ne7ermore / deeping-flow Goto Github PK

After I run the lstm-cnn-classfication, the output turns out that the val-loss rate is increased.
But in the result.png, the train-loss is decreased. How can I make the train-loss displayed in terminal? so that I can debug with it.

hi,
I tried to run deeping-flow/reinforced-translate/ but got an error:

Traceback (most recent call last):
File "train.py", line 65, in
rl = Reinforced(model, args)
File "/Users/zhengwenjia/Workspace/Temp/deeping-flow/reinforced-translate/model.py", line 176, in init
mask = pad_mask(model.tgt, EOS, [args.batch_size, args.max_len])
NameError: name 'EOS' is not defined

Hi，

https://github.com/ne7ermore/deeping-flow/blob/master/reinforced-translate/model.py#L175

mask = pad_mask(model.tgt, EOS, [args.batch_size, args.max_len])

这里是不是应该是 MC 采样的作为target？ 应该是s_words?

Thanks

corpus文件在哪儿？

Hi，

https://github.com/ne7ermore/deeping-flow/blob/master/reinforced-translate/model.py

b_words = model.sample(self.prev)
s_words, s_props = model.sample(self.prev, False)

rewards = self.compute_levenshtein(model.tgt, s_words)
baseline = self.compute_levenshtein(model.tgt, b_words)
advantage = rewards - baseline

其中的 b-words 和 s-words 都是 Tensor 类型， 而计算 compute_levenshtein 里面直接用的 for 循环展开，用的 eager 模式？

Thanks

Sorry, I have encountered a UnicodeDecodeError when I run the corpus.py with python3.6.
The problem is following:

Traceback (most recent call last):
File "corpus.py", line 136, in
Corpus()
File "corpus.py", line 69, in init
self.parse_data("data/train.txt", True)
File "corpus.py", line 75, in parse_data
for line in open(inf):
File "/home/guest/Users/Wangyue/virtualenv/tensorflow_py36/lib/python3.6/codecs.py", line 321, in decode
(result, consumed) = self._buffer_decode(data, self.errors, final)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xe4 in position 513: invalid continuation byte

Can you help me to solve this problem?
Thank you.