From aa29f27372ea7b879b93504cd7db5c43e0cd2800 Mon Sep 17 00:00:00 2001
From: Daniel <13357734+thedch@users.noreply.github.com>
Date: Thu, 12 Dec 2024 10:49:20 -0800
Subject: [PATCH 01/21] Update pong.py

---
 pufferlib/ocean/pong/pong.py | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/pufferlib/ocean/pong/pong.py b/pufferlib/ocean/pong/pong.py
index 6ec04ecd..9f1fe7d1 100644
--- a/pufferlib/ocean/pong/pong.py
+++ b/pufferlib/ocean/pong/pong.py
@@ -11,6 +11,7 @@
 import pufferlib
 from pufferlib.ocean.pong.cy_pong import CyPong
 
+
 class Pong(pufferlib.PufferEnv):
     def __init__(self, num_envs=1, render_mode=None,
             width=500, height=640, paddle_width=20, paddle_height=70,
@@ -18,8 +19,9 @@ def __init__(self, num_envs=1, render_mode=None,
             ball_initial_speed_x=10, ball_initial_speed_y=1,
             ball_speed_y_increment=3, ball_max_speed_y=13,
             max_score=21, frameskip=1, report_interval=1, buf=None):
-        self.single_observation_space = gymnasium.spaces.Box(low=0, high=1,
-            shape=(8,), dtype=np.float32)
+        self.single_observation_space = gymnasium.spaces.Box(
+            low=0, high=1, shape=(8,), dtype=np.float32,
+        )
         self.single_action_space = gymnasium.spaces.Discrete(3)
         self.render_mode = render_mode
         self.num_agents = num_envs
@@ -65,7 +67,7 @@ def test_performance(timeout=10, atn_cache=1024):
     env.reset()
     tick = 0
 
-    actions = np.random.randint(0, 2, (atn_cache, env.num_envs))
+    actions = np.random.randint(0, 2, (atn_cache, env.num_agents))
 
     import time
     start = time.time()
@@ -74,7 +76,8 @@ def test_performance(timeout=10, atn_cache=1024):
         env.step(atn)
         tick += 1
 
-    print(f'SPS: %f', env.num_envs * tick / (time.time() - start))
+    print('SPS: {env.num_agents * tick / (time.time() - start)})
+
 
 if __name__ == '__main__':
     test_performance()

From d06946e0806610d1696e95590687a0b0753da3a4 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Sat, 21 Dec 2024 21:12:55 +0000
Subject: [PATCH 02/21] Makes PufferLib compatible with Python3.12. Probably
 doesn't break anything important

---
 setup.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/setup.py b/setup.py
index 58f9501f..9d80dfaf 100644
--- a/setup.py
+++ b/setup.py
@@ -66,7 +66,7 @@
     'tensorboard==2.11.2',
     'torch',
     'tyro==0.8.6',
-    'wandb==0.13.7',
+    'wandb==0.19.1',
 ]
 
 ray = [
@@ -281,7 +281,7 @@
     },
     include_package_data=True,
     install_requires=[
-        'numpy==1.23.3',
+        'numpy==2.2.0',
         'opencv-python==3.4.17.63',
         'cython>=3.0.0',
         'rich',

From 952e935c4fb5aff12af64c39a0e7e0b9835ba281 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Sat, 21 Dec 2024 21:23:57 +0000
Subject: [PATCH 03/21] Fix bug

---
 pufferlib/ocean/pong/pong.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pufferlib/ocean/pong/pong.py b/pufferlib/ocean/pong/pong.py
index 9f1fe7d1..dc2685e0 100644
--- a/pufferlib/ocean/pong/pong.py
+++ b/pufferlib/ocean/pong/pong.py
@@ -76,7 +76,7 @@ def test_performance(timeout=10, atn_cache=1024):
         env.step(atn)
         tick += 1
 
-    print('SPS: {env.num_agents * tick / (time.time() - start)})
+    print(f'SPS: {env.num_agents * tick / (time.time() - start)}')
 
 
 if __name__ == '__main__':

From a5a054b94f2ad4ad8f00fa958310dba477a7cfb7 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Sat, 21 Dec 2024 21:36:54 +0000
Subject: [PATCH 04/21] Revert numpy

---
 setup.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/setup.py b/setup.py
index 9d80dfaf..983d339e 100644
--- a/setup.py
+++ b/setup.py
@@ -281,7 +281,7 @@
     },
     include_package_data=True,
     install_requires=[
-        'numpy==2.2.0',
+        'numpy==1.23.3',
         'opencv-python==3.4.17.63',
         'cython>=3.0.0',
         'rich',

From e5168f938e360bd7b1c244710dd3ec3f5ae1067a Mon Sep 17 00:00:00 2001
From: thatguy11325 <148832074+thatguy11325@users.noreply.github.com>
Date: Tue, 24 Dec 2024 09:45:05 -0500
Subject: [PATCH 05/21] add a barebones github action and status badge

---
 .github/workflows/install.yml | 27 +++++++++++++++++++++++++++
 README.md                     |  2 ++
 2 files changed, 29 insertions(+)
 create mode 100644 .github/workflows/install.yml

diff --git a/.github/workflows/install.yml b/.github/workflows/install.yml
new file mode 100644
index 00000000..da7fa6a5
--- /dev/null
+++ b/.github/workflows/install.yml
@@ -0,0 +1,27 @@
+name: install
+on:
+  push:
+  pull_request:
+
+jobs:
+  test:
+    name: test ${{ matrix.py }} - ${{ matrix.os }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        os:
+          - ubuntu-latest
+        py:
+          - "3.11"
+          - "3.10"
+    steps:
+      - name: Setup python for test ${{ matrix.py }}
+        uses: actions/setup-python@v4
+        with:
+          python-version: ${{ matrix.py }}
+      - uses: actions/checkout@v3
+      - name: Upgrade pip
+        run: python -m pip install -U pip
+      - name: Install pufferlib
+        run: pip3 install -e . 
\ No newline at end of file
diff --git a/README.md b/README.md
index b6614449..dc1bfaa8 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,8 @@
 ![figure](https://pufferai.github.io/source/resource/header.png)
 
 [![PyPI version](https://badge.fury.io/py/pufferlib.svg)](https://badge.fury.io/py/pufferlib)
+![PyPI - Python Version](https://img.shields.io/pypi/pyversions/pufferlib)
+![Github Actions](https://github.com/PufferAI/PufferLib/actions/workflows/install.yml/badge.svg)
 [![](https://dcbadge.vercel.app/api/server/spT4huaGYV?style=plastic)](https://discord.gg/spT4huaGYV)
 [![Twitter](https://img.shields.io/twitter/url/https/twitter.com/cloudposse.svg?style=social&label=Follow%20%40jsuarez5341)](https://twitter.com/jsuarez5341)
 

From 7fa25a8d2e499375570baffe29ee6d5581ad3192 Mon Sep 17 00:00:00 2001
From: thatguy11325 <148832074+thatguy11325@users.noreply.github.com>
Date: Tue, 24 Dec 2024 14:18:26 -0500
Subject: [PATCH 06/21] add macos

---
 .github/workflows/install.yml | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/.github/workflows/install.yml b/.github/workflows/install.yml
index da7fa6a5..5ebe7dc0 100644
--- a/.github/workflows/install.yml
+++ b/.github/workflows/install.yml
@@ -12,9 +12,12 @@ jobs:
       matrix:
         os:
           - ubuntu-latest
+          - macos-latest
         py:
           - "3.11"
           - "3.10"
+          - "3.9"
+          - "3.8"
     steps:
       - name: Setup python for test ${{ matrix.py }}
         uses: actions/setup-python@v4

From 4e793cf35d151248fb3c598fedc6555004ba82ad Mon Sep 17 00:00:00 2001
From: Joseph Suarez <sealsuarez@gmail.com>
Date: Tue, 24 Dec 2024 15:35:05 -0500
Subject: [PATCH 07/21] Update install.yml

---
 .github/workflows/install.yml | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/.github/workflows/install.yml b/.github/workflows/install.yml
index 5ebe7dc0..873d3c5c 100644
--- a/.github/workflows/install.yml
+++ b/.github/workflows/install.yml
@@ -17,7 +17,6 @@ jobs:
           - "3.11"
           - "3.10"
           - "3.9"
-          - "3.8"
     steps:
       - name: Setup python for test ${{ matrix.py }}
         uses: actions/setup-python@v4
@@ -27,4 +26,4 @@ jobs:
       - name: Upgrade pip
         run: python -m pip install -U pip
       - name: Install pufferlib
-        run: pip3 install -e . 
\ No newline at end of file
+        run: pip3 install -e . 

From c1dc5491553e997ecd28eacf1b5290759a599a3b Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Wed, 15 Jan 2025 02:08:26 +0000
Subject: [PATCH 08/21] Very dumb change. Prefix c functions with c_ to prevent
 Conda's garbage compiler from breaking

---
 config/default.ini                            |  1 +
 config/ocean/connect4.ini                     |  1 +
 config/ocean/go.ini                           |  1 +
 config/ocean/grid.ini                         |  1 +
 config/ocean/moba.ini                         |  1 +
 config/ocean/nmmo3.ini                        |  3 ++-
 config/ocean/snake.ini                        |  1 +
 config/ocean/trash_pickup.ini                 |  1 +
 demo.py                                       | 22 ++++++++++---------
 pufferlib/ocean/breakout/breakout.h           |  8 +++----
 pufferlib/ocean/breakout/cy_breakout.pyx      | 12 +++++-----
 pufferlib/ocean/connect4/connect4.h           |  8 +++----
 pufferlib/ocean/connect4/cy_connect4.pyx      | 12 +++++-----
 pufferlib/ocean/enduro/cy_enduro.pyx          |  8 +++----
 pufferlib/ocean/enduro/enduro.h               |  4 ++--
 pufferlib/ocean/go/cy_go.pyx                  | 16 ++++++--------
 pufferlib/ocean/go/go.h                       | 10 ++++-----
 pufferlib/ocean/moba/cy_moba.pyx              |  8 +++----
 pufferlib/ocean/moba/moba.h                   |  8 +++----
 pufferlib/ocean/nmmo3/cy_nmmo3.pyx            |  8 +++----
 pufferlib/ocean/nmmo3/nmmo3.h                 |  6 ++---
 pufferlib/ocean/pong/cy_pong.pyx              | 12 +++++-----
 pufferlib/ocean/pong/pong.h                   | 10 ++++-----
 pufferlib/ocean/rware/cy_rware.pyx            | 12 +++++-----
 pufferlib/ocean/rware/rware.h                 |  6 ++---
 pufferlib/ocean/snake/cy_snake.pyx            | 12 +++++-----
 pufferlib/ocean/snake/snake.h                 |  9 +++++---
 pufferlib/ocean/squared/cy_squared.pyx        | 13 +++++------
 pufferlib/ocean/squared/squared.h             | 10 ++++-----
 pufferlib/ocean/torch.py                      |  5 +----
 .../ocean/trash_pickup/cy_trash_pickup.pyx    | 13 +++++------
 pufferlib/ocean/trash_pickup/trash_pickup.h   | 16 ++++++++------
 .../ocean/tripletriad/cy_tripletriad.pyx      | 12 +++++-----
 pufferlib/ocean/tripletriad/tripletriad.h     |  8 +++----
 setup.py                                      | 22 ++++++++++++++-----
 35 files changed, 160 insertions(+), 140 deletions(-)

diff --git a/config/default.ini b/config/default.ini
index 6c41c811..6c923e0f 100644
--- a/config/default.ini
+++ b/config/default.ini
@@ -1,6 +1,7 @@
 [base]
 package = None
 env_name = None
+vec = native
 policy_name = Policy
 rnn_name = None
 max_suggestion_cost = 3600
diff --git a/config/ocean/connect4.ini b/config/ocean/connect4.ini
index 64964e82..2ad624b7 100644
--- a/config/ocean/connect4.ini
+++ b/config/ocean/connect4.ini
@@ -1,6 +1,7 @@
 [base]
 package = ocean
 env_name = puffer_connect4
+vec = multiprocessing
 policy_name = Policy
 rnn_name = Recurrent
 
diff --git a/config/ocean/go.ini b/config/ocean/go.ini
index 9790355f..a6397b0b 100644
--- a/config/ocean/go.ini
+++ b/config/ocean/go.ini
@@ -1,6 +1,7 @@
 [base]
 package = ocean
 env_name = puffer_go
+vec = multiprocessing
 policy_name = Go
 rnn_name = Recurrent
 
diff --git a/config/ocean/grid.ini b/config/ocean/grid.ini
index 95ec3f1f..0137d7a3 100644
--- a/config/ocean/grid.ini
+++ b/config/ocean/grid.ini
@@ -1,6 +1,7 @@
 [base]
 package = ocean
 env_name = puffer_grid
+vec = multiprocessing
 policy_name = Policy
 rnn_name = Recurrent
 
diff --git a/config/ocean/moba.ini b/config/ocean/moba.ini
index eaccece5..463739dd 100644
--- a/config/ocean/moba.ini
+++ b/config/ocean/moba.ini
@@ -1,6 +1,7 @@
 [base]
 package = ocean
 env_name = puffer_moba
+vec = multiprocessing
 policy_name = MOBA
 rnn_name = Recurrent
 
diff --git a/config/ocean/nmmo3.ini b/config/ocean/nmmo3.ini
index 3b4d8329..19b7c751 100644
--- a/config/ocean/nmmo3.ini
+++ b/config/ocean/nmmo3.ini
@@ -1,6 +1,7 @@
 [base]
 package = ocean
-env_name = nmmo3
+env_name = puffer_nmmo3
+vec = multiprocessing
 policy_name = NMMO3
 rnn_name = NMMO3LSTM
 
diff --git a/config/ocean/snake.ini b/config/ocean/snake.ini
index 4954254a..182acee6 100644
--- a/config/ocean/snake.ini
+++ b/config/ocean/snake.ini
@@ -1,6 +1,7 @@
 [base]
 package = ocean
 env_name = puffer_snake
+vec = multiprocessing
 rnn_name = Recurrent
 
 [env]
diff --git a/config/ocean/trash_pickup.ini b/config/ocean/trash_pickup.ini
index 9a07defa..c22eea6d 100644
--- a/config/ocean/trash_pickup.ini
+++ b/config/ocean/trash_pickup.ini
@@ -1,6 +1,7 @@
 [base]
 package = ocean
 env_name = trash_pickup puffer_trash_pickup 
+vec = multiprocessing
 policy_name = TrashPickup
 rnn_name = Recurrent
 
diff --git a/demo.py b/demo.py
index fbe93994..6a7192c0 100644
--- a/demo.py
+++ b/demo.py
@@ -199,13 +199,13 @@ def carbs_param(group, name, space, wandb_params, mmin=None, mmax=None,
         is_wandb_logging_enabled=False,
         resample_frequency=5,
         num_random_samples=len(param_spaces),
-        max_suggestion_cost=args['base']['max_suggestion_cost'],
+        max_suggestion_cost=args['max_suggestion_cost'],
         is_saved_on_every_observation=False,
     )
     carbs = CARBS(carbs_params, param_spaces)
 
     # GPUDrive doesn't let you reinit the vecenv, so we have to cache it
-    cache_vecenv = args['base']['env_name'] == 'gpudrive'
+    cache_vecenv = args['env_name'] == 'gpudrive'
 
     elos = {'model_random.pt': 1000}
     vecenv = {'vecenv': None} # can't reassign otherwise
@@ -293,7 +293,7 @@ def train(args, make_env, policy_cls, rnn_cls, wandb,
     elif args['vec'] == 'native':
         vec = pufferlib.environment.PufferEnv
     else:
-        raise ValueError(f'Invalid --vector (serial/multiprocessing/ray/native).')
+        raise ValueError(f'Invalid --vec (serial/multiprocessing/ray/native).')
 
     if vecenv is None:
         vecenv = pufferlib.vector.make(
@@ -360,8 +360,6 @@ def train(args, make_env, policy_cls, rnn_cls, wandb,
         default='puffer_squared', help='Name of specific environment to run')
     parser.add_argument('--mode', type=str, default='train',
         choices='train eval evaluate sweep sweep-carbs autotune profile'.split())
-    parser.add_argument('--vec', '--vector', '--vectorization', type=str,
-        default='native', choices=['serial', 'multiprocessing', 'ray', 'native'])
     parser.add_argument('--vec-overwork', action='store_true',
         help='Allow vectorization to use >1 worker/core. Not recommended.')
     parser.add_argument('--eval-model-path', type=str, default=None,
@@ -377,6 +375,7 @@ def train(args, make_env, policy_cls, rnn_cls, wandb,
     parser.add_argument('--wandb-group', type=str, default='debug')
     args = parser.parse_known_args()[0]
 
+
     file_paths = glob.glob('config/**/*.ini', recursive=True)
     for path in file_paths:
         p = configparser.ConfigParser()
@@ -394,7 +393,10 @@ def train(args, make_env, policy_cls, rnn_cls, wandb,
 
     for section in p.sections():
         for key in p[section]:
-            argparse_key = f'--{section}.{key}'.replace('_', '-')
+            if section == 'base':
+                argparse_key = f'--{key}'.replace('_', '-')
+            else:
+                argparse_key = f'--{section}.{key}'.replace('_', '-')
             parser.add_argument(argparse_key, default=p[section][key])
 
     # Late add help so you get a dynamic menu based on the env
@@ -416,7 +418,7 @@ def train(args, make_env, policy_cls, rnn_cls, wandb,
         except:
             prev[subkey] = value
 
-    package = args['base']['package']
+    package = args['package']
     module_name = f'pufferlib.environments.{package}'
     if package == 'ocean':
         module_name = 'pufferlib.ocean'
@@ -425,12 +427,12 @@ def train(args, make_env, policy_cls, rnn_cls, wandb,
     env_module = importlib.import_module(module_name)
 
     make_env = env_module.env_creator(env_name)
-    policy_cls = getattr(env_module.torch, args['base']['policy_name'])
+    policy_cls = getattr(env_module.torch, args['policy_name'])
     
-    rnn_name = args['base']['rnn_name']
+    rnn_name = args['rnn_name']
     rnn_cls = None
     if rnn_name is not None:
-        rnn_cls = getattr(env_module.torch, args['base']['rnn_name'])
+        rnn_cls = getattr(env_module.torch, args['rnn_name'])
 
     if args['baseline']:
         assert args['mode'] in ('train', 'eval', 'evaluate')
diff --git a/pufferlib/ocean/breakout/breakout.h b/pufferlib/ocean/breakout/breakout.h
index 76fb81dc..96e45fb8 100644
--- a/pufferlib/ocean/breakout/breakout.h
+++ b/pufferlib/ocean/breakout/breakout.h
@@ -437,7 +437,7 @@ void reset_round(Breakout* env) {
     env->ball_vx = 0.0;
     env->ball_vy = 0.0;
 }
-void reset(Breakout* env) {
+void c_reset(Breakout* env) {
     env->log = (Log){0};
     env->score = 0;
     env->num_balls = 5;
@@ -482,11 +482,11 @@ void step_frame(Breakout* env, int action) {
         env->dones[0] = 1;
         env->log.score = env->score;
         add_log(env->log_buffer, &env->log);
-        reset(env);
+        c_reset(env);
     }
 }
 
-void step(Breakout* env) {
+void c_step(Breakout* env) {
     env->dones[0] = 0;
     env->log.episode_length += 1;
     env->rewards[0] = 0.0;
@@ -523,7 +523,7 @@ Client* make_client(Breakout* env) {
     return client;
 }
 
-void render(Client* client, Breakout* env) {
+void c_render(Client* client, Breakout* env) {
     if (IsKeyDown(KEY_ESCAPE)) {
         exit(0);
     }
diff --git a/pufferlib/ocean/breakout/cy_breakout.pyx b/pufferlib/ocean/breakout/cy_breakout.pyx
index 484d9266..443fd896 100644
--- a/pufferlib/ocean/breakout/cy_breakout.pyx
+++ b/pufferlib/ocean/breakout/cy_breakout.pyx
@@ -55,9 +55,9 @@ cdef extern from "breakout.h":
 
     Client* make_client(Breakout* env)
     void close_client(Client* client)
-    void render(Client* client, Breakout* env)
-    void reset(Breakout* env)
-    void step(Breakout* env)
+    void c_render(Client* client, Breakout* env)
+    void c_reset(Breakout* env)
+    void c_step(Breakout* env)
 
 cdef class CyBreakout:
     cdef:
@@ -103,12 +103,12 @@ cdef class CyBreakout:
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def render(self):
         cdef Breakout* env = &self.envs[0]
@@ -119,7 +119,7 @@ cdef class CyBreakout:
             self.client = make_client(env)
             os.chdir(cwd)
 
-        render(self.client, env)
+        c_render(self.client, env)
 
     def close(self):
         if self.client != NULL:
diff --git a/pufferlib/ocean/connect4/connect4.h b/pufferlib/ocean/connect4/connect4.h
index 5fb2bd59..cbb0261b 100644
--- a/pufferlib/ocean/connect4/connect4.h
+++ b/pufferlib/ocean/connect4/connect4.h
@@ -276,7 +276,7 @@ void compute_observation(CConnect4* env) {
     }
 }
 
-void reset(CConnect4* env) {
+void c_reset(CConnect4* env) {
     env->log = (Log){0};
     env->dones[0] = NOT_DONE;
     env->player_pieces = 0;
@@ -294,13 +294,13 @@ void finish_game(CConnect4* env, float reward) {
     compute_observation(env);
 }
 
-void step(CConnect4* env) {
+void c_step(CConnect4* env) {
     env->log.episode_length += 1;
     env->rewards[0] = 0.0;
 
     if (env->dones[0] == DONE) {
         add_log(env->log_buffer, &env->log);
-        reset(env);
+        c_reset(env);
         return;
     }
 
@@ -359,7 +359,7 @@ Client* make_client(int width, int height) {
     return client;
 }
 
-void render(Client* client, CConnect4* env) {
+void c_render(Client* client, CConnect4* env) {
     if (IsKeyDown(KEY_ESCAPE)) {
         exit(0);
     }
diff --git a/pufferlib/ocean/connect4/cy_connect4.pyx b/pufferlib/ocean/connect4/cy_connect4.pyx
index abca0fb3..6eb39590 100644
--- a/pufferlib/ocean/connect4/cy_connect4.pyx
+++ b/pufferlib/ocean/connect4/cy_connect4.pyx
@@ -36,9 +36,9 @@ cdef extern from "connect4.h":
     void free_cconnect4(CConnect4* env)
     Client* make_client(float width, float height)
     void close_client(Client* client)
-    void render(Client* client, CConnect4* env)
-    void reset(CConnect4* env)
-    void step(CConnect4* env)
+    void c_render(Client* client, CConnect4* env)
+    void c_reset(CConnect4* env)
+    void c_step(CConnect4* env)
 
 cdef class CyConnect4:
     cdef:
@@ -75,12 +75,12 @@ cdef class CyConnect4:
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def render(self):
         cdef CConnect4* env = &self.envs[0]
@@ -91,7 +91,7 @@ cdef class CyConnect4:
             self.client = make_client(env.width, env.height)
             os.chdir(cwd)
 
-        render(self.client, env)
+        c_render(self.client, env)
 
     def close(self):
         if self.client != NULL:
diff --git a/pufferlib/ocean/enduro/cy_enduro.pyx b/pufferlib/ocean/enduro/cy_enduro.pyx
index 40e9ba6b..4baed1cf 100644
--- a/pufferlib/ocean/enduro/cy_enduro.pyx
+++ b/pufferlib/ocean/enduro/cy_enduro.pyx
@@ -52,7 +52,7 @@ cdef extern from "enduro.h":
     void free_logbuffer(LogBuffer* buffer)
     Log aggregate_and_clear(LogBuffer* logs)
     void init(Enduro* env, int seed, int env_index)
-    void reset(Enduro* env)
+    void c_reset(Enduro* env)
     void c_step(Enduro* env)
     void c_render(Client* client, Enduro* env)
     Client* make_client(Enduro* env)
@@ -103,15 +103,15 @@ cdef class CyEnduro:
             self.envs[i].log_buffer = self.logs
             self.envs[i].obs_size = observations.shape[1]
 
-            if i % 100 == 0:
-                print(f"Initializing environment #{i} with seed {unique_seed}")
+            #if i % 100 == 0:
+            #    print(f"Initializing environment #{i} with seed {unique_seed}")
 
             init(&self.envs[i], unique_seed, i)
 
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
diff --git a/pufferlib/ocean/enduro/enduro.h b/pufferlib/ocean/enduro/enduro.h
index 7e442e18..6369413a 100644
--- a/pufferlib/ocean/enduro/enduro.h
+++ b/pufferlib/ocean/enduro/enduro.h
@@ -475,7 +475,7 @@ void allocate(Enduro* env);
 void init(Enduro* env, int seed, int env_index);
 void free_allocated(Enduro* env);
 void reset_round(Enduro* env);
-void reset(Enduro* env);
+void c_reset(Enduro* env);
 unsigned char check_collision(Enduro* env, Car* car);
 int get_player_lane(Enduro* env);
 float get_car_scale(float y);
@@ -865,7 +865,7 @@ void reset_round(Enduro* env) {
 }
 
 // Reset all init vars; only called once after init
-void reset(Enduro* env) {
+void c_reset(Enduro* env) {
     // No random after first reset
     int reset_seed = (env->reset_count == 0) ? xorshift32(&env->rng_state) : 0;
 
diff --git a/pufferlib/ocean/go/cy_go.pyx b/pufferlib/ocean/go/cy_go.pyx
index 5b39e303..d8595c40 100644
--- a/pufferlib/ocean/go/cy_go.pyx
+++ b/pufferlib/ocean/go/cy_go.pyx
@@ -30,8 +30,6 @@ cdef extern from "go.h":
     int find(Group*)
     void union_groups(Group*, int, int)
 
-
-
     ctypedef struct CGo:
         float* observations
         int* actions
@@ -68,12 +66,12 @@ cdef extern from "go.h":
 
     void init(CGo* env)
     void free_initialized(CGo* env)
-    void reset(CGo* env)
-    void step(CGo* env)
+    void c_reset(CGo* env)
+    void c_step(CGo* env)
 
     Client* make_client(float width, float height)
     void close_client(Client* client)
-    void render(Client* client, CGo* env)
+    void c_render(Client* client, CGo* env)
     
 
 cdef class CyGo:
@@ -122,19 +120,19 @@ cdef class CyGo:
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def render(self):
         cdef CGo* env = &self.envs[0]
         if self.client == NULL:
             self.client = make_client(env.width,env.height)
 
-        render(self.client, &self.envs[0])
+        c_render(self.client, &self.envs[0])
 
     def close(self):
         if self.client != NULL:
@@ -144,4 +142,4 @@ cdef class CyGo:
 
     def log(self):
         cdef Log log = aggregate_and_clear(self.logs)
-        return log
\ No newline at end of file
+        return log
diff --git a/pufferlib/ocean/go/go.h b/pufferlib/ocean/go/go.h
index ca58ad53..afd51f04 100644
--- a/pufferlib/ocean/go/go.h
+++ b/pufferlib/ocean/go/go.h
@@ -648,7 +648,7 @@ void enemy_greedy_easy(CGo* env){
     enemy_random_move(env);
 }
 
-void reset(CGo* env) {
+void c_reset(CGo* env) {
     env->log = (Log){0};
     env->dones[0] = 0;
     env->score = 0;
@@ -687,10 +687,10 @@ void end_game(CGo* env){
     env->log.games_played++;
     env->log.episode_return += env->rewards[0];
     add_log(env->log_buffer, &env->log);
-    reset(env);
+    c_reset(env);
 }
 
-void step(CGo* env) {
+void c_step(CGo* env) {
     env->log.episode_length += 1;
     env->rewards[0] = 0.0;
     int action = (int)env->actions[0];
@@ -767,7 +767,7 @@ Client* make_client(int width, int height) {
     return client;
 }
 
-void render(Client* client, CGo* env) {
+void c_render(Client* client, CGo* env) {
     if (IsKeyDown(KEY_ESCAPE)) {
         exit(0);
     }
@@ -839,4 +839,4 @@ void render(Client* client, CGo* env) {
 void close_client(Client* client) {
     CloseWindow();
     free(client);
-}
\ No newline at end of file
+}
diff --git a/pufferlib/ocean/moba/cy_moba.pyx b/pufferlib/ocean/moba/cy_moba.pyx
index 13267bec..7f74b9d6 100644
--- a/pufferlib/ocean/moba/cy_moba.pyx
+++ b/pufferlib/ocean/moba/cy_moba.pyx
@@ -198,8 +198,8 @@ cdef extern from "moba.h":
  
     unsigned char* read_file(char* filename)
 
-    void reset(MOBA* env)
-    void step(MOBA* env)
+    void c_reset(MOBA* env)
+    void c_step(MOBA* env)
     void randomize_tower_hp(MOBA* env)
 
 cpdef entity_dtype():
@@ -267,12 +267,12 @@ cdef class CyMOBA:
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def render(self, int tick):
         if self.client == NULL:
diff --git a/pufferlib/ocean/moba/moba.h b/pufferlib/ocean/moba/moba.h
index 57717f9e..3243433c 100644
--- a/pufferlib/ocean/moba/moba.h
+++ b/pufferlib/ocean/moba/moba.h
@@ -1820,7 +1820,7 @@ MOBA* allocate_moba(MOBA* env) {
     return env;
 }
  
-void reset(MOBA* env) {
+void c_reset(MOBA* env) {
     //map->pids[:] = -1
     //randomize_tower_hp(env);
     
@@ -1890,7 +1890,7 @@ void reset(MOBA* env) {
     compute_observations(env);
 }
 
-void step(MOBA* env) {
+void c_step(MOBA* env) {
     for (int pid = 0; pid < NUM_ENTITIES; pid++) {
         Entity* entity = &env->entities[pid];
         entity->target_pid = -1;
@@ -1997,7 +1997,7 @@ void step(MOBA* env) {
         log.dire_carry = env->log[9];
         add_log(env->log_buffer, &log);
         if (do_reset) {
-            reset(env);
+            c_reset(env);
         }
     }
     compute_observations(env);
@@ -2261,7 +2261,7 @@ int render_game(GameRenderer* renderer, MOBA* env, int frame) {
         }
     }
     if (IsKeyDown(KEY_ESCAPE)) {
-        return 1;
+        exit(0);
     }
     if (HUMAN_CONTROL) {
         if (IsKeyDown(KEY_Q) || IsKeyPressed(KEY_Q)) {
diff --git a/pufferlib/ocean/nmmo3/cy_nmmo3.pyx b/pufferlib/ocean/nmmo3/cy_nmmo3.pyx
index 0c909194..65d03709 100644
--- a/pufferlib/ocean/nmmo3/cy_nmmo3.pyx
+++ b/pufferlib/ocean/nmmo3/cy_nmmo3.pyx
@@ -138,8 +138,8 @@ cdef extern from "nmmo3.h":
     int tick(Client* client, MMO* env, float delta)
 
     void init_mmo(MMO* env)
-    void reset(MMO* env, int seed)
-    void step(MMO* env)
+    void c_reset(MMO* env, int seed)
+    void c_step(MMO* env)
 
 cpdef entity_dtype():
     '''Make a dummy entity to get the dtype'''
@@ -226,13 +226,13 @@ cdef class Environment:
         cdef int i
         for i in range(self.num_envs):
             # TODO: Seed
-            reset(&self.envs[i], i+1)
+            c_reset(&self.envs[i], i+1)
             # Do I need to reset terrain here?
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def pids(self):
         ary = np.zeros((512, 512), dtype=np.intc)
diff --git a/pufferlib/ocean/nmmo3/nmmo3.h b/pufferlib/ocean/nmmo3/nmmo3.h
index c7794bc7..138ce1e1 100644
--- a/pufferlib/ocean/nmmo3/nmmo3.h
+++ b/pufferlib/ocean/nmmo3/nmmo3.h
@@ -1669,7 +1669,7 @@ void enemy_ai(MMO* env, int pid) {
     wander(env, pid);
 }
 
-void reset(MMO* env, int seed) {
+void c_reset(MMO* env, int seed) {
     srand(time(NULL));
     env->tick = 0;
 
@@ -1873,7 +1873,7 @@ void reset(MMO* env, int seed) {
     compute_all_obs(env);
 }
 
-void step(MMO* env) {
+void c_step(MMO* env) {
     env->tick += 1;
     int tick = env->tick;
 
@@ -2576,7 +2576,7 @@ void close_client(Client* client) {
     UnloadRenderTexture(client->ui_buffer);
     for (int i = 0; i < NUM_PLAYER_TEXTURES; i++) {
         for (int element = 0; element < 5; element++) {
-            UnloadTexture(client->players[i][element]);
+            UnloadTexture(client->players[element][i]);
         }
     }
     UnloadFont(client->font);
diff --git a/pufferlib/ocean/pong/cy_pong.pyx b/pufferlib/ocean/pong/cy_pong.pyx
index 75c652ee..5a4eff86 100644
--- a/pufferlib/ocean/pong/cy_pong.pyx
+++ b/pufferlib/ocean/pong/cy_pong.pyx
@@ -53,12 +53,12 @@ cdef extern from "pong.h":
     ctypedef struct Client
 
     void init(Pong* env)
-    void reset(Pong* env)
-    void step(Pong* env)
+    void c_reset(Pong* env)
+    void c_step(Pong* env)
 
     Client* make_client(Pong* env)
     void close_client(Client* client)
-    void render(Client* client, Pong* env)
+    void c_render(Client* client, Pong* env)
 
 cdef class CyPong:
     cdef:
@@ -113,12 +113,12 @@ cdef class CyPong:
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def render(self):
         cdef Pong* env = &self.envs[0]
@@ -129,7 +129,7 @@ cdef class CyPong:
             self.client = make_client(env)
             os.chdir(cwd)
 
-        render(self.client, env)
+        c_render(self.client, env)
 
     def close(self):
         if self.client != NULL:
diff --git a/pufferlib/ocean/pong/pong.h b/pufferlib/ocean/pong/pong.h
index e08498af..5ac28692 100644
--- a/pufferlib/ocean/pong/pong.h
+++ b/pufferlib/ocean/pong/pong.h
@@ -147,7 +147,7 @@ void reset_round(Pong* env) {
     env->n_bounces = 0;
 }
 
-void reset(Pong* env) {
+void c_reset(Pong* env) {
     env->log = (Log){0};
     reset_round(env);
     env->score_l = 0;
@@ -155,7 +155,7 @@ void reset(Pong* env) {
     compute_observations(env);
 }
 
-void step(Pong* env) {
+void c_step(Pong* env) {
     env->tick += 1;
     env->log.episode_length += 1;
     env->rewards[0] = 0;
@@ -213,7 +213,7 @@ void step(Pong* env) {
                 if (env->score_r == env->max_score) {
                     env->terminals[0] = 1;
                     add_log(env->log_buffer, &env->log);
-                    reset(env);
+                    c_reset(env);
                     return;
                 } else {
                     reset_round(env);
@@ -246,7 +246,7 @@ void step(Pong* env) {
                 if (env->score_l == env->max_score) {
                     env->terminals[0] = 1;
                     add_log(env->log_buffer, &env->log);
-                    reset(env);
+                    c_reset(env);
                     return;
                 } else {
                     reset_round(env);
@@ -302,7 +302,7 @@ void close_client(Client* client) {
     free(client);
 }
 
-void render(Client* client, Pong* env) {
+void c_render(Client* client, Pong* env) {
     if (IsKeyDown(KEY_ESCAPE)) {
         exit(0);
     }
diff --git a/pufferlib/ocean/rware/cy_rware.pyx b/pufferlib/ocean/rware/cy_rware.pyx
index b796929a..7364d125 100644
--- a/pufferlib/ocean/rware/cy_rware.pyx
+++ b/pufferlib/ocean/rware/cy_rware.pyx
@@ -51,9 +51,9 @@ cdef extern from "rware.h":
 
     Client* make_client(CRware* env)
     void close_client(Client* client)
-    void render(Client* client, CRware* env)
-    void reset(CRware* env)
-    void step(CRware* env)
+    void c_render(Client* client, CRware* env)
+    void c_reset(CRware* env)
+    void c_step(CRware* env)
 
 cdef class CyRware:
     cdef:
@@ -96,12 +96,12 @@ cdef class CyRware:
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def render(self):
         cdef CRware* env = &self.envs[0]
@@ -112,7 +112,7 @@ cdef class CyRware:
             self.client = make_client(env)
             os.chdir(cwd)
 
-        render(self.client, env)
+        c_render(self.client, env)
 
     def close(self):
         if self.client != NULL:
diff --git a/pufferlib/ocean/rware/rware.h b/pufferlib/ocean/rware/rware.h
index 14539946..be73651b 100644
--- a/pufferlib/ocean/rware/rware.h
+++ b/pufferlib/ocean/rware/rware.h
@@ -405,7 +405,7 @@ void compute_observations(CRware* env) {
     }
 }
 
-void reset(CRware* env) {
+void c_reset(CRware* env) {
      
 	env->dones[0] = 0;
     // set agents in center
@@ -720,7 +720,7 @@ void process_tree_movements(CRware* env, MovementGraph* graph) {
     }
 }
 
-void step(CRware* env) {
+void c_step(CRware* env) {
     memset(env->rewards, 0, env->num_agents * sizeof(float));
     MovementGraph* graph = env->movement_graph;
     for (int i = 0; i < env->num_agents; i++) {
@@ -778,7 +778,7 @@ Client* make_client(CRware* env) {
     return client;
 }
 
-void render(Client* client, CRware* env) {
+void c_render(Client* client, CRware* env) {
     if (IsKeyDown(KEY_ESCAPE)) {
         exit(0);
     }
diff --git a/pufferlib/ocean/snake/cy_snake.pyx b/pufferlib/ocean/snake/cy_snake.pyx
index 0f0cd5a8..d7ddb5a6 100644
--- a/pufferlib/ocean/snake/cy_snake.pyx
+++ b/pufferlib/ocean/snake/cy_snake.pyx
@@ -47,12 +47,12 @@ cdef extern from "snake.h":
         void compute_observations(CSnake* env)
         void spawn_snake(CSnake* env, int snake_id)
         void spawn_food(CSnake* env)
-        void reset(CSnake* env)
+        void c_reset(CSnake* env)
         void step_snake(CSnake* env, int i)
-        void step(CSnake* env)
+        void c_step(CSnake* env)
         ctypedef struct Client
         Client* make_client(int cell_size, int width, int height)
-        void render(Client* client, CSnake* env)
+        void c_render(Client* client, CSnake* env)
         void close_client(Client* client)
 
 cdef class CySnake:
@@ -100,19 +100,19 @@ cdef class CySnake:
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def render(self, cell_size=8):
         cdef CSnake* env = &self.envs[0]
         if self.client == NULL:
             self.client = make_client(cell_size, env.width, env.height)
 
-        render(self.client, env)
+        c_render(self.client, env)
 
     def close(self):
         if self.client != NULL:
diff --git a/pufferlib/ocean/snake/snake.h b/pufferlib/ocean/snake/snake.h
index 2c1d7474..48fa2509 100644
--- a/pufferlib/ocean/snake/snake.h
+++ b/pufferlib/ocean/snake/snake.h
@@ -194,7 +194,7 @@ void spawn_food(CSnake* env) {
     env->grid[idx] = FOOD;
 }
 
-void reset(CSnake* env) {
+void c_reset(CSnake* env) {
     env->window = 2*env->vision+1;
     env->obs_size = env->window*env->window;
 
@@ -300,7 +300,7 @@ void step_snake(CSnake* env, int i) {
     env->grid[next_r*env->width + next_c] = env->snake_colors[i];
 }
 
-void step(CSnake* env){
+void c_step(CSnake* env){
     for (int i = 0; i < env->num_snakes; i++)
         step_snake(env, i);
 
@@ -342,7 +342,10 @@ void close_client(Client* client) {
     free(client);
 }
 
-void render(Client* client, CSnake* env) {
+void c_render(Client* client, CSnake* env) {
+    if (IsKeyDown(KEY_ESCAPE)) {
+        exit(0);
+    }
     BeginDrawing();
     ClearBackground(COLORS[0]);
     int sz = client->cell_size;
diff --git a/pufferlib/ocean/squared/cy_squared.pyx b/pufferlib/ocean/squared/cy_squared.pyx
index 249f0376..d90f0dd6 100644
--- a/pufferlib/ocean/squared/cy_squared.pyx
+++ b/pufferlib/ocean/squared/cy_squared.pyx
@@ -14,12 +14,11 @@ cdef extern from "squared.h":
 
     ctypedef struct Client
 
-    void reset(Squared* env)
-    void step(Squared* env)
-
+    void c_reset(Squared* env)
+    void c_step(Squared* env)
     Client* make_client(Squared* env)
     void close_client(Client* client)
-    void render(Client* client, Squared* env)
+    void c_render(Client* client, Squared* env)
 
 cdef class CySquared:
     cdef:
@@ -48,19 +47,19 @@ cdef class CySquared:
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def render(self):
         cdef Squared* env = &self.envs[0]
         if self.client == NULL:
             self.client = make_client(env)
 
-        render(self.client, env)
+        c_render(self.client, env)
 
     def close(self):
         if self.client != NULL:
diff --git a/pufferlib/ocean/squared/squared.h b/pufferlib/ocean/squared/squared.h
index 74a7de6f..77cf6f50 100644
--- a/pufferlib/ocean/squared/squared.h
+++ b/pufferlib/ocean/squared/squared.h
@@ -38,7 +38,7 @@ void free_allocated(Squared* env) {
     free(env->terminals);
 }
 
-void reset(Squared* env) {
+void c_reset(Squared* env) {
     memset(env->observations, 0, env->size*env->size*sizeof(unsigned char));
     env->observations[env->size*env->size/2] = AGENT;
     env->r = env->size/2;
@@ -51,7 +51,7 @@ void reset(Squared* env) {
     env->observations[target_idx] = TARGET;
 }
 
-void step(Squared* env) {
+void c_step(Squared* env) {
     int action = env->actions[0];
     env->terminals[0] = 0;
     env->rewards[0] = 0;
@@ -75,7 +75,7 @@ void step(Squared* env) {
             || env->c >= env->size) {
         env->terminals[0] = 1;
         env->rewards[0] = -1.0;
-        reset(env);
+        c_reset(env);
         return;
     }
 
@@ -83,7 +83,7 @@ void step(Squared* env) {
     if (env->observations[pos] == TARGET) {
         env->terminals[0] = 1;
         env->rewards[0] = 1.0;
-        reset(env);
+        c_reset(env);
         return;
     }
 
@@ -111,7 +111,7 @@ void close_client(Client* client) {
     free(client);
 }
 
-void render(Client* client, Squared* env) {
+void c_render(Client* client, Squared* env) {
     if (IsKeyDown(KEY_ESCAPE)) {
         exit(0);
     }
diff --git a/pufferlib/ocean/torch.py b/pufferlib/ocean/torch.py
index e4d5c0cb..ce99d7dc 100644
--- a/pufferlib/ocean/torch.py
+++ b/pufferlib/ocean/torch.py
@@ -300,7 +300,6 @@ def decode_actions(self, flat_hidden, lookup, concat=None):
 class TrashPickup(nn.Module):
     def __init__(self, env, cnn_channels=32, hidden_size=128, **kwargs):
         super().__init__()
-        self.agent_sight_range = env.agent_sight_range
         self.network= nn.Sequential(
             pufferlib.pytorch.layer_init(
                 nn.Conv2d(5, cnn_channels, 5, stride=3)),
@@ -323,9 +322,7 @@ def forward(self, observations):
         return actions, value
 
     def encode_observations(self, observations):
-        crop_size = 2 * self.agent_sight_range + 1
-        observations = observations.view(-1, 5, crop_size, crop_size).float()
-        #observations = observations.view(-1, crop_size, crop_size, 5).permute(0, 3, 1, 2).float()
+        observations = observations.view(-1, 5, 11, 11).float()
         return self.network(observations), None
 
     def decode_actions(self, flat_hidden, lookup, concat=None):
diff --git a/pufferlib/ocean/trash_pickup/cy_trash_pickup.pyx b/pufferlib/ocean/trash_pickup/cy_trash_pickup.pyx
index 19d25f81..caf4cacf 100644
--- a/pufferlib/ocean/trash_pickup/cy_trash_pickup.pyx
+++ b/pufferlib/ocean/trash_pickup/cy_trash_pickup.pyx
@@ -37,10 +37,9 @@ cdef extern from "trash_pickup.h":
 
     Client* make_client(CTrashPickupEnv* env)
     void close_client(Client* client)
-    void render(Client* client, CTrashPickupEnv* env) 
-
-    void reset(CTrashPickupEnv* env)
-    void step(CTrashPickupEnv* env)
+    void c_render(Client* client, CTrashPickupEnv* env) 
+    void c_reset(CTrashPickupEnv* env)
+    void c_step(CTrashPickupEnv* env)
 
 cdef class CyTrashPickup:
     cdef:
@@ -83,19 +82,19 @@ cdef class CyTrashPickup:
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def render(self):
         cdef CTrashPickupEnv* env = &self.envs[0]
         if self.client == NULL:
             self.client = make_client(env)
 
