nextrec 0.4.6__py3-none-any.whl → 0.4.7__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (14) hide show
  1. nextrec/__version__.py +1 -1
  2. nextrec/basic/callback.py +399 -21
  3. nextrec/basic/model.py +274 -173
  4. nextrec/loss/loss_utils.py +73 -4
  5. nextrec/models/match/dssm.py +5 -4
  6. nextrec/models/match/dssm_v2.py +4 -3
  7. nextrec/models/match/mind.py +5 -4
  8. nextrec/models/match/sdm.py +5 -4
  9. nextrec/models/match/youtube_dnn.py +5 -4
  10. {nextrec-0.4.6.dist-info → nextrec-0.4.7.dist-info}/METADATA +30 -25
  11. {nextrec-0.4.6.dist-info → nextrec-0.4.7.dist-info}/RECORD +14 -14
  12. {nextrec-0.4.6.dist-info → nextrec-0.4.7.dist-info}/WHEEL +0 -0
  13. {nextrec-0.4.6.dist-info → nextrec-0.4.7.dist-info}/entry_points.txt +0 -0
  14. {nextrec-0.4.6.dist-info → nextrec-0.4.7.dist-info}/licenses/LICENSE +0 -0
nextrec/loss/loss_utils.py CHANGED
@@ -2,10 +2,12 @@
2
2
  Loss utilities for NextRec.
3
3
 
4
4
  Date: create on 27/10/2025
5
- Checkpoint: edit on 29/11/2025
5
+ Checkpoint: edit on 17/12/2025
6
6
  Author: Yang Zhou, zyaztec@gmail.com
7
7
  """
8
8
 
9
+ from typing import Literal
10
+
9
11
  import torch.nn as nn
10
12
 
11
13
  from nextrec.loss.listwise import (
@@ -30,14 +32,81 @@ VALID_TASK_TYPES = [
30
32
  "regression",
31
33
  ]
32
34
 
35
+ # Define all supported loss types
36
+ LossType = Literal[
37
+ # Pointwise losses
38
+ "bce",
39
+ "binary_crossentropy",
40
+ "weighted_bce",
41
+ "focal",
42
+ "focal_loss",
43
+ "cb_focal",
44
+ "class_balanced_focal",
45
+ "crossentropy",
46
+ "ce",
47
+ "mse",
48
+ "mae",
49
+ # Pairwise ranking losses
50
+ "bpr",
51
+ "hinge",
52
+ "triplet",
53
+ # Listwise ranking losses
54
+ "sampled_softmax",
55
+ "softmax",
56
+ "infonce",
57
+ "listnet",
58
+ "listmle",
59
+ "approx_ndcg",
60
+ ]
61
+
33
62
 
34
- def _build_cb_focal(kw):
63
+ def build_cb_focal(kw):
35
64
  if "class_counts" not in kw:
36
65
  raise ValueError("class_balanced_focal requires class_counts")
37
66
  return ClassBalancedFocalLoss(**kw)
38
67
 
39
68
 
40
- def get_loss_fn(loss=None, **kw):
69
+ def get_loss_fn(loss: LossType | nn.Module | None = None, **kw) -> nn.Module:
70
+ """
71
+ Get loss function by name or return the provided loss module.
72
+
73
+ Args:
74
+ loss: Loss function name or nn.Module instance. Supported options:
75
+
76
+ **Pointwise Losses:**
77
+ - "bce", "binary_crossentropy": Binary Cross-Entropy Loss
78
+ - "weighted_bce": Weighted Binary Cross-Entropy Loss
79
+ - "focal", "focal_loss": Focal Loss (for class imbalance)
80
+ - "cb_focal", "class_balanced_focal": Class-Balanced Focal Loss (requires class_counts parameter)
81
+ - "crossentropy", "ce": Cross-Entropy Loss for multi-class classification
82
+ - "mse": Mean Squared Error Loss
83
+ - "mae": Mean Absolute Error Loss
84
+
85
+ **Pairwise Ranking Losses:**
86
+ - "bpr": Bayesian Personalized Ranking Loss
87
+ - "hinge": Hinge Loss
88
+ - "triplet": Triplet Loss
89
+
90
+ **Listwise Ranking Losses:**
91
+ - "sampled_softmax", "softmax": Sampled Softmax Loss
92
+ - "infonce": InfoNCE Loss
93
+ - "listnet": ListNet Loss
94
+ - "listmle": ListMLE Loss
95
+ - "approx_ndcg": Approximate NDCG Loss
96
+
97
+ **kw: Additional keyword arguments passed to the loss function
98
+
99
+ Returns:
100
+ nn.Module: Loss function instance
101
+
102
+ Raises:
103
+ ValueError: If loss is None or unsupported type
104
+
105
+ Examples:
106
+ >>> loss_fn = get_loss_fn("bce")
107
+ >>> loss_fn = get_loss_fn("focal", alpha=0.25, gamma=2.0)
108
+ >>> loss_fn = get_loss_fn("cb_focal", class_counts=[100, 50, 200])
109
+ """
41
110
  if isinstance(loss, nn.Module):
42
111
  return loss
43
112
  if loss is None:
@@ -49,7 +118,7 @@ def get_loss_fn(loss=None, **kw):
49
118
  if loss in ["focal", "focal_loss"]:
50
119
  return FocalLoss(**kw)
51
120
  if loss in ["cb_focal", "class_balanced_focal"]:
52
- return _build_cb_focal(kw)
121
+ return build_cb_focal(kw)
53
122
  if loss in ["crossentropy", "ce"]:
54
123
  return nn.CrossEntropyLoss(**kw)
55
124
  if loss == "mse":
nextrec/models/match/dssm.py CHANGED
@@ -1,10 +1,10 @@
1
1
  """
2
2
  Date: create on 09/11/2025
3
- Author:
4
- Yang Zhou,zyaztec@gmail.com
3
+ Checkpoint: edit on 18/12/2025
4
+ Author: Yang Zhou, zyaztec@gmail.com
5
5
  Reference:
6
- [1] Huang P S, He X, Gao J, et al. Learning deep structured semantic models for web search using clickthrough data[C]
7
- //Proceedings of the 22nd ACM international conference on Information & Knowledge Management. 2013: 2333-2338.
6
+ [1] Huang P S, He X, Gao J, et al. Learning deep structured semantic models for web search using clickthrough data[C]
7
+ //Proceedings of the 22nd ACM international conference on Information & Knowledge Management. 2013: 2333-2338.
8
8
  """
9
9
 
10
10
  import torch
@@ -81,6 +81,7 @@ class DSSM(BaseMatchModel):
81
81
  dense_l1_reg=dense_l1_reg,
82
82
  embedding_l2_reg=embedding_l2_reg,
83
83
  dense_l2_reg=dense_l2_reg,
84
+ early_stop_patience=early_stop_patience,
84
85
  **kwargs,
85
86
  )
86
87
 
nextrec/models/match/dssm_v2.py CHANGED
@@ -1,9 +1,9 @@
1
1
  """
2
2
  Date: create on 09/11/2025
3
- Author:
4
- Yang Zhou,zyaztec@gmail.com
3
+ Checkpoint: edit on 18/12/2025
4
+ Author: Yang Zhou, zyaztec@gmail.com
5
5
  Reference:
6
- DSSM v2 - DSSM with pairwise training using BPR loss
6
+ DSSM v2 - DSSM with pairwise training using BPR loss
7
7
  """
8
8
 
