Skip to content

Program that classifies cuisine of the recipe given the ingredients

Notifications You must be signed in to change notification settings

lpiekarski/cuisine-classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Cuisine classification program

Program predicting the cuisine of the recipe given the ingredients list.

Run notes

Run the cuisine_classification.py script with options:

option long option argument description
-p --predict [FILE] predicts a cuisine given the json file
consisting of list of dictionaries containing key:
"ingredients" - list of ingredients
"id" - recipe id
the result is written into the file specified by
-out-file and is in .csv format
-h --help shows this message
-t --train [FILE] trains the neural network with the new data
[FILE] format should be .json list of dictionaries
containing keys:
"cuisine" - string
"ingredients" - list of ingredients
default file path is ./input/train.json
-e --epochs [NUM] changes the number of epochs executed during training
default is 10
-c --cuisines shows list of cuisines that the program is able to identify
-o --out-file [FILE] specifies the out file name for predictions

Training data examples

Additional informations

Current accuracy is ~77% but can go up to ~96% with more train time.

About

Program that classifies cuisine of the recipe given the ingredients

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages