tb_chainer/summary.py
# Copyright 2016 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""## Generation of summaries.
### Class for writing Summaries
@@FileWriter
@@FileWriterCache
### Summary Ops
@@tensor_summary
@@scalar
@@histogram
@@audio
@@image
@@merge
@@merge_all
## Utilities
@@get_summary_description
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import logging
import re as _re
import bisect
from six import StringIO
from six.moves import range
from PIL import Image
import numpy as np
import chainer.cuda
try:
import cupy
except ImportError:
print('Not found cupy.')
# pylint: disable=unused-import
from .src.summary_pb2 import Summary
from .src.summary_pb2 import HistogramProto
from .src.summary_pb2 import SummaryMetadata
from .src.tensor_pb2 import TensorProto
from .src.tensor_shape_pb2 import TensorShapeProto
_INVALID_TAG_CHARACTERS = _re.compile(r'[^-/\w\.]')
def _clean_tag(name):
# In the past, the first argument to summary ops was a tag, which allowed
# arbitrary characters. Now we are changing the first argument to be the node
# name. This has a number of advantages (users of summary ops now can
# take advantage of the tf name scope system) but risks breaking existing
# usage, because a much smaller set of characters are allowed in node names.
# This function replaces all illegal characters with _s, and logs a warning.
# It also strips leading slashes from the name.
if name is not None:
new_name = _INVALID_TAG_CHARACTERS.sub('_', name)
new_name = new_name.lstrip('/') # Remove leading slashes
if new_name != name:
logging.info(
'Summary name %s is illegal; using %s instead.' %
(name, new_name))
name = new_name
return name
def scalar(name, scalar, collections=None):
"""Outputs a `Summary` protocol buffer containing a single scalar value.
The generated Summary has a Tensor.proto containing the input Tensor.
Args:
name: A name for the generated node. Will also serve as the series name in
TensorBoard.
tensor: A real numeric Tensor containing a single value.
collections: Optional list of graph collections keys. The new summary op is
added to these collections. Defaults to `[GraphKeys.SUMMARIES]`.
Returns:
A scalar `Tensor` of type `string`. Which contains a `Summary` protobuf.
Raises:
ValueError: If tensor has the wrong shape or type.
"""
name = _clean_tag(name)
if not isinstance(scalar, float):
# try conversion, if failed then need handle by user.
scalar = float(scalar)
return Summary(value=[Summary.Value(tag=name, simple_value=scalar)])
def histogram(name, values, bins, collections=None):
# pylint: disable=line-too-long
"""Outputs a `Summary` protocol buffer with a histogram.
The generated
[`Summary`](https://www.tensorflow.org/code/tensorflow/core/framework/summary.proto)
has one summary value containing a histogram for `values`.
This op reports an `InvalidArgument` error if any value is not finite.
Args:
name: A name for the generated node. Will also serve as a series name in
TensorBoard.
values: A real numeric `Tensor`. Any shape. Values to use to
build the histogram.
collections: Optional list of graph collections keys. The new summary op is
added to these collections. Defaults to `[GraphKeys.SUMMARIES]`.
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer.
"""
name = _clean_tag(name)
hist = make_histogram(values.astype(float), bins)
return Summary(value=[Summary.Value(tag=name, histo=hist)])
def make_histogram(values, bins):
"""Convert values into a histogram proto using logic from histogram.cc."""
values = values.reshape(-1)
counts, limits = np.histogram(values, bins=bins)
limits = limits[1:]
sum_sq = values.dot(values)
return HistogramProto(min=values.min(),
max=values.max(),
num=len(values),
sum=values.sum(),
sum_squares=sum_sq,
bucket_limit=limits,
bucket=counts)
def image(tag, tensor):
"""Outputs a `Summary` protocol buffer with images.
The summary has up to `max_images` summary values containing images. The
images are built from `tensor` which must be 3-D with shape `[height, width,
channels]` and where `channels` can be:
* 1: `tensor` is interpreted as Grayscale.
* 3: `tensor` is interpreted as RGB.
* 4: `tensor` is interpreted as RGBA.
The `name` in the outputted Summary.Value protobufs is generated based on the
name, with a suffix depending on the max_outputs setting:
* If `max_outputs` is 1, the summary value tag is '*name*/image'.
* If `max_outputs` is greater than 1, the summary value tags are
generated sequentially as '*name*/image/0', '*name*/image/1', etc.
Args:
tag: A name for the generated node. Will also serve as a series name in
TensorBoard.
tensor: A 3-D `uint8` or `float32` `Tensor` of shape `[height, width,
channels]` where `channels` is 1, 3, or 4.
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer.
