nextrec 0.5.3__tar.gz → 0.5.5__tar.gz

{nextrec-0.5.3 → nextrec-0.5.5}/.gitignore

@@ -130,4 +130,6 @@ pypirc.template
 docs/rtd/_build/
 
 *.onnx
-artifacts/
+artifacts/
+
+node_modules/

{nextrec-0.5.3 → nextrec-0.5.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nextrec
-Version: 0.5.3
+Version: 0.5.5
 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
@@ -73,7 +73,7 @@ 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-0.5.3-orange.svg)
+![Version](https://img.shields.io/badge/Version-0.5.5-orange.svg)
 [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/zerolovesea/NextRec)
 
 中文文档 | [English Version](README_en.md)
@@ -259,11 +259,11 @@ nextrec --mode=predict --predict_config=path/to/predict_config.yaml
 
 预测结果固定保存到 `{checkpoint_path}/predictions/{name}.{save_data_format}`。
 
-> 截止当前版本0.5.3，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0.5.5，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0.5.3，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0.5.5，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 | 
 |------|------|

{nextrec-0.5.3 → nextrec-0.5.5}/README.md

@@ -8,7 +8,7 @@
 ![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-0.5.3-orange.svg)
+![Version](https://img.shields.io/badge/Version-0.5.5-orange.svg)
 [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/zerolovesea/NextRec)
 
 中文文档 | [English Version](README_en.md)
@@ -194,11 +194,11 @@ nextrec --mode=predict --predict_config=path/to/predict_config.yaml
 
 预测结果固定保存到 `{checkpoint_path}/predictions/{name}.{save_data_format}`。
 
-> 截止当前版本0.5.3，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0.5.5，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0.5.3，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0.5.5，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 | 
 |------|------|

{nextrec-0.5.3 → nextrec-0.5.5}/README_en.md

@@ -8,7 +8,7 @@
 ![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-0.5.3-orange.svg)
+![Version](https://img.shields.io/badge/Version-0.5.5-orange.svg)
 [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/zerolovesea/NextRec)
 
 English | [中文文档](README.md)
@@ -197,11 +197,11 @@ nextrec --mode=predict --predict_config=path/to/predict_config.yaml
 
 Prediction outputs are saved under `{checkpoint_path}/predictions/{name}.{save_data_format}`.
 
-> As of version 0.5.3, NextRec CLI supports single-machine training; distributed training features are currently under development.
+> As of version 0.5.5, NextRec CLI supports single-machine training; distributed training features are currently under development.
 
 ## Platform Compatibility
 
-The current version is 0.5.3. All models and test code have been validated on the following platforms. If you encounter compatibility issues, please report them in the issue tracker with your system version:
+The current version is 0.5.5. All models and test code have been validated on the following platforms. If you encounter compatibility issues, please report them in the issue tracker with your system version:
 
 | Platform | Configuration | 
 |----------|---------------|

{nextrec-0.5.3 → nextrec-0.5.5}/docs/rtd/conf.py

@@ -11,7 +11,7 @@ sys.path.insert(0, str(PROJECT_ROOT / "nextrec"))
 project = "NextRec"
 copyright = "2026, Yang Zhou"
 author = "Yang Zhou"
-release = "0.5.3"
+release = "0.5.5"
 
 extensions = [
     "myst_parser",

nextrec-0.5.5/nextrec/__version__.py

@@ -0,0 +1 @@
+__version__ = "0.5.5"

{nextrec-0.5.3 → nextrec-0.5.5}/nextrec/basic/model.py

@@ -2099,6 +2099,7 @@ class BaseModel(SummarySet, FeatureSet, nn.Module):
         ]
 
         ctx = mp.get_context("spawn")
+        error_queue = ctx.SimpleQueue()
         processes = []
         for rank in range(num_processes):
             process = ctx.Process(
@@ -2115,6 +2116,7 @@ class BaseModel(SummarySet, FeatureSet, nn.Module):
                     processor,
                     rank,
                     num_processes,
+                    error_queue,
                 ),
             )
             process.start()
@@ -2127,8 +2129,18 @@ class BaseModel(SummarySet, FeatureSet, nn.Module):
 
