nextrec 0.4.6__py3-none-any.whl → 0.4.8__py3-none-any.whl

nextrec/cli.py

@@ -14,9 +14,11 @@ Examples:
     nextrec --mode=predict --predict_config=nextrec_cli_preset/predict_config.yaml
 
 Date: create on 06/12/2025
+Checkpoint: edit on 18/12/2025
 Author: Yang Zhou, zyaztec@gmail.com
 """
 
+import sys
 import argparse
 import logging
 import pickle
@@ -44,6 +46,10 @@ from nextrec.utils.file import (
     read_yaml,
     resolve_file_paths,
 )
+from nextrec.utils.cli_utils import (
+    get_nextrec_version,
+    log_startup_info,
+)
 from nextrec.basic.loggers import setup_logger
 
 logger = logging.getLogger(__name__)
@@ -71,6 +77,9 @@ def train_model(train_config_path: str) -> None:
     artifact_root = Path(session_cfg.get("artifact_root", "nextrec_logs"))
     session_dir = artifact_root / session_id
     setup_logger(session_id=session_id)
+    logger.info(
+        f"[NextRec CLI] Training start | version={get_nextrec_version()} | session_id={session_id} | artifacts={session_dir.resolve()}"
+    )
 
     processor_path = session_dir / "processor.pkl"
     processor_path = Path(processor_path)
@@ -324,6 +333,9 @@ def predict_model(predict_config_path: str) -> None:
     artifact_root = Path(session_cfg.get("artifact_root", "nextrec_logs"))
     session_dir = Path(cfg.get("checkpoint_path") or (artifact_root / session_id))
     setup_logger(session_id=session_id)
+    logger.info(
+        f"[NextRec CLI] Predict start | version={get_nextrec_version()} | session_id={session_id} | checkpoint={session_dir.resolve()}"
+    )
 
     processor_path = Path(session_dir / "processor.pkl")
     if not processor_path.exists():
@@ -454,6 +466,13 @@ def predict_model(predict_config_path: str) -> None:
 
 def main() -> None:
     """Parse CLI arguments and dispatch to train or predict mode."""
+
+    root = logging.getLogger()
+    if not root.handlers:
+        handler = logging.StreamHandler(sys.stdout)
+        handler.setFormatter(logging.Formatter("%(message)s"))
+        root.addHandler(handler)
+
     parser = argparse.ArgumentParser(
         description="NextRec: Training and Prediction Pipeline",
         formatter_class=argparse.RawDescriptionHelpFormatter,
@@ -466,25 +485,32 @@ Examples:
   nextrec --mode=predict --predict_config=configs/predict_config.yaml
         """,
     )
+
     parser.add_argument(
         "--mode",
         choices=["train", "predict"],
-        required=True,
         help="Running mode: train or predict",
     )
     parser.add_argument("--train_config", help="Training configuration file path")
     parser.add_argument("--predict_config", help="Prediction configuration file path")
     args = parser.parse_args()
 
+    logger.info(get_nextrec_version())
+
+    if not args.mode:
+        parser.error("[NextRec CLI Error] --mode is required (train|predict)")
+
     if args.mode == "train":
         config_path = args.train_config
         if not config_path:
             parser.error("[NextRec CLI Error] train mode requires --train_config")
+        log_startup_info(logger, mode="train", config_path=config_path)
         train_model(config_path)
     else:
         config_path = args.predict_config
         if not config_path:
             parser.error("[NextRec CLI Error] predict mode requires --predict_config")
+        log_startup_info(logger, mode="predict", config_path=config_path)
         predict_model(config_path)
 
 

nextrec/loss/loss_utils.py

@@ -2,10 +2,12 @@
 Loss utilities for NextRec.
 
