当前位置：首页 > news >正文

tensorflow 模型计算中，预测错误；权重参数加载

news 2025/7/23 15:29:50

tensorflow 模型计算中，预测错误；权重参数加载

tensorflow 模型计算主要代码（正确代码）

linear1_kernel_initializer = tf.constant_initializer(numpy.transpose(data["linear1.weight"]))
linear1_bias_initializer = tf.constant_initializer(numpy.transpose(data["linear1.bias"]))
linear1 = layers.Dense(units=400, activation=tf.nn.relu, kernel_initializer=linear1_kernel_initializer, use_bias=True, bias_initializer=linear1_bias_initializer, input_shape=(48,))
linear2_kernel_initializer = tf.constant_initializer(numpy.transpose(data["linear2.weight"]))
linear2_bias_initializer = tf.constant_initializer(numpy.transpose(data["linear2.bias"]))
linear2 = layers.Dense(units=400, activation=tf.nn.relu, kernel_initializer=linear2_kernel_initializer, use_bias=True, bias_initializer=linear2_bias_initializer)
linear3_kernel_initializer = tf.constant_initializer(numpy.transpose(data["linear3.weight"]))
linear3_bias_initializer = tf.constant_initializer(numpy.transpose(data["linear3.bias"]))
linear3 = layers.Dense(units=2, activation=None, kernel_initializer=linear3_kernel_initializer, use_bias=True, bias_initializer=linear3_bias_initializer)
model = tf.keras.Sequential([linear1, linear2, linear3])input = numpy.ones((2, 48), dtype=float)
predict = model.predict(input)
print(predict[0:100,:])

原本权重参数采用以下代码

linear1_kernel_initializer = tf.constant_initializer(data["linear1.weight"])
linear1_bias_initializer = tf.constant_initializer((data["linear1.bias"])

但模型预测值与Matlab计算值有误。后经过测试定位到 layers.Dense 此处，然后创建 layers.Dense时设置use_bias=False参数，不去考虑偏差参数。改变初始权重参数方式：

input_size = 2
units_p = 3
data = numpy.array([1, 1, 2, 2, 2, 3], dtype=float)
linear1_kernel_initializer = tf.constant_initializer(data)
linear1 = layers.Dense(units=units_p, activation=None, kernel_initializer=linear1_kernel_initializer, use_bias=False, input_shape=(input_size,))
#变化data
data = numpy.array([1, 2, 3, 1, 2, 3], dtype=float)
#或者
data = numpy.array([1, 2, 3, 1, 2, 3], dtype=float).reshape(3, 2)

通过这样的方式，才发现 linear1_kernel_initializer = tf.constant_initializer(data) 中的 data 有问题，通过对预测结果的分析，发现 tf.constant_initializer() 会将传递过来的数据拉成一维，再根据 units 和 不同层 来变更数据矩阵大小，所以传入tf.constant_initializer()的数据只要总大小是对的就可以传入，而不需要shape一致。
所以，既然之前的数据预测结果有误，那就是数据排列有误，将 data 数据进行矩阵转置再传入到tf.constant_initializer() 函数中
问题成功解决。
同时我想说明的是，pytorch的torch.nn.Linear是W x + b 而 tensorflow 的 layers.Dense 是 x W + b。