Functions — github.com/AkaliKong/MiniOneRec

hub / github.com/AkaliKong/MiniOneRec / functions

Functions426 in github.com/AkaliKong/MiniOneRec

⨍ Functions 426◇ Types & classes 56 ↳ Endpoints 24

↓ 30 callersMethodencode

(self, s: str, bos: bool, eos: bool)

data.py:20

↓ 18 callersMethod__init__

(self, train_file, tokenizer, max_len=2048, sample=-1, test = False, seed=0, category="", K=4, dedup=False)

data.py:109

↓ 13 callersFunctionclean_text

Clean text by removing HTML and excessive whitespace.

data/amazon23_data_process.py:12

↓ 13 callersMethodget_inputs

(self)

data.py:56

↓ 12 callersMethodencode

(self, s: str, bos: bool, eos: bool)

ts_rec_data.py:21

↓ 10 callersMethodgenerate_prompt

(self, data_point)

data.py:80

↓ 10 callersFunctionset_color

(log, color, highlight=True)

rq/utils.py:10

↓ 9 callersMethod_make_response

(self, content, status_code=200)

tests/test_minimax_provider.py:125

↓ 8 callersFunctionget_hash

(x)

evaluate.py:24

↓ 6 callersMethod_make_response

(self, content, status_code=200)

tests/test_minimax_provider.py:60

↓ 6 callersMethodlog

(self, logs: dict[str, float], start_time: Optional[float] = None)

minionerec_trainer.py:1083

↓ 5 callersMethodget_indices

(self, xs, use_sk=False)

rq/models/rqvae.py:69

↓ 4 callersFunctionclean_text

Clean text by removing HTML tags and excessive whitespace

data/amazon18_data_process_gpr.py:15

↓ 4 callersFunctionclean_text

Clean text by removing HTML tags and excessive whitespace

data/amazon18_data_process.py:15

↓ 4 callersMethodencoder

(self, input_tensor)

utility.py:141

↓ 4 callersFunctionsemantic_tokens_to_id

Convert semantic tokens list to concatenated string with brackets preserved

convert_dataset_gpr.py:37

↓ 4 callersFunctionsemantic_tokens_to_id

Convert semantic tokens list to concatenated string with brackets preserved

convert_dataset.py:37

↓ 4 callersFunctionwrite_json_file

Write data to JSON file

data/amazon18_data_process_gpr.py:35

↓ 3 callersMethod_get_per_token_logps

(self, model, input_ids, attention_mask, logits_to_keep)

minionerec_trainer.py:627

↓ 3 callersMethod_save_checkpoint

(self, epoch, collision_rate=1, ckpt_file=None)

rq/trainer.py:154

↓ 3 callersFunctionevaluate_games

(model, test_data, device, topk, save_logits=False, eval_type="test")

↓ 3 callersMethodforward_eval

(self, states, len_states)

sasrec.py:114

↓ 3 callersMethodget_hash

(self, x)

minionerec_trainer.py:584

↓ 3 callersFunctionget_timestamp_start

Get timestamp for the start of a given year and month

data/amazon18_data_process_gpr.py:58

↓ 3 callersFunctionget_timestamp_start

Get timestamp for the start of a given year and month

data/amazon18_data_process.py:58

↓ 3 callersFunctionwrite_file

(path, data)

data/amazon23_data_process.py:370

↓ 3 callersFunctionwrite_json_file

(data, file_path)

data/amazon23_data_process.py:26

↓ 3 callersFunctionwrite_json_file

Write data to JSON file

data/amazon18_data_process.py:35

↓ 2 callersMethod__init__

(self, hidden_size, item_num, state_size, gru_layers=1)

sasrec.py:87

↓ 2 callersMethod_convert_to_semantic_ids

Convert item IDs to semantic ID format using index.json

data.py:1538

↓ 2 callersMethod_convert_to_semantic_ids

Convert item IDs to semantic ID format using index.json

data.py:1721

↓ 2 callersFunctionanalyze_codes

(codes, title="", verbose=True)

