OSDir


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Tensorflow Hub - cant seem to load module in Airflow


Ah, from that issue I see it's something to do with use unicode literals, though in local testing I can't get the error you see.

Could you add the following print statements to your dag, inside the python callable:

print(__import__("tensorflow_hub.compressed_module_resolver", fromlist=['_COMPRESSED_FORMAT_QUERY'])._COMPRESSED_FORMAT_QUERY[0].__class__)
print("https://tfhub.dev/google/nnlm-en-dim50/1 <https://tfhub.dev/google/nnlm-en-dim50/1>".__class__)


> On 19 Jun 2018, at 11:09, Andrew Maguire <andrewm4894@xxxxxxxxx> wrote:
> 
> There is a bit more info in the issue:
> 
> https://github.com/tensorflow/hub/issues/76
> 
> Seems like maybe Airflow is passing as bytes when it's being expected as a
> string..?
> 
> On Tue, Jun 19, 2018 at 11:02 AM Ash Berlin-Taylor <
> ash_airflowlist@xxxxxxxxxxxxxx> wrote:
> 
>> Welp, nothing useful in there :(
>> 
>> The error is coming form this line:
>> 
>>> embed = hub.Module("https://tfhub.dev/google/nnlm-en-dim50/1";)
>> 
>> I'm next to un-familar with TensorFlow - is it possible this model was
>> generated on Py3 but is being run on python2? Is that even a question that
>> makes sense?
>> 
>> You could always try
>> 
>> `embed = hub.Module(b"https://tfhub.dev/google/nnlm-en-dim50/1";)`
>> 
>> I'm not expecting that to make any difference but doesn't hurt to try.
>> 
>> -ash
>> 
>>> On 19 Jun 2018, at 10:47, Andrew Maguire <andrewm4894@xxxxxxxxx> wrote:
>>> 
>>> sure thing.
>>> 
>>> attached is a minimal example.
>>> 
>>> error i get is:
>>> 
>>> [2018-06-19 09:44:23,133] {cli.py:374} INFO - Running on host
>> airflow-worker-f796f6bd-7qzwc
>>> [2018-06-19 09:44:23,577] {models.py:1196} INFO - Dependencies all met
>> for <TaskInstance: dev_dag.encode_posts 2018-06-19 09:15:00 [queued]>
>>> [2018-06-19 09:44:23,813] {models.py:1196} INFO - Dependencies all met
>> for <TaskInstance: dev_dag.encode_posts 2018-06-19 09:15:00 [queued]>
>>> [2018-06-19 09:44:23,824] {models.py:1406} INFO -
>>> 
>> --------------------------------------------------------------------------------
>>> Starting attempt 1 of 6
>>> 
>> --------------------------------------------------------------------------------
>>> 
>>> [2018-06-19 09:44:24,188] {models.py:1427} INFO - Executing
>> <Task(PythonOperator): encode_posts> on 2018-06-19 09:15:00
>>> [2018-06-19 09:44:24,222] {base_task_runner.py:115} INFO - Running:
>> ['bash', '-c', u'airflow run dev_dag encode_posts 2018-06-19T09:15:00
>> --job_id 5753 --raw -sd DAGS_FOLDER/dev_dag.py']
>>> [2018-06-19 09:45:11,207] {base_task_runner.py:98} INFO - Subtask:
>> [2018-06-19 09:45:11,164] {__init__.py:45} INFO - Using executor
>> CeleryExecutor
>>> [2018-06-19 09:45:13,200] {base_task_runner.py:98} INFO - Subtask:
>> [2018-06-19 09:45:13,168] {models.py:189} INFO - Filling up the DagBag from
>> /home/airflow/gcs/dags/dev_dag.py
>>> [2018-06-19 09:45:13,222] {base_task_runner.py:98} INFO - Subtask:
>> /usr/local/lib/python2.7/site-packages/airflow/utils/helpers.py:351:
>> DeprecationWarning: Importing DummyOperator directly from <module
>> 'airflow.operators' from
>> '/usr/local/lib/python2.7/site-packages/airflow/operators/__init__.pyc'>
>> has been deprecated. Please import from '<module 'airflow.operators' from
>> '/usr/local/lib/python2.7/site-packages/airflow/operators/__init__.pyc'>.[operator_module]'
>> instead. Support for direct imports will be dropped entirely in Airflow 2.0.
>>> [2018-06-19 09:45:13,224] {base_task_runner.py:98} INFO - Subtask:
>> DeprecationWarning)