9
9
  import torch
@@ -77,6 +77,7 @@ class DSSM_v2(BaseMatchModel):
77
77
  dense_l1_reg=dense_l1_reg,
78
78
  embedding_l2_reg=embedding_l2_reg,
79
79
  dense_l2_reg=dense_l2_reg,
80
+ early_stop_patience=early_stop_patience,
80
81
  **kwargs,
81
82
  )
82
83
 
nextrec/models/match/mind.py CHANGED
@@ -1,10 +1,10 @@
1
1
  """
2
2
  Date: create on 09/11/2025
3
- Author:
4
- Yang Zhou,zyaztec@gmail.com
3
+ Checkpoint: edit on 18/12/2025
4
+ Author: Yang Zhou, zyaztec@gmail.com
5
5
  Reference:
6
- [1] Li C, Liu Z, Wu M, et al. Multi-interest network with dynamic routing for recommendation at Tmall[C]
7
- //Proceedings of the 28th ACM international conference on information and knowledge management. 2019: 2615-2623.
6
+ [1] Li C, Liu Z, Wu M, et al. Multi-interest network with dynamic routing for recommendation at Tmall[C]
7
+ //Proceedings of the 28th ACM international conference on information and knowledge management. 2019: 2615-2623.
8
8
  """
9
9
 
10
10
  import torch
@@ -237,6 +237,7 @@ class MIND(BaseMatchModel):
237
237
  dense_l1_reg=dense_l1_reg,
238
238
  embedding_l2_reg=embedding_l2_reg,
239
239
  dense_l2_reg=dense_l2_reg,
240
+ early_stop_patience=early_stop_patience,
240
241
  **kwargs,
241
242
  )
242
243
 
nextrec/models/match/sdm.py CHANGED
@@ -1,10 +1,10 @@
1
1
  """
2
2
  Date: create on 09/11/2025
3
- Author:
4
- Yang Zhou,zyaztec@gmail.com
3
+ Checkpoint: edit on 18/12/2025
4
+ Author: Yang Zhou, zyaztec@gmail.com
5
5
  Reference:
6
- [1] Ying H, Zhuang F, Zhang F, et al. Sequential recommender system based on hierarchical attention networks[C]
7
- //IJCAI. 2018: 3926-3932.
6
+ [1] Ying H, Zhuang F, Zhang F, et al. Sequential recommender system based on hierarchical attention networks[C]
7
+ //IJCAI. 2018: 3926-3932.
8
8
  """
9
9
 
10
10
  import torch
@@ -84,6 +84,7 @@ class SDM(BaseMatchModel):
84
84
  dense_l1_reg=dense_l1_reg,
85
85
  embedding_l2_reg=embedding_l2_reg,
86
86
  dense_l2_reg=dense_l2_reg,
87
+ early_stop_patience=early_stop_patience,
87
88
  **kwargs,
88
89
  )
89
90
 
nextrec/models/match/youtube_dnn.py CHANGED
@@ -1,10 +1,10 @@
1
1
  """
2
2
  Date: create on 09/11/2025
3
- Author:
4
- Yang Zhou,zyaztec@gmail.com
3
+ Checkpoint: edit on 18/12/2025
4
+ Author: Yang Zhou, zyaztec@gmail.com
5
5
  Reference:
6
- [1] Covington P, Adams J, Sargin E. Deep neural networks for youtube recommendations[C]
7
- //Proceedings of the 10th ACM conference on recommender systems. 2016: 191-198.
6
+ [1] Covington P, Adams J, Sargin E. Deep neural networks for youtube recommendations[C]
7
+ //Proceedings of the 10th ACM conference on recommender systems. 2016: 191-198.
8
8
  """
9
9
 
10
10
  import torch
@@ -81,6 +81,7 @@ class YoutubeDNN(BaseMatchModel):
81
81
  dense_l1_reg=dense_l1_reg,
82
82
  embedding_l2_reg=embedding_l2_reg,
83
83
  dense_l2_reg=dense_l2_reg,
84
+ early_stop_patience=early_stop_patience,
84
85
  **kwargs,
85
86
  )
86
87
 
