I ran this command:conda install -c maciejkula -c pytorch -c peterjc123 spotlight=0.1.4 but showed the error
conda install -c maciejkula -c pytorch -c peterjc123 spotlight=0.1.3 Fetching package metadata ........... Solving package specifications:

PackageNotFoundError: Packages missing in current channels:

• spotlight 0.1.3* -> pytorch 0.3.0 -> mkl >=2018

This PR aims to fix https://github.com/maciejkula/spotlight/issues/82

It includes,

• an Appveyor CI setup
• a fix of the randint overflow issue
• an attempt to fix dtype mismatch between IntTensor and LongTensor by casting them in forward as suggested here https://github.com/pytorch/pytorch/issues/145#issuecomment-255355000 . I must be missing something though as I still don't understand why this not an issue on Linux but only on Windows..

The latest Appveyor output can be seen here some of the failures will go away once https://github.com/maciejkula/spotlight/pull/83 is merged..

I ran into a very odd segmentation fault error. This could very well be a PyTorch bug, but I thought I'd bring it up here, first. I've produced a minimal example at the bottom of this issue.

So far, I know that the fault happens at the loss.backward() call in model.fit(). The fault only seems to happen under the combination of two conditions (that I can find, so far):

1. When sparse=True.
2. Pandas is imported at the top of the file

(BTW, I pass in an SGD optimizer because that seems to be the only one that works right now with sparse embeddings)

I'm using pandas version 0.20.3 from conda, the latest spotlight from master, and PyTorch 0.2.0 from conda. I'd love to know if others can reproduce this.

As I said, this could very well be a PyTorch bug, but, if others run into this, it'll be helpful to have this issue as a reference.

import pandas as pd
import numpy as np
import torch

from spotlight.interactions import Interactions
from spotlight.factorization.implicit import ImplicitFactorizationModel

user_ids = [2471, 5808, 3281, 4086, 6293, 8970, 11828, 3281]
item_ids = [1583, 57, 6963, 867, 8099, 10991, 24, 800]
num_users = 15274
num_items = 25655

train = Interactions(np.array(user_ids, dtype=np.int64),
                     np.array(item_ids, dtype=np.int64),
                     num_users=num_users,
                     num_items=num_items)

def optimizer_func(params, lr=0.01):
    return torch.optim.SGD(params, lr=lr)
  
RANDOM_STATE = np.random.RandomState(42)
model = ImplicitFactorizationModel(loss='bpr',
                                   embedding_dim=32,
                                   batch_size=4,
                                   n_iter=1,
                                   use_cuda=False,
                                   optimizer_func=optimizer_func,
                                   sparse=True,
                                   random_state=RANDOM_STATE)
# Fault
model.fit(train, verbose=True)

Hi,

I am using an environment where conda install is not possible. I have tried installing via pip but it doesn't work. Has anyone tried this? Would appreciate any tips.

Hi, I was looking at LightFM and saw item and user metadata being used for recommendations. This is really cool. Just wondering if such functionality is in the roadmap for spotlight?

I have a few questions about using Spotlight for an item-item problem involving graded implicit feedback, pardon me if there is a better forum for such questions, I wasn't able to find one.

I work on a system with feedback in the form of clicks (aka page view), likes and purchases. In this case obviously a purchase is substantially more desirable than a simple click.

Is there an obvious way to achieve this with Spotlight? Should I treat it as pure implicit and use the weights parameter to assign a greater weight to purchases than clicks? Or is it more appropriate to treat it as a ratings prediction problem where the "ratings" are really pseudo-ratings assigned by me?

Also, does Spotlight have any support for cold-start? Or support for predicting for a new user in production based on that user's (previously unseen) history of implicit feedback? Or would lightfm maybe be a better fit for all of this?

Finally, if deployed in production can Spotlight models predict at reasonably low latency? Perhaps <100ms?

thanks very much for Spotlight. It's well-documented and the code is a joy to read.

Using python 3.6 in conda environment. Getting the following error

conda install -c maciejkula -c soumith spotlight=0.1.2 Fetching package metadata .............

PackageNotFoundError: Package missing in current osx-64 channels:

• spotlight 0.1.2*

@maciejkula I guess we should remove the items found the training dataset before the negative sampling. Otherwise, it might make the learning less effective?

