Env - Flappy_puffer

config/flappy_puffer.ini

@@ -0,0 +1,19 @@
+[base]
+env_name = flappy_puffer
+
+[vec]
+total_agents = 256
+num_threads = 4
+
+[train]
+gpus = 1
+total_timesteps = 10_000_000
+learning_rate = 0.001
+minibatch_size = 1024
+horizon = 64
+
+[sweep]
+method = Protein
+metric = score
+max_runs = 1
+gpus = 0

ocean/flappy_puffer/binding.c

@@ -0,0 +1,28 @@
+#include "flappy_puffer.h"
+#include "flappy_puffer.c"
+#define Log             FPLog
+
+#define OBS_SIZE        FP_OBS_SIZE   
+#define NUM_ATNS        1            
+#define ACT_SIZES       {2}           
+#define OBS_TENSOR_T    FloatTensor   
+
+#define Env             FlappyPuffer
+#include "vecenv.h"
+
+void my_init(Env* env, Dict* kwargs) {
+    env->num_agents  = 1;  
+
+    void* seed_ptr = dict_get(kwargs, "seed");
+    env->rng = (seed_ptr != NULL)
+        ? (unsigned int)(uintptr_t)seed_ptr
+        : (unsigned int)time(NULL);
+    allocate_flappy(env);
+}
+
+void my_log(FPLog* log, Dict* out) {
+    dict_set(out, "perf",           log->perf);
+    dict_set(out, "score",          log->score);
+    dict_set(out, "episode_length", log->episode_length);
+    dict_set(out, "n_episodes",     log->n);
+}