Question

我在分布式模式下运行代码，而我的代码在异步模式下运行良好；但是代码无法在同步模式下运行。

opt = tf.train.MomentumOptimizer(learning_rate=lr_placeholder, momentum=0.9) opt=tf.train.SyncReplicasOptimizer(opt,replicas_to_aggregate=len(worker_hosts),total_num_replicas=len(worker_hosts),use_locking=True)
train_op = opt.minimize(full_loss, global_step=global_step)
val_op = validation_op(validation_step, vali_top1_error, vali_loss)
sync_replicas_hook = opt.make_session_run_hook(True)
init=tf.global_variables_initializer()
with training.MonitoredTrainingSession(master=server.target, is_chief=True,hooks=[sync_replicas_hook]) as sess:

回溯（最近通话最近）：文件“ /usr/local/python3/lib/python3.6/site-packages/tensorflow/python/client/session.py”，第1行 292，在_do_call中返回fn（* args）文件“ /usr/local/python3/lib/python3.6/site-packages/tensorflow/python/client/session.py”，第1行 277，在_run_fn选项中，feed_dict，fetch_list，target_list，run_metadata）文件“ /usr/local/python3/lib/python3.6/site-packages/tensorflow/python/client/session.py”，第1行 367，在_call_tf_sessionrun run_metadata中） tensorflow.python.framework.errors_impl.InvalidArgumentError：NodeDef缺少attr'reduction_type' 来自Op句柄：Ref（string）; attr = dtype：type，allowed = [DT_FLOAT，DT_DOUBLE，DT_INT32，DT_UINT8，DT_INT16，...，DT_UINT16，DT_COMPLEX128，DT_HALF，DT_UINT32，DT_UINT64]; attr = shape：shape; attr = container：string，default =“”; attr = shared_name：string，default =“”; attr = reduction_type：string，default =“ MEAN”，allowed = [“” MEAN“，” SUM“]; is_stateful = true>; NodeDef：{{node sync_replicas / conditional_accumulator}} = ConditionalAccumulator_class = [“ loc：@ sync_replicas / SetGlobalStep”]，container =“”，dtype = DT_FLOAT，shape = [3,3,3,16]，shared_name =“ conv0 / conv：0 / grad_accum“，_ device =” / job：ps /副本：0 / task：0 / device：CPU：0“ 在处理上述异常期间，发生了另一个异常：

Answer 1

tensorflow.python.framework.errors_impl.InvalidArgumentError：NodeDef缺少attr'reduction_type' 来自Op句柄：Ref（string）; attr = dtype：type，allowed = [DT_FLOAT，DT_DOUBLE，DT_INT32，DT_UINT8，DT_INT16，...，DT_UINT16，DT_COMPLEX128，DT_HALF，DT_UINT32，DT_UINT64]; attr = shape：shape; attr = container：string，default =“”; attr = shared_name：string，default =“”; attr = reduction_type：string，default =“ MEAN”，allowed = [“” MEAN“，” SUM“]; is_stateful = true>; NodeDef：{{node sync_replicas / conditional_accumulator}} = ConditionalAccumulator_class = [“ loc：@ sync_replicas / SetGlobalStep”]，container =“”，dtype = DT_FLOAT，shape = [3、3、3、16]，shared_name =“ conv0 / conv：0 / grad_accum“，_ device =” / job：ps /副本：0 / task：0 / device：CPU：0“

我在分布式模式下运行代码，而我的代码在异步模式下运行良好；但是代码无法在同步模式下运行

1 个答案: