NFSP solving problem Q-Loss

[emanueleiacca]

Hi, i'm sorry I don't know how to make a pull request, i am a noob in git :(
It has been 3 months since I'm working on RLCard for my bachelor degree thesis, rn I'm working on NFSP algorithm. I use Tensorboard to plot results and I was having problem to plot the loss, in the nfsp_agent.py code:
def feed(self, ts):
''' Feed data to inner RL agent

    Args:
        ts (list): A list of 5 elements that represent the transition.
    '''
    self._rl_agent.feed(ts)
    self.total_t += 1
    if self.total_t>0 and len(self._reservoir_buffer) >= self._min_buffer_size_to_learn and self.total_t%self._train_every == 0:
        sl_loss  = self.train_sl()
        print('\rINFO - Step {}, sl-loss: {}'.format(self.total_t, sl_loss), end='')

the function doesn't return the sl_loss, so this part should be added
return sl_loss
return None
Thank u very much for this wonderful project, I hope to help
Emanuele