Update README

README.md

@@ -19,7 +19,7 @@
 
 
 
-# Craftax
+# ⛏️ Craftax
 Craftax is an RL environment written entirely in <a href="https://github.com/google/jax">JAX</a>.  Craftax reimplements and significantly extends the game mechanics of <a href="https://danijar.com/project/crafter/">Crafter</a>, taking inspiration from roguelike games such as <a href="https://github.com/facebookresearch/nle">NetHack</a>.
 Craftax conforms to the <a href="https://github.com/RobertTLange/gymnax">gymnax</a> interface, allowing easy integration with existing JAX-based frameworks like <a href="https://chrislu.page/blog/meta-disco/">PureJaxRL</a> and [JaxUED](https://github.com/DramaCow/jaxued).
 
@@ -34,9 +34,9 @@ Craftax conforms to the <a href="https://github.com/RobertTLange/gymnax">gymnax<
   <img src="https://raw.githubusercontent.com/MichaelTMatthews/Craftax/main/images/mining.gif" width="200" />
 </p>
 
-# Basic Usage
+# 📜 Basic Usage
 Craftax conforms to the gymnax interface:
-```
+```python
 rng = jax.random.PRNGKey(0)
 rng, _rng = jax.random.split(rng)
 rngs = jax.random.split(_rng, 3)
@@ -55,7 +55,7 @@ action = env.action_space(env_params).sample(rngs[1])
 obs, state, reward, done, info = env.step(rngs[2], state, action, env_params)
 ```
 
-# Installation
+# ⬇️ Installation
 The latest Craftax release can be installed from PyPi:
 ```
 pip install craftax
@@ -80,7 +80,7 @@ For NVIDIA GPU the command is:
 pip install -U "jax[cuda12_pip]" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html
 ```
 
-# Play
+# 🎮 Play
 To play Craftax run:
 ```
 play_craftax
@@ -90,7 +90,8 @@ or to play Craftax-Classic run:
 play_craftax_classic
 ```
 Since Craftax runs entirely in JAX, it will take some time to compile the rendering and step functions - it might take around 30s to render the first frame and then another 20s to take the first action.  After this it should be very quick.  A tutorial for how to beat the game is present in `tutorial.md`.  The controls are printed out at the beginning of play.
-# Experiment
+
+# 📈 Experiment
 To run PPO with default hyperparameters run:
 ```
 python -m craftax.ppo
@@ -102,7 +103,7 @@ python -m craftax.ppo_rnn
 To use ICM or E3B with the default parameters use the `--train_icm` and `--use_e3b` flags.
 Use the `env_name` parameter to control which environment is used.  It can be set to  `"Craftax-Symbolic-v1"`, `"Craftax-Pixels-v1"`, `"Craftax-Classic-Symbolic-v1"` or `"Craftax-Classic-Pixels-v1"`
 
-# Gotchas
+# 🔪 Gotchas
 ### Optimistic Resets
 Craftax provides the option to use optimistic resets to improve performance, which means that (unlike regular gymnax environments) it **does not auto-reset** by default.
 This means that the environment should always be wrapped either in `OptimisticResetVecEnvWrapper` (for efficient resets) or `AutoResetEnvWrapper` (to recover the default gymnax auto-reset behaviour).  See `ppo.py` for correct usage of both wrappers.
@@ -113,7 +114,7 @@ We use a texture cache to avoid recreating the texture atlas every time Craftax
 export CRAFTAX_RELOAD_TEXTURES=true
 ```
 
-# Scoreboard
+# 📋 Scoreboard
 If you would like to add an algorithm please open a PR and provide a reference to the source of the results.
 We report reward as a % of the maximum (226).