Can we serialize/store/save the LightFM model?

[audiodude]

Maybe we can just pickle the model itself after it's been trained?

[siddz415]

user_ids = [0, 0, 1, 1, 2]
movie_ids = [0, 1, 1, 2, 2]
ratings = [5, 4, 5, 4, 5] # Only 4s and 5s for this example

Create sparse interaction matrix

interaction_matrix = coo_matrix((ratings, (user_ids, movie_ids)), shape=(3, 3))

Initialize LightFM model with WARP loss

model = LightFM(loss='warp')

Train the model

model.fit(interaction_matrix, epochs=30, num_threads=2)

Predict scores for User 0

scores = model.predict(0, np.arange(3))
top_items = np.argsort(-scores)
print("Top recommended items for User 0:", top_items)

Evaluate model

from lightfm.evaluation import precision_at_k, auc_score
precision = precision_at_k(model, interaction_matrix, k=5).mean()
auc = auc_score(model, interaction_matrix).mean()

print(f"Precision at k=5: {precision}, AUC Score: {auc}")
something like this?

[audiodude]

I think the pickleing would happen after your model.fit code.

There's also a question of how long it takes to fit the model. If it's under 10 seconds, it might not be worth it to deal with pickle.