initial commit

README.md

@@ -1 +1,36 @@
-# SASRec
+# SASRec: Self-Attentive Sequential Recommendation
+
+This is our TensorFlow implementation for the paper:
+
+Wang-Cheng Kang, Julian McAuley. *[Self-Attentive Sequential Recommendation.](https://cseweb.ucsd.edu/~jmcauley/pdfs/icdm18.pdf)* In Proceedings of IEEE International Conference on Data Mining (ICDM'18)
+
+Please cite our paper if you use the code or datasets.
+
+The code is tested under a Linux desktop (w/ GTX 1080 Ti GPU) with TensorFlow 1.2.
+
+## Datasets
+
+The preprocessed datasets are included in the repo (`e.g. data/Video.txt`), where each line contains an `user id` and 
+`item id` (starting from 1) meaning an interaction (sorted by timestamp).
+
+The data pre-processing script is also included. For example, you could download Amazon review data from *[here.](http://jmcauley.ucsd.edu/data/amazon/index.html)*, and run the script to produce the `txt` format data.  
+
+## Model Training
+
+To train our model on `Video` (with default hyper-parameters): 
+
+```
+python main.py --dataset=Video --train_dir=default 
+```
+
+or on `ml-1m`:
+
+```
+python main.py --dataset=ml-1m --train_dir=default --maxlen=200 --dropout_rate=0.2 
+``` 
+
+## Misc
+
+The implemention of self attention is modified based on *[this](https://github.com/Kyubyong/transformer)*
+
+The convergence curve on `ml-1m`, compared with CNN/RNN based approaches: ![](curve.png)