def image(name, data, step=None, max_outputs=3, description=None):
"""Write an image summary.
Arguments:
name: A name for this summary. The summary tag used for TensorBoard will
be this name prefixed by any active name scopes.
data: A `Tensor` representing pixel data with shape `[k, h, w, c]`,
where `k` is the number of images, `h` and `w` are the height and
width of the images, and `c` is the number of channels, which
should be 1, 2, 3, or 4 (grayscale, grayscale with alpha, RGB, RGBA).
Any of the dimensions may be statically unknown (i.e., `None`).
Floating point data will be clipped to the range [0,1).
step: Explicit `int64`-castable monotonic step value for this summary. If
omitted, this defaults to `tf.summary.experimental.get_step()`, which must
not be None.
max_outputs: Optional `int` or rank-0 integer `Tensor`. At most this
many images will be emitted at each step. When more than
`max_outputs` many images are provided, the first `max_outputs` many
images will be used and the rest silently discarded.
description: Optional long-form description for this summary, as a
constant `str`. Markdown is supported. Defaults to empty.
Returns:
True on success, or false if no summary was emitted because no default
summary writer was available.
Raises:
ValueError: if a default writer exists, but no step was provided and
`tf.summary.experimental.get_step()` is None.
"""
summary_metadata = metadata.create_summary_metadata(
display_name=None, description=description)
# TODO(https://github.com/tensorflow/tensorboard/issues/2109): remove fallback
summary_scope = (getattr(tf.summary.experimental, 'summary_scope', None)
or tf.summary.summary_scope)
with summary_scope(name, 'image_summary', values=[data, max_outputs,
step]) as (tag, _):
tf.debugging.assert_rank(data, 4)
tf.debugging.assert_non_negative(max_outputs)
images = tf.image.convert_image_dtype(data, tf.uint8, saturate=True)
limited_images = images[:max_outputs]
encoded_images = tf.map_fn(tf.image.encode_png,
limited_images,
dtype=tf.string,
name='encode_each_image')
# Workaround for map_fn returning float dtype for an empty elems input.
encoded_images = tf.cond(
tf.shape(input=encoded_images)[0] > 0, lambda: encoded_images,
lambda: tf.constant([], tf.string))
image_shape = tf.shape(input=images)
dimensions = tf.stack([
tf.as_string(image_shape[2], name='width'),
tf.as_string(image_shape[1], name='height')
],
name='dimensions')
tensor = tf.concat([dimensions, encoded_images], axis=0)
return tf.summary.write(tag=tag,
tensor=tensor,
step=step,
metadata=summary_metadata)
def lazy_tensor():
tf.debugging.assert_rank(data, 4)
tf.debugging.assert_non_negative(max_outputs)
images = tf.image.convert_image_dtype(data,
tf.uint8,
saturate=True)
limited_images = images[:max_outputs]
encoded_images = tf.map_fn(
tf.image.encode_png,
limited_images,
dtype=tf.string,
name="encode_each_image",
)
# Workaround for map_fn returning float dtype for an empty elems input.
encoded_images = tf.cond(
tf.shape(input=encoded_images)[0] > 0,
lambda: encoded_images,
lambda: tf.constant([], tf.string),
)
image_shape = tf.shape(input=images)
dimensions = tf.stack(
[
tf.as_string(image_shape[2], name="width"),
tf.as_string(image_shape[1], name="height"),
],
name="dimensions",
)
return tf.concat([dimensions, encoded_images], axis=0)
def lazy_tensor():
tf.debugging.assert_rank(data, 3)
tf.debugging.assert_non_negative(max_outputs)
limited_audio = data[:max_outputs]
encode_fn = functools.partial(
audio_ops.encode_wav, sample_rate=sample_rate
)
if lengths is not None:
tf.debugging.assert_rank(lengths, 1)
limited_lengths = lengths[:max_outputs]
def encode_with_length(datum_and_length):
datum, length = datum_and_length
return encode_fn(datum[:length])
encoded_audio = tf.map_fn(
encode_with_length,
(limited_audio, limited_lengths),
dtype=tf.string,
name="encode_each_audio",
)
else:
encoded_audio = tf.map_fn(
encode_fn,
limited_audio,
dtype=tf.string,
name="encode_each_audio",
)
