tf.estimator无法使用

[ddgetget]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

[qiufengyuyi]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

信息有点不太具体，不过根据我的经验，这种错误往往发生在input_fn或者模型的输入这块有点问题，建议在这块debug一下

[Chen-Dixi]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

请问找到问题原因了吗

[Chen-Dixi]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

请问找到问题原因了吗

现在跑通了

[Jorigorn]

请问是什么问题呢？ sh run_event_role.sh 后

aceback (most recent call last): File "run_event.py", line 73, in main() File "run_event.py", line 46, in main run_event_role_mrc(args) File "/data/ceph/dd/event_classification/ft_local/event_extraction_tiny/train_helper.py", line 182, in run_event_role_mrc tf.estimator.train_and_evaluate(estimator, train_spec, eval_spec) File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/training.py", line 473, in train_and_evaluate return executor.run() File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/training.py", line 613, in run return self.run_local() File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/training.py", line 714, in run_local saving_listeners=saving_listeners) File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 367, in train loss = self._train_model(input_fn, hooks, saving_listeners) File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1158, in _train_model return self._train_model_default(input_fn, hooks, saving_listeners) File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1188, in _train_model_default features, labels, ModeKeys.TRAIN, self.config) File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1146, in _call_model_fn model_fn_results = self._model_fn(features=features, **kwargs) File "/data/ceph/arikchen/event_classification/ft_local/event_extraction_tiny/models/bert_mrc.py", line 68, in model_fn start_labels,end_labels = labels File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 477, in iter "Tensor objects are only iterable when eager execution is " TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

[qiufengyuyi]

请问是什么问题呢？ sh run_event_role.sh 后

aceback (most recent call last): File "run_event.py", line 73, in main() File "run_event.py", line 46, in main run_event_role_mrc(args) File "/data/ceph/dd/event_classification/ft_local/event_extraction_tiny/train_helper.py", line 182, in run_event_role_mrc tf.estimator.train_and_evaluate(estimator, train_spec, eval_spec) File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/training.py", line 473, in train_and_evaluate return executor.run() File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/training.py", line 613, in run return self.run_local() File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/training.py", line 714, in run_local saving_listeners=saving_listeners) File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 367, in train loss = self._train_model(input_fn, hooks, saving_listeners) File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1158, in _train_model return self._train_model_default(input_fn, hooks, saving_listeners) File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1188, in _train_model_default features, labels, ModeKeys.TRAIN, self.config) File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1146, in _call_model_fn model_fn_results = self._model_fn(features=features, kwargs) File "/data/ceph/arikchen/event_classification/ft_local/event_extraction_tiny/models/bert_mrc.py", line 68, in model_fn start_labels,end_labels = labels File "/usr/local/app/.pyenv/versions/env-3.6.8/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 477, in iter** "Tensor objects are only iterable when eager execution is " TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

见楼上的回答，可能是input的数据格式有问题，需要debug一下，看具体模型对应的正确的输入数据格式

[veronic]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

请问找到问题原因了吗

现在跑通了

请问您是怎么调试成功的？ 我依照作者在知乎里的回答“一种是多分类形式，一种是两个类型label，分别预测起始位置，你这个问题应该是模型用的是预测起始位置，但是实际上加载的label是多分类，建议你看一下代码，换成加载预测起始位置的数据”，把“train_helper.py”的118行和123行分别改成“labels_start_train.npy”和“labels_start_dev.npy”但还是会报错。

已经难过到心有戚戚、食不下咽、日夜垂泪的地步。

[Chen-Dixi]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

请问找到问题原因了吗

现在跑通了

请问您是怎么调试成功的？ 我依照作者在知乎里的回答“一种是多分类形式，一种是两个类型label，分别预测起始位置，你这个问题应该是模型用的是预测起始位置，但是实际上加载的label是多分类，建议你看一下代码，换成加载预测起始位置的数据”，把“train_helper.py”的118行和123行分别改成“labels_start_train.npy”和“labels_start_dev.npy”但还是会报错。

已经难过到心有戚戚、食不下咽、日夜垂泪的地步。

试试我fork出来的代码，https://github.com/Chen-Dixi/event_extraction

[veronic]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

请问找到问题原因了吗

现在跑通了

请问您是怎么调试成功的？ 我依照作者在知乎里的回答“一种是多分类形式，一种是两个类型label，分别预测起始位置，你这个问题应该是模型用的是预测起始位置，但是实际上加载的label是多分类，建议你看一下代码，换成加载预测起始位置的数据”，把“train_helper.py”的118行和123行分别改成“labels_start_train.npy”和“labels_start_dev.npy”但还是会报错。 已经难过到心有戚戚、食不下咽、日夜垂泪的地步。

试试我fork出来的代码，https://github.com/Chen-Dixi/event_extraction。

谢谢！ 调通了，原来是少加了终止位置end_labels。 我终于又能吃得下饭了【泪目】

[chunxiaozheng]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

请问找到问题原因了吗

现在跑通了

请问您是怎么调试成功的？ 我依照作者在知乎里的回答“一种是多分类形式，一种是两个类型label，分别预测起始位置，你这个问题应该是模型用的是预测起始位置，但是实际上加载的label是多分类，建议你看一下代码，换成加载预测起始位置的数据”，把“train_helper.py”的118行和123行分别改成“labels_start_train.npy”和“labels_start_dev.npy”但还是会报错。 已经难过到心有戚戚、食不下咽、日夜垂泪的地步。

试试我fork出来的代码，https://github.com/Chen-Dixi/event_extraction。

谢谢！ 调通了，原来是少加了终止位置end_labels。 我终于又能吃得下饭了【泪目】

你好，可以请教一下怎么修改的嘛，我也遇到了这个问题

[chunxiaozheng]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

请问找到问题原因了吗

现在跑通了

你好，可以请教一下怎么调通的嘛，我看您分享的git地址404了

[veronic]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

请问找到问题原因了吗

现在跑通了

你好，可以请教一下怎么调通的嘛，我看您分享的git地址404了

https://github.com/Chen-Dixi/event_extraction 是这个地址，之前那个最后加了“。”所以404了

[chunxiaozheng]

TypeError: Tensor objects are only iterable when eager execution is enabled. To iterate over this tensor use tf.map_fn.

请问找到问题原因了吗

现在跑通了

你好，可以请教一下怎么调通的嘛，我看您分享的git地址404了

https://github.com/Chen-Dixi/event_extraction 是这个地址，之前那个最后加了“。”所以404了

好的，我学习一下，太感谢啦！