SPOClab-ca / COVFEFE

COre Variable Feature Extraction Feature Extractor
Apache License 2.0
30 stars 10 forks source link

Output Folder is empty #19

Closed anandhperumal closed 4 years ago

anandhperumal commented 4 years ago

Python command

python covfefe.py -i input_data -o output_folder -p lex

and in my input folder there is only one file with following lines

Hi How are you
I'm good

and in my output folder it creates lexicosyntactics folder but nothing is there inside that folder. Let me know if I'm missing anything

lucky-bai commented 4 years ago

This looks right to me. Are there any errors in the output when you run that command?

anandhperumal commented 4 years ago

Nope Where can I check the complete logs?

lucky-bai commented 4 years ago

The only logs are output to stdout. Can you paste the output when you run the command?

anandhperumal commented 4 years ago

Sorry for the delay. There is nothing image

lucky-bai commented 4 years ago

Does your input file have the extension .txt?

daniyall commented 4 years ago

+1 for checking the file extension. The fact that there were no warnings or progress bar makes me think that the file_finder didn't find any input files.

anandhperumal commented 4 years ago

@luckytoilet @daniyall Thanks my file didn't have txt extension. But now fixing that leads to me a new issue:

anandhpe_ualberta_ca@cpu8ram30:~/installs/GitHub/COVFEFE$ python3 covfefe.py -i input_data -o output_folder -p lex
  0%|                                                                                        | 0/1 [00:00<?, ?it/s]INFO:utils.logger:lexicosyntactic: Starting input_data/file1.txt
ERROR:utils.logger:lexicosyntactic: Failed with error 

===========================================================================
NLTK was unable to find the java file!
Use software specific configuration paramaters or set the JAVAHOME environment variable.
===========================================================================
INFO:utils.logger:lexicosyntactic: Done input_data/file1.txt -> output_folder/lexicosyntactic/file1.csv
100%|████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00,  1.86s/it]
anandhpe_ualberta_ca@cpu8ram30:~/installs/GitHub/COVFEFE$ cd output_folder/lexicosyntactic/
anandhpe_ualberta_ca@cpu8ram30:~/installs/GitHub/COVFEFE/output_folder/lexicosyntactic$ ls
lu_parses  rst_output  stanford_parses

There is no file1.csv and all other folders are empty

daniyall commented 4 years ago

Do you have JAVA_HOME set?

Looks similar to https://github.com/nltk/nltk/issues/1476

anandhperumal commented 4 years ago

Hi @daniyall, Yes image

daniyall commented 4 years ago

This error is because one of the scripts used by the lex features still uses python2. But it looks like calling python2 is not valid on your system. The immediate fix would be to create a link somewhere in your path that links python2 to python or to change https://github.com/SPOClab-ca/COVFEFE/blob/7ee3233aaf51d5aeea0852fd0f64b77bc6c63bc8/utils/lexicosyntactic/syntactic_features.py#L31 to say python instead of python2.

anandhperumal commented 4 years ago

@daniyall Thanks for the input image

Now there is some other issue, what is receptiviti and lexicosyntactic is supposed to be created as a ouput but then it's throwing error no such file or directory.

ERROR:utils.logger:lexicosyntactic: Failed with error [Errno 2] No such file or directory: '/home/anandhpe_ualberta_ca/installs/GitHub/COVFEFE/output_folder/lexicosyntactic/lu_parses/file1.txt'
daniyall commented 4 years ago

You can ignore the receptiviti error. The lexicosyntactic error looks a bit odd. Can you try deleting output_folder and running the pipeline again?

anandhperumal commented 4 years ago

@daniyall sorry for the delay tried deleting the output folder it didn't help image

daniyall commented 4 years ago

Is there anything inside the lu_parses folder?

anandhperumal commented 4 years ago

@daniyall nothing is there inside lu_parser, Nevermind I tried in another server it's working fine. I guess something is wrong with my env. Thanks anyway

kearnsdm commented 3 years ago

I have the same problem as @anandhperumal exactly. I also made the change he made, and it caused the same problem for me. In the subprocess.call in syntactic_features.py, it mentions 'python2' and 'analyzeText.py' ... where is the latter coming from? Is it necessary to reference python2? I'm working in a virtual env per the ReadMe, but it's not working.

daniyall commented 3 years ago

I believe some of the dependencies require python2 which is why its being called. Are you using Windows, Mac or Linux?

kearnsdm commented 3 years ago

Windows... I have Python 2 installed but I am using a virtual environment bc COVFEFE wants old versions of programs . Not sure if that could raise an issue

Get Outlook for iOShttps://aka.ms/o0ukef


From: daniyall notifications@github.com Sent: Monday, December 7, 2020 5:23:08 AM To: SPOClab-ca/COVFEFE COVFEFE@noreply.github.com Cc: kearnsdm devin.kearns@gmail.com; Comment comment@noreply.github.com Subject: Re: [SPOClab-ca/COVFEFE] Output Folder is empty (#19)

I believe some of the dependencies require python2 which is why its being called. Are you using Windows, Mac or Linux?

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/SPOClab-ca/COVFEFE/issues/19#issuecomment-739823798, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AB2536UQ5A5R66VADTFIFLDSTSUIZANCNFSM4JB7F3JA.

kearnsdm commented 3 years ago

It looks like it is just analyzeText that is needed from Python 2. Is there an analogue in Python 3 or another way to handle this?