When i try to import some modules:

from spotlight.datasets.movielens import get_movielens_dataset Traceback (most recent call last): File "", line 1, in ModuleNotFoundError: No module named 'spotlight.datasets'

from spotlight.factorization.explicit import ExplicitFactorizationModel Traceback (most recent call last): File "", line 1, in ModuleNotFoundError: No module named 'spotlight.factorization'

I have created a py 3.6 env.

Please let me know how to install it .

conda install -c maciejkula -c pytorch spotlight=0.1.4 is not working

conda install -c maciejkula -c pytorch -c peterjc123 spotlight=0.1.4 even this fails

Hi! Thank you for Spotlight.

I have a similar problem to the issue #80

My OS is Windows 10 x64.

First, i must admit i am new with Anaconda. I have a 'portable' installation Anaconda (with Anaconda3-5.1.0-Windows-x86_64.exe).

When i try:

conda install -c maciejkula -c pytorch spotlight=0.1.4

i get;

u:\Python\Anaconda3>conda install -c maciejkula -c pytorch spotlight=0.1.4
Solving environment: failed

PackagesNotFoundError: The following packages are not available from current channels:

  - spotlight=0.1.4
  - pytorch=0.3.1

Current channels:

  - https://conda.anaconda.org/maciejkula/win-64
  - https://conda.anaconda.org/maciejkula/noarch
  - https://conda.anaconda.org/pytorch/win-64
  - https://conda.anaconda.org/pytorch/noarch
  - https://repo.continuum.io/pkgs/main/win-64
  - https://repo.continuum.io/pkgs/main/noarch
  - https://repo.continuum.io/pkgs/free/win-64
  - https://repo.continuum.io/pkgs/free/noarch
  - https://repo.continuum.io/pkgs/r/win-64
  - https://repo.continuum.io/pkgs/r/noarch
  - https://repo.continuum.io/pkgs/pro/win-64
  - https://repo.continuum.io/pkgs/pro/noarch
  - https://repo.continuum.io/pkgs/msys2/win-64
  - https://repo.continuum.io/pkgs/msys2/noarch

u:\Python\Anaconda3>

I have not created an environment... Is it necessary?

Perhaps, is it necessary to add channels or so?

There is a small typo in tests/factorization/test_explicit.py.

Should read incompatible rather than imcompatible.

Semi-automated pull request generated by https://github.com/timgates42/meticulous/blob/master/docs/NOTE.md

I need to migrate a code that is using spotlight to Azure Machine Learning.

When calling the module I am constantly getting these type of errors.

My python version is 3.6.9 in Azure Machine Learning and I tried to call the module on my local pc as well, but I am getting the same error.

Any news on how to fix this?

After training the model, I add a new user or item to the Interactions dataset and then try to train the model, but I get the following error:ValueError: Maximum user id greater than number of users in model. Is there any way to add users and item to the dataset and train the model online after that?

Hello! Thank you for your great work! In your realisation of bpr loss you use the following formula:

loss = (1.0 - torch.sigmoid(positive_predictions - negative_predictions)).

In the original publication (bpr paper) the authors propose loss = log(sigmoid(positive - negative)) + regularisation. Is it okay?

My question is I'm thinking is there any reason we can simplify cross entropy loss into the below way instead of what [1] used in cross-entropy.

def pointwise_loss(positive_predictions, negative_predictions, mask=None):
    """
    Logistic loss function.
    Parameters
    ----------
    positive_predictions: tensor
        Tensor containing predictions for known positive items.
    negative_predictions: tensor
        Tensor containing predictions for sampled negative items.
    mask: tensor, optional
        A binary tensor used to zero the loss from some entries
        of the loss tensor.
    Returns
    -------
    loss, float
        The mean value of the loss function.
    """

    positives_loss = (1.0 - torch.sigmoid(positive_predictions))
    negatives_loss = torch.sigmoid(negative_predictions)

    loss = (positives_loss + negatives_loss)

    if mask is not None:
        mask = mask.float()
        loss = loss * mask
        return loss.sum() / mask.sum()

    return loss.mean()

[1].https://ml-cheatsheet.readthedocs.io/en/latest/logistic_regression.html