You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
@EdgarJHeller developed as a part of his Bachelor thesis a feature related to #86 (please read this issue first - it contains explanation of some prototype).
This feature suggests the second object for comparison (more specifically it suggests a list of such object). For instance, you type "python" and you do not know which objects to compare to (what are the alternatives for python). In this case, the correct suggestions will be words like "perl, ruby, php" or maybe even words in different senses like "cobra, alligator", etc.
The goal of this task is to deploy the feature developed by @EdgarJHeller in the real system. He managed to integrate the feature into the UI and test it and also pre-compute and index which have suggestions for the majority of words in English language. However, we did not have time to properly deploy it and this will be your task.
All the codes to make running the feature shall be currently available in the master branch of this repository. Though you may need also file with the pre-computed ElasticSearch index which stores pre-computed suggestions which can be dowloaded here: es-nodes.tar.gz
@EdgarJHeller managed to run it locally but it did not quite correctly was integrated into the docker. So please try first to make sure you have the system running locally well and then make sure you can run it (first locally on your machine) with Docker. I suggest using Linux or Mac for the tests. If you are developing on Windows it would be a good idea to try to deploy directly to some staging Linux server.
To start the system locally follow the instructions in the README https://github.com/uhh-lt/cam and ask questions in this thread if something doesn't work after you made a few trials to overcome the issue.
The text was updated successfully, but these errors were encountered:
Motivation
@EdgarJHeller developed as a part of his Bachelor thesis a feature related to #86 (please read this issue first - it contains explanation of some prototype).
This feature suggests the second object for comparison (more specifically it suggests a list of such object). For instance, you type "python" and you do not know which objects to compare to (what are the alternatives for python). In this case, the correct suggestions will be words like "perl, ruby, php" or maybe even words in different senses like "cobra, alligator", etc.
Below you will find a PDF with description of the used method and more motivation about it:
5heller-BA-thesis-final-draft.pdf
Implementation
The goal of this task is to deploy the feature developed by @EdgarJHeller in the real system. He managed to integrate the feature into the UI and test it and also pre-compute and index which have suggestions for the majority of words in English language. However, we did not have time to properly deploy it and this will be your task.
All the codes to make running the feature shall be currently available in the master branch of this repository. Though you may need also file with the pre-computed ElasticSearch index which stores pre-computed suggestions which can be dowloaded here:
es-nodes.tar.gz
Below also is a link to YouTube video showing how it shall work and a few screenshots:
https://www.youtube.com/watch?v=c60cdRh7ClM&feature=youtu.be
@EdgarJHeller managed to run it locally but it did not quite correctly was integrated into the docker. So please try first to make sure you have the system running locally well and then make sure you can run it (first locally on your machine) with Docker. I suggest using Linux or Mac for the tests. If you are developing on Windows it would be a good idea to try to deploy directly to some staging Linux server.
To start the system locally follow the instructions in the README https://github.com/uhh-lt/cam and ask questions in this thread if something doesn't work after you made a few trials to overcome the issue.
The text was updated successfully, but these errors were encountered: