擅长:python、mysql、java
<p>我挣扎了一会儿。给出的答案将向图中添加<code>assign</code>操作(因此,如果随后保存检查点,则不必要地增加<code>.meta</code>的大小)。更好的解决方案是使用<code>tf.keras.backend.set_value</code>。我们可以用原始的tensorflow来模拟:</p>
<pre><code> for x, value in zip(tf.global_variables(), values_npfmt):
if hasattr(x, '_assign_placeholder'):
assign_placeholder = x._assign_placeholder
assign_op = x._assign_op
else:
assign_placeholder = array_ops.placeholder(tf_dtype, shape=value.shape)
assign_op = x.assign(assign_placeholder)
x._assign_placeholder = assign_placeholder
x._assign_op = assign_op
get_session().run(assign_op, feed_dict={assign_placeholder: value})
</code></pre>