{nextrec-0.4.6.dist-info → nextrec-0.4.7.dist-info}/METADATA RENAMED
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: nextrec
3
- Version: 0.4.6
3
+ Version: 0.4.7
4
4
  Summary: A comprehensive recommendation library with match, ranking, and multi-task learning models
5
5
  Project-URL: Homepage, https://github.com/zerolovesea/NextRec
6
6
  Project-URL: Repository, https://github.com/zerolovesea/NextRec
@@ -65,9 +65,9 @@ Description-Content-Type: text/markdown
65
65
  ![Python](https://img.shields.io/badge/Python-3.10+-blue.svg)
66
66
  ![PyTorch](https://img.shields.io/badge/PyTorch-1.10+-ee4c2c.svg)
67
67
  ![License](https://img.shields.io/badge/License-Apache%202.0-green.svg)
68
- ![Version](https://img.shields.io/badge/Version-0.4.6-orange.svg)
68
+ ![Version](https://img.shields.io/badge/Version-0.4.7-orange.svg)
69
69
 
70
- 中文文档 | [English Version](README_en.md) |
70
+ 中文文档 | [English Version](README_en.md)
71
71
 
72
72
  **统一、高效、可扩展的推荐系统框架**
73
73
 
@@ -80,7 +80,7 @@ Description-Content-Type: text/markdown
80
80
  - [安装](#安装)
81
81
  - [架构](#架构)
82
82
  - [5分钟快速上手](#5分钟快速上手)
83
- - [命令行工具](#命令行工具)
83
+ - [命令行工具 NextRec-CLI](#命令行工具)
84
84
  - [兼容平台](#兼容平台)
85
85
  - [支持模型](#支持模型)
86
86
  - [贡献指南](#贡献指南)
@@ -90,35 +90,30 @@ Description-Content-Type: text/markdown
90
90
  NextRec是一个基于PyTorch的现代推荐系统框架，旨在为研究工程团队提供快速的建模、训练与评估流。框架内置丰富的模型库、数据处理工具和工程化训练组件。此外提供了易上手的接口，命令行工具及教程，推荐算法学习者能以最快速度了解模型架构，复现学术论文并进行训练和部署。
91
91
 
92
92
  ## Why NextRec
93
-
94
- - **统一的特征工程与数据流水线**：NextRec框架提供了统一的特征定义、可持久化的数据处理、并对批处理进行了优化，符合工业大数据Spark/Hive场景下，基于离线特征的模型训练推理流程。
95
93
  - **多场景推荐能力**：覆盖排序（CTR/CVR）、召回、多任务学习、生成式召回等推荐/营销模型，持续跟进业界进展。
96
- - **友好的工程体验**：支持各种格式数据(`csv/parquet/pathlike`)的流式预处理/分布式训练/推理，GPU加速与可视化指标监控，方便业务算法工程师和推荐算法学习者快速复现实验。
94
+ - **统一的特征工程与数据流水线**：NextRec框架提供了统一的特征定义、可持久化的数据处理、批处理优化，符合工业大数据Spark/Hive场景下，基于离线特征的模型训练推理流程。
95
+ - **友好的工程体验**：支持多种格式数据(`csv/parquet/pathlike`)的流式处理/分布式训练/推理与可视化指标监控，方便业务算法工程师和推荐算法学习者快速复现实验。
97
96
  - **灵活的命令行工具**：支持通过命令行和配置文件，一键启动训练和推理进程，方便快速实验迭代和敏捷部署。
98
97
  - **高效训练与评估**：内置多种优化器、学习率调度、早停、模型检查点与详细的日志管理，开箱即用。
99
98
 
