discover · compare · collections · requirements · contribute · feed · watching · stats · help with your tokens

princewen/tensorflow_practice · source

Snapshot 0 · 2026-11-04T12:47:25Z · Python ·

GitHub description at capture

tensorflow实战练习，包括强化学习、推荐系统、nlp等

README at capture

Tensroflow练习
======

相关数据集下载地址：链接:https://pan.baidu.com/s/1GMv7_3qruoVZBJMvN-afGA  密码:ako7
基于tf1.4

目录

1、基础<br>
[基本语法<br>](https://github.com/princewen/tensorflow_practice/blob/master/basic/basic.py)
[tensorBoard使用<br>](https://github.com/princewen/tensorflow_practice/blob/master/basic/tensorBoard.py)
[dropout<br>](https://github.com/princewen/tensorflow_practice/blob/master/basic/dropout.py)
[模型保存与重载<br>](https://github.com/princewen/tensorflow_practice/blob/master/basic/save2file.py)
[基本神经网络<br>](https://github.com/princewen/tensorflow_practice/blob/master/basic/first_nerual_network.py)
[卷积神经网络<br>](https://github.com/princewen/tensorflow_practice/blob/master/basic/CNN.py)
<br>
2、自然语言相关<br>
[static_RNN<br>](https://github.com/princewen/tensorflow_practice/blob/master/nlp/RNN_static_cell.py)
[dynamic_RNN<br>](https://github.com/princewen/tensorflow_practice/blob/master/nlp/RNN_dynamic_cell.py)
[LSTM<br>](https://github.com/princewen/tensorflow_practice/blob/master/nlp/LSTM.py)
[LSTM_regression<br>](https://github.com/princewen/tensorflow_practice/blob/master/nlp/LSTM_Regression.py)
[seq2seq<br>](https://github.com/princewen/tensorflow_practice/blob/master/nlp/basic_seq2seq.py)
[seq2seq_attention<br>](https://github.com/princewen/tensorflow_practice/tree/master/nlp/chat_bot_seq2seq_attention)
<br>
3、强化学习相关<br>
[Q-learning<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/my_q_learning_new)
[SARSA<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/SARSA)
[SARSA-lambda<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/sarsa_lambda)
[DQN<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/DQN-demo)
[Double DQN<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/Double-DQN-demo)
[Dueling DQN<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/Dueling%20DQN%20Demo)
[Prioritized Replay DQN<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/Prioritized_Replay_DQN_demo)
[Policy Gradient<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/Basic-Policy-Network)
[Actor-Critic<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/Basic-Actor-Critic)
[DDPG<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/Basic-DDPG)
[Pointer-Network<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/myPtrNetwork)
[MADDPG<br>](https://github.com/princewen/tensorflow_practice/tree/master/RL/Basic-MADDPG-Demo)
<br>
4、推荐系统<br>
[FM<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/recommendation-FM-demo)
[FFM<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/recommendation-FFM-Demo)
[DeepFM<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/Basic-DeepFM-model)
[Deep Cross Network<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/Basic-DCN-Demo)
[P NN<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/Basic-PNN-Demo)
[NFM<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/Basic-NFM-Demo)
[AFM<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/Basic-AFM-Demo)
[MLR<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/Basic-MLR-Demo)
[DIN<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/Basic-DIN-Demo)
[Bandit<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/Basic-Bandit-Demo)
[GBDT+LR<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/GBDT%2BLR-Demo)
[evaluation-metrics<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/Basic-Evaluation-metrics)
[NCF<br>](https://github.com/princewen/tensorflow_practice/tree/master/recommendation/Basic-NCF-Demo)
<br>
5、GAN<br>
[Basic GAN<br>](https://github.com/princewen/tensorflow_practice/blob/master/GAN/GAN.py)
[SeqGAN<br>](https://github.com/princewen/tensorflow_practice/tree/master/GAN/seqgan)