"""
tag = _clean_tag(tag)
assert isinstance(tensor, np.ndarray) or isinstance(tensor, cupy.ndarray), 'input tensor should be one of numpy.ndarray, cupy.ndarray'
if not isinstance(tensor, np.ndarray):
assert tensor.ndim<4 and tensor.ndim>1, 'input tensor should be 3 dimensional.'
if tensor.ndim==2:
tensor = cupy.expand_dims(tensor, 0)
tensor = chainer.cuda.to_cpu(cupy.transpose(tensor, (1,2,0)))
else:
if tensor.ndim==2:
tensor = np.expand_dims(tensor, 0)
tensor = np.transpose(tensor, (1,2,0))
tensor = tensor.astype(np.float32)
tensor = (tensor*255).astype(np.uint8)
image = make_image(tensor)
return Summary(value=[Summary.Value(tag=tag, image=image)])
def make_image(tensor):
"""Convert an numpy representation image to Image protobuf"""
height, width, channel = tensor.shape
image = Image.fromarray(tensor)
import io
output = io.BytesIO()
image.save(output, format='PNG')
image_string = output.getvalue()
output.close()
return Summary.Image(height=height,
width=width,
colorspace=channel,
encoded_image_string=image_string)
def video(tag, tensor, fps):
tag = _clean_tag(tag)
assert isinstance(tensor, np.ndarray) or isinstance(tensor, cupy.ndarray), 'input tensor should be one of numpy.ndarray, cupy.ndarray'
if isinstance(tensor, np.ndarray):
xp = np
else:
xp = cupy
assert tensor.ndim==5, 'input tensor should be 5 dimensional. (batch, channels, time, height, width)'
b, c, t, h, w = tensor.shape
if tensor.dtype == xp.uint8:
tensor = xp.float32(tensor) / 255.
def is_power2(num):
return num != 0 and ((num & (num - 1)) == 0)
# pad to power of 2
while not is_power2(tensor.shape[0]):
tensor = xp.concatenate((tensor, xp.zeros(shape=(1, c, t, h, w))), axis=0)
b = tensor.shape[0]
n_rows = 2**(int(xp.log(b) / xp.log(2)) // 2)
n_cols = b // n_rows
tensor = np.reshape(tensor, newshape=(n_rows, n_cols, c, t, h, w))
tensor = np.transpose(tensor, axes=(3, 0, 4, 1, 5, 2))
tensor = np.reshape(tensor, newshape=(t, n_rows * h, n_cols * w, c))
tensor = tensor.astype(xp.float32)
tensor = (tensor * 255).astype(xp.uint8)
tensor = chainer.cuda.to_cpu(tensor)
video = make_video(tensor, fps)
return Summary(value=[Summary.Value(tag=tag, image=video)])
def make_video(tensor, fps):
try:
import moviepy.editor as mpy
except ImportError:
print('add_video needs package moviepy')
return
import tempfile
t, h, w, c = tensor.shape
# encode sequence of images into gif string
clip = mpy.ImageSequenceClip(list(tensor), fps=fps)
with tempfile.NamedTemporaryFile() as f:
filename = f.name + '.gif'
clip.write_gif(filename, verbose=True)
with open(filename, 'rb') as f:
tensor_string = f.read()
return Summary.Image(height=h, width=w, colorspace=c, encoded_image_string=tensor_string)
def audio(tag, tensor, sample_rate=44100):
tensor = tensor.squeeze()
assert tensor.ndim==1, 'input tensor should be 1 dimensional.'
tensor_list = [int(32767.0*x) for x in tensor]
import io
import wave
import struct
fio = io.BytesIO()
Wave_write = wave.open(fio, 'wb')
Wave_write.setnchannels(1)
Wave_write.setsampwidth(2)
Wave_write.setframerate(sample_rate)
tensor_enc = b''
for v in tensor_list:
tensor_enc += struct.pack('<h', v)
Wave_write.writeframes(tensor_enc)
Wave_write.close()
audio_string = fio.getvalue()
fio.close()
audio = Summary.Audio(sample_rate=sample_rate, num_channels=1, length_frames=len(tensor_list), encoded_audio_string=audio_string, content_type='audio/wav')
return Summary(value=[Summary.Value(tag=tag, audio=audio)])
def text(tag, text):
import json
PluginData = [SummaryMetadata.PluginData(plugin_name='text')]
smd = SummaryMetadata(plugin_data=PluginData)
tensor = TensorProto(dtype='DT_STRING', string_val=[text.encode(encoding='utf_8')])
return Summary(value=[Summary.Value(node_name=tag, metadata=smd, tensor=tensor)])