Add an observation or memory to the agent's memory bank.

Summarize the feelings of the avatar for recommended item list.

Return random integers from `0` (inclusive) to `high` (exclusive).

Fetch related memories.

convert a list of interactions into a dictionary that maps each user to a list of their interactions input: df with columns ['user_id',

Get the hours passed between two datetime objects.

Evaluate `model`. Args: model: The model need to be evaluated. This model must have a method `predict_for_eval(se

generate one page items for one avatar

Return the combined score for a document.

Return the user's feeling about recent movies.

Return the user's feeling about recent movies.

Return the user's taste about recent movies.

Add documents to vectorstore.

async excute the simulation for one avatar avatar_id: the id of the simulated avatar

async validate the effectiveness of the model for one avatar avatar_id: the id of the simulated avatar

convert a scipy.sparse.csr_matrix to a dict, where the key is row number, and value is the non-empty index in each row.

The whole process of the simulation

document the full rankings of the items, according to a specific cf model

Return documents that are relevant to the query.

Return documents that are salient to the query.

Initialize the memory of the avatar

initialize all avatars

load additional information for the simulation

load the data for simulation

load the recommender for simulation

load the recommender args, which is saved when training the recommender

Pads sequences to the same length. Args: sequences (list): A list of lists, where each element is a sequence. value (int or float