#reward = max(data[t] - buying_price, 0)
#gain += data[t] - buying_price
if (gain > 0):
reward = 1
else:
reward = 0
print("Sell: " + formatPrice(data[t]) + " | Profit: " +
formatPrice(data[t] - buying_price) + "| Portfolio Value: " +
formatPrice(broker.portfolio) + "| Inventory Size:" +
str(len(broker.trades_list)))
elif action == 0: # hold
print("Hold | Portfolio Value: " + formatPrice(broker.portfolio) +
"| Inventory Size:" + str(len(broker.trades_list)))
reward = 0
done = True if t == data_size - 1 else False
next_state = getState(data, t + 1, state_size + 1)
broker.memory.append((state, action, reward, next_state, done))
state = next_state
if done:
print("_________________________________")
print("Epoch {0} Profit \n:".format(str(e)) + formatPrice(gain))
if len(broker.memory) > batch_size:
broker.expReplay(batch_size)
if e % 20 == 0:
broker.model.save("models/" + stock_name + "_ep_" + str(e))
