H. Adelson, "The Design
and Use of Steerable Filters", IEEE PAMI, 1991.
It evaluates the directional derivative of the input
image I, oriented at THETA degrees with respect to the
image rows. The standard deviation of the Gaussian kernel
is given by SIGMA (assumed to be equal to unity by default).
Args:
image: any input image (only one channel)
theta: orientation of filter [0, 2 * pi]
sigma (float, optional): standard derivation of Gaussian
filter_size (int, optional): filter support
Returns:
filtered image and the filter
"""
x = np.arange(-filter_size // 2 + 1, filter_size // 2 + 1)
# 1D Gaussian
g = np.array([np.exp(-(x**2) / (2 * sigma**2))])
# first-derivative of 1D Gaussian
gp = np.array([-(x / sigma) * np.exp(-(x**2) / (2 * sigma**2))])
ix = convolve2d(image, -gp, mode='same', boundary='fill', fillvalue=0)
ix = convolve2d(ix, g.T, mode='same', boundary='fill', fillvalue=0)
iy = convolve2d(image, g, mode='same', boundary='fill', fillvalue=0)
iy = convolve2d(iy, -gp.T, mode='same', boundary='fill', fillvalue=0)
output = np.cos(theta) * ix + np.sin(theta) * iy
# np.cos(theta) * np.matmul(g.T, gp) + np.sin(theta) * np.matmul(gp.T, g)
gt_filter = np.matmul(g.T, gp)
gt_filter = np.cos(theta) * gt_filter + np.sin(theta) * gt_filter.T
return output, gt_filter |
Assign `self.g_vars` to the parameters under scope `g_scope`,
and same with `self.d_vars`.
def collect_variables(self, g_scope='gen', d_scope='discrim'):
"""
Assign `self.g_vars` to the parameters under scope `g_scope`,
and same with `self.d_vars`.
"""
self.g_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, g_scope)
assert self.g_vars
self.d_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, d_scope)
assert self.d_vars |
Build standard GAN loss and set `self.g_loss` and `self.d_loss`.
D and G play two-player minimax game with value function V(G,D)
min_G max _D V(D, G) = IE_{x ~ p_data} [log D(x)] + IE_{z ~ p_fake} [log (1 - D(G(z)))]
Args:
logits_real (tf.Tensor): discrim logits from real samples
logits_fake (tf.Tensor): discrim logits from fake samples produced by generator
def build_losses(self, logits_real, logits_fake):
"""
Build standard GAN loss and set `self.g_loss` and `self.d_loss`.
D and G play two-player minimax game with value function V(G,D)
min_G max _D V(D, G) = IE_{x ~ p_data} [log D(x)] + IE_{z ~ p_fake} [log (1 - D(G(z)))]
Args:
logits_real (tf.Tensor): discrim logits from real samples
logits_fake (tf.Tensor): discrim logits from fake samples produced by generator
"""
with tf.name_scope("GAN_loss"):
score_real = tf.sigmoid(logits_real)
score_fake = tf.sigmoid(logits_fake)
tf.summary.histogram('score-real', score_real)
tf.summary.histogram('score-fake', score_fake)
with tf.name_scope("discrim"):
d_loss_pos = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(
logits=logits_real, labels=tf.ones_like(logits_real)), name='loss_real')
d_loss_neg = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(
logits=logits_fake, labels=tf.zeros_like(logits_fake)), name='loss_fake')
d_pos_acc = tf.reduce_mean(tf.cast(score_real > 0.5, tf.float32), name='accuracy_real')
d_neg_acc = tf.reduce_mean(tf.cast(score_fake < 0.5, tf.float32), name='accuracy_fake')
d_accuracy = tf.add(.5 * d_pos_acc, .5 * d_neg_acc, name='accuracy')
self.d_loss = tf.add(.5 * d_loss_pos, .5 * d_loss_neg, name='loss')
with tf.name_scope("gen"):
self.g_loss = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(
logits=logits_fake, labels=tf.ones_like(logits_fake)), name='loss')
g_accuracy = tf.reduce_mean(tf.cast(score_fake > 0.5, tf.float32), name='accuracy')
add_moving_summary(self.g_loss, self.d_loss, d_accuracy, g_accuracy) |
We need to set tower_func because it's a TowerTrainer,
and only TowerTrainer supports automatic graph creation for inference during training.
If we don't care about inference during training, using tower_func is
not needed. Just calling model.build_graph directly is OK.
def _build_gan_trainer(self, input, model):
"""
We need to set tower_func because it's a TowerTrainer,
and only TowerTrainer supports automatic graph creation for inference during training.
If we don't care about inference during training, using tower_func is
not needed. Just calling model.build_graph directly is OK.
"""
# Build the graph
self.tower_func = TowerFuncWrapper(model.build_graph, model.get_input_signature())
with TowerContext('', is_training=True):
self.tower_func(*input.get_input_tensors())
opt = model.get_optimizer()
# Define the training iteration
# by default, run one d_min after one g_min
with tf.name_scope('optimize'):
g_min = opt.minimize(model.g_loss, var_list=model.g_vars, name='g_op')
with tf.control_dependencies([g_min]):
d_min = opt.minimize(model.d_loss, var_list=model.d_vars, name='d_op')
self.train_op = d_min |
After applying this decorator:
1. data_format becomes tf.layers style
2. nl becomes activation
3. initializers are renamed
4. positional args are transformed to corresponding kwargs, according to args_names
5. kwargs are mapped to tf.layers names if needed, by name_mapping
def convert_to_tflayer_args(args_names, name_mapping):
"""
After applying this decorator:
1. data_format becomes tf.layers style
2. nl becomes activation
3. initializers are renamed
4. positional args are transformed to corresponding kwargs, according to args_names
5. kwargs are mapped to tf.layers names if needed, by name_mapping
"""
def decorator(func):
@functools.wraps(func)
def decorated_func(inputs, *args, **kwargs):
kwargs = map_common_tfargs(kwargs)
posarg_dic = {}
assert len(args) <= len(args_names), \
"Please use kwargs instead of positional args to call this model, " \
"except for the following arguments: {}".format(', '.join(args_names))
for pos_arg, name in zip(args, args_names):
posarg_dic[name] = pos_arg
ret = {}
for name, arg in six.iteritems(kwargs):
newname = name_mapping.get(name, None)
if newname is not None:
assert newname not in kwargs, \
"Argument {} and {} conflicts!".format(name, newname)
else:
newname = name
ret[newname] = arg
ret.update(posarg_dic) # Let pos arg overwrite kw arg, for argscope to work
return func(inputs, **ret)
return decorated_func
return decorator |
Args:
mapping(dict): an old -> new mapping for variable basename. e.g. {'kernel': 'W'}
Returns:
A context where the variables are renamed.
def rename_get_variable(mapping):
"""
Args:
mapping(dict): an old -> new mapping for variable basename. e.g. {'kernel': 'W'}
Returns:
A context where the variables are renamed.
"""
def custom_getter(getter, name, *args, **kwargs):
splits = name.split('/')
basename = splits[-1]
if basename in mapping:
basename = mapping[basename]
splits[-1] = basename
name = '/'.join(splits)
return getter(name, *args, **kwargs)
return custom_getter_scope(custom_getter) |
Apply a regularizer on trainable variables matching the regex, and print
the matched variables (only print once in multi-tower training).
In replicated mode, it will only regularize variables within the current tower.
If called under a TowerContext with `is_training==False`, this function returns a zero constant tensor.
Args:
regex (str): a regex to match variable names, e.g. "conv.*/W"
func: the regularization function, which takes a tensor and returns a scalar tensor.
E.g., ``tf.nn.l2_loss, tf.contrib.layers.l1_regularizer(0.001)``.
Returns:
tf.Tensor: a scalar, the total regularization cost.
Example:
.. code-block:: python
cost = cost + regularize_cost("fc.*/W", l2_regularizer(1e-5))
def regularize_cost(regex, func, name='regularize_cost'):
"""
Apply a regularizer on trainable variables matching the regex, and print
the matched variables (only print once in multi-tower training).
In replicated mode, it will only regularize variables within the current tower.
If called under a TowerContext with `is_training==False`, this function returns a zero constant tensor.
Args:
regex (str): a regex to match variable names, e.g. "conv.*/W"
func: the regularization function, which takes a tensor and returns a scalar tensor.
E.g., ``tf.nn.l2_loss, tf.contrib.layers.l1_regularizer(0.001)``.
Returns:
tf.Tensor: a scalar, the total regularization cost.
