Merge pull request #1 from Bharat123rox/patch-1

Create README.md

README.md

@@ -0,0 +1,27 @@
+## Self Driving Cars in Video Games using Deep Learning
+
+### Contents
+- [Installation](#Installation)
+- [Input](#Input)
+- [Usage](#Usage)
+
+
+### Installation
+This model is built in top of PyTorch, follow the installation instructions from PyTorch [here](https://pytorch.org/get-started/locally/).
+
+To install dependencies, do
+```bash
+pip install -r requirements.txt
+```
+
+### Input
+Currently, only the **Windows OS** is supported due to the `win32gui` requirements for screen grabbing / input capture.
+The `screen_capture.py`, `grabscreen.py`, `directkeys.py` and `getkeys.py` scripts handle screen capturing, image grabbing and keypress event recording.
+
+### Usage
+```bash
+python train_il.py --data <path-to-train-data> --model <name-of-model-architecture> --batch-size 
+<batch-size-for-train> --n-epochs <no-of-epochs> --device <cpu/cuda>
+```
+which trains the model for Imitation Learning.
+