Update ShareBottom

README.md

@@ -88,7 +88,7 @@ Click-through rate (CTR) prediction is a critical task for various industrial ap
 | 50  | NeurIPS'22          | [APG](./model_zoo/APG)               | [APG: Adaptive Parameter Generation Network for Click-Through Rate Prediction](https://arxiv.org/abs/2203.16218) :triangular_flag_on_post:**Alibaba**                                |    [:arrow_upper_right:](https://github.com/reczoo/BARS/tree/main/ranking/ctr/APG)                                                                                                   | `torch`       |
 | 51  | KDD'23        | [PPNet](./model_zoo/PEPNet)             | [PEPNet: Parameter and Embedding Personalized Network for Infusing with Personalized Prior Information](https://arxiv.org/abs/2302.01115) :triangular_flag_on_post:**KuaiShou**                          |    [:arrow_upper_right:](https://github.com/reczoo/BARS/tree/main/ranking/ctr/PPNet)                                                                                                   | `torch`       |
 |<tr><th colspan=6 align="center">:open_file_folder: **Multi-Task Modeling**</th></tr>|
-| 52  |     MachineLearn'97      | [SharedBottom](./model_zoo/multitask/SharedBottom)               | [Multitask Learning](https://link.springer.com/article/10.1023/A:1007379606734)                                                                                            |                                                                                                                 | `torch`       |
+| 52  |     Arxiv'17      | [ShareBottom](./model_zoo/multitask/ShareBottom)               | [An Overview of Multi-Task Learning in Deep Neural Networks](https://arxiv.org/abs/1706.05098)                                                                                            |                                                                                                                 | `torch`       |
 | 53  | KDD'18          | [MMoE](./model_zoo/multitask/MMOE)               | [Modeling Task Relationships in Multi-task Learning with Multi-Gate Mixture-of-Experts](https://dl.acm.org/doi/pdf/10.1145/3219819.3220007) :triangular_flag_on_post:**Google**                                                                                            |                                                                                                                 | `torch`       |
 | 54  | KDD'18          | [PLE](./model_zoo/multitask/PLE)               | [Progressive Layered Extraction (PLE): A Novel Multi-Task Learning (MTL) Model for Personalized Recommendations](https://dl.acm.org/doi/10.1145/3383313.3412236) :triangular_flag_on_post:**Tencent**                                                                                            |                                                                                                                 | `torch`       |
 |<tr><th colspan=6 align="center">:open_file_folder: **Multi-Domain Modeling**</th></tr>|

model_zoo/multitask/SharedBottom/README.md → model_zoo/multitask/ShareBottom/README.md

@@ -1,18 +1,18 @@
-## Shared-Bottom
+## ShareBottom
 
 | [Overview](#Overview) | [Configuration](#Configuration) | [Implementation](#Implementation) | 
 
 
 ### Overview
 
-Shared-Bottom is the classic multitasking model based on a shared bottom. The model is published in the following paper:
+Share-Bottom is the classic multitasking model based on a Share bottom. The model is published in the following paper:
 
 + Rich Caruana. [Multitask Learning](https://link.springer.com/article/10.1023/A:1007379606734), in Machine learning 1997.
 
 **Model structure:**
 
 <div align="center">
-    <img width="50%" src="https://cdn.jsdelivr.net/gh/xue-pai/FuxiCTR@main/docs/img/SharedBottom.jpg">
+    <img width="50%" src="https://cdn.jsdelivr.net/gh/reczoo/FuxiCTR@main/docs/img/ShareBottom.jpg">
 </div>
 
 This figure comes from the paper of [MMoE (KDD 2018)](https://www.kdd.org/kdd2018/accepted-papers/view/modeling-task-relationships-in-multi-task-learning-with-multi-gate-mixture-).
@@ -33,7 +33,7 @@ The `model_config.yaml` file contains all the model hyper-parameters as follows.
 
