geniusai-research / email-summarization Goto Github PK

I am using this model to summarize around 17k sentences and it is taking more than 10 minute for clustering.

On running the train module from this github repository

import train
train.trainer(list)

While running the above code. we are getting error in Theano. error file has been attached.

[error.txt](https://github.com/jatana-research/email-summarization/files/2414610/error.txt)
error2.txt

Thanks in advance.

I have been facing the memory error in my project. I am using an EC2 instance with 8GB of RAM.

Am facing issue in encode function of skipthought.py
Below line has thrown error:
uembedding[j,ind] = model['utable']['UNK']

KeyError: 'UNK'

am using below package version:
Theano== 0.7.0
numpy==1.15.1
keras== 2.2.2
python==2.7.15

Hi I followed your medium article, great article and thanks for sharing. But when I tried to execute the code it stops executing after the line given below which is in skipthought. py file and init_params(options) function.

params = get_layer(options['encoder'])[0](options, params, prefix='encoder',
                                              nin=options['dim_word'], dim=options['dim'])

I couldn't figure out the issue. Can you spot it?

FYI: I'm using Python 3.6.4 :: Anaconda custom and I've fixed the print statement and import errors.