def learn(self, obs, action, reward):
act_prob = self.model(obs)
# log_prob = layers.cross_entropy(act_prob, action)
log_prob = layers.reduce_sum(
-1.0 * layers.log(act_prob) * layers.one_hot(
action, act_prob.shape[1]),
dim=1)
cost = log_prob * reward
cost = layers.reduce_mean(cost)
optimizer = fluid.optimizer.Adam(self.lr)
optimizer.minimize(cost)
return cost
def learn(self, obs, action, reward):
""" 用policy gradient 算法更新policy model
"""
act_prob = self.model(obs) # 获取输出动作概率
# log_prob = layers.cross_entropy(act_prob, action) # 交叉熵
log_prob = layers.reduce_sum(-1.0 * layers.log(act_prob) *
layers.one_hot(action, act_prob.shape[1]),
dim=1)
cost = log_prob * reward
cost = layers.reduce_mean(cost)
print('====loss', cost)
optimizer = fluid.optimizer.Adam(self.lr)
optimizer.minimize(cost)
return cost
def learn(self, obs, action, reward):
"""
:param obs: [B,4]
:param action: [B,1]
:param reward: [B,]
:return:
"""
act_prob = self.model(obs) # [B,2]
# [B, 2] -> [B, ]
log_prob = layers.reduce_sum(
-1.0 * layers.log(act_prob) *
layers.one_hot(action, depth=act_prob.shape[1]),
dim=1,
keep_dim=False)
cost = log_prob * reward
cost = layers.reduce_mean(cost)
optimizer = fluid.optimizer.Adam(self.lr)
optimizer.minimize(cost)
return cost