Take another input.

[videetnimsarkar21]

How to take another article as an input? i have converted an article to bin and added it to data path.

it's giving. WARNING:tensorflow:Bucket input queue is empty when calling next_batch. Bucket queue size: 0, Input queue size: 0

[theamrzaki]

this is the code that i am using for model 4 to input 1 sentence i input a sentence , and i use an empty abstract , then i write this to a bin file , and i use the file name in flags.datapath

i have used this in model 4

    article = inputarticle
    abstract = ""
    file_name = 'my_test' + time.strftime("%Y%m%d-%H%M%S") + '.bin'
    with open(file_name, 'wb') as writer:
        # Write to tf.Example
        tf_example = example_pb2.Example()
        tf_example.features.feature['article'].bytes_list.value.extend([article.encode('utf-8')])
        tf_example.features.feature['abstract'].bytes_list.value.extend([abstract.encode('utf-8')])
        tf_example_str = tf_example.SerializeToString()
        str_len = len(tf_example_str)
        writer.write(struct.pack('q', str_len))
        writer.write(struct.pack('%ds' % str_len, tf_example_str))

please try this , and tell me if it working

[videetnimsarkar21]

attention_decoder timestep 99 of 100 W0708 22:36:27.830724 140576798738304 deprecation.py:323] From :113: py_func (from tensorflow.python.ops.script_ops) is deprecated and will be removed in a future version. Instructions for updating: tf.py_func is deprecated in TF V2. Instead, there are two options available in V2. - tf.py_function takes a python function which manipulates tf eager tensors instead of numpy arrays. It's easy to convert a tf eager tensor to an ndarray (just call tensor.numpy()) but having access to eager tensors means tf.py_functions can use accelerators such as GPUs as well as being differentiable using a gradient tape. - tf.numpy_function maintains the semantics of the deprecated tf.py_func (it is not differentiable, and manipulates numpy arrays). It drops the stateful argument making all functions stateful.

W0708 22:39:01.221755 140576798738304 deprecation.py:506] From /usr/local/lib/python3.6/dist-packages/tensorflow/python/training/adagrad.py:76: calling Constant.init (from tensorflow.python.ops.init_ops) with dtype is deprecated and will be removed in a future version. Instructions for updating: Call initializer instance with the dtype argument instead of passing it to the constructor I0708 22:39:01.943351 140576798738304 :447] Time to build graph: 228 seconds W0708 22:39:02.223653 140576798738304 deprecation.py:323] From :199: Supervisor.init (from tensorflow.python.training.supervisor) is deprecated and will be removed in a future version. Instructions for updating: Please switch to tf.train.MonitoredTrainingSession I0708 22:39:30.694431 140576798738304 session_manager.py:500] Running local_init_op. I0708 22:39:32.978297 140576798738304 session_manager.py:502] Done running local_init_op. I0708 22:40:53.739769 140576798738304 supervisor.py:737] Starting standard services. I0708 22:40:55.880370 140572232775424 supervisor.py:1117] Saving checkpoint to path logs/train/model.ckpt I0708 22:40:55.880727 140576798738304 supervisor.py:743] Starting queue runners. I0708 22:40:55.881700 140572713105152 supervisor.py:1099] global_step/sec: 0 I0708 22:40:55.885952 140576798738304 :241] Preparing or waiting for session... I0708 22:40:55.890155 140576798738304 :242] Created session. I0708 22:40:55.891672 140576798738304 :253] Starting run_training I0708 22:40:55.895927 140576798738304 :273] Starting Seq2Seq training... W0708 22:40:55.897850 140576798738304 :262] Bucket input queue is empty when calling next_batch. Bucket queue size: 0, Input queue size: 0 I0708 22:41:55.880472 140572232775424 supervisor.py:1117] Saving checkpoint to path logs/train/model.ckpt I0708 22:41:55.881977 140572713105152 supervisor.py:1099] global_step/sec: 0 I0708 22:42:55.880480 140572232775424 supervisor.py:1117] Saving checkpoint to path logs/train/model.ckpt I0708 22:42:55.882748 140572713105152 supervisor.py:1099] global_step/sec: 0

[theamrzaki]

are you working on the notebook on python 3 .. i think this is the problem , actually this code which come from https://github.com/yaserkl/RLSeq2Seq which is built on pythoon 2.7 https://github.com/yaserkl/RLSeq2Seq/issues/2

so this notebook now only works with python 2.7 i am actually trying to upgrade it to python 3 if i am successful if God Wills it , i would upload the new upgraded notebook here

please just confirm if this is the problem (working on python3) or if there is another problem that i need to solve

thanks