rxnn 0.2.36__py3-none-any.whl → 0.2.37__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
rxnn/training/rl.py CHANGED
@@ -112,7 +112,7 @@ class PPOAlgorithm(RlAlgorithm):
112
112
  advantages = torch.zeros_like(rewards, device=rewards.device)
113
113
  last_advantage = 0
114
114
  next_value = last_value
115
- next_done = last_done
115
+ next_done = last_done.float()
116
116
  dones = dones.float()
117
117
 
118
118
  for t in reversed(range(trajectory_len)):
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.3
2
2
  Name: rxnn
3
- Version: 0.2.36
3
+ Version: 0.2.37
4
4
  Summary: RxNN: Reactive Neural Networks Platform
5
5
  License: Apache-2.0
6
6
  Keywords: deep-learning,ai,machine-learning
@@ -19,7 +19,7 @@ rxnn/training/ddp.py,sha256=VsNBjn3cY-uUj8hbsW7oKvb0_ZKnXnJ2KgObm-Mr9i4,836
19
19
  rxnn/training/models.py,sha256=y-9XHedSheyK1AmLBp3ayulnUvAmDuJ3t0qVg8wHBRg,7463
20
20
  rxnn/training/mrl.py,sha256=fIrg1Er0aAK4TnyDRmJC1m7az9wdkhikxv0CBCrGT-c,55868
21
21
  rxnn/training/reward.py,sha256=B7nerPk9eNAv2i7umtNF88tVQVwijNNrchIrEITGHKk,11623
22
- rxnn/training/rl.py,sha256=DKB7oiePVMz-8Tp3jJOzxVlPZMd7HUEOJUurCJj9f68,5974
22
+ rxnn/training/rl.py,sha256=47wxFeUSHSqc1dKEEy8skTcNHDqNuthsYTGA-HeUbhg,5982
23
23
  rxnn/training/scheduler.py,sha256=LcjU35mEwz2U5x3U6tLfeeYlBqMxbFSxYzJYuXkWbSY,1408
24
24
  rxnn/training/tokenizer.py,sha256=umaLByMBx_NMrQElA45HLm9gkuzyKWDTFaKVd-CjXl0,8344
25
25
  rxnn/training/utils.py,sha256=Bw8nZLKIt7NQpUVCYkb_79kWKChVFOYgYXwODo4SvNc,5718
@@ -33,7 +33,7 @@ rxnn/transformers/moe.py,sha256=j6jEx6Ip0zttlUZKKn82azxo95lkLZs-H2GLSMD88hY,5859
33
33
  rxnn/transformers/positional.py,sha256=1PjcJybUzeQlIKJI4tahAGZcYgCRCL0otxs7mpsNuzM,4410
34
34
  rxnn/transformers/sampler.py,sha256=t6iiQTdLQ0TakUWnnhKkb5DKF2F_9-thXHBydDF3fxg,17389
35
35
  rxnn/utils.py,sha256=ihb6OTyDtPiocB_lOvnq7eOkjjpCkgs8wxvXUBNQ7mM,996
36
- rxnn-0.2.36.dist-info/LICENSE,sha256=C8coDFIUYuOcke4JLPwTqahQUCyXyGq6WOaigOkx8tY,11275
37
- rxnn-0.2.36.dist-info/METADATA,sha256=TZZFu-DI5dxsjYVR3qXL9UyTY_Z53Br9zlk3PtE2mlU,25960
38
- rxnn-0.2.36.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
39
- rxnn-0.2.36.dist-info/RECORD,,
36
+ rxnn-0.2.37.dist-info/LICENSE,sha256=C8coDFIUYuOcke4JLPwTqahQUCyXyGq6WOaigOkx8tY,11275
37
+ rxnn-0.2.37.dist-info/METADATA,sha256=GXBCyK-3ALJw6TpVk7rJ7Z2uyFSq8u8N-TYpmQaeUE8,25960
38
+ rxnn-0.2.37.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
39
+ rxnn-0.2.37.dist-info/RECORD,,
File without changes
File without changes