def train()

in agents/bc_diffusion.py [0:0]

• 8 lines of code
• 2 McCabe index (conditional complexity)

    def train(self, replay_buffer, iterations, batch_size=100):

        for it in range(iterations):
            # Sample replay buffer / batch
            state, action, next_state, reward, not_done = replay_buffer.sample(batch_size)

            loss = self.actor.loss(action, state)

            self.actor_optimizer.zero_grad()
            loss.backward()
            self.actor_optimizer.step()

        # Logging
        logger.record_tabular('Diffusion BC Loss', loss.item())