Example:
.. code-block:: python
cost = cost + regularize_cost("fc.*/W", l2_regularizer(1e-5))
"""
assert len(regex)
ctx = get_current_tower_context()
if not ctx.is_training:
# Currently cannot build the wd_cost correctly at inference,
# because ths vs_name used in inference can be '', therefore the
# variable filter will fail
return tf.constant(0, dtype=tf.float32, name='empty_' + name)
# If vars are shared, regularize all of them
# If vars are replicated, only regularize those in the current tower
if ctx.has_own_variables:
params = ctx.get_collection_in_tower(tfv1.GraphKeys.TRAINABLE_VARIABLES)
else:
params = tfv1.trainable_variables()
names = []
with tfv1.name_scope(name + '_internals'):
costs = []
for p in params:
para_name = p.op.name
if re.search(regex, para_name):
regloss = func(p)
assert regloss.dtype.is_floating, regloss
# Some variables may not be fp32, but it should
# be fine to assume regularization in fp32
if regloss.dtype != tf.float32:
regloss = tf.cast(regloss, tf.float32)
costs.append(regloss)
names.append(p.name)
if not costs:
return tf.constant(0, dtype=tf.float32, name='empty_' + name)
# remove tower prefix from names, and print
if len(ctx.vs_name):
prefix = ctx.vs_name + '/'
prefixlen = len(prefix)
def f(name):
if name.startswith(prefix):
return name[prefixlen:]
return name
names = list(map(f, names))
logger.info("regularize_cost() found {} variables to regularize.".format(len(names)))
_log_once("The following tensors will be regularized: {}".format(', '.join(names)))
return tf.add_n(costs, name=name) |
Get the cost from the regularizers in ``tf.GraphKeys.REGULARIZATION_LOSSES``.
If in replicated mode, will only regularize variables created within the current tower.
Args:
name (str): the name of the returned tensor
Returns:
tf.Tensor: a scalar, the total regularization cost.
def regularize_cost_from_collection(name='regularize_cost'):
"""
Get the cost from the regularizers in ``tf.GraphKeys.REGULARIZATION_LOSSES``.
If in replicated mode, will only regularize variables created within the current tower.
Args:
name (str): the name of the returned tensor
Returns:
tf.Tensor: a scalar, the total regularization cost.
"""
ctx = get_current_tower_context()
if not ctx.is_training:
# TODO Currently cannot build the wd_cost correctly at inference,
# because ths vs_name used in inference can be '', therefore the
# variable filter will fail
return tf.constant(0, dtype=tf.float32, name='empty_' + name)
# NOTE: this collection doesn't always grow with towers.
# It only grows with actual variable creation, but not get_variable call.
if ctx.has_own_variables: # be careful of the first tower (name='')
losses = ctx.get_collection_in_tower(tfv1.GraphKeys.REGULARIZATION_LOSSES)
else:
losses = tfv1.get_collection(tfv1.GraphKeys.REGULARIZATION_LOSSES)
if len(losses) > 0:
logger.info("regularize_cost_from_collection() found {} regularizers "
"in REGULARIZATION_LOSSES collection.".format(len(losses)))
def maploss(l):
assert l.dtype.is_floating, l
if l.dtype != tf.float32:
l = tf.cast(l, tf.float32)
return l
losses = [maploss(l) for l in losses]
reg_loss = tf.add_n(losses, name=name)
return reg_loss
else:
return tf.constant(0, dtype=tf.float32, name='empty_' + name) |
Same as `tf.layers.dropout`.
However, for historical reasons, the first positional argument is
interpreted as keep_prob rather than drop_prob.
Explicitly use `rate=` keyword arguments to ensure things are consistent.
def Dropout(x, *args, **kwargs):
"""
Same as `tf.layers.dropout`.
However, for historical reasons, the first positional argument is
interpreted as keep_prob rather than drop_prob.
Explicitly use `rate=` keyword arguments to ensure things are consistent.
"""
if 'is_training' in kwargs:
kwargs['training'] = kwargs.pop('is_training')
if len(args) > 0:
if args[0] != 0.5:
logger.warn(
"The first positional argument to tensorpack.Dropout is the probability to keep, rather than to drop. "
"This is different from the rate argument in tf.layers.Dropout due to historical reasons. "
"To mimic tf.layers.Dropout, explicitly use keyword argument 'rate' instead")
rate = 1 - args[0]
elif 'keep_prob' in kwargs:
assert 'rate' not in kwargs, "Cannot set both keep_prob and rate!"
rate = 1 - kwargs.pop('keep_prob')
elif 'rate' in kwargs:
rate = kwargs.pop('rate')
else:
rate = 0.5
if kwargs.get('training', None) is None:
kwargs['training'] = get_current_tower_context().is_training
if get_tf_version_tuple() <= (1, 12):
return tf.layers.dropout(x, rate=rate, **kwargs)
else:
return tf.nn.dropout(x, rate=rate if kwargs['training'] else 0.) |
Return a proper background image of background_shape, given img.
Args:
background_shape (tuple): a shape (h, w)
img: an image
Returns:
a background image
def fill(self, background_shape, img):
"""
Return a proper background image of background_shape, given img.
Args:
background_shape (tuple): a shape (h, w)
img: an image
Returns:
a background image
"""
background_shape = tuple(background_shape)
return self._fill(background_shape, img) |
Apply a function on the wrapped tensor.
Returns:
LinearWrap: ``LinearWrap(func(self.tensor(), *args, **kwargs))``.
def apply(self, func, *args, **kwargs):
"""
Apply a function on the wrapped tensor.
Returns:
LinearWrap: ``LinearWrap(func(self.tensor(), *args, **kwargs))``.
"""
ret = func(self._t, *args, **kwargs)
return LinearWrap(ret) |
Apply a function on the wrapped tensor. The tensor
will be the second argument of func.
This is because many symbolic functions
(such as tensorpack's layers) takes 'scope' as the first argument.
Returns:
LinearWrap: ``LinearWrap(func(args[0], self.tensor(), *args[1:], **kwargs))``.
def apply2(self, func, *args, **kwargs):
"""
Apply a function on the wrapped tensor. The tensor
will be the second argument of func.
This is because many symbolic functions
(such as tensorpack's layers) takes 'scope' as the first argument.
Returns:
LinearWrap: ``LinearWrap(func(args[0], self.tensor(), *args[1:], **kwargs))``.
"""
ret = func(args[0], self._t, *(args[1:]), **kwargs)
return LinearWrap(ret) |
Returns:
A context where the gradient of :meth:`tf.nn.relu` is replaced by
guided back-propagation, as described in the paper:
`Striving for Simplicity: The All Convolutional Net
<https://arxiv.org/abs/1412.6806>`_
def guided_relu():
"""
Returns:
A context where the gradient of :meth:`tf.nn.relu` is replaced by
guided back-propagation, as described in the paper:
`Striving for Simplicity: The All Convolutional Net
<https://arxiv.org/abs/1412.6806>`_
"""
from tensorflow.python.ops import gen_nn_ops # noqa
@tf.RegisterGradient("GuidedReLU")
def GuidedReluGrad(op, grad):
return tf.where(0. < grad,
gen_nn_ops._relu_grad(grad, op.outputs[0]),
tf.zeros(grad.get_shape()))
g = tf.get_default_graph()
with g.gradient_override_map({'Relu': 'GuidedReLU'}):
yield |
Produce a saliency map as described in the paper:
`Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
<https://arxiv.org/abs/1312.6034>`_.
The saliency map is the gradient of the max element in output w.r.t input.
Returns:
tf.Tensor: the saliency map. Has the same shape as input.
def saliency_map(output, input, name="saliency_map"):
"""
Produce a saliency map as described in the paper:
`Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
<https://arxiv.org/abs/1312.6034>`_.
The saliency map is the gradient of the max element in output w.r.t input.
Returns:
tf.Tensor: the saliency map. Has the same shape as input.
"""
max_outp = tf.reduce_max(output, 1)
saliency_op = tf.gradients(max_outp, input)[:][0]
return tf.identity(saliency_op, name=name) |
A wrapper around `tf.layers.Conv2D`.
Some differences to maintain backward-compatibility:
1. Default kernel initializer is variance_scaling_initializer(2.0).
2. Default padding is 'same'.
3. Support 'split' argument to do group conv. Note that this is not efficient.
Variable Names:
* ``W``: weights
* ``b``: bias
def Conv2D(
inputs,
filters,
kernel_size,
strides=(1, 1),
padding='same',
data_format='channels_last',
dilation_rate=(1, 1),
activation=None,
use_bias=True,
kernel_initializer=None,
bias_initializer=tf.zeros_initializer(),
kernel_regularizer=None,
bias_regularizer=None,
activity_regularizer=None,
split=1):
"""
A wrapper around `tf.layers.Conv2D`.