         for process in processes:
             if process.exitcode not in (0, None):
+                errors = []
+                try:
+                    while not error_queue.empty():
+                        errors.append(error_queue.get_nowait())
+                except Exception:
+                    pass
+                error_text = (
+                    "\n\n".join(errors) if errors else "No worker traceback captured."
+                )
                 raise RuntimeError(
-                    "[BaseModel-predict-streaming Error] One or more inference processes failed."
+                    "[BaseModel-predict-streaming Error] One or more inference processes failed.\n"
+                    + error_text
                 )
         # Merge part files
         existing_parts = [p for p in part_paths if p.exists()]
@@ -2827,22 +2839,32 @@ def predict_streaming_worker(
     processor: Any | None,
     shard_rank: int,
     shard_count: int,
+    error_queue: "mp.SimpleQueue[str] | None" = None,
 ) -> None:
-    model.eval()
-    model.predict_streaming(
-        data=data_path,
-        batch_size=batch_size,
-        save_path=save_path,
-        save_format=save_format,
-        stream_chunk_size=stream_chunk_size,
-        return_dataframe=False,
-        num_workers=num_workers,
-        prefetch_factor=prefetch_factor,
-        processor=processor,
-        num_processes=1,
-        shard_rank=shard_rank,
-        shard_count=shard_count,
-    )
+    try:
+        model.eval()
+        model.predict_streaming(
+            data=data_path,
+            batch_size=batch_size,
+            save_path=save_path,
+            save_format=save_format,
+            stream_chunk_size=stream_chunk_size,
+            return_dataframe=False,
+            num_workers=num_workers,
+            prefetch_factor=prefetch_factor,
+            processor=processor,
+            num_processes=1,
+            shard_rank=shard_rank,
+            shard_count=shard_count,
+        )
+    except Exception:
+        if error_queue is not None:
+            import traceback
+
+            error_queue.put(
+                f"[PredictWorker Error] rank={shard_rank}\n{traceback.format_exc()}"
+            )
+        raise
 
 
 class BaseMatchModel(BaseModel):

{nextrec-0.5.3 → nextrec-0.5.5}/nextrec/cli.py

@@ -25,7 +25,7 @@ import resource
 import sys
 import time
 from pathlib import Path
-from typing import Any, Dict, List
+from typing import Any, Dict
 
 import pandas as pd
 
@@ -137,7 +137,7 @@ def train_model(train_config_path: str) -> None:
     model_cfg = read_yaml(model_cfg_path)
 
     # Extract id_column from data config for GAUC metrics
-    id_column = data_cfg.get("id_column") or data_cfg.get("user_id_column")
+    id_column = data_cfg.get("id_column")
     id_columns = [id_column] if id_column else []
 
     log_cli_section("Data")
@@ -378,6 +378,7 @@ def train_model(train_config_path: str) -> None:
         sparse_features,
         sequence_features,
         target,
+        id_columns,
         device,
     )
 
@@ -585,19 +586,16 @@ def predict_model(predict_config_path: str) -> None:
         sparse_features=sparse_features,
         sequence_features=sequence_features,
         target=target_cols,
+        id_columns=id_columns,
         device=predict_cfg.get("device", "cpu"),
     )
-    model.id_columns = id_columns
+
     model.load_model(
         model_file, map_location=predict_cfg.get("device", "cpu"), verbose=True
     )
 
-    id_columns = []
-    if predict_cfg.get("id_column"):
-        id_columns = [predict_cfg["id_column"]]
-        model.id_columns = id_columns
-
     effective_id_columns = id_columns or model.id_columns
+
     log_cli_section("Features")
     log_kv_lines(
         [

{nextrec-0.5.3 → nextrec-0.5.5}/nextrec/utils/config.py

@@ -388,6 +388,7 @@ def build_model_instance(
     sparse_features: List["SparseFeature"],
     sequence_features: List["SequenceFeature"],
     target: List[str],
+    id_columns: list[str] | str | None,
     device: str,
 ) -> Any:
     """
@@ -400,6 +401,7 @@ def build_model_instance(
         sparse_features: List of sparse feature objects
         sequence_features: List of sequence feature objects
         target: List of target column names
+        id_columns: Identifier column name(s) for GAUC or ID passthrough
         device: Device string (e.g., 'cpu', 'cuda:0')
     """
     dense_map = {f.name: f for f in dense_features}
@@ -436,7 +438,7 @@ def build_model_instance(
         param.kind == inspect.Parameter.VAR_KEYWORD for param in sig_params.values()
     )
 
