How to improve accuracy about finetune_alexnet_with_tensorflow HOT 1 OPEN

    # Get number of input channels
    input_channels = int(x.get_shape()[-1])
    
    # Create lambda function for the convolution
    convolve = lambda i, k: tf.nn.conv2d(i, k,
                                     strides=[1, stride_y, stride_x, 1],
                                     padding=padding)
    with tf.variable_scope(name) as scope:
        # Create tf variables for the weights and biases of the conv layer
        w=tf.random_normal_initializer(mean=0.0, stddev=0.001, seed=None, dtype=tf.float32) 
        #b=tf.tf.constant_initializer(value)
        weights = tf.get_variable('weights', shape=[filter_height,
                                                filter_width,
                                                input_channels/groups,
                                                num_filters],
                                                initializer=w,
                                                trainable=True)
        biases = tf.get_variable('biases', shape=[num_filters],initializer=tf.ones_initializer(),trainable=True)
        
        
    if groups == 1:
        conv = convolve(x, weights)
    else:
         # Split input and weights and convolve them separately
         input_groups = tf.split(axis=3, num_or_size_splits=groups, value=x)
         weight_groups = tf.split(axis=3, num_or_size_splits=groups,
                             value=weights)
         output_groups = [convolve(i, k) for i, k in zip(input_groups, weight_groups)]

        # Concat the convolved output together again
         conv = tf.concat(axis=3, values=output_groups)#拼接张量
    # Add biases
    bias = tf.reshape(tf.nn.bias_add(conv, biases), tf.shape(conv))     
    # Apply relu function
    relu = tf.nn.relu(bias, name=scope.name)    
    return relu