forked from microsoft/dpu-utils
-
Notifications
You must be signed in to change notification settings - Fork 0
/
unsortedsegmentops.py
54 lines (43 loc) · 2.63 KB
/
unsortedsegmentops.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
import tensorflow as tf
SMALL_NUMBER = 1e-7
def unsorted_segment_logsumexp(scores, segment_ids, num_segments):
"""Perform an unsorted segment safe logsumexp."""
# Note: if a segment is empty, the smallest value for the score will be returned,
# which yields the correct behavior
max_per_segment = tf.unsorted_segment_max(data=scores,
segment_ids=segment_ids,
num_segments=num_segments)
scattered_log_maxes = tf.gather(params=max_per_segment,
indices=segment_ids)
recentered_scores = scores - scattered_log_maxes
exped_recentered_scores = tf.exp(recentered_scores)
per_segment_sums = tf.unsorted_segment_sum(exped_recentered_scores, segment_ids, num_segments)
per_segment_logs = tf.log(per_segment_sums)
return per_segment_logs + max_per_segment
def unsorted_segment_log_softmax(logits, segment_ids, num_segments):
"""Perform an unsorted segment safe log_softmax."""
# Note: if a segment is empty, the smallest value for the score will be returned,
# which yields the correct behavior
max_per_segment = tf.unsorted_segment_max(data=logits,
segment_ids=segment_ids,
num_segments=num_segments)
scattered_maxes = tf.gather(params=max_per_segment,
indices=segment_ids)
recentered_scores = logits - scattered_maxes
exped_recentered_scores = tf.exp(recentered_scores)
per_segment_sums = tf.unsorted_segment_sum(exped_recentered_scores, segment_ids, num_segments)
per_segment_normalization_consts = tf.log(per_segment_sums)
log_probs = recentered_scores - tf.gather(params=per_segment_normalization_consts, indices=segment_ids)
return log_probs
def unsorted_segment_softmax(logits, segment_ids, num_segments):
"""Perform a safe unsorted segment softmax."""
max_per_segment = tf.unsorted_segment_max(data=logits,
segment_ids=segment_ids,
num_segments=num_segments)
scattered_maxes = tf.gather(params=max_per_segment,
indices=segment_ids)
recentered_scores = logits - scattered_maxes
exped_recentered_scores = tf.exp(recentered_scores)
per_segment_sums = tf.unsorted_segment_sum(exped_recentered_scores, segment_ids, num_segments)
probs = exped_recentered_scores / (tf.gather(params=per_segment_sums, indices=segment_ids) + SMALL_NUMBER)
return probs