Some differences to maintain backward-compatibility:
1. Default kernel initializer is variance_scaling_initializer(2.0).
2. Default padding is 'same'.
3. Support 'split' argument to do group conv. Note that this is not efficient.
Variable Names:
* ``W``: weights
* ``b``: bias
"""
if kernel_initializer is None:
if get_tf_version_tuple() <= (1, 12):
kernel_initializer = tf.contrib.layers.variance_scaling_initializer(2.0)
else:
kernel_initializer = tf.keras.initializers.VarianceScaling(2.0, distribution='untruncated_normal')
dilation_rate = shape2d(dilation_rate)
if split == 1 and dilation_rate == [1, 1]:
# tf.layers.Conv2D has bugs with dilations (https://github.com/tensorflow/tensorflow/issues/26797)
with rename_get_variable({'kernel': 'W', 'bias': 'b'}):
layer = tf.layers.Conv2D(
filters,
kernel_size,
strides=strides,
padding=padding,
data_format=data_format,
dilation_rate=dilation_rate,
activation=activation,
use_bias=use_bias,
kernel_initializer=kernel_initializer,
bias_initializer=bias_initializer,
kernel_regularizer=kernel_regularizer,
bias_regularizer=bias_regularizer,
activity_regularizer=activity_regularizer,
_reuse=tf.get_variable_scope().reuse)
ret = layer.apply(inputs, scope=tf.get_variable_scope())
ret = tf.identity(ret, name='output')
ret.variables = VariableHolder(W=layer.kernel)
if use_bias:
ret.variables.b = layer.bias
else:
# group conv implementation
data_format = get_data_format(data_format, keras_mode=False)
in_shape = inputs.get_shape().as_list()
channel_axis = 3 if data_format == 'NHWC' else 1
in_channel = in_shape[channel_axis]
assert in_channel is not None, "[Conv2D] Input cannot have unknown channel!"
assert in_channel % split == 0
assert kernel_regularizer is None and bias_regularizer is None and activity_regularizer is None, \
"Not supported by group conv or dilated conv!"
out_channel = filters
assert out_channel % split == 0
assert dilation_rate == [1, 1] or get_tf_version_tuple() >= (1, 5), 'TF>=1.5 required for dilated conv.'
kernel_shape = shape2d(kernel_size)
filter_shape = kernel_shape + [in_channel / split, out_channel]
stride = shape4d(strides, data_format=data_format)
kwargs = dict(data_format=data_format)
if get_tf_version_tuple() >= (1, 5):
kwargs['dilations'] = shape4d(dilation_rate, data_format=data_format)
W = tf.get_variable(
'W', filter_shape, initializer=kernel_initializer)
if use_bias:
b = tf.get_variable('b', [out_channel], initializer=bias_initializer)
if split == 1:
conv = tf.nn.conv2d(inputs, W, stride, padding.upper(), **kwargs)
else:
conv = None
if get_tf_version_tuple() >= (1, 13):
try:
conv = tf.nn.conv2d(inputs, W, stride, padding.upper(), **kwargs)
except ValueError:
log_once("CUDNN group convolution support is only available with "
"https://github.com/tensorflow/tensorflow/pull/25818 . "
"Will fall back to a loop-based slow implementation instead!", 'warn')
if conv is None:
inputs = tf.split(inputs, split, channel_axis)
kernels = tf.split(W, split, 3)
outputs = [tf.nn.conv2d(i, k, stride, padding.upper(), **kwargs)
for i, k in zip(inputs, kernels)]
conv = tf.concat(outputs, channel_axis)
ret = tf.nn.bias_add(conv, b, data_format=data_format) if use_bias else conv
if activation is not None:
ret = activation(ret)
ret = tf.identity(ret, name='output')
ret.variables = VariableHolder(W=W)
if use_bias:
ret.variables.b = b
return ret |
A wrapper around `tf.layers.Conv2DTranspose`.
Some differences to maintain backward-compatibility:
1. Default kernel initializer is variance_scaling_initializer(2.0).
2. Default padding is 'same'
Variable Names:
* ``W``: weights
* ``b``: bias
def Conv2DTranspose(
inputs,
filters,
kernel_size,
strides=(1, 1),
padding='same',
data_format='channels_last',
activation=None,
use_bias=True,
kernel_initializer=None,
bias_initializer=tf.zeros_initializer(),
kernel_regularizer=None,
bias_regularizer=None,
activity_regularizer=None):
"""
A wrapper around `tf.layers.Conv2DTranspose`.
Some differences to maintain backward-compatibility:
1. Default kernel initializer is variance_scaling_initializer(2.0).
2. Default padding is 'same'
Variable Names:
* ``W``: weights
* ``b``: bias
"""
if kernel_initializer is None:
if get_tf_version_tuple() <= (1, 12):
kernel_initializer = tf.contrib.layers.variance_scaling_initializer(2.0)
else:
kernel_initializer = tf.keras.initializers.VarianceScaling(2.0, distribution='untruncated_normal')
if get_tf_version_tuple() <= (1, 12):
with rename_get_variable({'kernel': 'W', 'bias': 'b'}):
layer = tf.layers.Conv2DTranspose(
filters,
kernel_size,
strides=strides,
padding=padding,
data_format=data_format,
activation=activation,
use_bias=use_bias,
kernel_initializer=kernel_initializer,
bias_initializer=bias_initializer,
kernel_regularizer=kernel_regularizer,
bias_regularizer=bias_regularizer,
activity_regularizer=activity_regularizer,
_reuse=tf.get_variable_scope().reuse)
ret = layer.apply(inputs, scope=tf.get_variable_scope())
ret = tf.identity(ret, name='output')
ret.variables = VariableHolder(W=layer.kernel)
if use_bias:
ret.variables.b = layer.bias
else:
# Our own implementation, to avoid Keras bugs. https://github.com/tensorflow/tensorflow/issues/25946
assert kernel_regularizer is None and bias_regularizer is None and activity_regularizer is None, \
"Unsupported arguments due to Keras bug in TensorFlow 1.13"
data_format = get_data_format(data_format, keras_mode=False)
shape_dyn = tf.shape(inputs)
strides2d = shape2d(strides)
channels_in = inputs.shape[1 if data_format == 'NCHW' else 3]
if data_format == 'NCHW':
channels_in = inputs.shape[1]
out_shape_dyn = tf.stack(
[shape_dyn[0], filters,
shape_dyn[2] * strides2d[0],
shape_dyn[3] * strides2d[1]])
out_shape3_sta = [filters,
None if inputs.shape[2] is None else inputs.shape[2] * strides2d[0],
None if inputs.shape[3] is None else inputs.shape[3] * strides2d[1]]
else:
channels_in = inputs.shape[-1]
out_shape_dyn = tf.stack(
[shape_dyn[0],
shape_dyn[1] * strides2d[0],
shape_dyn[2] * strides2d[1],
filters])
out_shape3_sta = [None if inputs.shape[1] is None else inputs.shape[1] * strides2d[0],
None if inputs.shape[2] is None else inputs.shape[2] * strides2d[1],
filters]
kernel_shape = shape2d(kernel_size)
W = tf.get_variable('W', kernel_shape + [filters, channels_in], initializer=kernel_initializer)
if use_bias:
b = tf.get_variable('b', [filters], initializer=bias_initializer)
conv = tf.nn.conv2d_transpose(
inputs, W, out_shape_dyn,
shape4d(strides, data_format=data_format),
padding=padding.upper(),
data_format=data_format)
conv.set_shape(tf.TensorShape([None] + out_shape3_sta))
ret = tf.nn.bias_add(conv, b, data_format=data_format) if use_bias else conv
if activation is not None:
ret = activation(ret)
ret = tf.identity(ret, name='output')
ret.variables = VariableHolder(W=W)
if use_bias:
ret.variables.b = b
return ret |
Will setup the assign operator for that variable.
def setup_graph(self):
""" Will setup the assign operator for that variable. """
all_vars = tfv1.global_variables() + tfv1.local_variables()
for v in all_vars:
if v.name == self.var_name:
self.var = v
break
else:
raise ValueError("{} is not a variable in the graph!".format(self.var_name)) |
Returns:
The value to assign to the variable.
Note:
Subclasses will implement the abstract method
:meth:`_get_value_to_set`, which should return a new value to
set, or return None to do nothing.
def get_value_to_set(self):
"""
Returns:
The value to assign to the variable.
Note:
Subclasses will implement the abstract method
:meth:`_get_value_to_set`, which should return a new value to
set, or return None to do nothing.
"""
ret = self._get_value_to_set()
if ret is not None and ret != self._last_value:
if self.epoch_num != self._last_epoch_set: # Print this message at most once every epoch
if self._last_value is None:
logger.info("[HyperParamSetter] At global_step={}, {} is set to {:.6f}".format(
self.global_step, self.param.readable_name, ret))
else:
logger.info("[HyperParamSetter] At global_step={}, {} changes from {:.6f} to {:.6f}".format(
self.global_step, self.param.readable_name, self._last_value, ret))
self._last_epoch_set = self.epoch_num
self._last_value = ret
return ret |
Using schedule, compute the value to be set at a given point.
def _get_value_to_set_at_point(self, point):
"""
Using schedule, compute the value to be set at a given point.
