if agent.episodes > 1 and not agent.episodes % interval:
winrate = sum(agent.goal) / len(agent.goal)
avgsteps = sum(agent.last100) / len(agent.last100)
wstr += ',{}'.format(winrate)
astr += ',{}'.format(avgsteps)
print('Episode {}...'.format(agent.episodes))
w.write(wstr + '\n')
a.write(astr + '\n')
agent.reset()
w.write('Lambda\n')
a.write('Lambda\n')
w.write(top_label)
a.write(top_label)
print('Testing Lambdas...')
for lambd in tenths:
agent.lambd = lambd
wstr = astr = '{},0'.format(lambd)
for i in range(episodes):
agent.episode()
if agent.episodes > 1 and not agent.episodes % interval:
winrate = sum(agent.goal) / len(agent.goal)
avgsteps = sum(agent.last100) / len(agent.last100)
wstr += ',{}'.format(winrate)
astr += ',{}'.format(avgsteps)
print('Episode {}...'.format(agent.episodes))
w.write(wstr + '\n')
a.write(astr + '\n')
agent.reset()
w.write('Alpha\n')
a.write('Alpha\n')
w.write(top_label)
