stardut / text-generate-rnn Goto Github PK

按照
text1 = '两个黄鹂鸣翠柳'
text2 = '<《春》 '
的输入文本变量，即使在训练完所有数据以后，还会存在生成的古诗中有“”和“》”这两种特殊符号的出现，
eg：
两个黄鹂鸣翠柳园花头大空》 友事调龙地，临临坐作灯。
《春》 *途居》 无安思不惺，不爲歌许子。故次客无声，此心尽谁得。
后期效果没有readme.md里面显示的那么好。
请问1.你有什么好的建议或者修改方法吗？2.net.py文件是用来建立网络的吗，实际作用是什么呢，运行应该是在紧挨哪个py文件运行之后？
本人深度学习基础薄弱，所以问的问题也很浅，望有空答复，期待您的回复！

自己训练感觉要好久的

ub16c9@ub16c9-gpu:/media/ub16c9/fcd84300-9270-4bbd-896a-5e04e79203b7/ub16_prj/Gen/Text-Generate-RNN$ python3.6 sample.py
loading preprocessed files
reading: data/vocab.pkl
words size: 11103
load data done
Traceback (most recent call last):
File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/client/session.py", line 1322, in _do_call
return fn(*args)
File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/client/session.py", line 1307, in _run_fn
options, feed_dict, fetch_list, target_list, run_metadata)
File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/client/session.py", line 1409, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.DataLossError: Unable to open table file model/model_loss_4.7890.ckpt-262000.meta: Data loss: not an sstable (bad magic number): perhaps your file is in a different file format and you need to use a different restore operator?
[[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]
[[Node: save/RestoreV2/_3 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_8_save/RestoreV2", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

ub16c9@ub16c9-gpu:/media/ub16c9/fcd84300-9270-4bbd-896a-5e04e79203b7/ub16_prj/Gen/Text-Generate-RNN$ tree
.
├── data
│   ├── poetry.txt
│   ├── tang
│   ├── tensor.npy
│   └── vocab.pkl
├── data.py
├── LICENSE
├── model
│   ├── checkpoint
│   ├── model
│   ├── model_loss_1.7149.ckpt-261000.data-00000-of-00001
│   ├── model_loss_1.7149.ckpt-261000.index
│   ├── model_loss_1.7149.ckpt-261000.meta
│   ├── model_loss_2.6208.ckpt-263000.data-00000-of-00001
│   ├── model_loss_2.6208.ckpt-263000.index
│   ├── model_loss_2.6208.ckpt-263000.meta
│   ├── model_loss_3.8258.ckpt-260000.data-00000-of-00001
│   ├── model_loss_3.8258.ckpt-260000.index
│   ├── model_loss_3.8258.ckpt-260000.meta
│   ├── model_loss_4.5588.ckpt-264000.data-00000-of-00001
│   ├── model_loss_4.5588.ckpt-264000.index
│   ├── model_loss_4.5588.ckpt-264000.meta
│   ├── model_loss_4.7890.ckpt-262000.data-00000-of-00001
│   ├── model_loss_4.7890.ckpt-262000.index
│   └── model_loss_4.7890.ckpt-262000.meta
├── net.py

I liked your model . i have concern .does it generate sentence ?

eg
input : fever
output : i am having fever since last night