 Date: create on 27/10/2025
-Checkpoint: edit on 29/11/2025
+Checkpoint: edit on 17/12/2025
 Author: Yang Zhou, zyaztec@gmail.com
 """
 
+from typing import Literal
+
 import torch.nn as nn
 
 from nextrec.loss.listwise import (
@@ -30,14 +32,81 @@ VALID_TASK_TYPES = [
     "regression",
 ]
 
+# Define all supported loss types
+LossType = Literal[
+    # Pointwise losses
+    "bce",
+    "binary_crossentropy",
+    "weighted_bce",
+    "focal",
+    "focal_loss",
+    "cb_focal",
+    "class_balanced_focal",
+    "crossentropy",
+    "ce",
+    "mse",
+    "mae",
+    # Pairwise ranking losses
+    "bpr",
+    "hinge",
+    "triplet",
+    # Listwise ranking losses
+    "sampled_softmax",
+    "softmax",
+    "infonce",
+    "listnet",
+    "listmle",
+    "approx_ndcg",
+]
+
 
-def _build_cb_focal(kw):
+def build_cb_focal(kw):
     if "class_counts" not in kw:
         raise ValueError("class_balanced_focal requires class_counts")
     return ClassBalancedFocalLoss(**kw)
 
 
-def get_loss_fn(loss=None, **kw):
+def get_loss_fn(loss: LossType | nn.Module | None = None, **kw) -> nn.Module:
+    """
+    Get loss function by name or return the provided loss module.
+
+    Args:
+        loss: Loss function name or nn.Module instance. Supported options:
+
+            **Pointwise Losses:**
+            - "bce", "binary_crossentropy": Binary Cross-Entropy Loss
+            - "weighted_bce": Weighted Binary Cross-Entropy Loss
+            - "focal", "focal_loss": Focal Loss (for class imbalance)
+            - "cb_focal", "class_balanced_focal": Class-Balanced Focal Loss (requires class_counts parameter)
+            - "crossentropy", "ce": Cross-Entropy Loss for multi-class classification
+            - "mse": Mean Squared Error Loss
+            - "mae": Mean Absolute Error Loss
+
+            **Pairwise Ranking Losses:**
+            - "bpr": Bayesian Personalized Ranking Loss
+            - "hinge": Hinge Loss
+            - "triplet": Triplet Loss
+
+            **Listwise Ranking Losses:**
+            - "sampled_softmax", "softmax": Sampled Softmax Loss
+            - "infonce": InfoNCE Loss
+            - "listnet": ListNet Loss
+            - "listmle": ListMLE Loss
+            - "approx_ndcg": Approximate NDCG Loss
+
+        **kw: Additional keyword arguments passed to the loss function
+
+    Returns:
+        nn.Module: Loss function instance
+
+    Raises:
+        ValueError: If loss is None or unsupported type
+
+    Examples:
+        >>> loss_fn = get_loss_fn("bce")
+        >>> loss_fn = get_loss_fn("focal", alpha=0.25, gamma=2.0)
+        >>> loss_fn = get_loss_fn("cb_focal", class_counts=[100, 50, 200])
+    """
     if isinstance(loss, nn.Module):
         return loss
     if loss is None:
@@ -49,7 +118,7 @@ def get_loss_fn(loss=None, **kw):
     if loss in ["focal", "focal_loss"]:
         return FocalLoss(**kw)
     if loss in ["cb_focal", "class_balanced_focal"]:
-        return _build_cb_focal(kw)
+        return build_cb_focal(kw)
     if loss in ["crossentropy", "ce"]:
         return nn.CrossEntropyLoss(**kw)
     if loss == "mse":

nextrec/models/match/dssm.py

@@ -1,10 +1,10 @@
 """
 Date: create on 09/11/2025
-Author:
-    Yang Zhou,zyaztec@gmail.com
+Checkpoint: edit on 18/12/2025
+Author: Yang Zhou, zyaztec@gmail.com
 Reference:
-    [1] Huang P S, He X, Gao J, et al. Learning deep structured semantic models for web search using clickthrough data[C]
-        //Proceedings of the 22nd ACM international conference on Information & Knowledge Management. 2013: 2333-2338.
+[1] Huang P S, He X, Gao J, et al. Learning deep structured semantic models for web search using clickthrough data[C]
+//Proceedings of the 22nd ACM international conference on Information & Knowledge Management. 2013: 2333-2338.
 """
 
 import torch
@@ -81,6 +81,7 @@ class DSSM(BaseMatchModel):
             dense_l1_reg=dense_l1_reg,
             embedding_l2_reg=embedding_l2_reg,
             dense_l2_reg=dense_l2_reg,
+            early_stop_patience=early_stop_patience,
             **kwargs,
         )
 

nextrec/models/match/dssm_v2.py

@@ -1,9 +1,9 @@
 """
 Date: create on 09/11/2025
-Author:
-    Yang Zhou,zyaztec@gmail.com
+Checkpoint: edit on 18/12/2025
+Author: Yang Zhou, zyaztec@gmail.com
 Reference:
-    DSSM v2 - DSSM with pairwise training using BPR loss
+DSSM v2 - DSSM with pairwise training using BPR loss
 """
 
 import torch
@@ -77,6 +77,7 @@ class DSSM_v2(BaseMatchModel):
             dense_l1_reg=dense_l1_reg,
             embedding_l2_reg=embedding_l2_reg,
             dense_l2_reg=dense_l2_reg,
+            early_stop_patience=early_stop_patience,
             **kwargs,
         )
 

nextrec/models/match/mind.py

@@ -1,10 +1,10 @@
 """
 Date: create on 09/11/2025
-Author:
-    Yang Zhou,zyaztec@gmail.com
+Checkpoint: edit on 18/12/2025
+Author: Yang Zhou, zyaztec@gmail.com
 Reference:
-    [1] Li C, Liu Z, Wu M, et al. Multi-interest network with dynamic routing for recommendation at Tmall[C]
-        //Proceedings of the 28th ACM international conference on information and knowledge management. 2019: 2615-2623.
+[1] Li C, Liu Z, Wu M, et al. Multi-interest network with dynamic routing for recommendation at Tmall[C]
+//Proceedings of the 28th ACM international conference on information and knowledge management. 2019: 2615-2623.
 """
 
 import torch
@@ -237,6 +237,7 @@ class MIND(BaseMatchModel):
             dense_l1_reg=dense_l1_reg,
             embedding_l2_reg=embedding_l2_reg,
             dense_l2_reg=dense_l2_reg,
+            early_stop_patience=early_stop_patience,
             **kwargs,
         )
 

nextrec/models/match/sdm.py

@@ -1,10 +1,10 @@
 """
 Date: create on 09/11/2025
-Author:
-    Yang Zhou,zyaztec@gmail.com
+Checkpoint: edit on 18/12/2025
+Author: Yang Zhou, zyaztec@gmail.com
 Reference:
-    [1] Ying H, Zhuang F, Zhang F, et al. Sequential recommender system based on hierarchical attention networks[C]
-        //IJCAI. 2018: 3926-3932.
+[1] Ying H, Zhuang F, Zhang F, et al. Sequential recommender system based on hierarchical attention networks[C]
+//IJCAI. 2018: 3926-3932.
 """
 
 import torch
@@ -84,6 +84,7 @@ class SDM(BaseMatchModel):
             dense_l1_reg=dense_l1_reg,
             embedding_l2_reg=embedding_l2_reg,
             dense_l2_reg=dense_l2_reg,
+            early_stop_patience=early_stop_patience,
             **kwargs,
         )
 

nextrec/models/match/youtube_dnn.py

@@ -1,10 +1,10 @@
 """
 Date: create on 09/11/2025
-Author:
-    Yang Zhou,zyaztec@gmail.com
+Checkpoint: edit on 18/12/2025
+Author: Yang Zhou, zyaztec@gmail.com
 Reference:
-    [1] Covington P, Adams J, Sargin E. Deep neural networks for youtube recommendations[C]
-        //Proceedings of the 10th ACM conference on recommender systems. 2016: 191-198.
+[1] Covington P, Adams J, Sargin E. Deep neural networks for youtube recommendations[C]
+//Proceedings of the 10th ACM conference on recommender systems. 2016: 191-198.
 """
 
 import torch
@@ -81,6 +81,7 @@ class YoutubeDNN(BaseMatchModel):
             dense_l1_reg=dense_l1_reg,
             embedding_l2_reg=embedding_l2_reg,
             dense_l2_reg=dense_l2_reg,
+            early_stop_patience=early_stop_patience,
             **kwargs,
         )
 

nextrec/utils/cli_utils.py

@@ -0,0 +1,58 @@
+"""
+CLI utilities for NextRec.
+
+This module provides small helpers used by command-line entrypoints, such as
+printing startup banners and resolving the installed package version.
+"""
+
+from __future__ import annotations
+
+import logging
+import os
+import platform
+import sys
+from datetime import datetime
+
+
+def get_nextrec_version() -> str:
+    """
+    Best-effort version resolver for NextRec.
+
+    Prefer in-repo `nextrec.__version__`, fall back to installed package metadata.
+    """
+    try:
+        from nextrec import __version__  # type: ignore
+
+        if __version__:
+            return str(__version__)
+    except Exception:
+        pass
+
+    try:
+        from importlib.metadata import version
+
+        return version("nextrec")
+    except Exception:
+        return "unknown"
+
+
+def log_startup_info(
+    logger: logging.Logger, *, mode: str, config_path: str | None
+) -> None:
+    """Log a short, user-friendly startup banner."""
+    version = get_nextrec_version()
+    now = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+
+    lines = [
+        "NextRec CLI",
+        f"- Version: {version}",
+        f"- Time: {now}",
+        f"- Mode: {mode}",
+        f"- Config: {config_path or '(not set)'}",
+        f"- Python: {platform.python_version()} ({sys.executable})",
+        f"- Platform: {platform.system()} {platform.release()} ({platform.machine()})",
+        f"- Workdir: {os.getcwd()}",
+        f"- Command: {' '.join(sys.argv)}",
+    ]
+    for line in lines:
+        logger.info(line)

nextrec/utils/config.py

@@ -62,7 +62,7 @@ def select_features(
 
 
 def register_processor_features(
-    processor: DataProcessor,
+    processor: "DataProcessor",
     feature_cfg: Dict[str, Any],
     dense_names: List[str],
     sparse_names: List[str],
@@ -129,6 +129,7 @@ def build_feature_objects(
         sparse_names: List of sparse feature names
         sequence_names: List of sequence feature names
     """
+    from nextrec.basic.features import DenseFeature, SparseFeature, SequenceFeature
 
     dense_cfg = feature_cfg.get("dense", {}) or {}
     sparse_cfg = feature_cfg.get("sparse", {}) or {}
@@ -360,9 +361,9 @@ def load_model_class(model_cfg: Dict[str, Any], base_dir: Path) -> type:
 def build_model_instance(
     model_cfg: Dict[str, Any],
     model_cfg_path: Path,
-    dense_features: List[DenseFeature],
-    sparse_features: List[SparseFeature],
-    sequence_features: List[SequenceFeature],
+    dense_features: List["DenseFeature"],
+    sparse_features: List["SparseFeature"],
+    sequence_features: List["SequenceFeature"],
     target: List[str],
     device: str,
 ) -> Any:

{nextrec-0.4.6.dist-info → nextrec-0.4.8.dist-info}/METADATA

@@ -1,12 +1,13 @@
 Metadata-Version: 2.4
 Name: nextrec
-Version: 0.4.6
+Version: 0.4.8
 Summary: A comprehensive recommendation library with match, ranking, and multi-task learning models
 Project-URL: Homepage, https://github.com/zerolovesea/NextRec
 Project-URL: Repository, https://github.com/zerolovesea/NextRec
 Project-URL: Documentation, https://github.com/zerolovesea/NextRec/blob/main/README.md
 Project-URL: Issues, https://github.com/zerolovesea/NextRec/issues
-Author-email: zerolovesea <zyaztec@gmail.com>
+Author: Yang Zhou
+Author-email: zyaztec@gmail.com
 License-File: LICENSE
 Keywords: ctr,deep-learning,match,pytorch,ranking,recommendation
 Classifier: Development Status :: 3 - Alpha
@@ -65,9 +66,9 @@ Description-Content-Type: text/markdown
 ![Python](https://img.shields.io/badge/Python-3.10+-blue.svg)
 ![PyTorch](https://img.shields.io/badge/PyTorch-1.10+-ee4c2c.svg)
 ![License](https://img.shields.io/badge/License-Apache%202.0-green.svg)
-![Version](https://img.shields.io/badge/Version-0.4.6-orange.svg)
+![Version](https://img.shields.io/badge/Version-0.4.8-orange.svg)
 
-中文文档 | [English Version](README_en.md) | 
+中文文档 | [English Version](README_en.md)
 
 **统一、高效、可扩展的推荐系统框架**
 
@@ -80,7 +81,7 @@ Description-Content-Type: text/markdown
 - [安装](#安装)
 - [架构](#架构)
 - [5分钟快速上手](#5分钟快速上手)
-- [命令行工具](#命令行工具)
+- [命令行工具 NextRec-CLI](#命令行工具)
 - [兼容平台](#兼容平台)
 - [支持模型](#支持模型)
 - [贡献指南](#贡献指南)
@@ -90,35 +91,30 @@ Description-Content-Type: text/markdown
 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。
 
 ## Why NextRec
-
-- **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。
 - **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。
-- **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。
+- **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。
+- **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。
 - **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。
 - **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。
 
 ## NextRec近期进展
 
-- **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
+- **12/12/2025** 在v0.4.8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
 - **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)
-- **03/12/2025** NextRec获得100颗🌟，感谢大家的支持
+- **03/12/2025** NextRec获得了100颗🌟！感谢大家的支持
 - **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)
 - **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
 - **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
 
 ## 架构
 
-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。
-
-![NextRec架构](assets/nextrec_diagram_zh.png)
+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。
 
-> 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。
-
----
+![NextRec架构](assets/nextrec_diagram.png)
 
 ## 安装
 
-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：
+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：
 
 ```bash
 git clone https://github.com/zerolovesea/NextRec.git
@@ -245,11 +241,11 @@ nextrec --mode=train --train_config=path/to/train_config.yaml
 nextrec --mode=predict --predict_config=path/to/predict_config.yaml
 ```
 
-> 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。
+> 截止当前版本0.4.8，NextRec CLI支持单机训练，分布式训练相关功能尚在开发中。
 
 ## 兼容平台
 
-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：
+当前最新版本为0.4.8，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：
 
 | 平台 | 配置 | 
 |------|------|
@@ -273,7 +269,7 @@ nextrec --mode=predict --predict_config=path/to/predict_config.yaml
 | [PNN](nextrec/models/ranking/pnn.py) | Product-based Neural Networks for User Response Prediction | ICDM 2016 | 已支持 |
 | [AutoInt](nextrec/models/ranking/autoint.py) | AutoInt: Automatic Feature Interaction Learning | CIKM 2019 | 已支持 |
 | [DCN](nextrec/models/ranking/dcn.py) | Deep & Cross Network for Ad Click Predictions | ADKDD 2017 | 已支持 |
-| [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 |
+| [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 |
 | [DIN](nextrec/models/ranking/din.py) | Deep Interest Network for Click-Through Rate Prediction | KDD 2018 | 已支持 |
 | [DIEN](nextrec/models/ranking/dien.py) | Deep Interest Evolution Network for Click-Through Rate Prediction | AAAI 2019 | 已支持 |
 | [MaskNet](nextrec/models/ranking/masknet.py) | MaskNet: Introducing Feature-wise Gating Blocks for High-dimensional Sparse Recommendation Data | 2020 | 已支持 |
@@ -337,21 +333,15 @@ nextrec --mode=predict --predict_config=path/to/predict_config.yaml
 - 实际行为
 - 环境信息（Python 版本、PyTorch 版本等）
 