-        render(self.client, env)
+        c_render(self.client, env)
 
     def close(self):
         if self.client != NULL:
diff --git a/pufferlib/ocean/trash_pickup/trash_pickup.h b/pufferlib/ocean/trash_pickup/trash_pickup.h
index aa460cc7..ba7c441a 100644
--- a/pufferlib/ocean/trash_pickup/trash_pickup.h
+++ b/pufferlib/ocean/trash_pickup/trash_pickup.h
@@ -167,8 +167,6 @@ void compute_observations(CTrashPickupEnv* env) {
 // Local crop version
 void compute_observations(CTrashPickupEnv* env) {
     int sight_range = env->agent_sight_range;
-    int num_cell_types = 4;  // EMPTY, TRASH, BIN, AGENT
-
     char* obs = env->observations;
 
     int obs_dim = 2*env->agent_sight_range + 1;
@@ -354,7 +352,7 @@ bool is_episode_over(CTrashPickupEnv* env) {
     return true;
 }
 
-void reset(CTrashPickupEnv* env) {
+void c_reset(CTrashPickupEnv* env) {
     env->current_step = 0;
     env->total_episode_reward = 0;
 
@@ -383,7 +381,7 @@ void initialize_env(CTrashPickupEnv* env) {
     env->entities = (Entity*)calloc(env->num_agents + env->num_bins + env->num_trash, sizeof(Entity));
     env->total_num_obs = env->num_agents * ((((env->agent_sight_range * 2 + 1) * (env->agent_sight_range * 2 + 1)) * 5));
 
-    reset(env);
+    c_reset(env);
 }
 
 void allocate(CTrashPickupEnv* env) {
@@ -397,7 +395,7 @@ void allocate(CTrashPickupEnv* env) {
     initialize_env(env);
 }
 
-void step(CTrashPickupEnv* env) {
+void c_step(CTrashPickupEnv* env) {
     // Reset reward for each agent
     memset(env->rewards, 0, sizeof(float) * env->num_agents);
     memset(env->dones, 0, sizeof(unsigned char) * env->num_agents);
@@ -427,7 +425,7 @@ void step(CTrashPickupEnv* env) {
 
         add_log(env->log_buffer, &log);
 
-        reset(env);
+        c_reset(env);
     }
 
     compute_observations(env);
@@ -479,7 +477,11 @@ Client* make_client(CTrashPickupEnv* env) {
 }
 
 // Render the TrashPickup environment
-void render(Client* client, CTrashPickupEnv* env) {
+void c_render(Client* client, CTrashPickupEnv* env) {
+    if (IsKeyDown(KEY_ESCAPE)) {
+        exit(0);
+    }
+
     BeginDrawing();
     ClearBackground(PUFF_BACKGROUND);
 
diff --git a/pufferlib/ocean/tripletriad/cy_tripletriad.pyx b/pufferlib/ocean/tripletriad/cy_tripletriad.pyx
index 903659c0..561f2352 100644
--- a/pufferlib/ocean/tripletriad/cy_tripletriad.pyx
+++ b/pufferlib/ocean/tripletriad/cy_tripletriad.pyx
@@ -43,9 +43,9 @@ cdef extern from "tripletriad.h":
 
     Client* make_client(float width, float height)
     void close_client(Client* client)
-    void render(Client* client, CTripleTriad* env)
-    void reset(CTripleTriad* env)
-    void step(CTripleTriad* env)
+    void c_render(Client* client, CTripleTriad* env)
+    void c_reset(CTripleTriad* env)
+    void c_step(CTripleTriad* env)
 
 cdef class CyTripleTriad:
     cdef:
@@ -82,19 +82,19 @@ cdef class CyTripleTriad:
     def reset(self):
         cdef int i
         for i in range(self.num_envs):
-            reset(&self.envs[i])
+            c_reset(&self.envs[i])
 
     def step(self):
         cdef int i
         for i in range(self.num_envs):
-            step(&self.envs[i])
+            c_step(&self.envs[i])
 
     def render(self):
         cdef CTripleTriad* env = &self.envs[0]
         if self.client == NULL:
             self.client = make_client(env.width, env.height)
 
-        render(self.client, env)
+        c_render(self.client, env)
 
     def close(self):
         if self.client != NULL:
diff --git a/pufferlib/ocean/tripletriad/tripletriad.h b/pufferlib/ocean/tripletriad/tripletriad.h
index 2fdc5fb8..48e0894f 100644
--- a/pufferlib/ocean/tripletriad/tripletriad.h
+++ b/pufferlib/ocean/tripletriad/tripletriad.h
@@ -294,7 +294,7 @@ void compute_observations(CTripleTriad* env) {
     }
 }
 
-void reset(CTripleTriad* env) {
+void c_reset(CTripleTriad* env) {
     env->log = (Log){0};
     env->game_over = 0;
     for(int i=0; i< 2; i++) {
@@ -490,7 +490,7 @@ void check_card_conversions(CTripleTriad* env, int card_placement, int player) {
     }
 }
 
-void step(CTripleTriad* env) {
+void c_step(CTripleTriad* env) {
     env->log.episode_length += 1;
     env->rewards[0] = 0.0;
     int action = env->actions[0];
@@ -499,7 +499,7 @@ void step(CTripleTriad* env) {
         env->log.score = env->score[0];
         add_log(env->log_buffer, &env->log);
         //printf("Log: %f, %f, %f\n", env->log.episode_return, env->log.episode_length, env->log.score);
-        reset(env);
+        c_reset(env);
         return;
     }
     // select a card if the card is in the range of 1-5 and the card is not placed
@@ -575,7 +575,7 @@ Client* make_client(int width, int height) {
     return client;
 }
 
-void render(Client* client, CTripleTriad* env) {
+void c_render(Client* client, CTripleTriad* env) {
     if (IsKeyDown(KEY_ESCAPE)) {
         exit(0);
     }
diff --git a/setup.py b/setup.py
index 1094c7a8..dd9bc793 100644
--- a/setup.py
+++ b/setup.py
@@ -278,11 +278,23 @@
     library_dirs=['raylib/lib'],
     libraries=["raylib"],
     runtime_library_dirs=["raylib/lib"],
-    extra_compile_args=['-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION', '-DPLATFORM_DESKTOP', '-O2', '-Wno-alloc-size-larger-than'],#, '-g'],
-    extra_link_args=[rpath_arg]
-
+    extra_compile_args=['-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION', '-DPLATFORM_DESKTOP', '-O2', '-Wno-alloc-size-larger-than', '-fwrapv'],#, '-g'],
+    extra_link_args=[rpath_arg, '-Bsymbolic-functions', '-O2', '--enable-new-dtags', '-fwrapv']
 ) for path in extension_paths]
- 
+
+# Prevent Conda from injecting garbage compile flags
+from distutils.sysconfig import get_config_vars
+cfg_vars = get_config_vars()
+for key in ('CC', 'CXX', 'LDSHARED'):
+    if cfg_vars[key]:
+        cfg_vars[key] = cfg_vars[key].replace('-B /root/anaconda3/compiler_compat', '')
+        cfg_vars[key] = cfg_vars[key].replace('-pthread', '')
+        cfg_vars[key] = cfg_vars[key].replace('-fno-strict-overflow', '')
+
+for key, value in cfg_vars.items():
+    if value and '-fno-strict-overflow' in str(value):
+        cfg_vars[key] = value.replace('-fno-strict-overflow', '')
+
 setup(
     name="pufferlib",
     description="PufferAI Library"
@@ -295,7 +307,7 @@
     },
     include_package_data=True,
     install_requires=[
-        'numpy==1.23.3',
+        'numpy>=1.23.3',
         'opencv-python==3.4.17.63',
         'cython>=3.0.0',
         'rich',

From 25ccb9305f2b3f1ecd9d580a9eb7787580c2e7f7 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Wed, 15 Jan 2025 14:14:12 +0000
Subject: [PATCH 09/21] Link path fixes for package build

---
 MANIFEST.in          |  5 +++++
 pufferlib/version.py |  2 +-
 setup.py             | 21 +++++++++++++--------
 3 files changed, 19 insertions(+), 9 deletions(-)

diff --git a/MANIFEST.in b/MANIFEST.in
index 451161a7..a9a48075 100644
--- a/MANIFEST.in
+++ b/MANIFEST.in
@@ -3,4 +3,9 @@ global-include *.pxd
 global-include *.h
 global-include *.py
 recursive-include pufferlib/resources *
+recursive-exclude experiments *
+recursive-exclude wandb *
+recursive-exclude tests *
+recursive-exclude raylib *
+recursive-exclude raylib_wasm *
 
diff --git a/pufferlib/version.py b/pufferlib/version.py
index e7c12d28..13ce17d8 100644
--- a/pufferlib/version.py
+++ b/pufferlib/version.py
@@ -1 +1 @@
-__version__ = '2.0.3'
+__version__ = '2.0.6'
diff --git a/setup.py b/setup.py
index dd9bc793..a99edfd2 100644
--- a/setup.py
+++ b/setup.py
@@ -1,4 +1,4 @@
-from setuptools import find_packages, setup, Extension
+from setuptools import find_packages, find_namespace_packages, setup, Extension
 from Cython.Build import cythonize
 import numpy
 import os
@@ -9,7 +9,7 @@
 	
 #  python3 setup.py built_ext --inplace
 
-VERSION = '2.0.3'
+VERSION = '2.0.6'
 
 RAYLIB_BASE = 'https://github.com/raysan5/raylib/releases/download/5.0/'
 RAYLIB_NAME = 'raylib-5.0_macos' if platform.system() == "Darwin" else 'raylib-5.0_linux_amd64'
@@ -266,8 +266,10 @@
     # So @loader_path/../../raylib/lib is common.
     rpath_arg = '-Wl,-rpath,@loader_path/../../raylib/lib'
 elif system == 'Linux':
-    # On Linux, $ORIGIN works
-    rpath_arg = '-Wl,-rpath,$ORIGIN/raylib/lib'
+    # TODO: Check if anything moves packages around after they are installed.
+    # That would break this linking. Rel path doesn't work outside the pufferlib dir
+    raylib_dir = os.path.abspath("raylib/lib")
+    rpath_arg = f"-Wl,-rpath,{raylib_dir}"
 else:
     raise ValueError(f'Unsupported system: {system}')
 
@@ -277,7 +279,7 @@
     include_dirs=[numpy.get_include(), 'raylib/include'],
     library_dirs=['raylib/lib'],
     libraries=["raylib"],
-    runtime_library_dirs=["raylib/lib"],
+    runtime_library_dirs=['raylib/lib'],
     extra_compile_args=['-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION', '-DPLATFORM_DESKTOP', '-O2', '-Wno-alloc-size-larger-than', '-fwrapv'],#, '-g'],
     extra_link_args=[rpath_arg, '-Bsymbolic-functions', '-O2', '--enable-new-dtags', '-fwrapv']
 ) for path in extension_paths]
@@ -301,9 +303,12 @@
     "PufferAI's library of RL tools and utilities",
     long_description_content_type="text/markdown",
     version=VERSION,
-    packages=find_packages(),
+    packages=find_namespace_packages() + find_packages(),
     package_data={
-        "pufferlib": ["raylib/lib/libraylib.so.500", "raylib/lib/libraylib.so"]
+        "pufferlib": [
+            "raylib/lib/libraylib.so.500",
+            "raylib/lib/libraylib.so"
+        ]
     },
     include_package_data=True,
     install_requires=[
@@ -347,7 +352,7 @@
        #annotate=True,
        #compiler_directives={'profile': True},# annotate=True
     ),
-    include_dirs=[numpy.get_include(), 'raylib-5.0_linux_amd64/include'],
+    include_dirs=[numpy.get_include(), 'raylib/include'],
     python_requires=">=3.9",
     license="MIT",
     author="Joseph Suarez",