>>> [2018-06-19 09:45:27,573] {base_task_runner.py:98} INFO - Subtask:
>> [2018-06-19 09:45:27,571] {dev_dag.py:52} INFO - ... begin - get module
>> from tf-hub ...
>>> [2018-06-19 09:45:28,228] {base_task_runner.py:98} INFO - Subtask:
>> Traceback (most recent call last):
>>> [2018-06-19 09:45:28,230] {base_task_runner.py:98} INFO - Subtask:
>> File "/usr/local/bin/airflow", line 27, in <module>
>>> [2018-06-19 09:45:28,232] {base_task_runner.py:98} INFO - Subtask:
>> args.func(args)
>>> [2018-06-19 09:45:28,232] {base_task_runner.py:98} INFO - Subtask:
>> File "/usr/local/lib/python2.7/site-packages/airflow/bin/cli.py", line
>> 392, in run
>>> [2018-06-19 09:45:28,233] {base_task_runner.py:98} INFO - Subtask:
>> pool=args.pool,
>>> [2018-06-19 09:45:28,233] {base_task_runner.py:98} INFO - Subtask:
>> File "/usr/local/lib/python2.7/site-packages/airflow/utils/db.py", line
>> 50, in wrapper
>>> [2018-06-19 09:45:28,234] {base_task_runner.py:98} INFO - Subtask:
>> result = func(*args, **kwargs)
>>> [2018-06-19 09:45:28,234] {base_task_runner.py:98} INFO - Subtask:
>> File "/usr/local/lib/python2.7/site-packages/airflow/models.py", line
>> 1492, in _run_raw_task
>>> [2018-06-19 09:45:28,250] {base_task_runner.py:98} INFO - Subtask:
>> result = task_copy.execute(context=context)
>>> [2018-06-19 09:45:28,250] {base_task_runner.py:98} INFO - Subtask:
>> File
>> "/usr/local/lib/python2.7/site-packages/airflow/operators/python_operator.py",
>> line 89, in execute
>>> [2018-06-19 09:45:28,251] {base_task_runner.py:98} INFO - Subtask:
>> return_value = self.execute_callable()
>>> [2018-06-19 09:45:28,251] {base_task_runner.py:98} INFO - Subtask:
>> File
>> "/usr/local/lib/python2.7/site-packages/airflow/operators/python_operator.py",
>> line 94, in execute_callable
>>> [2018-06-19 09:45:28,252] {base_task_runner.py:98} INFO - Subtask:
>> return self.python_callable(*self.op_args, **self.op_kwargs)
>>> [2018-06-19 09:45:28,252] {base_task_runner.py:98} INFO - Subtask:
>> File "/home/airflow/gcs/dags/dev_dag.py", line 53, in fn_encode_posts
>>> [2018-06-19 09:45:28,253] {base_task_runner.py:98} INFO - Subtask:
>> embed = hub.Module("https://tfhub.dev/google/nnlm-en-dim50/1 <
>> https://tfhub.dev/google/nnlm-en-dim50/1>")
>>> [2018-06-19 09:45:28,253] {base_task_runner.py:98} INFO - Subtask:
>> File "/usr/local/lib/python2.7/site-packages/tensorflow_hub/module.py",
>> line 105, in __init__
>>> [2018-06-19 09:45:28,276] {base_task_runner.py:98} INFO - Subtask:
>> self._spec = as_module_spec(spec)
>>> [2018-06-19 09:45:28,277] {base_task_runner.py:98} INFO - Subtask:
>> File "/usr/local/lib/python2.7/site-packages/tensorflow_hub/module.py",
>> line 31, in as_module_spec
>>> [2018-06-19 09:45:28,278] {base_task_runner.py:98} INFO - Subtask:
>> return native_module.load_module_spec(spec)
>>> [2018-06-19 09:45:28,278] {base_task_runner.py:98} INFO - Subtask:
>> File
>> "/usr/local/lib/python2.7/site-packages/tensorflow_hub/native_module.py",
>> line 99, in load_module_spec
>>> [2018-06-19 09:45:28,280] {base_task_runner.py:98} INFO - Subtask:
>> path = compressed_module_resolver.get_default().get_module_path(path)
>>> [2018-06-19 09:45:28,280] {base_task_runner.py:98} INFO - Subtask:
>> File "/usr/local/lib/python2.7/site-packages/tensorflow_hub/resolver.py",
>> line 385, in get_module_path
>>> [2018-06-19 09:45:28,295] {base_task_runner.py:98} INFO - Subtask:
>> return self._get_module_path(handle)
>>> [2018-06-19 09:45:28,296] {base_task_runner.py:98} INFO - Subtask:
>> File "/usr/local/lib/python2.7/site-packages/tensorflow_hub/resolver.py",
>> line 467, in _get_module_path
>>> [2018-06-19 09:45:28,297] {base_task_runner.py:98} INFO - Subtask:
