lstm: fix dataset and plots

README.md

@@ -1,32 +1,37 @@
-ts_vae_lstm
-================
+
 
 <!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->
 
-This file will become your README and also the index of your
-documentation.
+## TS VAE-LSTM
 
-## Install
+> Implementation of the paper [Anomaly Detection for Time Series Using
+> VAE-LSTM Hybrid Model](https://ieeexplore.ieee.org/document/9053558)
 
-``` sh
+This is a work in progress.
+
+### TODO
 
+- [ ] Separate training from notebooks
+- [ ] use a squared term if the absolute element-wise error falls below
+  delta and a delta-scaled L1 term otherwise (Huber)
+- [x] Use `.project-env` to manage paths
+
+``` sh
 pip install ts_vae_lstm
 ```
 
-## CUDA
+### Env variables
 
-Download the driver and cuda version compiled for the driver.
-```bash
-sudo mhwd -i pci video-nvidia-470xx
-sudo pacman -U https://archive.archlinux.org/packages/c/cuda/cuda-11.4.2-1-x86_64.pkg.tar.zst
+``` bash
+BASEDIR='<your-base-path>/ts_vae-lstm'
+MODELDIR=${BASEDIR}/models
 ```
 
-## How to use
+### CUDA
 
-Fill me in please! Don’t forget code examples:
+Download the driver and cuda version compiled for the driver.
 
-``` python
-1+1
+``` bash
+sudo mhwd -i pci video-nvidia-470xx
+sudo pacman -U https://archive.archlinux.org/packages/c/cuda/cuda-11.4.2-1-x86_64.pkg.tar.zst
 ```
-
-    2