From dcff24311459fe55a809d0e063100b37fdbbdb9d Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Wed, 15 Jan 2025 19:06:30 +0000
Subject: [PATCH 10/21] Switch to static linking to avoid much pain and
 suffering

---
 MANIFEST.in |  6 +++---
 setup.py    | 60 +++++++++++++++++++++++++++--------------------------
 2 files changed, 34 insertions(+), 32 deletions(-)

diff --git a/MANIFEST.in b/MANIFEST.in
index a9a48075..5d7e29df 100644
--- a/MANIFEST.in
+++ b/MANIFEST.in
@@ -6,6 +6,6 @@ recursive-include pufferlib/resources *
 recursive-exclude experiments *
 recursive-exclude wandb *
 recursive-exclude tests *
-recursive-exclude raylib *
-recursive-exclude raylib_wasm *
-
+include raylib-5.0_linux_amd64/lib/libraylib.a
+include raylib-5.0_macos/lib/libraylib.a
+recursive-exclude raylib-5.0_webassembly * 
diff --git a/setup.py b/setup.py
index a99edfd2..b9de6e8e 100644
--- a/setup.py
+++ b/setup.py
@@ -12,31 +12,35 @@
 VERSION = '2.0.6'
 
 RAYLIB_BASE = 'https://github.com/raysan5/raylib/releases/download/5.0/'
+
 RAYLIB_NAME = 'raylib-5.0_macos' if platform.system() == "Darwin" else 'raylib-5.0_linux_amd64'
-RAYLIB_WASM_URL = RAYLIB_BASE + 'raylib-5.0_webassembly.zip'
-RAYLIB_URL = RAYLIB_BASE + RAYLIB_NAME + '.tar.gz'
 
-if not os.path.exists('raylib'):
-    print("Raylib not found, downloading...")
-    urllib.request.urlretrieve(RAYLIB_URL, 'raylib.tar.gz')
-    with tarfile.open('raylib.tar.gz', 'r') as tar_ref:
+RAYLIB_LINUX = 'raylib-5.0_linux_amd64'
+RAYLIB_LINUX_URL = RAYLIB_BASE + RAYLIB_LINUX + '.tar.gz'
+if not os.path.exists(RAYLIB_LINUX):
+    urllib.request.urlretrieve(RAYLIB_LINUX_URL, RAYLIB_LINUX + '.tar.gz')
+    with tarfile.open(RAYLIB_LINUX + '.tar.gz', 'r') as tar_ref:
         tar_ref.extractall()
-        os.rename(RAYLIB_NAME, 'raylib')
 
-    os.remove('raylib.tar.gz')
+    os.remove(RAYLIB_LINUX + '.tar.gz')
 
-if not os.path.exists('raylib_wasm'):
-    print("Raylib WASM not found, downloading...")
-    urllib.request.urlretrieve(RAYLIB_WASM_URL, 'raylib.zip')
-    with zipfile.ZipFile('raylib.zip', 'r') as zip_ref:
-        zip_ref.extractall()
-        os.rename('raylib-5.0_webassembly', 'raylib_wasm')
+RAYLIB_MACOS = 'raylib-5.0_macos'
+RAYLIB_MACOS_URL = RAYLIB_BASE + RAYLIB_MACOS + '.tar.gz'
+if not os.path.exists(RAYLIB_MACOS):
+    urllib.request.urlretrieve(RAYLIB_MACOS_URL, RAYLIB_MACOS + '.tar.gz')
+    with tarfile.open(RAYLIB_MACOS + '.tar.gz', 'r') as tar_ref:
+        tar_ref.extractall()
 
-    os.remove('raylib.zip')
-    
-#import os
-#os.environ['CFLAGS'] = '-O3 -march=native -Wall'
+    os.remove(RAYLIB_MACOS + '.tar.gz')
+
+RAYLIB_WASM = 'raylib-5.0_webassembly'
+RAYLIB_WASM_URL = RAYLIB_BASE + RAYLIB_WASM + '.zip'
+if not os.path.exists(RAYLIB_WASM):
+    urllib.request.urlretrieve(RAYLIB_WASM_URL, RAYLIB_WASM + '.zip')
+    with zipfile.ZipFile(RAYLIB_WASM + '.zip', 'r') as zip_ref:
+        zip_ref.extractall()
 
+    os.remove(RAYLIB_WASM + '.zip')
 
 # Default Gym/Gymnasium/PettingZoo versions
 # Gym:
@@ -264,24 +268,23 @@
     # The extension “.so” is typically in pufferlib/ocean/...,
     # and “raylib/lib” is (maybe) two directories up from ocean/<env>.
     # So @loader_path/../../raylib/lib is common.
-    rpath_arg = '-Wl,-rpath,@loader_path/../../raylib/lib'
+    RAYLIB_INCLUDE = f'{RAYLIB_MACOS}/include'
+    RAYLIB_LIB = f'{RAYLIB_MACOS}/lib'
 elif system == 'Linux':
     # TODO: Check if anything moves packages around after they are installed.
     # That would break this linking. Rel path doesn't work outside the pufferlib dir
-    raylib_dir = os.path.abspath("raylib/lib")
-    rpath_arg = f"-Wl,-rpath,{raylib_dir}"
+    RAYLIB_INCLUDE = f'{RAYLIB_LINUX}/include'
+    RAYLIB_LIB = f'{RAYLIB_LINUX}/lib'
 else:
     raise ValueError(f'Unsupported system: {system}')
 
 extensions = [Extension(
     path.replace('/', '.'),
     [path + '.pyx'],
-    include_dirs=[numpy.get_include(), 'raylib/include'],
-    library_dirs=['raylib/lib'],
-    libraries=["raylib"],
-    runtime_library_dirs=['raylib/lib'],
+    include_dirs=[numpy.get_include(), RAYLIB_INCLUDE],
     extra_compile_args=['-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION', '-DPLATFORM_DESKTOP', '-O2', '-Wno-alloc-size-larger-than', '-fwrapv'],#, '-g'],
-    extra_link_args=[rpath_arg, '-Bsymbolic-functions', '-O2', '--enable-new-dtags', '-fwrapv']
+    extra_link_args=['-Bsymbolic-functions', '-O2', '--enable-new-dtags', '-fwrapv'],
+    extra_objects=[f'{RAYLIB_LIB}/libraylib.a']
 ) for path in extension_paths]
 
 # Prevent Conda from injecting garbage compile flags
@@ -306,8 +309,7 @@
     packages=find_namespace_packages() + find_packages(),
     package_data={
         "pufferlib": [
-            "raylib/lib/libraylib.so.500",
-            "raylib/lib/libraylib.so"
+            f'{RAYLIB_LIB}/libraylib.a',
         ]
     },
     include_package_data=True,
@@ -352,7 +354,7 @@
        #annotate=True,
        #compiler_directives={'profile': True},# annotate=True
     ),
-    include_dirs=[numpy.get_include(), 'raylib/include'],
+    include_dirs=[numpy.get_include(), RAYLIB_INCLUDE],
     python_requires=">=3.9",
     license="MIT",
     author="Joseph Suarez",

From ec53ba040571a95b2df0169935c87b801ebf7f09 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Wed, 15 Jan 2025 19:11:48 +0000
Subject: [PATCH 11/21] Remove arg for mac

---
 setup.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/setup.py b/setup.py
index b9de6e8e..537a45c1 100644
--- a/setup.py
+++ b/setup.py
@@ -283,7 +283,7 @@
     [path + '.pyx'],
     include_dirs=[numpy.get_include(), RAYLIB_INCLUDE],
     extra_compile_args=['-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION', '-DPLATFORM_DESKTOP', '-O2', '-Wno-alloc-size-larger-than', '-fwrapv'],#, '-g'],
-    extra_link_args=['-Bsymbolic-functions', '-O2', '--enable-new-dtags', '-fwrapv'],
+    extra_link_args=['-Bsymbolic-functions', '-O2', '-fwrapv'],
     extra_objects=[f'{RAYLIB_LIB}/libraylib.a']
 ) for path in extension_paths]
 

From 770e994290d78b739b11d43851021260f85e8382 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Wed, 15 Jan 2025 19:16:47 +0000
Subject: [PATCH 12/21] Add conda to ci

---
 .github/workflows/install.yml | 27 +++++++++++++++++++++++----
 1 file changed, 23 insertions(+), 4 deletions(-)

diff --git a/.github/workflows/install.yml b/.github/workflows/install.yml
index 873d3c5c..1f13317a 100644
--- a/.github/workflows/install.yml
+++ b/.github/workflows/install.yml
@@ -1,11 +1,12 @@
 name: install
+
 on:
   push:
   pull_request:
 
 jobs:
   test:
-    name: test ${{ matrix.py }} - ${{ matrix.os }}
+    name: test ${{ matrix.py }} - ${{ matrix.os }} - ${{ matrix.env }}
     runs-on: ${{ matrix.os }}
     strategy:
       fail-fast: false
@@ -17,13 +18,31 @@ jobs:
           - "3.11"
           - "3.10"
           - "3.9"
+        env:
+          - pip
+          - conda
     steps:
-      - name: Setup python for test ${{ matrix.py }}
+      - name: Checkout code
+        uses: actions/checkout@v3
+
+      - name: Setup Conda or pip
+        if: matrix.env == 'conda'
+        uses: conda-incubator/setup-miniconda@v2
+        with:
+          python-version: ${{ matrix.py }}
+          activate-environment: test-env
+          environment-file: environment.yml
+          auto-update-conda: true
+
+      - name: Setup Python for pip
+        if: matrix.env == 'pip'
         uses: actions/setup-python@v4
         with:
           python-version: ${{ matrix.py }}
-      - uses: actions/checkout@v3
+
       - name: Upgrade pip
+        if: matrix.env == 'pip'
         run: python -m pip install -U pip
+
       - name: Install pufferlib
-        run: pip3 install -e . 
+        run: pip install -e .

From e881766ca0c03f5dcd07de1ebe5e9c74b8035461 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <sealsuarez@gmail.com>
Date: Wed, 15 Jan 2025 14:17:45 -0500
Subject: [PATCH 13/21] Add conda to ci

---
 .github/workflows/install.yml | 27 +++++++++++++++++++++++----
 1 file changed, 23 insertions(+), 4 deletions(-)

diff --git a/.github/workflows/install.yml b/.github/workflows/install.yml
index 873d3c5c..1f13317a 100644
--- a/.github/workflows/install.yml
+++ b/.github/workflows/install.yml
@@ -1,11 +1,12 @@
 name: install
+
 on:
   push:
   pull_request:
 
 jobs:
   test:
-    name: test ${{ matrix.py }} - ${{ matrix.os }}
+    name: test ${{ matrix.py }} - ${{ matrix.os }} - ${{ matrix.env }}
     runs-on: ${{ matrix.os }}
     strategy:
       fail-fast: false
@@ -17,13 +18,31 @@ jobs:
           - "3.11"
           - "3.10"
           - "3.9"
+        env:
+          - pip
+          - conda
     steps:
-      - name: Setup python for test ${{ matrix.py }}
+      - name: Checkout code
+        uses: actions/checkout@v3
+
+      - name: Setup Conda or pip
+        if: matrix.env == 'conda'
+        uses: conda-incubator/setup-miniconda@v2
+        with:
+          python-version: ${{ matrix.py }}
+          activate-environment: test-env
+          environment-file: environment.yml
+          auto-update-conda: true
+
+      - name: Setup Python for pip
+        if: matrix.env == 'pip'
         uses: actions/setup-python@v4
         with:
           python-version: ${{ matrix.py }}
-      - uses: actions/checkout@v3
+
       - name: Upgrade pip
+        if: matrix.env == 'pip'
         run: python -m pip install -U pip
+
       - name: Install pufferlib
-        run: pip3 install -e . 
+        run: pip install -e .

