def samples(self):
"""Return all stored samples."""
with self._mutex:
if self._visual_range_up_to_date:
return list(self._samples)
# calc visual range
visual_range = _VisualRange()
max_count = 0
for sample in self._samples:
histogram = sample.value
if histogram.count == 0:
# ignore empty tensor
continue
max_count = max(histogram.count, max_count)
visual_range.update(histogram.max, histogram.min)
if visual_range.max == visual_range.min and not max_count:
logger.info("Max equals to min. Count is zero.")
bins = calc_histogram_bins(max_count)
# update visual range
logger.debug(
"Visual histogram: min %s, max %s, bins %s, max_count %s.",
visual_range.min, visual_range.max, bins, max_count)
for sample in self._samples:
histogram = sample.value
histogram.set_visual_range(visual_range.max, visual_range.min,
bins)
self._visual_range_up_to_date = True
return list(self._samples)
def __init__(self, buckets, max_val, min_val, count):
self._visual_max = max_val
self._visual_min = min_val
self._count = count
self._original_buckets = buckets
# default bin number
self._visual_bins = calc_histogram_bins(count)
# Note that tuple is immutable, so sharing tuple is often safe.
self._re_sampled_buckets = ()
def __init__(self, histogram_message: Summary.Histogram):
self._msg = histogram_message
original_buckets = [
Bucket(bucket.left, bucket.width, bucket.count)
for bucket in self._msg.buckets
]
# Ensure buckets are sorted from min to max.
original_buckets.sort(key=lambda bucket: bucket.left)
self._original_buckets = tuple(original_buckets)
self._count = sum(bucket.count for bucket in self._original_buckets)
self._max = _mask_invalid_number(histogram_message.max)
self._min = _mask_invalid_number(histogram_message.min)
self._visual_max = self._max
self._visual_min = self._min
# default bin number
self._visual_bins = calc_histogram_bins(self._count)
# Note that tuple is immutable, so sharing tuple is often safe.
self._re_sampled_buckets = ()
def calc_original_buckets(np_value, stats):
"""
Calculate buckets from tensor data.
Args:
np_value (numpy.ndarray): An numpy.ndarray of tensor data.
stats (Statistics): An instance of Statistics about tensor data.
Returns:
list, a list of bucket about tensor data.
Raises:
ParamValueError, If np_value or stats is None.
"""
if np_value is None or stats is None:
raise ParamValueError("Invalid input. np_value or stats is None.")
valid_count = stats.count - stats.nan_count - stats.neg_inf_count - stats.pos_inf_count
if not valid_count:
return []
bins = calc_histogram_bins(valid_count)
first_edge, last_edge = stats.min, stats.max
if not first_edge < last_edge:
first_edge -= 0.5
last_edge += 0.5
bins = np.linspace(first_edge, last_edge, bins + 1, dtype=np_value.dtype)
hists, edges = np.histogram(np_value, bins=bins)
buckets = []
for hist, edge1, edge2 in zip(hists, edges, edges[1:]):
bucket = Bucket(edge1, edge2 - edge1, hist)
buckets.append(bucket)
return buckets