开发者社区> 问答> 正文

错误:参数必须为密集张量:range(2,3)-形状为[1],但需要[]

我正在尝试在Python中运行一个用于使用TensorFlow进行自组织地图(SOM)的代码。我从这里获得了代码,但是当我运行它时,出现了一个错误:

错误:参数必须为密集张量:range(2,3)-形状为1,但需要[]

我认为相关的代码是:

s = SOM( (3,), 30, num_training, sess )
接```  
着:
```js
class SOM:
    def __init__(self, input_shape, map_size_n, num_expected_iterations, session):
    input_shape = tuple([i for i in input_shape if i is not None])

要么:

def initialize_graph(self):
    self.weights = tf.Variable( tf.random_uniform((self.n*self.n, )+self.input_shape, 0.0, 1.0) ) 

    self.input_placeholder = tf.placeholder(tf.float32, (None,)+self.input_shape)
    self.current_iteration = tf.placeholder(tf.float32)

    ## Compute the current iteration's neighborhood sigma and learning rate alpha:
    self.sigma_tmp = self.sigma * tf.exp( - self.current_iteration/self.timeconst_sigma  )
    self.sigma2 = 2.0*tf.multiply(self.sigma_tmp, self.sigma_tmp)

    self.alpha_tmp = self.alpha * tf.exp( - self.current_iteration/self.timeconst_alpha  )


    self.input_placeholder_ = tf.expand_dims(self.input_placeholder, 1)
    self.input_placeholder_ = tf.tile(self.input_placeholder_, (1,self.n*self.n,1) )

    self.diff = self.input_placeholder_ - self.weights
    self.diff_sq = tf.square(self.diff)
    self.diff_sum = tf.reduce_sum( self.diff_sq, axis=range(2, 2+len(self.input_shape)) )

    # Get the index of the best matching unit
    self.bmu_index = tf.argmin(self.diff_sum, 1)

    self.bmu_dist = tf.reduce_min(self.diff_sum, 1)
    self.bmu_activity = tf.exp( -self.bmu_dist/self.sigma_act )

    self.diff = tf.squeeze(self.diff)

    self.diff_2 = tf.placeholder(tf.float32, (self.n*self.n,)+self.input_shape)
    self.dist_sliced = tf.placeholder(tf.float32, (self.n*self.n,))

    self.distances = tf.exp(-self.dist_sliced / self.sigma2 )
    self.lr_times_neigh = tf.multiply( self.alpha_tmp, self.distances )
    for i in range(len(self.input_shape)):
        self.lr_times_neigh = tf.expand_dims(self.lr_times_neigh, -1)
    self.lr_times_neigh = tf.tile(self.lr_times_neigh, (1,)+self.input_shape )

    self.delta_w = self.lr_times_neigh * self.diff_2

    self.update_weights = tf.assign_add(self.weights, self.delta_w)

展开
收起
祖安文状元 2020-02-24 09:56:10 548 0
0 条回答
写回答
取消 提交回答
问答地址:
问答排行榜
最热
最新

相关电子书

更多
重新定义计算的边界 立即下载
低代码开发师(初级)实战教程 立即下载
阿里巴巴DevOps 最佳实践手册 立即下载