From f63ed9b0f89cc49b29769768c0810e800776c81f Mon Sep 17 00:00:00 2001
From: Joseph Suarez <sealsuarez@gmail.com>
Date: Wed, 15 Jan 2025 14:19:56 -0500
Subject: [PATCH 14/21] Fix conda ci

---
 .github/workflows/install.yml | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/.github/workflows/install.yml b/.github/workflows/install.yml
index 1f13317a..9e8f15fd 100644
--- a/.github/workflows/install.yml
+++ b/.github/workflows/install.yml
@@ -25,13 +25,12 @@ jobs:
       - name: Checkout code
         uses: actions/checkout@v3
 
-      - name: Setup Conda or pip
+      - name: Setup Conda
         if: matrix.env == 'conda'
         uses: conda-incubator/setup-miniconda@v2
         with:
           python-version: ${{ matrix.py }}
           activate-environment: test-env
-          environment-file: environment.yml
           auto-update-conda: true
 
       - name: Setup Python for pip
@@ -41,7 +40,6 @@ jobs:
           python-version: ${{ matrix.py }}
 
       - name: Upgrade pip
-        if: matrix.env == 'pip'
         run: python -m pip install -U pip
 
       - name: Install pufferlib

From e18bc65cffe883e50d7e54e88b5a2d0981b20728 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <sealsuarez@gmail.com>
Date: Wed, 15 Jan 2025 14:22:35 -0500
Subject: [PATCH 15/21] Fix miniconda version

---
 .github/workflows/install.yml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.github/workflows/install.yml b/.github/workflows/install.yml
index 9e8f15fd..8541dbee 100644
--- a/.github/workflows/install.yml
+++ b/.github/workflows/install.yml
@@ -30,6 +30,7 @@ jobs:
         uses: conda-incubator/setup-miniconda@v2
         with:
           python-version: ${{ matrix.py }}
+          miniconda-version: "latest"
           activate-environment: test-env
           auto-update-conda: true
 

From 076696a5a70527e6fb974b807d6c5d84ffe9bd59 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Tue, 4 Feb 2025 21:06:02 +0000
Subject: [PATCH 16/21] nmmo3 obs space

---
 pufferlib/ocean/nmmo3/nmmo3.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/pufferlib/ocean/nmmo3/nmmo3.py b/pufferlib/ocean/nmmo3/nmmo3.py
index 3648e4df..93a3bcfd 100644
--- a/pufferlib/ocean/nmmo3/nmmo3.py
+++ b/pufferlib/ocean/nmmo3/nmmo3.py
@@ -138,8 +138,8 @@ def __init__(self, width=4*[512], height=4*[512], num_envs=4,
         self.prof_goal_mask = np.array([0, 0, 0, 1, 0, 0, 1, 1, 1, 1])
         self.tick = 0
 
-        self.single_observation_space = gymnasium.spaces.Box(low=-1,
-            high=2**32-1, shape=(11*15*10+47+10,), dtype=np.uint8)
+        self.single_observation_space = gymnasium.spaces.Box(low=0,
+            high=255, shape=(11*15*10+47+10,), dtype=np.uint8)
         self.single_action_space = gymnasium.spaces.Discrete(26)
         self.render_mode = 'human'
 

From 0cb9a319ad527450ebf98392714e663528e915d3 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Tue, 4 Feb 2025 21:19:50 +0000
Subject: [PATCH 17/21] Update nmmo c to c_step, c_reset

---
 pufferlib/ocean/nmmo3/nmmo3.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/pufferlib/ocean/nmmo3/nmmo3.c b/pufferlib/ocean/nmmo3/nmmo3.c
index 10057231..a7a5ef78 100644
--- a/pufferlib/ocean/nmmo3/nmmo3.c
+++ b/pufferlib/ocean/nmmo3/nmmo3.c
@@ -165,7 +165,7 @@ void demo(int num_players) {
     };
     allocate_mmo(&env);
 
-    reset(&env, 42);
+    c_reset(&env, 42);
 
     // Must reset before making client
     Client* client = make_client(&env);
@@ -183,7 +183,7 @@ void demo(int num_players) {
                 env.actions[0] = human_action;
             }
 
-            step(&env);
+            c_step(&env);
             //printf("Reward: %f\n\tDeath: %f\n\tProf: %f\n\tComb: %f\n\tItem: %f\n", env.rewards[0].death, env.rewards[0].death, env.rewards[0].prof_lvl, env.rewards[0].comb_lvl, env.rewards[0].item_atk_lvl);
             human_action = ATN_NOOP;
         } else {
@@ -222,13 +222,13 @@ void test_mmonet_performance(int num_players, int timeout) {
         .y_window = 5,
     };
     allocate_mmo(&env);
-    reset(&env, 42);
+    c_reset(&env, 42);
 
     int start = time(NULL);
     int num_steps = 0;
     while (time(NULL) - start < timeout) {
         forward(net, env.obs, env.actions);
-        step(&env);
+        c_step(&env);
         num_steps++;
     }
 
@@ -434,7 +434,7 @@ void test_performance(int num_players, int timeout) {
         .y_window = 5,
     };
     allocate_mmo(&env);
-    reset(&env, 0);
+    c_reset(&env, 0);
 
     int start = time(NULL);
     int num_steps = 0;
@@ -442,7 +442,7 @@ void test_performance(int num_players, int timeout) {
         for (int i = 0; i < num_players; i++) {
             env.actions[i] = rand() % 23;
         }
-        step(&env);
+        c_step(&env);
         num_steps++;
     }
 

From bda11c6f0763472393f4532054df8d64d09e479e Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Tue, 4 Feb 2025 21:27:56 +0000
Subject: [PATCH 18/21] Fix raylib for linux for now

---
 scripts/build_ocean.sh | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/scripts/build_ocean.sh b/scripts/build_ocean.sh
index 563ca529..b65625ca 100755
--- a/scripts/build_ocean.sh
+++ b/scripts/build_ocean.sh
@@ -42,10 +42,10 @@ fi
 
 FLAGS=(
     -Wall
-    -I./raylib/include 
+    -I./raylib-5.0_linux_amd64/include 
     -I./pufferlib
     "$SRC_DIR/$ENV.c" -o "$ENV"
-    ./raylib/lib/libraylib.a
+    ./raylib-5.0_linux_amd64/lib/libraylib.a
     -lm
     -lpthread
     -DPLATFORM_DESKTOP

From 542236e5aecc0a4c64401d310aae5ec33a5af94a Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Wed, 5 Feb 2025 17:37:16 +0000
Subject: [PATCH 19/21] Fix build from previous conda support changes

---
 pufferlib/ocean/moba/moba.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/pufferlib/ocean/moba/moba.c b/pufferlib/ocean/moba/moba.c
index 81a95798..e0a0c519 100644
--- a/pufferlib/ocean/moba/moba.c
+++ b/pufferlib/ocean/moba/moba.c
@@ -125,11 +125,11 @@ void demo() {
 
     GameRenderer* renderer = init_game_renderer(32, 41, 23);
 
-    reset(&env);
+    c_reset(&env);
     int frame = 0;
     while (!WindowShouldClose()) {
         if (frame % 12 == 0) {
-            step(&env);
+            c_step(&env);
             forward(net, env.observations, env.actions);
         }
         render_game(renderer, &env, frame);
@@ -157,7 +157,7 @@ void test_performance(float test_time) {
     };
     allocate_moba(&env);
 
-    reset(&env);
+    c_reset(&env);
     int start = time(NULL);
     int i = 0;
     while (time(NULL) - start < test_time) {
@@ -169,7 +169,7 @@ void test_performance(float test_time) {
             env.actions[6*j + 4] = rand()%2;
             env.actions[6*j + 5] = rand()%2;
         }
-        step(&env);
+        c_step(&env);
         i++;
     }
     int end = time(NULL);
@@ -192,11 +192,11 @@ void test_bugs(float test_time) {
     };
     allocate_moba(&env);
 
-    reset(&env);
+    c_reset(&env);
     int start = time(NULL);
     int i = 0;
     while (time(NULL) - start < test_time) {
-        step(&env);
+        c_step(&env);
         forward(net, env.observations, env.actions);
         i++;
     }

From 33f4ee969dcd445c18a806d7cb64c2079e7689f1 Mon Sep 17 00:00:00 2001
From: Joseph Suarez <jsuarez@puffer.ai>
Date: Wed, 5 Feb 2025 17:48:19 +0000
Subject: [PATCH 20/21] Fix c builds after conda support

---
 pufferlib/ocean/breakout/breakout.c         | 10 +++++-----
 pufferlib/ocean/connect4/connect4.c         | 10 +++++-----
 pufferlib/ocean/enduro/enduro.c             |  4 ++--
 pufferlib/ocean/go/go.c                     | 10 +++++-----
 pufferlib/ocean/pong/pong.c                 |  6 +++---
 pufferlib/ocean/rware/rware.c               | 10 +++++-----
 pufferlib/ocean/snake/snake.c               | 10 +++++-----
 pufferlib/ocean/squared/squared.c           |  6 +++---
 pufferlib/ocean/trash_pickup/trash_pickup.c | 12 ++++++------
 pufferlib/ocean/tripletriad/tripletriad.c   |  6 +++---
 10 files changed, 42 insertions(+), 42 deletions(-)

diff --git a/pufferlib/ocean/breakout/breakout.c b/pufferlib/ocean/breakout/breakout.c
index 9203ddee..e2897d14 100644
--- a/pufferlib/ocean/breakout/breakout.c
+++ b/pufferlib/ocean/breakout/breakout.c
@@ -20,7 +20,7 @@ void demo() {
         .brick_cols = 18,
     };
     allocate(&env);
-    reset(&env);
+    c_reset(&env);
  
     Client* client = make_client(&env);
 
@@ -35,8 +35,8 @@ void demo() {
             forward_linearlstm(net, env.observations, env.actions);
         }
 
-        step(&env);
-        render(client, &env);
+        c_step(&env);
+        c_render(client, &env);
     }
     free_linearlstm(net);
     free(weights);
@@ -60,13 +60,13 @@ void performance_test() {
         .brick_cols = 18,
     };
     allocate(&env);
-    reset(&env);
+    c_reset(&env);
 
     long start = time(NULL);
     int i = 0;
     while (time(NULL) - start < test_time) {
         env.actions[0] = rand() % 4;
-        step(&env);
+        c_step(&env);
         i++;
     }
     long end = time(NULL);
diff --git a/pufferlib/ocean/connect4/connect4.c b/pufferlib/ocean/connect4/connect4.c
index a5d7ddc2..fa08b7de 100644
--- a/pufferlib/ocean/connect4/connect4.c
+++ b/pufferlib/ocean/connect4/connect4.c
@@ -15,7 +15,7 @@ void interactive() {
         .piece_height = 96,
     };
     allocate_cconnect4(&env);
-    reset(&env);
+    c_reset(&env);
  
