Could not open model file for reading - english_proj_tagger.model

[AyushiAggarwal]

Using Python27, CentOS:

Upon loading the TurboParser pipeline and running the following commands from the python shell to check that the parser in TurboParser works fine, I get an error(given below).

Code in python shell:

import nlp_pipeline 
pipe = nlp_pipeline.NLPPipeline() 
print pipe.parse_conll("I used statistics to solve my problem", "EN") 

ERROR: Check failed: fs Could not open model file for reading: models/english_proj/english_proj_tagger.model

A full stack trace of the error is given in the image below.

I found the tagger file from this link - http://www.ark.cs.cmu.edu/turboParser/sample/models/english_proj_tagger.tar.gz

Solutions tried:

1. The first thing I tried was changing the permission of the file to u+x. It appears the issue is not with the access permission of the file since this did not eliminate this error.
2. It could be that the method calling the file is erroneous in opening the file. But I checked and found that the file path is correctly specified in the nlp_pipeline.config file. Looking at the load_tagger_model method that invokes the file was also did not help me figure out a solution to this problem.

Any suggestions towards debugging this or getting another file that would work for TurboParser or another route to making the parser component of TurboParser work would be very much appreciated.

Thanks!

P.S - This issue is probably the same as the one titled 'Which are the current models?' because the error message is the same as that posted in that thread, however being unsure about what is causing the error in the first place, I am creating a separate thread.