----
-
 ## 许可证
 
 本项目采用 [Apache 2.0 许可证](./LICENSE)。
 
----
-
 ## 联系方式
 
 - **GitHub Issues**: [提交问题](https://github.com/zerolovesea/NextRec/issues)
 - **邮箱**: zyaztec@gmail.com
 
----
-
 ## 致谢
 
 NextRec 的开发受到以下优秀项目的启发：
@@ -362,6 +352,22 @@ NextRec 的开发受到以下优秀项目的启发：
 
 感谢开源社区的所有贡献者！
 
+
+## 引用
+
+如果您在研究或工作中使用了本框架，欢迎引用本项目：
+
+```bibtex
+@misc{nextrec,
+    title = {NextRec},
+    author = {Yang Zhou},
+    year = {2025},
+    publisher = {GitHub},
+    journal = {GitHub repository},
+    howpublished = {\url{https://github.com/zerolovesea/NextRec}},
+    note = {A unified, efficient, and extensible PyTorch-based recommendation library}
+}
+```
 ---
 
 <div align="center">

{nextrec-0.4.6.dist-info → nextrec-0.4.8.dist-info}/RECORD

@@ -1,14 +1,14 @@
 nextrec/__init__.py,sha256=_M3oUqyuvQ5k8Th_3wId6hQ_caclh7M5ad51XN09m98,235
-nextrec/__version__.py,sha256=bbBpXE_PBbo_SaI807mDML0QJywD0_ufCDPgAMlDHaE,22
-nextrec/cli.py,sha256=b6tv7ZO7UBRVR6IfyqVP24JEcdu9-2_vV5MlfWcQucM,18468
+nextrec/__version__.py,sha256=40-PUZPRIakJU2yYWQcwTYvSJA6iewqiG8XylhxuAQk,22
+nextrec/cli.py,sha256=ItikUEFDju7c_3tDhYQuNPbhvFK6jij9zAujR_MYYK8,19388
 nextrec/basic/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nextrec/basic/activation.py,sha256=uzTWfCOtBSkbu_Gk9XBNTj8__s241CaYLJk6l8nGX9I,2885
-nextrec/basic/callback.py,sha256=YPkuSmy3WV8cXj8YmLKxwNP2kULpkUlJQf8pV8CkNYQ,1037
+nextrec/basic/callback.py,sha256=lELZhkB1TlAT_5My7Ce41u7Rox73rbyaulykWCk-svo,14108
 nextrec/basic/features.py,sha256=kfgU3gCORjYD7ylmKJDrWIrRIL97GcKWCDWeULU5N7g,4339
 nextrec/basic/layers.py,sha256=jDT6nnpCxdgzKMdhU-apxPgjHm-AHujRxjgBum4mBm0,29017
 nextrec/basic/loggers.py,sha256=p9wNmLuRYyvHsOzP0eNOYSlV3hrTDjrt6ggrH_r4RE0,6243
 nextrec/basic/metrics.py,sha256=EicHCyyr2j45J-emHf5KifQN6fnb6IV6gqEBu7ioANQ,26066
-nextrec/basic/model.py,sha256=03tt8DB4P5Pgw_t7a2vUK1RaX4bGn3lBVVCVm1Ccaso,97063
+nextrec/basic/model.py,sha256=tYpQAbeW99aXTrsmryoWe9sA8ihG8pHNytRcU_49-pQ,100764
 nextrec/basic/session.py,sha256=UOG_-EgCOxvqZwCkiEd8sgNV2G1sm_HbzKYVQw8yYDI,4483
 nextrec/data/__init__.py,sha256=auT_PkbgU9pUCt7KQl6H2ajcUorRhSyHa8NG3wExcG8,1197
 nextrec/data/batch_utils.py,sha256=FAJiweuDyAIzX7rICVmcxMofdFs2-7RLinovwB-lAYM,2878
@@ -18,7 +18,7 @@ nextrec/data/dataloader.py,sha256=L4VBpWUZrxozFBV54nhJAAC-ZX5Hg6zFwIwpGnguJ9c,18
 nextrec/data/preprocessor.py,sha256=BxoD6GHEre86i-TbxPi58Uwmg_G7oLkiER6f7VfmVHo,41583
 nextrec/loss/__init__.py,sha256=mO5t417BneZ8Ysa51GyjDaffjWyjzFgPXIQrrggasaQ,827
 nextrec/loss/listwise.py,sha256=UT9vJCOTOQLogVwaeTV7Z5uxIYnngGdxk-p9e97MGkU,5744
-nextrec/loss/loss_utils.py,sha256=dFbVB9NAZdBDY-fnWkPvXrvCGL2Bz4I4DvpBlzz0X8w,2579
+nextrec/loss/loss_utils.py,sha256=8lGJpiJ7GYplUdQqJL5XzHIiQr98g5pRb2cTYdS5VLA,4646
 nextrec/loss/pairwise.py,sha256=X9yg-8pcPt2IWU0AiUhWAt3_4W_3wIF0uSdDYTdoPFY,3398
 nextrec/loss/pointwise.py,sha256=o9J3OznY0hlbDsUXqn3k-BBzYiuUH5dopz8QBFqS_kQ,7343
 nextrec/models/generative/__init__.py,sha256=RU4DRibbfK802rzSXIgf3mSwYsSsYLKJiPzcXqprbvY,350
@@ -26,11 +26,11 @@ nextrec/models/generative/hstu.py,sha256=ro0ifzvlS0SQaS1PMVDjvGDBgql5RnWEPGCu_iR
 nextrec/models/generative/rqvae.py,sha256=yxZ9KNEjRu0yxBjeNtCa2Y0ElB97diV_Z92CS9hbhzE,29255
 nextrec/models/generative/tiger.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nextrec/models/match/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-nextrec/models/match/dssm.py,sha256=suvle-7EF-P-FK3KAhoPG1FRCLvqbiu6HA8USFGf7kk,7854
-nextrec/models/match/dssm_v2.py,sha256=TOTMEdIC6APIcQDonXLoOnO_wTD_UIrqq5M-ptEUATg,6878
-nextrec/models/match/mind.py,sha256=so7XkuCHr5k5UBhEB65GL0JavFOjLGLYeN9Nuc4eNKA,15020
-nextrec/models/match/sdm.py,sha256=MGEpLe1-UZ8kiHhR7-Q6zW-d9NnOm0ptHQWYVzh7m_Y,10488
-nextrec/models/match/youtube_dnn.py,sha256=DxMn-WLaLGAWRy5qhpRszUugbpPxOMUsWEuh7QEAWQw,7214
+nextrec/models/match/dssm.py,sha256=W6oBOEMGT-xYvhd-qzmGpYqxm9B-fwIQaXE7nEw2gfI,7923
+nextrec/models/match/dssm_v2.py,sha256=I2eWWyzFd7nIiShOB7VPxNMLoqST5IY98WxjDyvAGa8,6955
+nextrec/models/match/mind.py,sha256=semg4EEThD4l7K8fW1wuRF3i9Ex-smJVI4CNORVajvA,15089
+nextrec/models/match/sdm.py,sha256=kwDjDkFpoTmicfKScXjhJLe500upbg8eDXICQLMW03k,10557
+nextrec/models/match/youtube_dnn.py,sha256=-XuaM7lh02qkLCwJF3KgWxOQGn61p92MnYA100LWG4o,7283
 nextrec/models/multi_task/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nextrec/models/multi_task/esmm.py,sha256=tQg_jE51VDTyc-F0auviyP8CI9uzYQ_KjybbCAXWp1s,6491
