0.1.2 snapshot (#36)

* Update setup.py * Feature/set sim params (#28) * Updated version number to 0.1.1+snapshot. * Fix/rename (#19) * update requirements and readme * update installation instructions * Update README.md * rename baseline notebooks * update baseline notebook; rename -> * Fix/update requirements (#22) * update requirements and readme * update installation instructions * Update README.md * Feature/specify mode (#23) * add options for evaluation * reformatting * Update README.md * Update README.md (#24) * Update README.md * Feature/refactor tasks (#25) Feature/refactor tasks * set sim parameters upon startup and add error checking * clear notebook * remove navigation notebook * Fix typo in docs, * Fix `%matplotlib notebook line` * Feature/decode goseek observations (#29) * add function to decode observations into imgs + pose * update notebooK * Feature/update docs (#30) * add function to decode observations into imgs + pose * update notebooK * update notebook docs * updated docs * Update goseek_full_perception.py * Feature/update typehints (#32) * add function to decode observations into imgs + pose * update notebooK * update notebook docs * updated docs * update typehints * update docstring * Feature/update tesse install (#34) * Update setup.py * Update README.md * Delete requirements.txt * Update setup.py * Fix/remove segmentation from reward (#35) * option to change scene and random seen on reset; add custom message to signal episode reset * moved scene change and random seed setting to gym env * remove segmentation requirement from reward; refactor and add docs * remove extra line * update docstring * minor comment update * Update README.md * Update README.md * Update requirements.txt * Change required tesse-interface version * Update required tesse version to 0.1.0 * remove objects during reset * added install_requires to setup.py * remove requirements.txt * Update README.md * add required python version * Update README.md * Update README.md * Update README.md * remove `+snapshot`
MIT-TESSE · Feb 27, 2020 · 402fc85 · 402fc85
1 parent d87aa6a
commit 402fc85
Show file tree

Hide file tree

Showing 11 changed files with 347 additions and 467 deletions.
diff --git a/README.md b/README.md
@@ -3,7 +3,7 @@
 Provides a Python interface for reinforcement learning using the TESSE Unity environment and the OpenAI Gym toolkit.
 
 
-Treasure Hunt |  Navigation
+GOSEEK |  Navigation
 :----------:|:---------------:
 ![](docs/hunt-1.gif) | ![](docs/nav-1.gif)
 
@@ -14,17 +14,16 @@ Using [Anaconda](https://www.anaconda.com/distribution/#download-section) or [mi
 
 1. If using conda, create a new environment
 ```sh
-conda create -n tesse_gym python=3.7
-conda activate tesse_gym
+conda create -n goseek python=3.7 ipython jupyter numpy scipy
+conda activate goseek
 ```
 
-2. Clone this repository, install dependencies, then install tesse-gym. *NOTE*: This requires access to [tesse-interface](https://github.mit.edu/TESS/tesse-interface). 
+2. Clone and install this repository. 
 ```sh
-git clone git@github.mit.edu:TESS/tesse-gym.git
+git clone https://github.com/MIT-TESSE/tesse-gym.git
 cd tesse-gym
 
-pip install -r requirements.txt
-python setup.py develop
+python setup.py install
 
 cd ..
 ```
@@ -40,34 +39,38 @@ More specifically, the agent can select from one of four actions at each decisio
 
 ### Quick Start
 
-See the [example notebook](baselines/goseek-ppo.ipynb) to train an agent.
+To run the example PPO baseline, install [Stable Baselines](https://stable-baselines.readthedocs.io/en/master/) and a version of [Tensorflow](https://www.tensorflow.org/) between v0.8.0 and v1.14.0 (see the [Stable Baselines docs](https://stable-baselines.readthedocs.io/en/master/guide/install.html#prerequisites) for details).
 
-### Challenge Details 
+__Note__: Currently, we've tested Python 3.7 Tensorflow installation with Anaconda against Cuda 10.0-10.2 (run `nvcc -V` to check your Cuda version).
 
-See the [GOSEEK Challenge](https://github.mit.edu/TESS/goseek-challenge) landing page for details on setup, evaluation, and submission.
+For Cuda 10.0, we'd recommend installing `tensorflow-gpu v1.13.1`:
 
-## Other Tasks
+```sh
+conda activate goseek 
+conda install tensorflow-gpu==1.13.1
+```
 
-### Navigation
+For Cuda 10.1 and 10.2, we'd recommend installing `tensorflow-gpu v1.14`:
 
-The agent must move throughout it's environment without collisions. See  the [example notebook](baselines/navigation.ipynb) to get started.
+```sh
+conda activate goseek 
+conda install tensorflow-gpu==1.14
+```
 
-### New Tasks
-At a minimum, a new task will inherit `tess_gym.TesseGym` and impliment the following:
+Then, install [Stable Baselines](https://stable-baselines.readthedocs.io/en/master/)
 
-```python
-class CustomTask(TesseGym):
-    @property
-    def action_space(self):
-        pass
-
-    def apply_action(self, action):
-        pass
-
-    def compute_reward(self, observation, action):
-        pass
+```sh
+conda activate goseek 
+pip install stable-baselines
 ```
 
+Then, see the [example notebook](baselines/goseek-ppo.ipynb) to train an agent.
+
+### Challenge Details 
+
+See the [GOSEEK Challenge](../../../goseek-challenge) landing page for details on setup, evaluation, and submission.
+
+
 ### Disclaimer
 
 DISTRIBUTION STATEMENT A. Approved for public release. Distribution is unlimited.