This program analyzes the datasets available at kaggle and outputs the following three results
- Posts Score Average
- Memes liked by students
- Download meme with best average sentiment
Datasets should be located at the datasets
directory, the default expected names are complete-posts.csv
and complete-comments.csv
The meme will be downloaded in the results
directory
To run the program with docker simply run make docker-compose-up
The program allows the following customizations:
-
Student Nodes Replicas: The nodes that calculate student memes can have multiple replicas this is done via the
STUDENT_MEME_CALCULATOR_REPLICAS
environment variable that is both located atlocal.env
and thedocker-compose.yaml
-
Student Memes Repeat: The systems prints the URLs for the student memes. They might be duplicated. Allowing duplicates can be toggled by changing
STUDENT_MEME_ALLOW_REPEATS
in the docker compose -
Input files and chunks: For both the posts and comments we can configure where the program should be looking for the files and how many lines in a single chunk. Variables:
POSTS_FILE_PATH
,POSTS_FILE_CHUNK
,COMMENT_FILE_PATH
,COMMENT_FILE_CHUNK
. In case of changing the paths, the volumes should also be updated. -
Meme Result: The path of where the meme will be downloaded and how many bytes per request are downloaded is also configurable
MEME_RESULT_PATH
,MEME_RESULT_BUFFER