 nextrec/models/multi_task/mmoe.py,sha256=qFWKdCE_VSGpVrMgx0NOO-HtLRNGdVxCWdkMfoEgjLA,8583
@@ -52,7 +52,8 @@ nextrec/models/ranking/pnn.py,sha256=twwixy26mfAVaI9AqNnMLdwOG-WtDga60xsNiyJrFjI
 nextrec/models/ranking/widedeep.py,sha256=Xm2klmKBOoSKWCBQN7FhwLStu0BHSTOgAJ9kwLmtiFY,5077
 nextrec/models/ranking/xdeepfm.py,sha256=kcPLoNC1940YxRMgWZS4mSxIXlwtc_HfNDIae_uYrsU,8156
 nextrec/utils/__init__.py,sha256=zqU9vjRUpVzJepcvdbxboik68K5jnMR40kdVjr6tpXY,2599
-nextrec/utils/config.py,sha256=69Vj55kujc36iG9Tpx9Ljtjpw3agHr1D17ONI9JRXdo,19811
+nextrec/utils/cli_utils.py,sha256=mOI2yKZwCE3lVH5RCKs3WnIKyXYiLI08vHFT8duBRrQ,1497
+nextrec/utils/config.py,sha256=wx2Y-oxQqWhwOlW6DdG6ExLXDf-QDRnj8EQYDjrVhuY,19903
 nextrec/utils/device.py,sha256=DtgmrJnVJQKtgtVUbm0SW0vZ5Le0R9HU8TsvqPnRLZc,2453
 nextrec/utils/distributed.py,sha256=tIkgUjzEjR_FHOm9ckyM8KddkCfxNSogP-rdHcVGhuk,4782
 nextrec/utils/embedding.py,sha256=akAEc062MG2cD7VIOllHaqtwzAirQR2gq5iW7oKpGAU,1449
@@ -63,8 +64,8 @@ nextrec/utils/model.py,sha256=dYl1XfIZt6aVjNyV2AAhcArwFRMcEAKrjG_pr8AVHs0,1163
 nextrec/utils/optimizer.py,sha256=eX8baIvWOpwDTGninbyp6pQfzdHbIL62GTi4ldpYcfM,2337
 nextrec/utils/synthetic_data.py,sha256=V6lxEQCU-yJXMBCoCAeKZZWBMYXpZhvbyfiI8jCmvqs,18112
 nextrec/utils/tensor.py,sha256=Z6MBpSuQpHw4kGjeKxG0cXZMpRBCM45zTKhk9WolyiM,2220
-nextrec-0.4.6.dist-info/METADATA,sha256=oRz_rA3u71FBZyhPVuJTgLPX6wAcpen7EAy9ByVuhFU,19379
-nextrec-0.4.6.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
-nextrec-0.4.6.dist-info/entry_points.txt,sha256=NN-dNSdfMRTv86bNXM7d3ZEPW2BQC6bRi7QP7i9cIps,45
-nextrec-0.4.6.dist-info/licenses/LICENSE,sha256=2fQfVKeafywkni7MYHyClC6RGGC3laLTXCNBx-ubtp0,1064
-nextrec-0.4.6.dist-info/RECORD,,
+nextrec-0.4.8.dist-info/METADATA,sha256=25dlzG2TNFZCvm_V6ZAnqaMTL6qwNNt7xMSRsmaQ_Nk,19460
+nextrec-0.4.8.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
+nextrec-0.4.8.dist-info/entry_points.txt,sha256=NN-dNSdfMRTv86bNXM7d3ZEPW2BQC6bRi7QP7i9cIps,45
+nextrec-0.4.8.dist-info/licenses/LICENSE,sha256=2fQfVKeafywkni7MYHyClC6RGGC3laLTXCNBx-ubtp0,1064
+nextrec-0.4.8.dist-info/RECORD,,