     Client* client = make_client(env.width, env.height);
     float observations[42] = {0};
@@ -43,10 +43,10 @@ void interactive() {
 
         tick = (tick + 1) % 60;
         if (env.actions[0] >= 0 && env.actions[0] <= 6) {
-            step(&env);
+            c_step(&env);
         }
 
-        render(client, &env);
+        c_render(client, &env);
     }
     free_linearlstm(net);
     free(weights);
@@ -63,13 +63,13 @@ void performance_test() {
         .piece_height = 96,
     };
     allocate_cconnect4(&env);
-    reset(&env);
+    c_reset(&env);
  
     long start = time(NULL);
     int i = 0;
     while (time(NULL) - start < test_time) {
         env.actions[0] = rand() % 7;
-        step(&env);
+        c_step(&env);
         i++;
     }
     long end = time(NULL);
diff --git a/pufferlib/ocean/enduro/enduro.c b/pufferlib/ocean/enduro/enduro.c
index fbdb3e74..1bef4a4e 100644
--- a/pufferlib/ocean/enduro/enduro.c
+++ b/pufferlib/ocean/enduro/enduro.c
@@ -47,7 +47,7 @@ int demo() {
 
     unsigned int seed = 0;
     init(&env, seed, 0);
-    reset(&env);
+    c_reset(&env);
 
     while (!WindowShouldClose()) {
         if (IsKeyDown(KEY_LEFT_SHIFT)) {
@@ -78,7 +78,7 @@ void perftest(float test_time) {
 
     unsigned int seed = 12345;
     init(&env, seed, 0);
-    reset(&env);
+    c_reset(&env);
 
     int start = time(NULL);
     int i = 0;
diff --git a/pufferlib/ocean/go/go.c b/pufferlib/ocean/go/go.c
index 22e9bed3..371af40d 100644
--- a/pufferlib/ocean/go/go.c
+++ b/pufferlib/ocean/go/go.c
@@ -133,7 +133,7 @@ void demo(int grid_size) {
     Weights* weights = load_weights("resources/go_weights.bin", 254867);
     GoNet* net = init_gonet(weights, 1, grid_size);
     allocate(&env);
-    reset(&env);
+    c_reset(&env);
  
     Client* client = make_client(env.width, env.height);
     int tick = 0;
@@ -147,7 +147,7 @@ void demo(int grid_size) {
             if (IsKeyDown(KEY_LEFT_SHIFT)) {
                 env.actions[0] = human_action;
             }
-            step(&env);
+            c_step(&env);
             if (IsKeyDown(KEY_LEFT_SHIFT)) {
                 env.actions[0] = -1;
             }
@@ -187,7 +187,7 @@ void demo(int grid_size) {
                 }
             }
         }
-        render(client, &env);
+        c_render(client, &env);
     }
     close_client(client);
     free_allocated(&env);
@@ -209,13 +209,13 @@ void performance_test() {
 	.reward_move_valid = 0.1
     };
     allocate(&env);
-    reset(&env);
+    c_reset(&env);
 
     long start = time(NULL);
     int i = 0;
     while (time(NULL) - start < test_time) {
         env.actions[0] = rand() % (env.grid_size)*(env.grid_size);
-        step(&env);
+        c_step(&env);
         i++;
     }
     long end = time(NULL);
diff --git a/pufferlib/ocean/pong/pong.c b/pufferlib/ocean/pong/pong.c
index 43f57d2e..9a1fd947 100644
--- a/pufferlib/ocean/pong/pong.c
+++ b/pufferlib/ocean/pong/pong.c
@@ -26,7 +26,7 @@ int main() {
 
     Client* client = make_client(&env);
 
-    reset(&env);
+    c_reset(&env);
     while (!WindowShouldClose()) {
         // User can take control of the paddle
         if (IsKeyDown(KEY_LEFT_SHIFT)) {
@@ -37,8 +37,8 @@ int main() {
             forward_linearlstm(net, env.observations, env.actions);
         }
 
-        step(&env);
-        render(client, &env);
+        c_step(&env);
+        c_render(client, &env);
     }
     free_linearlstm(net);
     free(weights);
diff --git a/pufferlib/ocean/rware/rware.c b/pufferlib/ocean/rware/rware.c
index 77a030bc..d24d63ca 100644
--- a/pufferlib/ocean/rware/rware.c
+++ b/pufferlib/ocean/rware/rware.c
@@ -36,7 +36,7 @@ void demo(int map_choice) {
     LinearLSTM* net = make_linearlstm(weights, env.num_agents, 27, 5);
 
     allocate(&env);
-    reset(&env);
+    c_reset(&env);
     Client* client = make_client(&env);
 
     int tick = 0;
@@ -54,7 +54,7 @@ void demo(int map_choice) {
                 env.actions[env.human_agent_idx] = human_action;
             }
 
-            step(&env);
+            c_step(&env);
 
             if (IsKeyDown(KEY_LEFT_SHIFT)) {
                 env.actions[env.human_agent_idx] = NOOP;
@@ -82,7 +82,7 @@ void demo(int map_choice) {
             }
         }
 
-        render(client,&env);
+        c_render(client,&env);
     }
     close_client(client);
     free_allocated(&env);
@@ -99,13 +99,13 @@ void performance_test() {
 	.reward_type = 2
     };
     allocate(&env);
-    reset(&env);
+    c_reset(&env);
 
     long start = time(NULL);
     int i = 0;
     while (time(NULL) - start < test_time) {
         env.actions[0] = rand() % 5;
-        step(&env);
+        c_step(&env);
         i++;
     }
     long end = time(NULL);
diff --git a/pufferlib/ocean/snake/snake.c b/pufferlib/ocean/snake/snake.c
index 4b4eaebb..76b791dc 100644
--- a/pufferlib/ocean/snake/snake.c
+++ b/pufferlib/ocean/snake/snake.c
@@ -16,7 +16,7 @@ int demo() {
         .reward_death = -1.0f,
     };
     allocate_csnake(&env);
-    reset(&env);
+    c_reset(&env);
 
     Weights* weights = load_weights("resources/snake_weights.bin", 148357);
     LinearLSTM* net = make_linearlstm(weights, env.num_snakes, env.obs_size, 4);
@@ -35,8 +35,8 @@ int demo() {
             }
             forward_linearlstm(net, net->obs, env.actions);
         }
-        step(&env);
-        render(client, &env);
+        c_step(&env);
+        c_render(client, &env);
     }
     free_linearlstm(net);
     free(weights);
@@ -59,7 +59,7 @@ void test_performance(float test_time) {
         .reward_death = -1.0f,
     };
     allocate_csnake(&env);
-    reset(&env);
+    c_reset(&env);
 
     int start = time(NULL);
     int i = 0;
@@ -67,7 +67,7 @@ void test_performance(float test_time) {
         for (int j = 0; j < env.num_snakes; j++) {
             env.actions[j] = rand()%4;
         }
-        step(&env);
+        c_step(&env);
         i++;
     }
     int end = time(NULL);
diff --git a/pufferlib/ocean/squared/squared.c b/pufferlib/ocean/squared/squared.c
index 34332090..98a46eeb 100644
--- a/pufferlib/ocean/squared/squared.c
+++ b/pufferlib/ocean/squared/squared.c
@@ -10,7 +10,7 @@ int main() {
 
     Client* client = make_client(&env);
 
-    reset(&env);
+    c_reset(&env);
     while (!WindowShouldClose()) {
         if (IsKeyDown(KEY_LEFT_SHIFT)) {
             env.actions[0] = 0;
@@ -22,8 +22,8 @@ int main() {
             env.actions[0] = NOOP;
             //forward_linearlstm(net, env.observations, env.actions);
         }
-        step(&env);
-        render(client, &env);
+        c_step(&env);
+        c_render(client, &env);
     }
     //free_linearlstm(net);
     //free(weights);
diff --git a/pufferlib/ocean/trash_pickup/trash_pickup.c b/pufferlib/ocean/trash_pickup/trash_pickup.c
index b8021719..f5aa295b 100644
--- a/pufferlib/ocean/trash_pickup/trash_pickup.c
+++ b/pufferlib/ocean/trash_pickup/trash_pickup.c
@@ -14,7 +14,7 @@ void demo(int grid_size, int num_agents, int num_trash, int num_bins, int max_st
         .do_human_control = true
     };
 
-    bool use_pretrained_model = true;
+    bool use_pretrained_model = false;
 
     Weights* weights;
     ConvLSTM* net;
@@ -28,7 +28,7 @@ void demo(int grid_size, int num_agents, int num_trash, int num_bins, int max_st
     allocate(&env);
     Client* client = make_client(&env);
 
-    reset(&env);
+    c_reset(&env);
 
     int tick = 0;
     while (!WindowShouldClose()) {
@@ -64,12 +64,12 @@ void demo(int grid_size, int num_agents, int num_trash, int num_bins, int max_st
             }
 
             // Step the environment and render the grid
-            step(&env);
+            c_step(&env);
             
         }
         tick++;
 
-        render(client, &env);
+        c_render(client, &env);
     }
 
     free_convlstm(net);
@@ -91,7 +91,7 @@ void performance_test() {
         .agent_sight_range = 5
     };
     allocate(&env);
-    reset(&env);
+    c_reset(&env);
 
     long start = time(NULL);
     int i = 0;
@@ -100,7 +100,7 @@ void performance_test() {
         for (int e = 0; e < env.num_agents; e++) {
             env.actions[e] = rand() % 4;
         }
-        step(&env);
+        c_step(&env);
         i += inc;
     }
     long end = time(NULL);
diff --git a/pufferlib/ocean/tripletriad/tripletriad.c b/pufferlib/ocean/tripletriad/tripletriad.c
index 3dd64a5f..7bb6b350 100644
--- a/pufferlib/ocean/tripletriad/tripletriad.c
+++ b/pufferlib/ocean/tripletriad/tripletriad.c
@@ -14,7 +14,7 @@ int main() {
         .num_cards = 10,
     };
     allocate_ctripletriad(&env);
-    reset(&env); 
+    c_reset(&env); 
     Client* client = make_client(env.width, env.height);
 
     int tick = 0;
@@ -59,10 +59,10 @@ int main() {
 
         tick = (tick + 1) % 45;
         if (env.actions[0] != NOOP) {
-            step(&env);
+            c_step(&env);
         }
 
-        render(client, &env);
+        c_render(client, &env);
     }
     free_linearlstm(net);
     free(weights);

From 447acb7cd8ab54c5bf5df1d6795d5470ab9c9e98 Mon Sep 17 00:00:00 2001
From: murtazarang <frenchfrywithpepper@gmail.com>
Date: Fri, 7 Feb 2025 16:40:28 -0600
Subject: [PATCH 21/21] PZ observation space (Box) fix

Previous check was made against PZ dictionary, and not the individual agent space.
---
 pufferlib/emulation.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pufferlib/emulation.py b/pufferlib/emulation.py
index 8b8089b7..b14a6b67 100644
--- a/pufferlib/emulation.py
+++ b/pufferlib/emulation.py
@@ -256,7 +256,7 @@ def __init__(self, env=None, env_creator=None, env_args=[], buf=None, env_kwargs
         self.num_agents = len(self.possible_agents)
 
         set_buffers(self, buf)
-        if isinstance(self.env.observation_space, pufferlib.spaces.Box):
+        if isinstance(self.env_single_observation_space, pufferlib.spaces.Box):
             self.obs_struct = self.observations
         else:
             self.obs_struct = self.observations.view(self.obs_dtype)