update README.md

README.md

@@ -12,6 +12,9 @@ Note: The Humanoid experiment depends on [Mujoco](http://www.mujoco.org/). Pleas
 
 The article describing these papers can be found [here](https://eng.uber.com/deep-neuroevolution/)
 
+## Visual Inspector for NeuroEvolution (VINE)
+The folder `./visual_inspector` contains implementations of VINE, i.e., Visual Inspector for NeuroEvolution, an interactive data visualization tool for neuroevolution. Refer to `README.md` in that folder for further instructions on running and customizing your visualization. An article describing this visualization tool can be found [here](https://eng.uber.com/vine/).
+
 ## How to run locally
 
 clone repo
@@ -76,6 +79,3 @@ python -m scripts.viz 'Humanoid-v1' <YOUR_H5_FILE>
 ### extra folder
 The extra folder holds the XML specification file for the  Humanoid
 Locomotion with Deceptive Trap domain used in https://arxiv.org/abs/1712.06560. Use this XML file in gym to recreate the environment.
-
-### Visual Inspector for NeuroEvolution (VINE)
-The folder `./visual_inspector` contains implementations of VINE, i.e., Visual Inspector for NeuroEvolution, an interactive data visualization tool for neuroevolution.

visual_inspector/README.md

@@ -1,6 +1,6 @@
 ## Visual Inspector for NeuroEvolution (VINE)
 
-This repo contains implementations of VINE, i.e., Visual Inspector for NeuroEvolution, an interactive data visualization tool for neuroevolution.
+This repo contains implementations of VINE, i.e., Visual Inspector for NeuroEvolution, an interactive data visualization tool for neuroevolution. An article describing this visualization tool can be found [here](https://eng.uber.com/vine/).
 
 ### Dependencies that need to be downloaded by end-user from a third party
 
@@ -17,11 +17,11 @@ __Example 1__: visualize the sample Mujoco Humanoid 2D BC (i.e., final x-y locat
 ```
 python -m main_mujoco 90 99 sample_data/mujoco/final_xy_bc/
 ```
-This will bring up the GUI which consists of two interrelated plots: a pseudo-offspring cloud plot, and a fitness plot.
+This will bring up the GUI which consists of two interrelated plots: a pseudo-offspring cloud plot, and a fitness plot, similar to Figure 2 of the [article](https://eng.uber.com/vine/), which is described in detail there.
 
-__Example 2__: click `Movie` button on the GUI to generate a visualization of the moving cloud, which can be saved as a movie clip by checking `save movie` checkbox.
+__Example 2__: click `Movie` button on the GUI to generate a visualization of the moving cloud similar to Figure 3 of the [article](https://eng.uber.com/vine/), which can be saved as a movie clip by checking `save movie` checkbox.
 
-__Example 3__: right click any point of the pseudo-offspring cloud to view videos of the corresponding agent’s deterministic and stochastic behaviors (only available for Generation 97 in `sample_data`).
+__Example 3__: right click any point of the pseudo-offspring cloud to view videos of the corresponding agent’s deterministic and stochastic behaviors (only available for Generation 97 in `sample_data`). Follow the steps (all "clicks" are right click) illustrated in Figure 5 of the [article](https://eng.uber.com/vine/).
 
 
 To see HELP for the complete description of all available options (e.g., multiple BCs, and hi-dimensional BCs):
@@ -45,7 +45,7 @@ python -m process_bc --help
 
 ### Create and visualize your own data
 
-1. Choose proper behavior characterizations (BCs).
+1. Choose proper behavior characterizations (BCs) (refer to the [article](https://eng.uber.com/vine/) for examples).
 2. Moderately modify your GA or ES code that dump out the BCs during neuroevolution.
    Examples of BC choices and modified version of GA and ES, namely, `es_modified.py`, `ga_modified.py`, are privode in `../es_distributed` for your references.
 3. If applicable, using dimensionality reduction (see above) to reduce hi-dimensional BCs to 2D.