chenhaokun / tpgr

python implementation of the TPGR

TPGR

Python implementation of the TPGR. To run the model, one should make configuration in config file and run main.py. All parameters in config file are explained in the readme file.

tpgr's Issues

报错

请问对比实验的代码哪里可以找到啊

Expected date for code release?

Read the paper, very interesting work! Could you say when would you be sharing your implementation?

Why you use that reward?

Hi,

I have read your paper. But I don't know why do you use that reward mapping function? Could you please explain it?

环境

您的方法与对比算法使用的env和reward是一样的吗

请问你们多久公布源代码

一直在关注你们这片文章，我也从事这方面的研究，看到你们延后了代码公布时间，我很想学习你们的模型和对比实验中DDPG-KNN的实现细节，非常感谢

Replication of the experiment

I'm trying to replicate the results obtained in the paper using the code in the repository but I'm able to run it only on the demo data reported in the repository. How can I use the Movielens and Netflix dataset in your code? Is there a way to format them in a proper way in order to use them with your code?

evaluate error.

evalue error as follows, reward nan, precision nan, recal nan, f1 nan.

qs means: 0.70035
training step: 10
train average reward over step: nan, precision@32: nan, recall@32: nan, f1@32: nan
test average reward over step: nan, precision@32: nan, recall@32: nan, f1@32: nan
average rmse over train and test: nan
log: evaluated

time: Tue Jun 25 17:19:48 2019
qs means: 0.72152
training step: 11
train average reward over step: nan, precision@32: nan, recall@32: nan, f1@32: nan
test average reward over step: nan, precision@32: nan, recall@32: nan, f1@32: nan
average rmse over train and test: nan
log: evaluated

time: Tue Jun 25 17:21:05 2019
qs means: 0.73909
training step: 12
train average reward over step: nan, precision@32: nan, recall@32: nan, f1@32: nan
test average reward over step: nan, precision@32: nan, recall@32: nan, f1@32: nan
average rmse over train and test: nan
log: evaluated

time: Tue Jun 25 17:22:29 2019
qs means: 0.75385

请问你们可以一同公布对比实验中 DDPG-KNN 模型的代码吗

请问你们可以一同公布对比实验中 DDPG-KNN 模型的代码吗