lyst lightfm issues - Githubissues

lyst / lightfm

A Python implementation of LightFM, a hybrid recommendation algorithm.

Apache License 2.0

4.73k stars 691 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

how interactions are defined?

#561 ynorouzz opened 3 years ago
4
No user feature matrix defined but still get ValueError: The user feature matrix specifies more features than there are estimated feature embeddings

#560 syl-kim opened 3 years ago
1
Illegal hardware instruction (core dumped) when installing via conda-forge

#559 SimonCW closed 3 years ago
9
Deleting user or product from a pretrained model

#558 sgaseretto opened 4 years ago
2
How the model calculates the bias?

#557 bivar opened 4 years ago
0
Item cold-start prediction

#556 ZaruhiNavasardyan opened 4 years ago
4
Model input: CSR vs COO sparse matrices ? (Different results)

#555 jimzer opened 4 years ago
1
Issues with pre-trained embedding as item features

#554 mohammad1234 closed 3 years ago
0
ValueError: Test interactions matrix and train interactions matrix share 95786 interactions. This will cause incorrect evaluation, check your data split.

#553 kholoudsaa opened 4 years ago
1
Question building binary recommender system

#552 brk030 opened 4 years ago
3
Including item features seems to reduce performance.

#551 jemmott opened 4 years ago
12
Update Hybrid Recommender Tutorial

#550 Valenzione closed 4 years ago
1
hybrid model inference with user features only no user id

#549 AbDsML opened 4 years ago
0
ModuleNotFoundError: No module named 'lightfm._lightfm_fast_no_openmp' Docker usage fails on macOS:

#548 tandav opened 4 years ago
4
documentation page is down

#547 alimirferdos closed 4 years ago
1
Social network use case: Handle user to user interactions (items = users, reciprocal)

#546 kdevo opened 4 years ago
0
user bias term for WARP loss

#545 tianpunchh opened 4 years ago
0
movielens example: recall@k vs. precision@k vs. auc

#544 deckikwok opened 4 years ago
2
Different latent vectors for same (test)users

#543 momassimo opened 4 years ago
8
Update readme development instructions

#542 matthewhughes934 closed 4 years ago
1
CI improvements

#541 iserko closed 4 years ago
2
how can we give recommendations for new user if we only have user and item interaction data for new user

#540 namansahore closed 4 years ago
1
How to print train loss at each epoch?

#539 ness001 opened 4 years ago
0
What's a good way to "weight" the importance of item/user features in LightFM?

#538 freytheviking opened 4 years ago
2
Can take dot product because user and item are in the same latent space?

#537 SantiDu closed 4 years ago
1
How to normalize numerical features when combining numerical and categorical features?

#536 SantiDu closed 4 years ago
1
Item similarity without using `interactions` data?

#535 econti opened 4 years ago
0
Normalize explicit ratings for weights? Train more models with different types of interactions?

#534 SantiDu closed 4 years ago
1
Predict with ndarrays/int only

#533 guidopetri closed 4 years ago
4
Predicting using a pandas Series tries to allocate gigantic array

#532 guidopetri closed 4 years ago
2
Question about black formatting

#531 npow opened 4 years ago
0
Interpreting negative scores from lightfm.LightFM.predict()

#530 freytheviking closed 4 years ago
6
Evaluating precision@k/recall@k for users with <k "known positives"

#529 keryums opened 4 years ago
1
Prediction for Users with multiple items

#528 dawnzaw opened 4 years ago
2
Typo fix - Build dataset section.

#527 wbashir closed 4 years ago
1
Illegal Instruction (core dumped)

#526 bradfox2 closed 4 years ago
1
Same prediction when using user features and item features

#525 eveTu opened 4 years ago
0
Getting top users for an item

#524 chakra1166 opened 4 years ago
2
"No module named '_lightfm_fast_no_openmp'" on a Macbook pro-2017

#523 hixtc opened 4 years ago
4
Turn on OpenMP compiler option in Windows.

#522 nchaly opened 4 years ago
2
Do large negative user biases mean anything?

#521 trygvebw opened 4 years ago
3
Getting different recommendations

#520 yashkan27 opened 4 years ago
0
how to deal with fit partial option in products recommendations system of e-commerce to new data points

#519 surajPowerWeave opened 4 years ago
2
User - Item interaction weights optimisations

#518 Gaurav470 closed 4 years ago
1
Pre-processing support

#517 zkid18 closed 4 years ago
1
Predictions using only item features and no user features

#516 chakra1166 closed 4 years ago
1
The expected workflow for mapping the external id's while doing predict

#515 zkid18 closed 4 years ago
4
Handle the item and user features in Dataset

#514 zkid18 opened 4 years ago
0
How to predict

#513 AjiNugroho closed 4 years ago
3
LightFM.predict returning np.float64 instead of np.float32

#512 PaschalisSk closed 3 years ago
1

Previous Next