100
99
  ## NextRec近期进展
101
100
 
102
- - **12/12/2025** 在v0.4.6中加入了生成式召回中常见的[RQ-VAE](/nextrec/models/generative/rqvae.py)模块。配套的[数据集](/dataset/ecommerce_task.csv)和[代码](tutorials/notebooks/zh/使用RQ-VAE构建语义ID.ipynby)已经同步在仓库中
101
+ - **12/12/2025** 在v0.4.7中加入了[RQ-VAE](/nextrec/models/generative/rqvae.py)模块。配套的[数据集](/dataset/ecommerce_task.csv)和[代码](tutorials/notebooks/zh/使用RQ-VAE构建语义ID.ipynby)已经同步在仓库中
103
102
  - **07/12/2025** 发布了NextRec CLI命令行工具，它允许用户根据配置文件进行一键训练和推理，我们提供了相关的[教程](/nextrec_cli_preset/NextRec-CLI_zh.md)和[教学代码](/nextrec_cli_preset)
104
- - **03/12/2025** NextRec获得100颗🌟，感谢大家的支持
103
+ - **03/12/2025** NextRec获得了100颗🌟！感谢大家的支持
105
104
  - **06/12/2025** 在v0.4.1中支持了单机多卡的分布式DDP训练，并且提供了配套的[代码](tutorials/distributed)
106
105
  - **23/11/2025** 在v0.2.2中对basemodel进行了逻辑上的大幅重构和流程统一，并且对listwise/pairwise/pointwise损失进行了统一
107
106
  - **11/11/2025** NextRec v0.1.0发布，我们提供了10余种Ranking模型，4种多任务模型和4种召回模型，以及统一的训练/日志/指标管理系统
108
107
 
109
108
  ## 架构
110
109
 
111
- NextRec采用模块化工程设计，核心组件包括：Feature Spec驱动的Embedding架构；模型基类BaseModel；独立Layer模块；支持训练和推理的统一的DataLoader；开箱即用的模型库等。
112
-
113
- ![NextRec架构](assets/nextrec_diagram_zh.png)
110
+ NextRec采用模块化工程设计，核心组件包括：统一特征驱动的BaseModel架构；独立Layer模块；支持训练推理的统一的DataLoader；命令行工具NextCLI等。
114
111
 
115
- > 项目的架构借鉴了一些优秀的开源推荐算法库，例如DataWhaleChina社区的[torch-rechub](https://github.com/datawhalechina/torch-rechub)。torch-rechub在开发架构和模型实现上相对成熟，本人也参与了其中一小部分的维护，欢迎感兴趣的开发者前往了解。
116
-
117
- ---
112
+ ![NextRec架构](assets/nextrec_diagram.png)
118
113
 
119
114
  ## 安装
120
115
 
121
- 开发者可以通过`pip install nextrec`快速安装NextRec的最新版本，环境要求为Python 3.10+。如果需要执行示例代码，则需要先拉取仓库：
116
+ 开发者可以通过`pip install nextrec`快速安装NextRec的最新版本，环境要求为Python 3.10+（对于需要使用CUDA加速的开发者，建议安装对应版本的pytorch）。如果需要执行示例代码，则需要先拉取仓库：
122
117
 
123
118
  ```bash
124
119
  git clone https://github.com/zerolovesea/NextRec.git
@@ -245,11 +240,11 @@ nextrec --mode=train --train_config=path/to/train_config.yaml
245
240
  nextrec --mode=predict --predict_config=path/to/predict_config.yaml
246
241
  ```
247
242
 
248
- > 截止当前版本0.4.6，NextRec CLI支持单机训练，分布式训练相关功能尚在开发中。
243
+ > 截止当前版本0.4.7，NextRec CLI支持单机训练，分布式训练相关功能尚在开发中。
249
244
 
250
245
  ## 兼容平台
251
246
 
252
- 当前最新版本为0.4.6，所有模型和测试代码均已在以下平台通过验证，如果开发者在使用中遇到兼容问题，请在issue区提出错误报告及系统版本：
247
+ 当前最新版本为0.4.7，所有模型和测试代码均已在以下平台通过验证，如果开发者在使用中遇到兼容问题，请在issue区提出错误报告及系统版本：
253
248
 
254
249
  | 平台 | 配置 |
255
250
  |------|------|
@@ -273,7 +268,7 @@ nextrec --mode=predict --predict_config=path/to/predict_config.yaml
273
268
  | [PNN](nextrec/models/ranking/pnn.py) | Product-based Neural Networks for User Response Prediction | ICDM 2016 | 已支持 |
274
269
  | [AutoInt](nextrec/models/ranking/autoint.py) | AutoInt: Automatic Feature Interaction Learning | CIKM 2019 | 已支持 |
275
270
  | [DCN](nextrec/models/ranking/dcn.py) | Deep & Cross Network for Ad Click Predictions | ADKDD 2017 | 已支持 |
276
- | [DCN v2](nextrec/models/ranking/dcn_v2.py) | DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems | KDD 2021 | 开发中 |
271
+ | [DCN v2](nextrec/models/ranking/dcn_v2.py) | DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems | KDD 2021 | 已支持 |
277
272
  | [DIN](nextrec/models/ranking/din.py) | Deep Interest Network for Click-Through Rate Prediction | KDD 2018 | 已支持 |
278
273
  | [DIEN](nextrec/models/ranking/dien.py) | Deep Interest Evolution Network for Click-Through Rate Prediction | AAAI 2019 | 已支持 |
279
274
  | [MaskNet](nextrec/models/ranking/masknet.py) | MaskNet: Introducing Feature-wise Gating Blocks for High-dimensional Sparse Recommendation Data | 2020 | 已支持 |
@@ -337,21 +332,15 @@ nextrec --mode=predict --predict_config=path/to/predict_config.yaml
337
332
  - 实际行为
338
333
  - 环境信息（Python 版本、PyTorch 版本等）
339
334
 
340
- ---
341
-
342
335
  ## 许可证
343
336
 
344
337
  本项目采用 [Apache 2.0 许可证](./LICENSE)。
345
338
 
346
- ---
347
-
348
339
  ## 联系方式
349
340
 
350
341
  - **GitHub Issues**: [提交问题](https://github.com/zerolovesea/NextRec/issues)
351
342
  - **邮箱**: zyaztec@gmail.com
352
343
 
353
- ---
354
-
355
344
  ## 致谢
356
345
 
357
346
  NextRec 的开发受到以下优秀项目的启发：
@@ -362,6 +351,22 @@ NextRec 的开发受到以下优秀项目的启发：
362
351
 
363
352
  感谢开源社区的所有贡献者！
364
353
 
354
+
355
+ ## 引用
356
+
357
+ 如果您在研究或工作中使用了本框架，欢迎引用本项目：
358
+
359
+ ```bibtex
360
+ @misc{nextrec,
361
+ title = {NextRec},
362
+ author = {Yang Zhou},
363
+ year = {2025},
364
+ publisher = {GitHub},
365
+ journal = {GitHub repository},
366
+ howpublished = {\url{https://github.com/zerolovesea/NextRec}},
367
+ note = {A unified, efficient, and extensible PyTorch-based recommendation library}
368
+ }
369
+ ```
365
370
  ---
366
371
 
367
372
  <div align="center">
{nextrec-0.4.6.dist-info → nextrec-0.4.7.dist-info}/RECORD RENAMED
@@ -1,14 +1,14 @@
1
1
  nextrec/__init__.py,sha256=_M3oUqyuvQ5k8Th_3wId6hQ_caclh7M5ad51XN09m98,235
2
- nextrec/__version__.py,sha256=bbBpXE_PBbo_SaI807mDML0QJywD0_ufCDPgAMlDHaE,22
2
+ nextrec/__version__.py,sha256=MHGyAIWXVeovtteWUUSzLq9UGWJLLooUZCXB9KGpNK8,22
3
3
  nextrec/cli.py,sha256=b6tv7ZO7UBRVR6IfyqVP24JEcdu9-2_vV5MlfWcQucM,18468
4
4
  nextrec/basic/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
5
5
  nextrec/basic/activation.py,sha256=uzTWfCOtBSkbu_Gk9XBNTj8__s241CaYLJk6l8nGX9I,2885
6
- nextrec/basic/callback.py,sha256=YPkuSmy3WV8cXj8YmLKxwNP2kULpkUlJQf8pV8CkNYQ,1037
6
+ nextrec/basic/callback.py,sha256=lELZhkB1TlAT_5My7Ce41u7Rox73rbyaulykWCk-svo,14108
7
7
  nextrec/basic/features.py,sha256=kfgU3gCORjYD7ylmKJDrWIrRIL97GcKWCDWeULU5N7g,4339
8
8
  nextrec/basic/layers.py,sha256=jDT6nnpCxdgzKMdhU-apxPgjHm-AHujRxjgBum4mBm0,29017
9
9
  nextrec/basic/loggers.py,sha256=p9wNmLuRYyvHsOzP0eNOYSlV3hrTDjrt6ggrH_r4RE0,6243
10
10
  nextrec/basic/metrics.py,sha256=EicHCyyr2j45J-emHf5KifQN6fnb6IV6gqEBu7ioANQ,26066
11
- nextrec/basic/model.py,sha256=03tt8DB4P5Pgw_t7a2vUK1RaX4bGn3lBVVCVm1Ccaso,97063
11
+ nextrec/basic/model.py,sha256=y1SI-isXTGbI9_1J54RMy7u6Hw9PXgjkiMHX38JxUaQ,100515
12
12
  nextrec/basic/session.py,sha256=UOG_-EgCOxvqZwCkiEd8sgNV2G1sm_HbzKYVQw8yYDI,4483
13
13
  nextrec/data/__init__.py,sha256=auT_PkbgU9pUCt7KQl6H2ajcUorRhSyHa8NG3wExcG8,1197
14
14
  nextrec/data/batch_utils.py,sha256=FAJiweuDyAIzX7rICVmcxMofdFs2-7RLinovwB-lAYM,2878
@@ -18,7 +18,7 @@ nextrec/data/dataloader.py,sha256=L4VBpWUZrxozFBV54nhJAAC-ZX5Hg6zFwIwpGnguJ9c,18
18
18
  nextrec/data/preprocessor.py,sha256=BxoD6GHEre86i-TbxPi58Uwmg_G7oLkiER6f7VfmVHo,41583
19
19
  nextrec/loss/__init__.py,sha256=mO5t417BneZ8Ysa51GyjDaffjWyjzFgPXIQrrggasaQ,827
20
20
  nextrec/loss/listwise.py,sha256=UT9vJCOTOQLogVwaeTV7Z5uxIYnngGdxk-p9e97MGkU,5744
21
- nextrec/loss/loss_utils.py,sha256=dFbVB9NAZdBDY-fnWkPvXrvCGL2Bz4I4DvpBlzz0X8w,2579
21
+ nextrec/loss/loss_utils.py,sha256=8lGJpiJ7GYplUdQqJL5XzHIiQr98g5pRb2cTYdS5VLA,4646
22
22
  nextrec/loss/pairwise.py,sha256=X9yg-8pcPt2IWU0AiUhWAt3_4W_3wIF0uSdDYTdoPFY,3398
23
23
  nextrec/loss/pointwise.py,sha256=o9J3OznY0hlbDsUXqn3k-BBzYiuUH5dopz8QBFqS_kQ,7343
24
24
  nextrec/models/generative/__init__.py,sha256=RU4DRibbfK802rzSXIgf3mSwYsSsYLKJiPzcXqprbvY,350
@@ -26,11 +26,11 @@ nextrec/models/generative/hstu.py,sha256=ro0ifzvlS0SQaS1PMVDjvGDBgql5RnWEPGCu_iR
26
26
  nextrec/models/generative/rqvae.py,sha256=yxZ9KNEjRu0yxBjeNtCa2Y0ElB97diV_Z92CS9hbhzE,29255
