#python#gym#gym_environment#reinforcement_learning#reinforcement_learning_agent#reinforcement_learning_environments#rl_environment#rl_training
NeMo Gym helps you build and run reinforcement‑learning training environments for large language models, letting you develop, test, and collect verified rollouts separately from the training loop and integrate with your preferred RL framework and model endpoints (OpenAI, vLLM, etc.). It includes ready resource servers, datasets, and patterns for multi‑step, multi‑turn, and tool‑using scenarios, runs on a typical dev machine (no GPU required), and is early-stage with evolving APIs and docs. Benefit: you can generate high‑quality, verifiable training data faster and plug it into existing training pipelines to improve model behavior.
https://github.com/NVIDIA-NeMo/Gym
🚀 Вышел стабильный JavaScript движок от Google для Android
Новая стабильная библиотека Jetpack JavaScript Engine позволит разработчикам выполнять JS код в изолированной и ограниченной среде.
class MainActivity : ComponentActivity() {
// Теперь nullable, без lateinit
private var jsSandbox: JavaScriptSandbox? = null
private var jsIsolate: JavaScriptIsolate? = null
override fun onCreate(savedInstanceState: Bundle?) {
super.onCreate(savedInstanceState)
if (!JavaScriptSandbox.isSupported()) {
Log.e("JS", "JavaScriptSandbox не поддерживается")
return
}
lifecycleScope.launch {
// Создаём и сохраняем в nullable-поле
jsSandbox = JavaScriptSandbox
.createConnectedInstanceAsync(applicationContext)
.await()
jsIsolate = jsSandbox?.createIsolate()
// При выполнении гарантируем, что jsIsolate != null
val result: String = jsIsolate
?.evaluateJavaScriptAsync(JS_SCRIPT_SCRING)
?.await()
?: "Ошибка: isolate не инициализирован"
Log.d("JS", "Результат выполнения: $result")
}
}
override fun onDestroy() {
super.onDestroy()
// Закрываем только если не null
jsIsolate?.close()
jsSandbox?.close()
}
}
#jetpack#js