"""
laste, lastv = None, None
for e, v in self.schedule:
if e == point:
return v # meet the exact boundary, return directly
if e > point:
break
laste, lastv = e, v
if laste is None or laste == e:
# hasn't reached the first scheduled point, or reached the end of all scheduled points
return None
if self.interp is None:
# If no interpolation, nothing to do.
return None
v = (point - laste) * 1. / (e - laste) * (v - lastv) + lastv
return v |
This function should build the model which takes the input variables
and return cost at the end
def build_graph(self, image, label):
"""This function should build the model which takes the input variables
and return cost at the end"""
# In tensorflow, inputs to convolution function are assumed to be
# NHWC. Add a single channel here.
image = tf.expand_dims(image, 3)
image = image * 2 - 1 # center the pixels values at zero
# The context manager `argscope` sets the default option for all the layers under
# this context. Here we use 32 channel convolution with shape 3x3
with argscope(Conv2D, kernel_size=3, activation=tf.nn.relu, filters=32):
logits = (LinearWrap(image)
.Conv2D('conv0')
.MaxPooling('pool0', 2)
.Conv2D('conv1')
.Conv2D('conv2')
.MaxPooling('pool1', 2)
.Conv2D('conv3')
.FullyConnected('fc0', 512, activation=tf.nn.relu)
.Dropout('dropout', rate=0.5)
.FullyConnected('fc1', 10, activation=tf.identity)())
# a vector of length B with loss of each sample
cost = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=logits, labels=label)
cost = tf.reduce_mean(cost, name='cross_entropy_loss') # the average cross-entropy loss
correct = tf.cast(tf.nn.in_top_k(predictions=logits, targets=label, k=1), tf.float32, name='correct')
accuracy = tf.reduce_mean(correct, name='accuracy')
# This will monitor training error & accuracy (in a moving average fashion). The value will be automatically
# 1. written to tensosrboard
# 2. written to stat.json
# 3. printed after each epoch
train_error = tf.reduce_mean(1 - correct, name='train_error')
summary.add_moving_summary(train_error, accuracy)
# Use a regex to find parameters to apply weight decay.
# Here we apply a weight decay on all W (weight matrix) of all fc layers
# If you don't like regex, you can certainly define the cost in any other methods.
wd_cost = tf.multiply(1e-5,
regularize_cost('fc.*/W', tf.nn.l2_loss),
name='regularize_loss')
total_cost = tf.add_n([wd_cost, cost], name='total_cost')
summary.add_moving_summary(cost, wd_cost, total_cost)
# monitor histogram of all weight (of conv and fc layers) in tensorboard
summary.add_param_summary(('.*/W', ['histogram', 'rms']))
# the function should return the total cost to be optimized
return total_cost |
Convert a caffe parameter name to a tensorflow parameter name as
defined in the above model
def name_conversion(caffe_layer_name):
""" Convert a caffe parameter name to a tensorflow parameter name as
defined in the above model """
# beginning & end mapping
NAME_MAP = {'bn_conv1/beta': 'conv0/bn/beta',
'bn_conv1/gamma': 'conv0/bn/gamma',
'bn_conv1/mean/EMA': 'conv0/bn/mean/EMA',
'bn_conv1/variance/EMA': 'conv0/bn/variance/EMA',
'conv1/W': 'conv0/W', 'conv1/b': 'conv0/b',
'fc1000/W': 'linear/W', 'fc1000/b': 'linear/b'}
if caffe_layer_name in NAME_MAP:
return NAME_MAP[caffe_layer_name]
s = re.search('([a-z]+)([0-9]+)([a-z]+)_', caffe_layer_name)
if s is None:
s = re.search('([a-z]+)([0-9]+)([a-z]+)([0-9]+)_', caffe_layer_name)
layer_block_part1 = s.group(3)
layer_block_part2 = s.group(4)
assert layer_block_part1 in ['a', 'b']
layer_block = 0 if layer_block_part1 == 'a' else int(layer_block_part2)
else:
layer_block = ord(s.group(3)) - ord('a')
layer_type = s.group(1)
layer_group = s.group(2)
layer_branch = int(re.search('_branch([0-9])', caffe_layer_name).group(1))
assert layer_branch in [1, 2]
if layer_branch == 2:
layer_id = re.search('_branch[0-9]([a-z])/', caffe_layer_name).group(1)
layer_id = ord(layer_id) - ord('a') + 1
TYPE_DICT = {'res': 'conv{}', 'bn': 'conv{}/bn'}
layer_type = TYPE_DICT[layer_type].format(layer_id if layer_branch == 2 else 'shortcut')
tf_name = caffe_layer_name[caffe_layer_name.index('/'):]
tf_name = 'group{}/block{}/{}'.format(
int(layer_group) - 2, layer_block, layer_type) + tf_name
return tf_name |
Args:
custom_getter: the same as in :func:`tf.get_variable`
Returns:
The current variable scope with a custom_getter.
def custom_getter_scope(custom_getter):
"""
Args:
custom_getter: the same as in :func:`tf.get_variable`
Returns:
The current variable scope with a custom_getter.
"""
scope = tf.get_variable_scope()
if get_tf_version_tuple() >= (1, 5):
with tf.variable_scope(
scope, custom_getter=custom_getter,
auxiliary_name_scope=False):
yield
else:
ns = tf.get_default_graph().get_name_scope()
with tf.variable_scope(
scope, custom_getter=custom_getter):
with tf.name_scope(ns + '/' if ns else ''):
yield |
Use fn to map the output of any variable getter.
Args:
fn (tf.Variable -> tf.Tensor)
Returns:
The current variable scope with a custom_getter that maps
all the variables by fn.
Example:
.. code-block:: python
with varreplace.remap_variables(lambda var: quantize(var)):
x = FullyConnected('fc', x, 1000) # fc/{W,b} will be quantized
def remap_variables(fn):
"""
Use fn to map the output of any variable getter.
Args:
fn (tf.Variable -> tf.Tensor)
Returns:
The current variable scope with a custom_getter that maps
all the variables by fn.
Example:
.. code-block:: python
with varreplace.remap_variables(lambda var: quantize(var)):
x = FullyConnected('fc', x, 1000) # fc/{W,b} will be quantized
"""
def custom_getter(getter, *args, **kwargs):
v = getter(*args, **kwargs)
return fn(v)
return custom_getter_scope(custom_getter) |
Return a context to freeze variables,
by wrapping ``tf.get_variable`` with a custom getter.
It works by either applying ``tf.stop_gradient`` on the variables,
or by keeping them out of the ``TRAINABLE_VARIABLES`` collection, or
both.
Example:
.. code-block:: python
with varreplace.freeze_variable(stop_gradient=False, skip_collection=True):
x = FullyConnected('fc', x, 1000) # fc/* will not be trained
Args:
stop_gradient (bool): if True, variables returned from `get_variable`
will be wrapped with `tf.stop_gradient` and therefore has no
gradient when used later.
Note that the created variables may still have gradient when accessed
by other approaches (e.g. by name, or by collection).
Also note that this makes `tf.get_variable` returns a Tensor instead of a Variable,
which may break existing code.
Therefore, it's recommended to use the `skip_collection` option instead.
skip_collection (bool): if True, do not add the variable to
``TRAINABLE_VARIABLES`` collection, but to ``MODEL_VARIABLES``
collection. As a result they will not be trained by default.
def freeze_variables(stop_gradient=True, skip_collection=False):
"""
Return a context to freeze variables,
by wrapping ``tf.get_variable`` with a custom getter.
It works by either applying ``tf.stop_gradient`` on the variables,
or by keeping them out of the ``TRAINABLE_VARIABLES`` collection, or
both.
Example:
.. code-block:: python
with varreplace.freeze_variable(stop_gradient=False, skip_collection=True):
x = FullyConnected('fc', x, 1000) # fc/* will not be trained
Args:
stop_gradient (bool): if True, variables returned from `get_variable`
will be wrapped with `tf.stop_gradient` and therefore has no
gradient when used later.
Note that the created variables may still have gradient when accessed
by other approaches (e.g. by name, or by collection).
Also note that this makes `tf.get_variable` returns a Tensor instead of a Variable,
which may break existing code.
Therefore, it's recommended to use the `skip_collection` option instead.
skip_collection (bool): if True, do not add the variable to
``TRAINABLE_VARIABLES`` collection, but to ``MODEL_VARIABLES``
collection. As a result they will not be trained by default.
"""
def custom_getter(getter, *args, **kwargs):
trainable = kwargs.get('trainable', True)
name = args[0] if len(args) else kwargs.get('name')
if skip_collection:
kwargs['trainable'] = False
v = getter(*args, **kwargs)
if skip_collection:
tf.add_to_collection(tf.GraphKeys.MODEL_VARIABLES, v)
if trainable and stop_gradient:
v = tf.stop_gradient(v, name='freezed_' + name)
return v
return custom_getter_scope(custom_getter) |
Load a caffe model. You must be able to ``import caffe`` to use this
function.
Args:
model_desc (str): path to caffe model description file (.prototxt).
model_file (str): path to caffe model parameter file (.caffemodel).
Returns:
dict: the parameters.
def load_caffe(model_desc, model_file):
"""
Load a caffe model. You must be able to ``import caffe`` to use this
function.
Args:
model_desc (str): path to caffe model description file (.prototxt).
model_file (str): path to caffe model parameter file (.caffemodel).
Returns:
dict: the parameters.
"""
with change_env('GLOG_minloglevel', '2'):
import caffe
caffe.set_mode_cpu()
net = caffe.Net(model_desc, model_file, caffe.TEST)
param_dict = CaffeLayerProcessor(net).process()
logger.info("Model loaded from caffe. Params: " +
", ".join(sorted(param_dict.keys())))
return param_dict |
Get caffe protobuf.
Returns:
The imported caffe protobuf module.
def get_caffe_pb():
"""
Get caffe protobuf.
Returns:
The imported caffe protobuf module.
"""
dir = get_dataset_path('caffe')
caffe_pb_file = os.path.join(dir, 'caffe_pb2.py')
if not os.path.isfile(caffe_pb_file):
download(CAFFE_PROTO_URL, dir)
assert os.path.isfile(os.path.join(dir, 'caffe.proto'))
if sys.version_info.major == 3:
cmd = "protoc --version"
version, ret = subproc_call(cmd, timeout=3)
if ret != 0:
sys.exit(1)
try:
version = version.decode('utf-8')
version = float('.'.join(version.split(' ')[1].split('.')[:2]))
assert version >= 2.7, "Require protoc>=2.7 for Python3"
except Exception:
logger.exception("protoc --version gives: " + str(version))
raise
cmd = 'cd {} && protoc caffe.proto --python_out .'.format(dir)
ret = os.system(cmd)
assert ret == 0, \
"Command `{}` failed!".format(cmd)
assert os.path.isfile(caffe_pb_file), caffe_pb_file
import imp
return imp.load_source('caffepb', caffe_pb_file) |
Run some sanity checks, and populate some configs from others
def finalize_configs(is_training):
"""
Run some sanity checks, and populate some configs from others
"""
_C.freeze(False) # populate new keys now
_C.DATA.NUM_CLASS = _C.DATA.NUM_CATEGORY + 1 # +1 background
_C.DATA.BASEDIR = os.path.expanduser(_C.DATA.BASEDIR)
if isinstance(_C.DATA.VAL, six.string_types): # support single string (the typical case) as well
_C.DATA.VAL = (_C.DATA.VAL, )
assert _C.BACKBONE.NORM in ['FreezeBN', 'SyncBN', 'GN', 'None'], _C.BACKBONE.NORM
if _C.BACKBONE.NORM != 'FreezeBN':
assert not _C.BACKBONE.FREEZE_AFFINE
assert _C.BACKBONE.FREEZE_AT in [0, 1, 2]
_C.RPN.NUM_ANCHOR = len(_C.RPN.ANCHOR_SIZES) * len(_C.RPN.ANCHOR_RATIOS)
assert len(_C.FPN.ANCHOR_STRIDES) == len(_C.RPN.ANCHOR_SIZES)
# image size into the backbone has to be multiple of this number
_C.FPN.RESOLUTION_REQUIREMENT = _C.FPN.ANCHOR_STRIDES[3] # [3] because we build FPN with features r2,r3,r4,r5
if _C.MODE_FPN:
size_mult = _C.FPN.RESOLUTION_REQUIREMENT * 1.
_C.PREPROC.MAX_SIZE = np.ceil(_C.PREPROC.MAX_SIZE / size_mult) * size_mult
assert _C.FPN.PROPOSAL_MODE in ['Level', 'Joint']
assert _C.FPN.FRCNN_HEAD_FUNC.endswith('_head')
assert _C.FPN.MRCNN_HEAD_FUNC.endswith('_head')
assert _C.FPN.NORM in ['None', 'GN']
if _C.FPN.CASCADE:
# the first threshold is the proposal sampling threshold
assert _C.CASCADE.IOUS[0] == _C.FRCNN.FG_THRESH
assert len(_C.CASCADE.BBOX_REG_WEIGHTS) == len(_C.CASCADE.IOUS)
if is_training:
train_scales = _C.PREPROC.TRAIN_SHORT_EDGE_SIZE
if isinstance(train_scales, (list, tuple)) and train_scales[1] - train_scales[0] > 100:
# don't autotune if augmentation is on
os.environ['TF_CUDNN_USE_AUTOTUNE'] = '0'
os.environ['TF_AUTOTUNE_THRESHOLD'] = '1'
assert _C.TRAINER in ['horovod', 'replicated'], _C.TRAINER
# setup NUM_GPUS
if _C.TRAINER == 'horovod':
import horovod.tensorflow as hvd
ngpu = hvd.size()
if ngpu == hvd.local_size():
logger.warn("It's not recommended to use horovod for single-machine training. "
"Replicated trainer is more stable and has the same efficiency.")
else:
assert 'OMPI_COMM_WORLD_SIZE' not in os.environ
ngpu = get_num_gpu()
assert ngpu > 0, "Has to train with GPU!"
assert ngpu % 8 == 0 or 8 % ngpu == 0, "Can only train with 1,2,4 or >=8 GPUs, but found {} GPUs".format(ngpu)
else:
# autotune is too slow for inference
os.environ['TF_CUDNN_USE_AUTOTUNE'] = '0'
ngpu = get_num_gpu()
if _C.TRAIN.NUM_GPUS is None:
_C.TRAIN.NUM_GPUS = ngpu
else:
if _C.TRAINER == 'horovod':
assert _C.TRAIN.NUM_GPUS == ngpu
else:
assert _C.TRAIN.NUM_GPUS <= ngpu
_C.freeze()
logger.info("Config: ------------------------------------------\n" + str(_C)) |
Convert to a nested dict.
def to_dict(self):
"""Convert to a nested dict. """
return {k: v.to_dict() if isinstance(v, AttrDict) else v
for k, v in self.__dict__.items() if not k.startswith('_')} |
Update from command line args.
def update_args(self, args):
"""Update from command line args. """
for cfg in args:
keys, v = cfg.split('=', maxsplit=1)
keylist = keys.split('.')
dic = self
for i, k in enumerate(keylist[:-1]):
assert k in dir(dic), "Unknown config key: {}".format(keys)
dic = getattr(dic, k)
key = keylist[-1]
oldv = getattr(dic, key)
if not isinstance(oldv, str):
v = eval(v)
setattr(dic, key, v) |
Get a corresponding model loader by looking at the file name.
Returns:
SessInit: either a :class:`DictRestore` (if name ends with 'npy/npz') or
:class:`SaverRestore` (otherwise).
def get_model_loader(filename):
"""
Get a corresponding model loader by looking at the file name.
Returns:
SessInit: either a :class:`DictRestore` (if name ends with 'npy/npz') or
:class:`SaverRestore` (otherwise).
"""
assert isinstance(filename, six.string_types), filename
filename = os.path.expanduser(filename)
if filename.endswith('.npy'):
assert tf.gfile.Exists(filename), filename
return DictRestore(np.load(filename, encoding='latin1').item())
elif filename.endswith('.npz'):
assert tf.gfile.Exists(filename), filename
obj = np.load(filename)
return DictRestore(dict(obj))
else:
return SaverRestore(filename) |
return a set of strings
def _read_checkpoint_vars(model_path):
""" return a set of strings """
reader = tf.train.NewCheckpointReader(model_path)
reader = CheckpointReaderAdapter(reader) # use an adapter to standardize the name
ckpt_vars = reader.get_variable_to_shape_map().keys()
return reader, set(ckpt_vars) |
Args:
layers (list or layer): layer or list of layers to apply the arguments.
Returns:
a context where all appearance of these layer will by default have the
arguments specified by kwargs.
Example:
.. code-block:: python
with argscope(Conv2D, kernel_shape=3, nl=tf.nn.relu, out_channel=32):
x = Conv2D('conv0', x)
x = Conv2D('conv1', x)
x = Conv2D('conv2', x, out_channel=64) # override argscope
def argscope(layers, **kwargs):
"""
Args:
layers (list or layer): layer or list of layers to apply the arguments.
Returns:
a context where all appearance of these layer will by default have the
arguments specified by kwargs.
Example:
.. code-block:: python
with argscope(Conv2D, kernel_shape=3, nl=tf.nn.relu, out_channel=32):
x = Conv2D('conv0', x)
x = Conv2D('conv1', x)
x = Conv2D('conv2', x, out_channel=64) # override argscope
"""
if not isinstance(layers, list):
layers = [layers]
# def _check_args_exist(l):
# args = inspect.getargspec(l).args
# for k, v in six.iteritems(kwargs):
# assert k in args, "No argument {} in {}".format(k, l.__name__)
for l in layers:
assert hasattr(l, 'symbolic_function'), "{} is not a registered layer".format(l.__name__)
# _check_args_exist(l.symbolic_function)
new_scope = copy.copy(get_arg_scope())
for l in layers:
new_scope[l.__name__].update(kwargs)
_ArgScopeStack.append(new_scope)
yield
del _ArgScopeStack[-1] |
Decorator for function to support argscope
Example:
.. code-block:: python
from mylib import myfunc
myfunc = enable_argscope_for_function(myfunc)
Args:
func: A function mapping one or multiple tensors to one or multiple
tensors.
log_shape (bool): Specify whether the first input resp. output tensor
shape should be printed once.