-    init_kwargs: Dict[str, Any] = dict(params_cfg)
+    init_kwargs = dict(params_cfg)
 
     # Explicit bindings (model_config.feature_bindings) take priority
     for param_name, binding in feature_bindings_cfg.items():
@@ -517,6 +519,8 @@ def build_model_instance(
 
     if accepts("target"):
         init_kwargs.setdefault("target", target)
+    if accepts("id_columns") or accepts_var_kwargs:
+        init_kwargs.setdefault("id_columns", id_columns)
     if accepts("device"):
         init_kwargs.setdefault("device", device)
 

{nextrec-0.5.3 → nextrec-0.5.5}/nextrec_cli_preset/feature_config.yaml

@@ -7,44 +7,44 @@
 dense:
   user_active_days_7:
     processor_config: {type: numeric, scaler: standard}
-    embedding_config: {name: user_active_days_7, input_dim: 1, embedding_dim: 8, use_projection: false}
+    embedding_config: {input_dim: 1, proj_dim: 8, use_projection: false}
   user_ctr:
     processor_config: {type: numeric, scaler: standard}
-    embedding_config: {name: user_ctr, input_dim: 1, embedding_dim: 8, use_projection: false}
+    embedding_config: {input_dim: 1, proj_dim: 8, use_projection: false}
   item_price:
     processor_config: {type: numeric, scaler: standard}
-    embedding_config: {name: item_price, input_dim: 1, embedding_dim: 8, use_projection: false}
+    embedding_config: {input_dim: 1, proj_dim: 8, use_projection: false}
   item_popularity:
     processor_config: {type: numeric, scaler: standard}
-    embedding_config: {name: item_popularity, input_dim: 1, embedding_dim: 8, use_projection: false}
+    embedding_config: {input_dim: 1, proj_dim: 8, use_projection: false}
   time_since_last_click:
     processor_config: {type: numeric, scaler: standard}
-    embedding_config: {name: time_since_last_click, input_dim: 1, embedding_dim: 8, use_projection: false}
+    embedding_config: {input_dim: 1, proj_dim: 8, use_projection: false}
 
 sparse:
   user_id:
     processor_config: {type: sparse, encode_method: hash, hash_size: 100000, min_freq: 1}
-    embedding_config: {name: user_id, embedding_dim: 8, padding_idx: 0}
+    embedding_config: {name: user_id, embedding_name: user_id, embedding_dim: 8, padding_idx: 0}
   item:
     processor_config: {type: sparse, encode_method: hash, hash_size: 5000, min_freq: 1}
-    embedding_config: {name: item, embedding_dim: 8, padding_idx: 0}
+    embedding_config: {name: item, embedding_name: item, embedding_dim: 8, padding_idx: 0}
   gender:
     processor_config: {type: sparse, encode_method: hash, hash_size: 10, min_freq: 1}
-    embedding_config: {name: gender, embedding_dim: 8, padding_idx: 0}
+    embedding_config: {name: gender, embedding_name: gender, embedding_dim: 8, padding_idx: 0}
   city:
     processor_config: {type: sparse, encode_method: hash, hash_size: 20, min_freq: 1}
-    embedding_config: {name: city, embedding_dim: 8, padding_idx: 0}
+    embedding_config: {name: city, embedding_name: city, embedding_dim: 8, padding_idx: 0}
   device:
     processor_config: {type: sparse, encode_method: hash, hash_size: 10, min_freq: 1}
-    embedding_config: {name: device, embedding_dim: 8, padding_idx: 0}
+    embedding_config: {name: device, embedding_name: device, embedding_dim: 8, padding_idx: 0}
   channel:
     processor_config: {type: sparse, encode_method: hash, hash_size: 10, min_freq: 1}
-    embedding_config: {name: channel, embedding_dim: 8, padding_idx: 0}
+    embedding_config: {name: channel, embedding_name: channel, embedding_dim: 8, padding_idx: 0}
   age_bucket:
     processor_config: {type: sparse, encode_method: hash, hash_size: 20, min_freq: 1}
-    embedding_config: {name: age_bucket, embedding_dim: 8, padding_idx: 0}
+    embedding_config: {name: age_bucket, embedding_name: age_bucket, embedding_dim: 8, padding_idx: 0}
 