rq/rqkmeans_faiss.py:206

↓ 2 callersFunctioncheck_path

(path)

data/amazon23_data_process.py:22

↓ 2 callersFunctionclean_text

(raw_text)

rq/text2emb/utils.py:302

↓ 2 callersFunctioncreate_user_features

Create user features (U-Token)

data/amazon18_data_process_gpr.py:476

↓ 2 callersFunctiondelete_file

(filename)

rq/utils.py:35

↓ 2 callersMethodgenerate_prompt

(self, data_point)

ts_rec_data.py:55

↓ 2 callersFunctionget_first_nbits

(rq)

rq/rqkmeans_faiss.py:293

↓ 2 callersMethodget_history

(self, row)

ts_rec_data.py:61

↓ 2 callersMethodget_inputs

(self)

ts_rec_data.py:126

↓ 2 callersMethodget_inputs

(self)

ts_rec_data.py:407

↓ 2 callersMethodget_inputs

(self)

ts_rec_data.py:637

↓ 2 callersFunctionget_rq_codebooks

(rq)

rq/rqkmeans_faiss.py:88

↓ 2 callersFunctionget_timestamp_start

Return epoch seconds for start of a month.

data/amazon23_data_process.py:36

↓ 2 callersFunctionget_timestamp_start

(year, month)

data/process.py:13

↓ 2 callersFunctionload_json

(file)

rq/text2emb/utils.py:297

↓ 2 callersFunctionpairwise_sq_dists_batch

X: (B,d) C: (K,d)

rq/rqkmeans_faiss.py:18

↓ 2 callersFunctionparse_sid

↓ 2 callersFunctionset_seed

(seed)

rl.py:20

↓ 2 callersFunctionwrite_remap_index

Write index mapping to file

data/amazon18_data_process_gpr.py:41

↓ 2 callersFunctionwrite_remap_index

(index_map, file_path)

data/amazon23_data_process.py:31

↓ 2 callersFunctionwrite_remap_index

Write index mapping to file

data/amazon18_data_process.py:41

↓ 1 callersMethod__init__

(self, input_size, output_size, emb_size, clusters_k=10)

utility.py:154

↓ 1 callersMethod__init__

(self, hidden_size, num_units, num_heads, dropout_rate)

SASRecModules_ori.py:27

↓ 1 callersMethod_build_optimizer

(self)

rq/trainer.py:49

↓ 1 callersMethod_check_nan

(self, loss)

rq/trainer.py:93

↓ 1 callersMethod_generate_train_loss_output

(self, epoch_idx, s_time, e_time, loss, recon_loss)

rq/trainer.py:174

↓ 1 callersMethod_get_scheduler

(self)

rq/trainer.py:83

↓ 1 callersMethod_load_data

(self)

sft.py:38

↓ 1 callersMethod_load_data

(self)

ts_rec_sft.py:34

↓ 1 callersMethod_load_data

(self)

sft_gpr.py:38

↓ 1 callersMethod_move_model_to_vllm

(self)

minionerec_trainer.py:638

↓ 1 callersMethod_prepare_inputs

(self, inputs: dict[str, Union[torch.Tensor, Any]])

minionerec_trainer.py:665

↓ 1 callersMethod_prepare_preference_data

Prepare data directly from training samples

data.py:1511

↓ 1 callersMethod_prepare_sequence_data

Prepare sequence prediction data from training samples

data.py:1694

↓ 1 callersMethod_process_description

Process description according to the requirements: 1. If description is empty, use title 2. If description is a list, select

ts_rec_data.py:484

↓ 1 callersMethod_process_description

Process description according to the requirements: 1. If description is empty, use title 2. If description is a list, select

data.py:1181

↓ 1 callersMethod_train_epoch

(self, train_data, epoch_idx)

rq/trainer.py:98

↓ 1 callersMethod_valid_epoch

(self, valid_data)

rq/trainer.py:128

↓ 1 callersFunctionactivation_layer

(activation_name="relu", emb_dim=None)