<br>
<br>
<br>

推荐阅读
==============
1、基础<br>
[TensorFlow基础知识点总结<br>](https://www.jianshu.com/p/ce213e6b2dc0)
[用tensorboard来看看我们的网络流吧<br>](https://www.jianshu.com/p/41466470b347)
[使用dropout来避免过拟合吧<br>](https://www.jianshu.com/p/4f1b525ddf86)
[使用Tensorflow实现第一个神经网络吧！<br>](https://www.jianshu.com/p/596a30d46f34y)
[实现CNN对mnist手写数字分类<br>](https://www.jianshu.com/p/49ab6568e472)
<br>
2、自然语言相关<br>
[使用简单的RNN观测数字中的规律<br>](https://www.jianshu.com/p/3ccc1eb5fda2)
[更进一步，使用LSTM实现对手写数字识别<br>](https://www.jianshu.com/p/d25baccde6bc)
[简单的Seq2Seq实现作对联<br>](https://www.jianshu.com/p/83443b2baf27)
[使用Seq2Seq+attention model实现简单的Chatbot<br>](https://www.jianshu.com/p/aab40f439012)
<br>
3、强化学习相关<br>
[实战深度强化学习DQN-理论和实践<br>](https://www.jianshu.com/p/10930c371cac)
[DQN三大改进(一)-Double DQN<br>](https://www.jianshu.com/p/fae51b5fe000)
[DQN三大改进(二)-Prioritised replay<br>](https://www.jianshu.com/p/db14fdc67d2c)
[DQN三大改进(三)-Dueling Network<br>](https://www.jianshu.com/p/b421c85796a2)
[深度强化学习-Policy Gradient基本实现<br>](https://www.jianshu.com/p/2ccbab48414b)
[深度强化学习-Actor-Critic算法原理和实现<br>](https://www.jianshu.com/p/6fe18d0d8822)
[深度强化学习-DDPG算法原理和实现<br>](https://www.jianshu.com/p/6fe18d0d8822)
[Pointer-network理论及tensorflow实战<br>](https://www.jianshu.com/p/2ad389e91467)
[探秘多智能体强化学习-MADDPG算法原理及简单实现<br>](https://www.jianshu.com/p/4e4e35d80137)
<br>
4、推荐系统<br>
[推荐系统遇上深度学习(一)--FM模型理论和实践<br>](https://www.jianshu.com/p/152ae633fb00)
[推荐系统遇上深度学习(二)--FFM模型理论和实践<br>](https://www.jianshu.com/p/781cde3d5f3d)
[推荐系统遇上深度学习(三)--DeepFM模型理论和实践<br>](https://www.jianshu.com/p/6f1c2643d31b)
[推荐系统遇上深度学习(四)--多值离散特征的embedding解决方案<br>](https://www.jianshu.com/p/4a7525c018b2)
[推荐系统遇上深度学习(五)--Deep&Cross Network模型理论和实践<br>](https://www.jianshu.com/p/77719fc252fa)
[推荐系统遇上深度学习(六)--PNN模型理论和实践<br>](https://www.jianshu.com/p/be784ab4abc2)
[推荐系统遇上深度学习(七)--NFM模型理论和实践<br>](https://www.jianshu.com/p/4e65723ee632)
[推荐系统遇上深度学习(八)--AFM模型理论和实践<br>](https://www.jianshu.com/p/83d3b2a1e55d)
[推荐系统遇上深度学习(九)--评价指标AUC原理及实践<br>](https://www.jianshu.com/p/4dde15a56d44)
[推荐系统遇上深度学习(十)--GBDT+LR融合方案实战<br>](https://www.jianshu.com/p/96173f2c2fb4)
[推荐系统遇上深度学习(十一)--神经协同过滤NCF原理及实战<br>](https://www.jianshu.com/p/6173dbde4f53)
[推荐系统遇上深度学习(十二)--推荐系统中的EE问题及基本Bandit算法<br>](https://www.jianshu.com/p/95b2de50ce44)
[推荐系统遇上深度学习(十三)--linUCB方法浅析及实现<br>](https://www.jianshu.com/p/e0e843d78e3c)
[推荐系统遇上深度学习(十四)--《DRN:A Deep Reinforcement Learning Framework for News Recommendation》<br>](https://www.jianshu.com/p/c0384b213320)
[推荐系统遇上深度学习(十五)--强化学习在京东推荐中的探索<br>](https://www.jianshu.com/p/b9113332e33e)
[推荐系统遇上深度学习(十六)--详解推荐系统中的常用评测指标<br>](https://www.jianshu.com/p/665f9f168eff)
[推荐系统遇上深度学习(十七)--探秘阿里之MLR算法浅析及实现<br>](https://www.jianshu.com/p/627fc0d755b2)
[推荐系统遇上深度学习(十八)--探秘阿里之深度兴趣网络(DIN)浅析及实现<br>](https://www.jianshu.com/p/73b6f5d00f46)
[推荐系统遇上深度学习(十九)--探秘阿里之完整空间多任务模型ESSM<br>](https://www.jianshu.com/p/35f00299c059)
[推荐系统遇上深度学习(二十)--贝叶斯个性化排序(BPR)算法原理及实战<br>](https://www.jianshu.com/p/ba1936ee0b69)
[推荐系统遇上深度学习(二十一)--阶段性回顾<br>](https://www.jianshu.com/p/99e8f24ec7df)
[推荐系统遇上深度学习(二十二)--DeepFM升级版XDeepFM模型强势来袭！<br>](https://www.jianshu.com/p/b4128bc79df0)
[推荐系统遇上深度学习(二十三)--大一统信息检索模型IRGAN在推荐领域的应用<br>](https://www.jianshu.com/p/d151b52e57f9)
[推荐系统遇上深度学习(二十四)--深度兴趣进化网络DIEN原理及实战！<br>](https://www.jianshu.com/p/6742d10b89a8)
[推荐系统遇上深度学习(二十五)--当知识图谱遇上个性化推荐<br>](https://www.jianshu.com/p/6a5e796499e8)
[推荐系统遇上深度学习(二十六)--知识图谱与推荐系统结合之DKN模型原理及实现<br>](https://www.jianshu.com/p/2e3cade31098)
[推荐系统遇上深度学习(二十七)--知识图谱与推荐系统结合之RippleNet模型原理及实现<br>](https://www.jianshu.com/p/c5ffaf7ed449)
<br>
5、GAN<br>
[听说GAN很高大上，其实就这么简单<br>](https://www.jianshu.com/p/5f638f493b7a)
[对抗思想与强化学习的碰撞-SeqGAN模型原理和代码解析<br>](https://www.jianshu.com/p/de4e913e0580)