f52e199a4f
update for new version of torch
2020-06-08 10:34:41 +08:00
7f0e442d8b
循环的最后一个元素和下一次循环的第一个元素时间不连续
...
steps = np.linspace(start, end, TIME_STEP, dtype=np.float32)
其中默认参数endpoint=True，也就是说这一次循环的最后一个元素和下一次循环的第一个元素是一致的，按照RNN传递的state的特点，应该保证输入是关于时间连续的，所以不应该有相邻的时间是一致的情况，这可能不是想要的，所以我认为这里是一个小bug。
2018-11-30 21:49:23 +08:00
759aa01de7
Merge pull request #62 from keineahnung2345/502-squeeze
...
502 - remove squeeze for 1-D tensor
2018-11-13 10:33:47 +08:00
818b4b4411
Merge pull request #61 from keineahnung2345/403-comment
...
403 - move the comment to right place
2018-11-13 10:33:26 +08:00
0b3aa238ba
Merge pull request #60 from keineahnung2345/402-squeeze
...
402 - remove squeeze for 1-D array
2018-11-13 10:32:26 +08:00
5d6d4f056e
502 - remove squeeze for 1-D tensor
2018-11-13 10:05:44 +08:00
256559c12a
403 - move the comment to right place
2018-11-13 09:34:19 +08:00
7411f602fb
402 - remove squeeze for 1-D array
2018-11-13 09:13:44 +08:00
61d6dc6757
supplement another method
2018-11-12 14:24:09 +08:00
2748b02bd6
fill up the missing view() in forward() in comment
2018-11-12 14:18:49 +08:00
a9ef65ed9a
[Bug fix] 401_CNN.py
...
1. fix typo
2. remove .squeeze for 1-D tensor
2018-11-09 17:26:27 +08:00
0f4219c72c
Merge pull request #53 from keineahnung2345/302-squeeze
...
302 - remove .squeeze() from 1-D numpy array
2018-11-08 19:43:20 +08:00
867aa1d1b2
Merge pull request #52 from keineahnung2345/update-203-activation
...
update to torch 0.4
2018-11-08 19:42:52 +08:00
5b1d191946
302 - remove .squeeze() from 1-D numpy array
...
prediction.data.numpy() is already 1-D, so the .squeeze() is unnecessary.
2018-11-08 14:14:35 +08:00
59487ef2b4
update to torch 0.4
...
Warning message from old version:
/opt/conda/lib/python3.6/site-packages/torch/nn/functional.py:1047: UserWarning: nn.functional.sigmoid is deprecated. Use torch.sigmoid instead.
warnings.warn("nn.functional.sigmoid is deprecated. Use torch.sigmoid instead.")
2018-11-07 16:11:19 +08:00
906cf71b6f
update for new version of torch
2018-11-07 16:08:38 +08:00
99158fd0ca
Fixing error accurate and error tensor to numpy
2018-07-14 22:35:34 +08:00
d053113d3a
fixed for windows
2018-07-05 16:38:07 +08:00
f2ebd8a37b
update accuracy function
2018-06-29 10:09:52 +08:00
79380fb90e
fix typo
2018-06-14 22:20:51 +08:00
921b69a582
update to torch 0.4
2018-05-30 01:39:53 +08:00
7e7c9bb383
update to torch 0.4
2018-05-30 00:46:22 +08:00
e19eb5d060
Remove deprecated code
...
The use of Variable() explicitly is deprecated as Autograd now supports tensors directly.
2018-05-29 21:02:10 +05:30
c711da855a
update to torch 0.4
2018-05-10 08:48:03 +10:00
68396f5972
fix bug for windows
2018-05-04 11:39:52 +10:00
46c516fc10
fix typo
2018-03-08 00:22:06 +11:00
ebf9bfd250
update for torch 0.3
2018-01-04 16:56:37 +11:00
f65e27f71e
update for torch 0.3
2018-01-02 15:05:18 +11:00
5df56c99dd
remove random seed
2017-12-15 14:21:23 +11:00
895ad9a26c
mnist dataset download setting
2017-11-22 08:13:22 +08:00
2ae0c0fa6a
fix typo
2017-11-08 22:45:10 +08:00
9318a69f08
fix action shape problem
2017-10-30 09:58:09 +11:00
a7b14b8091
fix action shape
2017-10-27 09:06:10 +11:00
ce4a8286fd
update for torch 0.2
2017-09-05 12:57:40 +10:00
3b23f6bced
update for torch 0.2
2017-09-05 12:57:25 +10:00
e7599c0f14
update
2017-08-09 01:51:16 +10:00
1b5fa9f7aa
update
2017-06-27 17:00:01 +10:00
528d7ccd41
Fix up test code
2017-06-19 21:49:39 +02:00
51f1c938f3
update
2017-06-18 00:18:42 +10:00
dcc4e7cb84
update
2017-06-17 23:50:57 +10:00
3fc6f4e4a0
update
2017-06-11 15:49:22 +10:00
cc831bf4ef
update
2017-05-29 16:04:38 +10:00
b2922fd3f3
update
2017-05-24 13:30:46 +10:00
9fb0f5b6c6
Update 502_GPU.py
...
There are several little mistakes in this file which will make you fail to debug successfully.
2017-05-24 08:59:21 +08:00
d87529ed89
update
2017-05-23 10:37:51 +10:00
30c0070497
update
2017-05-13 18:37:25 +10:00
e95d798e68
update
2017-05-11 16:58:23 +10:00
3e3b0e9d10
update
2017-05-11 16:53:32 +10:00
53b13ff245
update
2017-05-11 13:56:13 +10:00
2523e0c1b1
update
2017-05-11 12:06:47 +10:00