 sequence:
   hist_item_seq:
     processor_config: {type: sequence, encode_method: hash, hash_size: 5000, min_freq: 1, max_len: 30, pad_value: 0, truncate: post, separator: ','}
-    embedding_config: {name: hist_item_seq, vocab_size: 5000, max_len: 30, combiner: mean, embedding_dim: 8, padding_idx: 0, shared_with: item}
+    embedding_config: {name: hist_item_seq, embedding_name: hist_item_seq, vocab_size: 5000, max_len: 30, combiner: mean, embedding_dim: 8, padding_idx: 0}

{nextrec-0.5.3 → nextrec-0.5.5}/nextrec_cli_preset/predict_config_template.yaml

@@ -30,7 +30,7 @@ predict:
   # ----- Data Format Settings -----
   source_data_format: parquet              # Input data format: csv, parquet, etc.
                                            # Use 'auto' for automatic detection
-  
+
   # ----- Identification Settings -----
   id_column: user_id                       # Column name for user/sample IDs
                                            # Will be included in prediction output

{nextrec-0.5.3 → nextrec-0.5.5}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "nextrec"
-version = "0.5.3"
+version = "0.5.5"
 description = "A comprehensive recommendation library with match, ranking, and multi-task learning models"
 readme = "README.md"
 requires-python = ">=3.10"

nextrec-0.5.5/ui/nextrec-ui/Dockerfile

@@ -0,0 +1,11 @@
+FROM node:20-alpine
+
+WORKDIR /app
+
+COPY package*.json ./
+RUN npm install
+
+COPY . .
+
+EXPOSE 15173
+CMD ["npm", "run", "dev", "--", "--host", "0.0.0.0", "--port", "15173"]

nextrec-0.5.5/ui/nextrec-ui/docker-compose.yml

@@ -0,0 +1,7 @@
+services:
+  nextrec-ui:
+    build: .
+    ports:
+      - "15173:15173"
+    environment:
+      - NODE_ENV=development

nextrec-0.5.5/ui/nextrec-ui/index.html

@@ -0,0 +1,16 @@
+<!doctype html>
+<html lang="zh">
+  <head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>NextRec Studio</title>
+    <link rel="icon" type="image/png" href="/src/assets/logo.png" />
+    <link rel="preconnect" href="https://fonts.googleapis.com">
+    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
+    <link href="https://fonts.googleapis.com/css2?family=Space+Grotesk:wght@400;500;600;700&family=IBM+Plex+Mono:wght@400;500&display=swap" rel="stylesheet">
+  </head>
+  <body>
+    <div id="app"></div>
+    <script type="module" src="/src/main.js"></script>
+  </body>
+</html>