27
27
  nextrec/models/generative/tiger.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
28
28
  nextrec/models/match/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
29
- nextrec/models/match/dssm.py,sha256=suvle-7EF-P-FK3KAhoPG1FRCLvqbiu6HA8USFGf7kk,7854
30
- nextrec/models/match/dssm_v2.py,sha256=TOTMEdIC6APIcQDonXLoOnO_wTD_UIrqq5M-ptEUATg,6878
31
- nextrec/models/match/mind.py,sha256=so7XkuCHr5k5UBhEB65GL0JavFOjLGLYeN9Nuc4eNKA,15020
32
- nextrec/models/match/sdm.py,sha256=MGEpLe1-UZ8kiHhR7-Q6zW-d9NnOm0ptHQWYVzh7m_Y,10488
33
- nextrec/models/match/youtube_dnn.py,sha256=DxMn-WLaLGAWRy5qhpRszUugbpPxOMUsWEuh7QEAWQw,7214
29
+ nextrec/models/match/dssm.py,sha256=W6oBOEMGT-xYvhd-qzmGpYqxm9B-fwIQaXE7nEw2gfI,7923
30
+ nextrec/models/match/dssm_v2.py,sha256=I2eWWyzFd7nIiShOB7VPxNMLoqST5IY98WxjDyvAGa8,6955
31
+ nextrec/models/match/mind.py,sha256=semg4EEThD4l7K8fW1wuRF3i9Ex-smJVI4CNORVajvA,15089
32
+ nextrec/models/match/sdm.py,sha256=kwDjDkFpoTmicfKScXjhJLe500upbg8eDXICQLMW03k,10557
33
+ nextrec/models/match/youtube_dnn.py,sha256=-XuaM7lh02qkLCwJF3KgWxOQGn61p92MnYA100LWG4o,7283
34
34
  nextrec/models/multi_task/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
35
35
  nextrec/models/multi_task/esmm.py,sha256=tQg_jE51VDTyc-F0auviyP8CI9uzYQ_KjybbCAXWp1s,6491
36
36
  nextrec/models/multi_task/mmoe.py,sha256=qFWKdCE_VSGpVrMgx0NOO-HtLRNGdVxCWdkMfoEgjLA,8583
@@ -63,8 +63,8 @@ nextrec/utils/model.py,sha256=dYl1XfIZt6aVjNyV2AAhcArwFRMcEAKrjG_pr8AVHs0,1163
63
63
  nextrec/utils/optimizer.py,sha256=eX8baIvWOpwDTGninbyp6pQfzdHbIL62GTi4ldpYcfM,2337
64
64
  nextrec/utils/synthetic_data.py,sha256=V6lxEQCU-yJXMBCoCAeKZZWBMYXpZhvbyfiI8jCmvqs,18112
65
65
  nextrec/utils/tensor.py,sha256=Z6MBpSuQpHw4kGjeKxG0cXZMpRBCM45zTKhk9WolyiM,2220
66
- nextrec-0.4.6.dist-info/METADATA,sha256=oRz_rA3u71FBZyhPVuJTgLPX6wAcpen7EAy9ByVuhFU,19379
67
- nextrec-0.4.6.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
68
- nextrec-0.4.6.dist-info/entry_points.txt,sha256=NN-dNSdfMRTv86bNXM7d3ZEPW2BQC6bRi7QP7i9cIps,45
69
- nextrec-0.4.6.dist-info/licenses/LICENSE,sha256=2fQfVKeafywkni7MYHyClC6RGGC3laLTXCNBx-ubtp0,1064
70
- nextrec-0.4.6.dist-info/RECORD,,
66
+ nextrec-0.4.7.dist-info/METADATA,sha256=FBGofRoDIlk-Y4TE1-QNFaLgHSM1SVOWlm9fvThE-QM,19456
67
+ nextrec-0.4.7.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
68
+ nextrec-0.4.7.dist-info/entry_points.txt,sha256=NN-dNSdfMRTv86bNXM7d3ZEPW2BQC6bRi7QP7i9cIps,45
69
+ nextrec-0.4.7.dist-info/licenses/LICENSE,sha256=2fQfVKeafywkni7MYHyClC6RGGC3laLTXCNBx-ubtp0,1064
70
+ nextrec-0.4.7.dist-info/RECORD,,