reccomendFor empty array

[gsmeros]

No description provided.

[guymorita]

How many ratings were there? Has user 1 already rated everything?

—
Sent from Mailbox

On Sat, Oct 10, 2015 at 8:47 PM, George Smeros notifications@github.com
wrote:

var raccoon = require('raccoon');
var mysql = require('mysql');
var connection = mysql.createConnection({
...
});
connection.connect(function(err) {
if (err) {
console.error('error connecting: ' + err.stack);
return;
}
console.log('connected as id ' + connection.threadId);
});
raccoon.connect(6379, '127.0.0.1');
connection.query('SELECT * FROM ratings' , function (error, results, fields) {
for (var i = 0; i < results.length; i++) {
if (results[i].rating > 2) {
raccoon.liked(results[i].userId, results[i].movieId, function () {
console.log("t");
});
} else {
raccoon.disliked(results[i].userId, results[i].movieId, function () {
});
}
}
raccoon.recommendFor(1, 1 , function(results){
console.log(results);
});
});

there is a user 1 but I get an empty array

Reply to this email directly or view it on GitHub:
#35

[gsmeros]

i am looping a query and insert ratings . if i use then the users ratings and then recommend it works fine! but it seems that I have problems with async because I get out of memory and node keeps running 100%cpu. my code :

connection.query('SELECT * FROM ratings', function (error, results, fields) {
results.forEach(function(user){
if (user.rating < 3)
raccoon.disliked(user.userId, user.movieId, function() {});
else
raccoon.liked(user.userId, user.movieId, function() {});
});
connection.end();

if it is a loop , where can I put the recommendFor ?

[gsmeros]

I also tried this :
{connection.query('SELECT * FROM ratings, function (error, results, fields) {
async.parallel([
function(done) {
console.log("Updating Engine");
results.forEach(function(user){
if (user.rating < 3)
raccoon.disliked(user.userId, user.movieId, function() {});
else
raccoon.liked(user.userId, user.movieId, function() {});
});
done(null);
}], function (err) {
response.send("finished");
});
});}

[gsmeros]

Also the redis server keys * :

1. "movie:2788:liked"
2. "movie:3880:disliked"
3. "movie:1226:liked"
4. "movie:592:liked"
5. "movie:3508:liked"
6. "movie:2353:liked"
7. "movie:317:liked"

[gsmeros]

sorry for the trouble. I manage to fix mostly everything with alternative code. but it seems that my database ( 100.000 ratings from movielens ) is too big to uploaded into redis. It crashes after 13-14k entries

[guymorita]

it looks like every set should be able to hold 2^32,
http://redis.io/topics/faq

sure that's it? what was the fix? thanks for keeping me in the loop!

On Tue, Oct 13, 2015 at 10:11 AM, George Smeros notifications@github.com
wrote:

sorry for the trouble. I manage to fix mostly everything with alternative
code. but it seems that my database ( 100.000 ratings from movielens ) is
too big to uploaded into redis. It crashes after 13-14k entries

—
Reply to this email directly or view it on GitHub
#35 (comment)
.

Guy Morita

T 206.240.5846 l guymorita@gmail.com guymorita@gmail.com

LinkedIn http://www.linkedin.com/in/guymorita l Github
https://github.com/guymorita l Developer Blog
http://guymorita.tumblr.com l Svbtle http://alchemist.svbtle.com

[gsmeros]

If I use this if statement and then try to recommendFor in the callback it doesnt work. Redis is still storing in the background. So what I did i am just trying to put everything up from my training dataset into redis.(Stops at 13k keys) then with every like/dislike of the user I return the updated recommendset. I think maybe Redis can handle the request but node is crashing while it tries to store. Probably a memory issue in my end , but the macbook I am using has no problems...

[gsmeros]

I used a remote server and in a 100.000 ratings table (movie lens) I got 13.8k keys into redis. is this correct ? does this count movieIds only ?

[joshbedo]

Same thing happened to me, I simply put code in my test that was displayed in the example. I logged into my redis server and made sure that the keys do exist. Here is the code i have.

NOTE: I'm connecting to a docker container by the docker machine ip address so its a remote connection

var expect = require('chai').expect;
var raccoon = require('raccoon');
raccoon.connect(6379, '192.168.99.100');

describe('this doesnt work', function() {

  before(function(done) {
    raccoon.flush();
    done();
  });

  before(function(done) {
    raccoon.liked('garyId', 'movieId');
    raccoon.liked('garyId', 'movie2Id');
    raccoon.liked('chrisId', 'movieId');
    done();
  });

  it('should show movie2', function(done) {
    raccoon.recommendFor('chrisId', 10, function(results) {
      console.log(results, 'results'); // empty array returned
      expect(true).to.eq(false);
      done();
    })
  });
})

[guymorita]

@joshbedo I updated the readme to show that the liked /disliked calls are async. They now use promises. So if you call recommendFor before they resolve, you won't get any recommendations.

[guymorita]

@gsmeros I'm going to try the Movielens database as well and I'll let you know what the results are!

For every user there are:
liked lists
disliked lists
similarity sets
recommendation sets

For every item there are:
liked lists
disliked lists

Others include:
most liked set
most disliked set
overall score