nextrec-0.5.5/ui/nextrec-ui/src/App.vue

@@ -0,0 +1,152 @@
+<template>
+  <div class="app-shell">
+    <aside class="sidebar" ref="sidebarRef">
+      <div class="brand">
+        <img :src="logoUrl" alt="NextRec Logo" />
+      </div>
+      <nav class="nav">
+        <RouterLink to="/train">{{ t('训练配置', 'Training') }}</RouterLink>
+        <RouterLink to="/feature">{{ t('特征配置', 'Features') }}</RouterLink>
+        <RouterLink to="/model">{{ t('模型配置', 'Models') }}</RouterLink>
+        <RouterLink to="/predict">{{ t('推理配置', 'Prediction') }}</RouterLink>
+      </nav>
+      <div class="lang-switch">
+        <button class="lang-pill" :class="{ active: store.ui.lang === 'zh' }" @click="store.ui.lang = 'zh'">中文</button>
+        <button class="lang-pill" :class="{ active: store.ui.lang === 'en' }" @click="store.ui.lang = 'en'">EN</button>
+      </div>
+    </aside>
+    <main class="main" ref="mainRef">
+      <div class="hero">
+        <h1>NextRec Studio</h1>
+        <p>为训练、特征、模型、推理提供标准化的 YAML 文件配置。</p>
+        <div v-if="bundleError" class="alert">{{ bundleError }}</div>
+        <div class="actions">
+          <button class="primary" @click="downloadBundle" :disabled="!!bundleError">打包下载 4 个 YAML</button>
+        </div>
+      </div>
+      <RouterView />
+    </main>
+  </div>
+</template>
+
+<script setup>
+import { computed, ref, watch } from 'vue';
+import { RouterLink, RouterView } from 'vue-router';
+import { store } from './store/configStore.js';
+import { dumpYaml, parseYaml } from './utils/yaml.js';
+import { buildTrainConfig } from './utils/buildTrainConfig.js';
+import { downloadZip } from './utils/zip.js';
+import logoUrl from './assets/logo.png';
+
+const mainRef = ref(null);
+const sidebarRef = ref(null);
+const lang = computed(() => store.ui.lang);
+const t = (zh, en) => (lang.value === 'zh' ? zh : en);
+
+watch(
+  () => store.ui.lang,
+  () => {
+    if (!mainRef.value) return;
+    mainRef.value.classList.remove('lang-flash');
+    // force reflow to restart animation
+    void mainRef.value.offsetWidth;
+    mainRef.value.classList.add('lang-flash');
+    if (sidebarRef.value) {
+      sidebarRef.value.classList.remove('lang-flash');
+      void sidebarRef.value.offsetWidth;
+      sidebarRef.value.classList.add('lang-flash');
+    }
+  }
+);
+
+function buildFeatureSection(items) {
+  const section = {};
+  for (const item of items) {
+    if (!item.name) {
+      continue;
+    }
+    section[item.name] = {
+      processor_config: { ...item.processor },
+      embedding_config: { ...item.embedding }
+    };
+  }
+  return section;
+}
+
+const trainYamlResult = computed(() => {
+  try {
+    return { text: buildTrainConfig(store.train), error: '' };
+  } catch (err) {
+    return { text: '', error: `训练配置解析失败: ${err.message}` };
+  }
+});
+
+const featureYamlResult = computed(() => {
+  try {
+    const obj = {
+      dense: buildFeatureSection(store.feature.dense),
+      sparse: buildFeatureSection(store.feature.sparse),
+      sequence: buildFeatureSection(store.feature.sequence)
+    };
+    return { text: dumpYaml(obj), error: '' };
+  } catch (err) {
+    return { text: '', error: `特征配置解析失败: ${err.message}` };
+  }
+});
+
+const modelYamlResult = computed(() => {
+  try {
+    const params = parseYaml(store.model.paramsText) || {};
+    const obj = {
+      model: store.model.model,
+      params
+    };
+    return { text: dumpYaml(obj), error: '' };
+  } catch (err) {
+    return { text: '', error: `模型配置解析失败: ${err.message}` };
+  }
+});
+
+const predictYaml = computed(() => {
+  const output = {
+    checkpoint_path: store.predict.checkpoint_path,
+    predict: { ...store.predict.predict }
+  };
+
+  if (store.predict.model_config) {
+    output.model_config = store.predict.model_config;
+  }
+
+  if (store.predict.session.id) {
+    output.session = { id: store.predict.session.id };
+  }
+
+  return dumpYaml(output);
+});
+
+const bundleError = computed(() => {
+  if (trainYamlResult.value.error) {
+    return trainYamlResult.value.error;
+  }
+  if (featureYamlResult.value.error) {
+    return featureYamlResult.value.error;
+  }
+  if (modelYamlResult.value.error) {
+    return modelYamlResult.value.error;
+  }
+  return '';
+});
+
+async function downloadBundle() {
+  if (bundleError.value) {
+    return;
+  }
+  await downloadZip('nextrec-configs.zip', {
+    'train_config.yaml': trainYamlResult.value.text,
+    'feature_config.yaml': featureYamlResult.value.text,
+    'model_config.yaml': modelYamlResult.value.text,
+    'predict_config.yaml': predictYaml.value
+  });
+}
+
+</script>

nextrec-0.5.5/ui/nextrec-ui/src/main.js

@@ -0,0 +1,6 @@
+import { createApp } from 'vue';
+import App from './App.vue';
+import router from './router/index.js';
+import './styles.css';
+
+createApp(App).use(router).mount('#app');