Remarks:
If the function ``func`` returns multiple input or output tensors,
only the first input/output tensor shape is displayed during logging.
Returns:
The decorated function.
def enable_argscope_for_function(func, log_shape=True):
"""Decorator for function to support argscope
Example:
.. code-block:: python
from mylib import myfunc
myfunc = enable_argscope_for_function(myfunc)
Args:
func: A function mapping one or multiple tensors to one or multiple
tensors.
log_shape (bool): Specify whether the first input resp. output tensor
shape should be printed once.
Remarks:
If the function ``func`` returns multiple input or output tensors,
only the first input/output tensor shape is displayed during logging.
Returns:
The decorated function.
"""
assert callable(func), "func should be a callable"
@wraps(func)
def wrapped_func(*args, **kwargs):
actual_args = copy.copy(get_arg_scope()[func.__name__])
actual_args.update(kwargs)
out_tensor = func(*args, **actual_args)
in_tensor = args[0]
ctx = get_current_tower_context()
name = func.__name__ if 'name' not in kwargs else kwargs['name']
if log_shape:
if ('tower' not in ctx.ns_name.lower()) or ctx.is_main_training_tower:
# we assume the first parameter is the most interesting
if isinstance(out_tensor, tuple):
out_tensor_descr = out_tensor[0]
else:
out_tensor_descr = out_tensor
logger.info('%20s: %20s -> %20s' %
(name, in_tensor.shape.as_list(),
out_tensor_descr.shape.as_list()))
return out_tensor
# argscope requires this property
wrapped_func.symbolic_function = None
return wrapped_func |
Overwrite all functions of a given module to support argscope.
Note that this function monkey-patches the module and therefore could
have unexpected consequences.
It has been only tested to work well with ``tf.layers`` module.
Example:
.. code-block:: python
import tensorflow as tf
enable_argscope_for_module(tf.layers)
Args:
log_shape (bool): print input/output shapes of each function.
def enable_argscope_for_module(module, log_shape=True):
"""
Overwrite all functions of a given module to support argscope.
Note that this function monkey-patches the module and therefore could
have unexpected consequences.
It has been only tested to work well with ``tf.layers`` module.
Example:
.. code-block:: python
import tensorflow as tf
enable_argscope_for_module(tf.layers)
Args:
log_shape (bool): print input/output shapes of each function.
"""
if is_tfv2() and module == tf.layers:
module = tf.compat.v1.layers
for name, obj in getmembers(module):
if isfunction(obj):
setattr(module, name, enable_argscope_for_function(obj,
log_shape=log_shape)) |
Generate tensor for TensorBoard (casting, clipping)
Args:
name: name for visualization operation
*imgs: multiple tensors as list
scale_func: scale input tensors to fit range [0, 255]
Example:
visualize_tensors('viz1', [img1])
visualize_tensors('viz2', [img1, img2, img3], max_outputs=max(30, BATCH))
def visualize_tensors(name, imgs, scale_func=lambda x: (x + 1.) * 128., max_outputs=1):
"""Generate tensor for TensorBoard (casting, clipping)
Args:
name: name for visualization operation
*imgs: multiple tensors as list
scale_func: scale input tensors to fit range [0, 255]
Example:
visualize_tensors('viz1', [img1])
visualize_tensors('viz2', [img1, img2, img3], max_outputs=max(30, BATCH))
"""
xy = scale_func(tf.concat(imgs, axis=2))
xy = tf.cast(tf.clip_by_value(xy, 0, 255), tf.uint8, name='viz')
tf.summary.image(name, xy, max_outputs=30) |
img: an RGB image of shape (s, 2s, 3).
:return: [input, output]
def split_input(img):
"""
img: an RGB image of shape (s, 2s, 3).
:return: [input, output]
"""
# split the image into left + right pairs
s = img.shape[0]
assert img.shape[1] == 2 * s
input, output = img[:, :s, :], img[:, s:, :]
if args.mode == 'BtoA':
input, output = output, input
if IN_CH == 1:
input = cv2.cvtColor(input, cv2.COLOR_RGB2GRAY)[:, :, np.newaxis]
if OUT_CH == 1:
output = cv2.cvtColor(output, cv2.COLOR_RGB2GRAY)[:, :, np.newaxis]
return [input, output] |
return a (b, 1) logits
def discriminator(self, inputs, outputs):
""" return a (b, 1) logits"""
l = tf.concat([inputs, outputs], 3)
with argscope(Conv2D, kernel_size=4, strides=2, activation=BNLReLU):
l = (LinearWrap(l)
.Conv2D('conv0', NF, activation=tf.nn.leaky_relu)
.Conv2D('conv1', NF * 2)
.Conv2D('conv2', NF * 4)
.Conv2D('conv3', NF * 8, strides=1, padding='VALID')
.Conv2D('convlast', 1, strides=1, padding='VALID', activation=tf.identity)())
return l |
A simple print Op that might be easier to use than :meth:`tf.Print`.
Use it like: ``x = print_stat(x, message='This is x')``.
def print_stat(x, message=None):
""" A simple print Op that might be easier to use than :meth:`tf.Print`.
Use it like: ``x = print_stat(x, message='This is x')``.
"""
if message is None:
message = x.op.name
lst = [tf.shape(x), tf.reduce_mean(x)]
if x.dtype.is_floating:
lst.append(rms(x))
return tf.Print(x, lst + [x], summarize=20,
message=message, name='print_' + x.op.name) |
Returns:
root mean square of tensor x.
def rms(x, name=None):
"""
Returns:
root mean square of tensor x.
"""
if name is None:
name = x.op.name + '/rms'
with tfv1.name_scope(None): # name already contains the scope
return tf.sqrt(tf.reduce_mean(tf.square(x)), name=name)
return tf.sqrt(tf.reduce_mean(tf.square(x)), name=name) |
`Peek Signal to Noise Ratio <https://en.wikipedia.org/wiki/Peak_signal-to-noise_ratio>`_.
.. math::
PSNR = 20 \cdot \log_{10}(MAX_p) - 10 \cdot \log_{10}(MSE)
Args:
prediction: a :class:`tf.Tensor` representing the prediction signal.
ground_truth: another :class:`tf.Tensor` with the same shape.
maxp: maximum possible pixel value of the image (255 in in 8bit images)
Returns:
A scalar tensor representing the PSNR
def psnr(prediction, ground_truth, maxp=None, name='psnr'):
"""`Peek Signal to Noise Ratio <https://en.wikipedia.org/wiki/Peak_signal-to-noise_ratio>`_.
.. math::
PSNR = 20 \cdot \log_{10}(MAX_p) - 10 \cdot \log_{10}(MSE)
Args:
prediction: a :class:`tf.Tensor` representing the prediction signal.
ground_truth: another :class:`tf.Tensor` with the same shape.
maxp: maximum possible pixel value of the image (255 in in 8bit images)
Returns:
A scalar tensor representing the PSNR
"""
maxp = float(maxp)
def log10(x):
with tf.name_scope("log10"):
numerator = tf.log(x)
denominator = tf.log(tf.constant(10, dtype=numerator.dtype))
return numerator / denominator
mse = tf.reduce_mean(tf.square(prediction - ground_truth))
if maxp is None:
psnr = tf.multiply(log10(mse), -10., name=name)
else:
psnr = tf.multiply(log10(mse), -10.)
psnr = tf.add(tf.multiply(20., log10(maxp)), psnr, name=name)
return psnr |
Args:
anchor: coordinate of the center
def get_gaussian_weight(self, anchor):
"""
Args:
anchor: coordinate of the center
"""
ret = np.zeros(self.shape, dtype='float32')
y, x = np.mgrid[:self.shape[0], :self.shape[1]]
y = y.astype('float32') / ret.shape[0] - anchor[0]
x = x.astype('float32') / ret.shape[1] - anchor[1]
g = np.exp(-(x**2 + y ** 2) / self.sigma)
# cv2.imshow(" ", g)
# cv2.waitKey()
return g |
Pad tensor in H, W
Remarks:
TensorFlow uses "ceil(input_spatial_shape[i] / strides[i])" rather than explicit padding
like Caffe, pyTorch does. Hence, we need to pad here beforehand.
Args:
x (tf.tensor): incoming tensor
p (int, optional): padding for H, W
Returns:
tf.tensor: padded tensor
def pad(x, p=3):
"""Pad tensor in H, W
Remarks:
TensorFlow uses "ceil(input_spatial_shape[i] / strides[i])" rather than explicit padding
like Caffe, pyTorch does. Hence, we need to pad here beforehand.
Args:
x (tf.tensor): incoming tensor
p (int, optional): padding for H, W
Returns:
tf.tensor: padded tensor
"""
return tf.pad(x, [[0, 0], [0, 0], [p, p], [p, p]]) |
Correlation Cost Volume computation.