 | Params                 | Type            | Default                   | Description                                                                                                                                                                                                       |
 | ---------------------- | --------------- | ------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| model                  | str             | "SharedBottom"                     | model name,  which should be same with model class name                                                                                                                                                           |
+| model                  | str             | "ShareBottom"                     | model name,  which should be same with model class name                                                                                                                                                           |
 | dataset_id             | str             | "TBD"                     | dataset_id to be determined                                                                                                                                                                                       |
 | loss                   | list             | ["binary_crossentropy","binary_crossentropy"]     | loss function for each task                                                                                                                                                                                                     |
 | metrics                | list            | ['logloss', 'AUC']        | a list of metrics for evaluation                                                                                                                                                                                  |
@@ -76,7 +76,7 @@ The `model_config.yaml` file contains all the model hyper-parameters as follows.
 │   ├── dataset_config.yaml       # 数据集配置文件
 │   └── model_config.yaml         # 模型配置文件
 ├── src                           # 模型代码文件夹
-│   └── SharedBottom.py           # 模型代码
+│   └── ShareBottom.py           # 模型代码
 ├── fuxictr_version.py            # fuxictr加载及版本检查文件
 ├── README.md                     # 使用说明
 └── run_expid.py                  # 执行脚本文件
@@ -95,6 +95,6 @@ The model is tested with the following dependencies.
 Running the model on the tiny data:
 
 ```
-python run_expid.py --expid SharedBottom_test --gpu 0 
+python run_expid.py --expid ShareBottom_test --gpu 0 
 ```
 

model_zoo/multitask/SharedBottom/config/model_config.yaml → model_zoo/multitask/ShareBottom/config/model_config.yaml

@@ -12,16 +12,16 @@ Base:
     feature_specs: null
     feature_config: null
 
-SharedBottom_test:
+ShareBottom_test:
     model: SharedBottom
     dataset_id: tiny_mtl
-    loss: ['binary_crossentropy','binary_crossentropy']
+    loss: ['binary_crossentropy', 'binary_crossentropy']
     metrics: ['logloss', 'AUC']
-    task: ['binary_classification','binary_classification']
+    task: ['binary_classification', 'binary_classification']
     num_tasks: 2
     optimizer: adam
     learning_rate: 1.e-3
-    bottom_hidden_units: [512,256,128]
+    bottom_hidden_units: [512, 256, 128]
     tower_hidden_units: [128, 64]
     hidden_activations: relu
     net_regularizer: 0
@@ -30,8 +30,8 @@ SharedBottom_test:
     net_dropout: 0
     batch_size: 128
     embedding_dim: 128
-    epochs: 50
+    epochs: 1
     shuffle: True
     seed: 2023
     monitor: 'AUC'
-    monitor_mode: 'max'
+    monitor_mode: 'max'

model_zoo/multitask/SharedBottom/run_expid.py → model_zoo/multitask/ShareBottom/run_expid.py

@@ -39,7 +39,7 @@
     '''
     parser = argparse.ArgumentParser()
     parser.add_argument('--config', type=str, default='./config/', help='The config directory.')
-    parser.add_argument('--expid', type=str, default='SharedBottom_test', help='The experiment id to run.')
+    parser.add_argument('--expid', type=str, default='ShareBottom_test', help='The experiment id to run.')
     parser.add_argument('--gpu', type=int, default=-1, help='The gpu index, -1 for cpu')
     args = vars(parser.parse_args())
 

model_zoo/multitask/SharedBottom/src/SharedBottom.py → model_zoo/multitask/ShareBottom/src/ShareBottom.py

@@ -22,7 +22,7 @@
 from fuxictr.pytorch.layers import FeatureEmbedding, MLP_Block
 
 
-class SharedBottom(MultiTaskModel):
+class ShareBottom(MultiTaskModel):
     def __init__(self,
                  feature_map,
                  model_id="SharedBottom",
@@ -40,7 +40,7 @@ def __init__(self,
                  embedding_regularizer=None,
                  net_regularizer=None,
                  **kwargs):
-        super(SharedBottom, self).__init__(feature_map,
+        super(ShareBottom, self).__init__(feature_map,
                                            task=task,
                                            loss_weight=loss_weight,
                                            num_tasks=num_tasks,

model_zoo/multitask/ShareBottom/src/__init__.py

@@ -0,0 +1 @@
+from .ShareBottom import *

model_zoo/multitask/__init__.py

@@ -1,2 +1,3 @@
 from .MMoE.src import MMoE
-from .SharedBottom.src import SharedBottom
+from .ShareBottom.src import ShareBottom
+from .PLE.src import PLE