>> return resolver.get_module_path(handle)
>>> [2018-06-19 09:45:28,297] {base_task_runner.py:98} INFO - Subtask:
>> File "/usr/local/lib/python2.7/site-packages/tensorflow_hub/resolver.py",
>> line 385, in get_module_path
>>> [2018-06-19 09:45:28,298] {base_task_runner.py:98} INFO - Subtask:
>> return self._get_module_path(handle)
>>> [2018-06-19 09:45:28,299] {base_task_runner.py:98} INFO - Subtask:
>> File
>> "/usr/local/lib/python2.7/site-packages/tensorflow_hub/compressed_module_resolver.py",
>> line 105, in _get_module_path
>>> [2018-06-19 09:45:28,342] {base_task_runner.py:98} INFO - Subtask:
>> self._lock_file_timeout_sec())
>>> [2018-06-19 09:45:28,343] {base_task_runner.py:98} INFO - Subtask:
>> File "/usr/local/lib/python2.7/site-packages/tensorflow_hub/resolver.py",
>> line 313, in atomic_download
>>> [2018-06-19 09:45:28,343] {base_task_runner.py:98} INFO - Subtask:
>> download_fn(handle, tmp_dir)
>>> [2018-06-19 09:45:28,344] {base_task_runner.py:98} INFO - Subtask:
>> File
>> "/usr/local/lib/python2.7/site-packages/tensorflow_hub/compressed_module_resolver.py",
>> line 86, in download
>>> [2018-06-19 09:45:28,345] {base_task_runner.py:98} INFO - Subtask:
>> request = url.Request(_append_compressed_format_query(handle))
>>> [2018-06-19 09:45:28,345] {base_task_runner.py:98} INFO - Subtask:
>> File
>> "/usr/local/lib/python2.7/site-packages/tensorflow_hub/compressed_module_resolver.py",
>> line 62, in _append_compressed_format_query
>>> [2018-06-19 09:45:28,346] {base_task_runner.py:98} INFO - Subtask:
>> return urlparse.urlunparse(parsed)
>>> [2018-06-19 09:45:28,346] {base_task_runner.py:98} INFO - Subtask:
>> File
>> "/usr/local/lib/python2.7/site-packages/future/backports/urllib/parse.py",
>> line 387, in urlunparse
>>> [2018-06-19 09:45:28,368] {base_task_runner.py:98} INFO - Subtask:
>> _coerce_args(*components))
>>> [2018-06-19 09:45:28,370] {base_task_runner.py:98} INFO - Subtask:
>> File
>> "/usr/local/lib/python2.7/site-packages/future/backports/urllib/parse.py",
>> line 115, in _coerce_args
>>> [2018-06-19 09:45:28,371] {base_task_runner.py:98} INFO - Subtask:
>> raise TypeError("Cannot mix str and non-str arguments")
>>> [2018-06-19 09:45:28,373] {base_task_runner.py:98} INFO - Subtask:
>> TypeError: Cannot mix str and non-str arguments
>>> 
>>> I'm running this on Google cloud composer which is airflow 1.9 i
>> believe.
>>> 
>>> Cheers,
>>> Andy
>>> 
>>> On Tue, Jun 19, 2018 at 10:06 AM Ash Berlin-Taylor <
>> ash_airflowlist@xxxxxxxxxxxxxx <mailto:ash_airflowlist@xxxxxxxxxxxxxx>>
>> wrote:
>>> There's nothing directly in Airflow itself that would cause this kind of
>> issue that I can think of.
>>> 
>>> It depends on what the PythonOperator you are using in the DAG does
>> really. Can you share that code?
>>> 
>>> -ash
>>> 
>>>> On 19 Jun 2018, at 10:01, Andrew Maguire <andrewm4894@xxxxxxxxx
>> <mailto:andrewm4894@xxxxxxxxx>> wrote:
>>>> 
>>>> Hi All,
>>>> 
>>>> Just wondering if anyone might have a deeper insight into what if
>> anything
>>>> airflow related might be causing this issue
>>>> <https://github.com/tensorflow/hub/issues/76 <
>> https://github.com/tensorflow/hub/issues/76>>.
>>>> 
>>>> When i try load a tensorflow hub module within an airflow operator i
>> get
>>>> the error in that issue.
>>>> 
>>>> Works fine if i just run the python script myself.
>>>> 
>>>> Best i could figure out was something airflow was doing didn't agree
>> with
>>>> something tensorflow hub was expecting. And i'm not really sure if
>> there is
>>>> anything i could do to resolve.
>>>> 
>>>> Cheers,
>>>> Andy
>>> 
>>> <dev_dag.py>
>> 
>>