This is a fallback Python-only implementation, specialized just for FlowNet2.
It takes a lot of memory and is slow.
If you know to compile a custom op yourself, it's better to use the cuda implementation here:
https://github.com/PatWie/tensorflow-recipes/tree/master/OpticalFlow/user_ops
def correlation(ina, inb,
kernel_size, max_displacement,
stride_1, stride_2,
pad, data_format):
"""
Correlation Cost Volume computation.
This is a fallback Python-only implementation, specialized just for FlowNet2.
It takes a lot of memory and is slow.
If you know to compile a custom op yourself, it's better to use the cuda implementation here:
https://github.com/PatWie/tensorflow-recipes/tree/master/OpticalFlow/user_ops
"""
assert pad == max_displacement
assert kernel_size == 1
assert data_format == 'NCHW'
assert max_displacement % stride_2 == 0
assert stride_1 == 1
D = int(max_displacement / stride_2 * 2) + 1 # D^2 == number of correlations per spatial location
b, c, h, w = ina.shape.as_list()
inb = tf.pad(inb, [[0, 0], [0, 0], [pad, pad], [pad, pad]])
res = []
for k1 in range(0, D):
start_h = k1 * stride_2
for k2 in range(0, D):
start_w = k2 * stride_2
s = tf.slice(inb, [0, 0, start_h, start_w], [-1, -1, h, w])
ans = tf.reduce_mean(ina * s, axis=1, keepdims=True)
res.append(ans)
res = tf.concat(res, axis=1) # ND^2HW
return res |
Resize input tensor with unkown input-shape by a factor
Args:
x (tf.Tensor): tensor NCHW
factor (int, optional): resize factor for H, W
Note:
Differences here against Caffe have huge impacts on the
quality of the predictions.
Returns:
tf.Tensor: resized tensor NCHW
def resize(x, mode, factor=4):
"""Resize input tensor with unkown input-shape by a factor
Args:
x (tf.Tensor): tensor NCHW
factor (int, optional): resize factor for H, W
Note:
Differences here against Caffe have huge impacts on the
quality of the predictions.
Returns:
tf.Tensor: resized tensor NCHW
"""
assert mode in ['bilinear', 'nearest'], mode
shp = tf.shape(x)[2:] * factor
# NCHW -> NHWC
x = tf.transpose(x, [0, 2, 3, 1])
if mode == 'bilinear':
x = tf.image.resize_bilinear(x, shp, align_corners=True)
else:
# better approximation of what Caffe is doing
x = tf.image.resize_nearest_neighbor(x, shp, align_corners=False)
# NHWC -> NCHW
return tf.transpose(x, [0, 3, 1, 2]) |
Architecture in Table 4 of FlowNet 2.0.
Args:
x: NCHW tensor, where C=11 is the concatenation of 7 items of [3, 2, 2, 1, 1, 1, 1] channels.
def flownet2_fusion(self, x):
"""
Architecture in Table 4 of FlowNet 2.0.
Args:
x: NCHW tensor, where C=11 is the concatenation of 7 items of [3, 2, 2, 1, 1, 1, 1] channels.
"""
with argscope([tf.layers.conv2d], activation=lambda x: tf.nn.leaky_relu(x, 0.1),
padding='valid', strides=2, kernel_size=3,
data_format='channels_first'), \
argscope([tf.layers.conv2d_transpose], padding='same', activation=tf.identity,
data_format='channels_first', strides=2, kernel_size=4):
conv0 = tf.layers.conv2d(pad(x, 1), 64, name='conv0', strides=1)
x = tf.layers.conv2d(pad(conv0, 1), 64, name='conv1')
conv1 = tf.layers.conv2d(pad(x, 1), 128, name='conv1_1', strides=1)
x = tf.layers.conv2d(pad(conv1, 1), 128, name='conv2')
conv2 = tf.layers.conv2d(pad(x, 1), 128, name='conv2_1', strides=1)
flow2 = tf.layers.conv2d(pad(conv2, 1), 2, name='predict_flow2', strides=1, activation=tf.identity)
flow2_up = tf.layers.conv2d_transpose(flow2, 2, name='upsampled_flow2_to_1')
x = tf.layers.conv2d_transpose(conv2, 32, name='deconv1', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat1 = tf.concat([conv1, x, flow2_up], axis=1, name='concat1')
interconv1 = tf.layers.conv2d(pad(concat1, 1), 32, strides=1, name='inter_conv1', activation=tf.identity)
flow1 = tf.layers.conv2d(pad(interconv1, 1), 2, name='predict_flow1', strides=1, activation=tf.identity)
flow1_up = tf.layers.conv2d_transpose(flow1, 2, name='upsampled_flow1_to_0')
x = tf.layers.conv2d_transpose(concat1, 16, name='deconv0', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat0 = tf.concat([conv0, x, flow1_up], axis=1, name='concat0')
interconv0 = tf.layers.conv2d(pad(concat0, 1), 16, strides=1, name='inter_conv0', activation=tf.identity)
flow0 = tf.layers.conv2d(pad(interconv0, 1), 2, name='predict_flow0', strides=1, activation=tf.identity)
return tf.identity(flow0, name='flow2') |
Architecture in Table 3 of FlowNet 2.0.
Args:
x: concatenation of two inputs, of shape [1, 2xC, H, W]
def flownet2_sd(self, x):
"""
Architecture in Table 3 of FlowNet 2.0.
Args:
x: concatenation of two inputs, of shape [1, 2xC, H, W]
"""
with argscope([tf.layers.conv2d], activation=lambda x: tf.nn.leaky_relu(x, 0.1),
padding='valid', strides=2, kernel_size=3,
data_format='channels_first'), \
argscope([tf.layers.conv2d_transpose], padding='same', activation=tf.identity,
data_format='channels_first', strides=2, kernel_size=4):
x = tf.layers.conv2d(pad(x, 1), 64, name='conv0', strides=1)
x = tf.layers.conv2d(pad(x, 1), 64, name='conv1')
conv1 = tf.layers.conv2d(pad(x, 1), 128, name='conv1_1', strides=1)
x = tf.layers.conv2d(pad(conv1, 1), 128, name='conv2')
conv2 = tf.layers.conv2d(pad(x, 1), 128, name='conv2_1', strides=1)
x = tf.layers.conv2d(pad(conv2, 1), 256, name='conv3')
conv3 = tf.layers.conv2d(pad(x, 1), 256, name='conv3_1', strides=1)
x = tf.layers.conv2d(pad(conv3, 1), 512, name='conv4')
conv4 = tf.layers.conv2d(pad(x, 1), 512, name='conv4_1', strides=1)
x = tf.layers.conv2d(pad(conv4, 1), 512, name='conv5')
conv5 = tf.layers.conv2d(pad(x, 1), 512, name='conv5_1', strides=1)
x = tf.layers.conv2d(pad(conv5, 1), 1024, name='conv6')
conv6 = tf.layers.conv2d(pad(x, 1), 1024, name='conv6_1', strides=1)
flow6 = tf.layers.conv2d(pad(conv6, 1), 2, name='predict_flow6', strides=1, activation=tf.identity)
flow6_up = tf.layers.conv2d_transpose(flow6, 2, name='upsampled_flow6_to_5')
x = tf.layers.conv2d_transpose(conv6, 512, name='deconv5', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat5 = tf.concat([conv5, x, flow6_up], axis=1, name='concat5')
interconv5 = tf.layers.conv2d(pad(concat5, 1), 512, strides=1, name='inter_conv5', activation=tf.identity)
flow5 = tf.layers.conv2d(pad(interconv5, 1), 2, name='predict_flow5', strides=1, activation=tf.identity)
flow5_up = tf.layers.conv2d_transpose(flow5, 2, name='upsampled_flow5_to_4')
x = tf.layers.conv2d_transpose(concat5, 256, name='deconv4', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat4 = tf.concat([conv4, x, flow5_up], axis=1, name='concat4')
interconv4 = tf.layers.conv2d(pad(concat4, 1), 256, strides=1, name='inter_conv4', activation=tf.identity)
flow4 = tf.layers.conv2d(pad(interconv4, 1), 2, name='predict_flow4', strides=1, activation=tf.identity)
flow4_up = tf.layers.conv2d_transpose(flow4, 2, name='upsampled_flow4_to_3')
x = tf.layers.conv2d_transpose(concat4, 128, name='deconv3', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat3 = tf.concat([conv3, x, flow4_up], axis=1, name='concat3')
interconv3 = tf.layers.conv2d(pad(concat3, 1), 128, strides=1, name='inter_conv3', activation=tf.identity)
flow3 = tf.layers.conv2d(pad(interconv3, 1), 2, name='predict_flow3', strides=1, activation=tf.identity)
flow3_up = tf.layers.conv2d_transpose(flow3, 2, name='upsampled_flow3_to_2')
x = tf.layers.conv2d_transpose(concat3, 64, name='deconv2', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat2 = tf.concat([conv2, x, flow3_up], axis=1, name='concat2')
interconv2 = tf.layers.conv2d(pad(concat2, 1), 64, strides=1, name='inter_conv2', activation=tf.identity)
flow2 = tf.layers.conv2d(pad(interconv2, 1), 2, name='predict_flow2', strides=1, activation=tf.identity)
return resize(flow2 / DISP_SCALE, mode='nearest') |
Architecture of FlowNetSimple in Figure 2 of FlowNet 1.0.