rq/models/layers.py:45

↓ 1 callersMethodaggregate

(self, input_tensor)

utility.py:129

↓ 1 callersFunctionanalyze_codes

Analyze code distribution and collision rate

rq/rqkmeans_constrained.py:149

↓ 1 callersFunctionanalyze_duplication

(codes_df)

rq/generate_indices_plus.py:152

↓ 1 callersFunctionapply_rqkmeans_plus_strategy

(model, codebook_path, device)

rq/rqkmeans_plus.py:24

↓ 1 callersFunctionbalanced_kmeans_level_constrained

Balanced K-means implemented with k-means-constrained

rq/rqkmeans_constrained.py:26

↓ 1 callersFunctionbuild_interaction_list_amazon23

Build interaction sequences (history_len=10). Same logic as your original json2csv version but using: - user_id - rating

data/amazon23_data_process.py:265

↓ 1 callersFunctionbuild_item_features_amazon23

Convert Amazon23 metadata to amazon18-style: - title - description (list → str) - features - categories -

data/amazon23_data_process.py:390

↓ 1 callersFunctionbuild_review_data_amazon23

Build review_data: key = (uid, iid, timestamp) value = {review, title}

data/amazon23_data_process.py:457

↓ 1 callersFunctioncalcu_propensity_score

(buffer)

sasrec.py:401

↓ 1 callersFunctioncalculate_hit_games_cuda

(prediction, topk_list, target, hit_all, ndcg_all)

sasrec.py:280

↓ 1 callersMethodcenter_distance_for_constraint

(distances)

rq/models/vq.py:52

↓ 1 callersFunctioncheck_collision

(all_indices_str)

rq/generate_indices.py:18

↓ 1 callersFunctioncheck_collision

(all_indices_str)

rq/models/generate_indices.py:18

↓ 1 callersFunctioncheck_path

Create directory if it doesn't exist

data/amazon18_data_process_gpr.py:30

↓ 1 callersFunctioncheck_path

Create directory if it doesn't exist

data/amazon18_data_process.py:30

↓ 1 callersMethodcompute_loss

(self, model, inputs, return_outputs=False, num_items_in_batch=None)

minionerec_trainer.py:1035

↓ 1 callersMethodcompute_loss

(self, out, quant_loss, xs=None)

rq/models/rqvae.py:74

↓ 1 callersFunctioncompute_residuals_upto_level

(rq, data, codes, upto_level, codebooks=None)

rq/rqkmeans_faiss.py:96

↓ 1 callersFunctionconvert_interactions_amazon23

Convert interactions to amazon18-style: - user2index - item2index - user2items[user_index] = [item_index...] - interactions list

data/amazon23_data_process.py:221

↓ 1 callersFunctionconvert_interactions_to_csv

Convert interaction data to MiniOneRec CSV format using semantic IDs

convert_dataset_gpr.py:54

↓ 1 callersFunctionconvert_interactions_to_csv

Convert interaction data to MiniOneRec CSV format using semantic IDs

convert_dataset.py:54

↓ 1 callersFunctionconvert_inters2dict_amazon18_style

Convert interactions to dict format like amazon18_data_process

data/amazon18_data_process_gpr.py:180

↓ 1 callersFunctionconvert_inters2dict_amazon18_style

Convert interactions to dict format like amazon18_data_process

data/amazon18_data_process.py:180

↓ 1 callersFunctionconvert_ms_to_sec

Amazon23 sort_timestamp is in milliseconds. Convert to seconds.

data/amazon23_data_process.py:41

↓ 1 callersFunctionconvert_to_atomic_files_json2csv_style

Convert interaction list to train/valid/test files using 8:1:1 split like json2csv

data/amazon18_data_process_gpr.py:286

↓ 1 callersFunctionconvert_to_atomic_files_json2csv_style

Convert interaction list to train/valid/test files using 8:1:1 split like json2csv

data/amazon18_data_process.py:281

next →1–100 of 426, ranked by callers