# Workaround for map_fn returning float dtype for an empty elems input.
encoded_audio = tf.cond(
tf.shape(input=encoded_audio)[0] > 0,
lambda: encoded_audio,
lambda: tf.constant([], tf.string),
)
limited_labels = tf.tile([""], tf.shape(input=limited_audio)[:1])
return tf.transpose(a=tf.stack([encoded_audio, limited_labels]))
def lazy_tensor():
tf.debugging.assert_rank(data, 3)
tf.debugging.assert_non_negative(max_outputs)
limited_audio = data[:max_outputs]
encode_fn = functools.partial(audio_ops.encode_wav,
sample_rate=sample_rate)
encoded_audio = tf.map_fn(encode_fn,
limited_audio,
dtype=tf.string,
name='encode_each_audio')
# Workaround for map_fn returning float dtype for an empty elems input.
encoded_audio = tf.cond(
tf.shape(input=encoded_audio)[0] > 0, lambda: encoded_audio,
lambda: tf.constant([], tf.string))
limited_labels = tf.tile([''], tf.shape(input=limited_audio)[:1])
return tf.transpose(a=tf.stack([encoded_audio, limited_labels]))
def audio(name,
data,
sample_rate,
step,
max_outputs=3,
encoding=None,
description=None):
"""Write an audio summary.
Arguments:
name: A name for this summary. The summary tag used for TensorBoard will
be this name prefixed by any active name scopes.
data: A `Tensor` representing audio data with shape `[k, t, c]`,
where `k` is the number of audio clips, `t` is the number of
frames, and `c` is the number of channels. Elements should be
floating-point values in `[-1.0, 1.0]`. Any of the dimensions may
be statically unknown (i.e., `None`).
sample_rate: An `int` or rank-0 `int32` `Tensor` that represents the
sample rate, in Hz. Must be positive.
step: Required `int64`-castable monotonic step value.
max_outputs: Optional `int` or rank-0 integer `Tensor`. At most this
many audio clips will be emitted at each step. When more than
`max_outputs` many clips are provided, the first `max_outputs`
many clips will be used and the rest silently discarded.
encoding: Optional constant `str` for the desired encoding. Only "wav"
is currently supported, but this is not guaranteed to remain the
default, so if you want "wav" in particular, set this explicitly.
description: Optional long-form description for this summary, as a
constant `str`. Markdown is supported. Defaults to empty.
Returns:
True on success, or false if no summary was emitted because no default
summary writer was available.
"""
# TODO(nickfelt): get encode_wav() exported in the public API.
from tensorflow.python.ops import gen_audio_ops
if encoding is None:
encoding = 'wav'
if encoding != 'wav':
raise ValueError('Unknown encoding: %r' % encoding)
summary_metadata = metadata.create_summary_metadata(
display_name=None,
description=description,
encoding=metadata.Encoding.Value('WAV'))
inputs = [data, sample_rate, max_outputs, step]
with tf.summary.summary_scope(name, 'audio_summary',
values=inputs) as (tag, _):
tf.debugging.assert_rank(data, 3)
tf.debugging.assert_non_negative(max_outputs)
limited_audio = data[:max_outputs]
encode_fn = functools.partial(gen_audio_ops.encode_wav,
sample_rate=sample_rate)
encoded_audio = tf.map_fn(encode_fn,
limited_audio,
dtype=tf.string,
name='encode_each_audio')
# Workaround for map_fn returning float dtype for an empty elems input.
encoded_audio = tf.cond(
tf.shape(input=encoded_audio)[0] > 0, lambda: encoded_audio,
lambda: tf.constant([], tf.string))
limited_labels = tf.tile([''], tf.shape(input=limited_audio)[:1])
tensor = tf.transpose(a=tf.stack([encoded_audio, limited_labels]))
return tf.summary.write(tag=tag,
tensor=tensor,
step=step,
metadata=summary_metadata)