Args:
x: 2CHW if standalone==True, else NCHW where C=12 is a concatenation
of 5 tensors of [3, 3, 3, 2, 1] channels.
standalone: If True, this model is used to predict flow from two inputs.
If False, this model is used as part of the FlowNet2.
def graph_structure(self, x, standalone=True):
"""
Architecture of FlowNetSimple in Figure 2 of FlowNet 1.0.
Args:
x: 2CHW if standalone==True, else NCHW where C=12 is a concatenation
of 5 tensors of [3, 3, 3, 2, 1] channels.
standalone: If True, this model is used to predict flow from two inputs.
If False, this model is used as part of the FlowNet2.
"""
if standalone:
x = tf.concat(tf.split(x, 2, axis=0), axis=1)
with argscope([tf.layers.conv2d], activation=lambda x: tf.nn.leaky_relu(x, 0.1),
padding='valid', strides=2, kernel_size=3,
data_format='channels_first'), \
argscope([tf.layers.conv2d_transpose], padding='same', activation=tf.identity,
data_format='channels_first', strides=2, kernel_size=4):
x = tf.layers.conv2d(pad(x, 3), 64, kernel_size=7, name='conv1')
conv2 = tf.layers.conv2d(pad(x, 2), 128, kernel_size=5, name='conv2')
x = tf.layers.conv2d(pad(conv2, 2), 256, kernel_size=5, name='conv3')
conv3 = tf.layers.conv2d(pad(x, 1), 256, name='conv3_1', strides=1)
x = tf.layers.conv2d(pad(conv3, 1), 512, name='conv4')
conv4 = tf.layers.conv2d(pad(x, 1), 512, name='conv4_1', strides=1)
x = tf.layers.conv2d(pad(conv4, 1), 512, name='conv5')
conv5 = tf.layers.conv2d(pad(x, 1), 512, name='conv5_1', strides=1)
x = tf.layers.conv2d(pad(conv5, 1), 1024, name='conv6')
conv6 = tf.layers.conv2d(pad(x, 1), 1024, name='conv6_1', strides=1)
flow6 = tf.layers.conv2d(pad(conv6, 1), 2, name='predict_flow6', strides=1, activation=tf.identity)
flow6_up = tf.layers.conv2d_transpose(flow6, 2, name='upsampled_flow6_to_5', use_bias=False)
x = tf.layers.conv2d_transpose(conv6, 512, name='deconv5', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat5 = tf.concat([conv5, x, flow6_up], axis=1, name='concat5')
flow5 = tf.layers.conv2d(pad(concat5, 1), 2, name='predict_flow5', strides=1, activation=tf.identity)
flow5_up = tf.layers.conv2d_transpose(flow5, 2, name='upsampled_flow5_to_4', use_bias=False)
x = tf.layers.conv2d_transpose(concat5, 256, name='deconv4', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat4 = tf.concat([conv4, x, flow5_up], axis=1, name='concat4')
flow4 = tf.layers.conv2d(pad(concat4, 1), 2, name='predict_flow4', strides=1, activation=tf.identity)
flow4_up = tf.layers.conv2d_transpose(flow4, 2, name='upsampled_flow4_to_3', use_bias=False)
x = tf.layers.conv2d_transpose(concat4, 128, name='deconv3', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat3 = tf.concat([conv3, x, flow4_up], axis=1, name='concat3')
flow3 = tf.layers.conv2d(pad(concat3, 1), 2, name='predict_flow3', strides=1, activation=tf.identity)
flow3_up = tf.layers.conv2d_transpose(flow3, 2, name='upsampled_flow3_to_2', use_bias=False)
x = tf.layers.conv2d_transpose(concat3, 64, name='deconv2', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat2 = tf.concat([conv2, x, flow3_up], axis=1, name='concat2')
flow2 = tf.layers.conv2d(pad(concat2, 1), 2, name='predict_flow2', strides=1, activation=tf.identity)
return tf.identity(flow2, name='flow2') |
Architecture of FlowNetCorr in Figure 2 of FlowNet 1.0.
Args:
x: 2CHW.
def graph_structure(self, x1x2):
"""
Architecture of FlowNetCorr in Figure 2 of FlowNet 1.0.
Args:
x: 2CHW.
"""
with argscope([tf.layers.conv2d], activation=lambda x: tf.nn.leaky_relu(x, 0.1),
padding='valid', strides=2, kernel_size=3,
data_format='channels_first'), \
argscope([tf.layers.conv2d_transpose], padding='same', activation=tf.identity,
data_format='channels_first', strides=2, kernel_size=4):
# extract features
x = tf.layers.conv2d(pad(x1x2, 3), 64, kernel_size=7, name='conv1')
conv2 = tf.layers.conv2d(pad(x, 2), 128, kernel_size=5, name='conv2')
conv3 = tf.layers.conv2d(pad(conv2, 2), 256, kernel_size=5, name='conv3')
conv2a, _ = tf.split(conv2, 2, axis=0)
conv3a, conv3b = tf.split(conv3, 2, axis=0)
corr = correlation(conv3a, conv3b,
kernel_size=1,
max_displacement=20,
stride_1=1,
stride_2=2,
pad=20, data_format='NCHW')
corr = tf.nn.leaky_relu(corr, 0.1)
conv_redir = tf.layers.conv2d(conv3a, 32, kernel_size=1, strides=1, name='conv_redir')
in_conv3_1 = tf.concat([conv_redir, corr], axis=1, name='in_conv3_1')
conv3_1 = tf.layers.conv2d(pad(in_conv3_1, 1), 256, name='conv3_1', strides=1)
x = tf.layers.conv2d(pad(conv3_1, 1), 512, name='conv4')
conv4 = tf.layers.conv2d(pad(x, 1), 512, name='conv4_1', strides=1)
x = tf.layers.conv2d(pad(conv4, 1), 512, name='conv5')
conv5 = tf.layers.conv2d(pad(x, 1), 512, name='conv5_1', strides=1)
x = tf.layers.conv2d(pad(conv5, 1), 1024, name='conv6')
conv6 = tf.layers.conv2d(pad(x, 1), 1024, name='conv6_1', strides=1)
flow6 = tf.layers.conv2d(pad(conv6, 1), 2, name='predict_flow6', strides=1, activation=tf.identity)
flow6_up = tf.layers.conv2d_transpose(flow6, 2, name='upsampled_flow6_to_5')
x = tf.layers.conv2d_transpose(conv6, 512, name='deconv5', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
# return flow6
concat5 = tf.concat([conv5, x, flow6_up], axis=1, name='concat5')
flow5 = tf.layers.conv2d(pad(concat5, 1), 2, name='predict_flow5', strides=1, activation=tf.identity)
flow5_up = tf.layers.conv2d_transpose(flow5, 2, name='upsampled_flow5_to_4')
x = tf.layers.conv2d_transpose(concat5, 256, name='deconv4', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat4 = tf.concat([conv4, x, flow5_up], axis=1, name='concat4')
flow4 = tf.layers.conv2d(pad(concat4, 1), 2, name='predict_flow4', strides=1, activation=tf.identity)
flow4_up = tf.layers.conv2d_transpose(flow4, 2, name='upsampled_flow4_to_3')
x = tf.layers.conv2d_transpose(concat4, 128, name='deconv3', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat3 = tf.concat([conv3_1, x, flow4_up], axis=1, name='concat3')
flow3 = tf.layers.conv2d(pad(concat3, 1), 2, name='predict_flow3', strides=1, activation=tf.identity)
flow3_up = tf.layers.conv2d_transpose(flow3, 2, name='upsampled_flow3_to_2')
x = tf.layers.conv2d_transpose(concat3, 64, name='deconv2', activation=lambda x: tf.nn.leaky_relu(x, 0.1))
concat2 = tf.concat([conv2a, x, flow3_up], axis=1, name='concat2')
flow2 = tf.layers.conv2d(pad(concat2, 1), 2, name='predict_flow2', strides=1, activation=tf.identity)
return tf.identity(flow2, name='flow2') |
Will not modify img
def draw_annotation(img, boxes, klass, is_crowd=None):
"""Will not modify img"""
labels = []
assert len(boxes) == len(klass)
if is_crowd is not None:
assert len(boxes) == len(is_crowd)
for cls, crd in zip(klass, is_crowd):
clsname = cfg.DATA.CLASS_NAMES[cls]
if crd == 1:
clsname += ';Crowd'
labels.append(clsname)
else:
for cls in klass:
labels.append(cfg.DATA.CLASS_NAMES[cls])
img = viz.draw_boxes(img, boxes, labels)
return img |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.