How to do inference on a single sentence

[parulsingh23]

I am running this model in a Google Colab notebook. However, after following the steps in the ReadMe file to train the model, I'm not sure how I can apply the model to provide an inference on a given sentence. So for example, what command would I provide to execute the model on a negative sentence of "I do not like this food" to generate a positive sentiment of "I like this food". Or is there no such command and would I need to place my input (negative sentiment) sentence in a file and call a function?

I had tried creating a directory called 'evaluatingWithNegative' located in 'data/evaluatingWithNegative' (adjacent to the Yelp dataset). In that directory, I placed a file named "x.0" where I put the sentence "i do not like this food ." I then ran the commands

%cd classifier
!python textcnn.py --mode evaluate

And besides some warning statements, it outputted:

Saving classifier result at: /content/DualRL/tmp/cls_result_yelp.txt
Test accuracy: 96.40

I then ran

%cd nmt
!python nmt.py --mode inference --nmt_direction 0-1

Which outputted:

WARNING: The TensorFlow contrib module will not be included in TensorFlow 2.0.
For more information, please see:
  * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md
  * https://github.com/tensorflow/addons
If you depend on functionality not listed there, please file an issue.

A=0, B=1
Vocabulary size:src:9353
Use x'->y to update model f(x->y)
WARNING:tensorflow:From ../utils/data.py:61: group_by_window (from tensorflow.contrib.data.python.ops.grouping) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.data.experimental.group_by_window(...)`.
WARNING:tensorflow:From ../utils/data.py:23: to_int64 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.cast instead.
WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/tensorflow/python/data/ops/dataset_ops.py:1419: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.
Prepare for model saver
('Model save path:', '/content/DualRL/tmp/model/yelp/nmt_template/0-1/')
Adopt bilstm as encoder and decoder
RNN Decoder CopyBridge
WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/function.py:1007: calling create_op (from tensorflow.python.framework.ops) with compute_shapes is deprecated and will be removed in a future version.
Instructions for updating:
Shapes are always computed; don't use the compute_shapes as it has no effect.
WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/opennmt/utils/cell.py:47: __init__ (from tensorflow.python.ops.rnn_cell_impl) is deprecated and will be removed in a future version.
Instructions for updating:
This class is equivalent as tf.keras.layers.LSTMCell, and will be replaced by that in Tensorflow 2.0.
WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/opennmt/encoders/rnn_encoder.py:133: bidirectional_dynamic_rnn (from tensorflow.python.ops.rnn) is deprecated and will be removed in a future version.
Instructions for updating:
Please use `keras.layers.Bidirectional(keras.layers.RNN(cell))`, which is equivalent to this API
WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/rnn.py:443: dynamic_rnn (from tensorflow.python.ops.rnn) is deprecated and will be removed in a future version.
Instructions for updating:
Please use `keras.layers.RNN(cell)`, which is equivalent to this API
WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/rnn.py:626: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.cast instead.
/content/DualRL/tmp/model/yelp/nmt_template/0-1/
('Loading nmt model from', '/content/DualRL/tmp/model/yelp/nmt_template/0-1/')
WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/tensorflow/python/training/saver.py:1266: checkpoint_exists (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version.
Instructions for updating:
Use standard file APIs to check for files with this prefix.
('Error! Loading nmt model from', '/content/DualRL/tmp/model/yelp/nmt_template/0-1/')
('Again! Loading nmt model from', u'/content/DualRL/tmp/model/yelp/nmt_template//0-1/-2300')
Adopt bilstm as encoder and decoder
RNN Decoder CopyBridge
Reuse parameters.
Result save path:/content/DualRL/tmp/output/yelp_template/test.0-1.tsf, /content/DualRL/tmp/output/yelp_template/test.0.tsf
INFERENCE---Total N batch:17	Generate probs:-11.8581542969	Cost time:5.07563018799

But I am very confused if I am following the right commands to get an inference, and where I can find the output 'positive' sentiment sentence.