Ludwig: Prediction on csv without the predictor labels.

[tommylees112]

When you want to predict a variable, would it not make sense for the csv file not to have the prediction labels? Does it need that extra column?

This would be the case for all kaggle competitions / competitions that focus on predicting an output variable.

What's the best way to deal with this sort of error?

---

looking at the data from the (Kaggle Titanic Link)[https://www.kaggle.com/c/titanic/] I ran an experiment:

ludwig experiment --data_csv data/train.csv --model_definition_file model_definition.yaml

And then get the trained model, I want to use that to predict the classes from the 'test.csv'.

The issue is that the test.csv (by design) doesn't have 'survived' labels. Ludwig throws an error

(ludwig) ➜  testing_ludwig git:(master) ✗ ludwig predict --data_csv data/test.csv -m results/experiment_run_1/model
 _         _        _
| |_  _ __| |_ __ _(_)__ _
| | || / _` \ V  V / / _` |
|_|\_,_\__,_|\_/\_/|_\__, |
                     |___/
ludwig v0.1.0 - Predict

Dataset type: generic
Dataset path: data/test.csv
Model path: results/experiment_run_1/model
Output path: results_0

Loading metadata from: results/experiment_run_1/model/train_set_metadata.json
/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/features/numerical_feature.py:63: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.
  np.float32).as_matrix()
Traceback (most recent call last):
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/pandas/core/indexes/base.py", line 2656, in get_loc
    return self._engine.get_loc(key)
  File "pandas/_libs/index.pyx", line 108, in pandas._libs.index.IndexEngine.get_loc
  File "pandas/_libs/index.pyx", line 132, in pandas._libs.index.IndexEngine.get_loc
  File "pandas/_libs/hashtable_class_helper.pxi", line 1601, in pandas._libs.hashtable.PyObjectHashTable.get_item
  File "pandas/_libs/hashtable_class_helper.pxi", line 1608, in pandas._libs.hashtable.PyObjectHashTable.get_item
KeyError: 'Survived'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/Users/tommylees/miniconda3/envs/ludwig/bin/ludwig", line 10, in <module>
    sys.exit(main())
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/cli.py", line 86, in main
    CLI()
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/cli.py", line 64, in __init__
    getattr(self, args.command)()
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/cli.py", line 73, in predict
    predict.cli(sys.argv[2:])
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/predict.py", line 379, in cli
    full_predict(**vars(args))
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/predict.py", line 86, in full_predict
    only_predictions
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/data/preprocessing.py", line 674, in preprocess_for_prediction
    train_set_metadata=train_set_metadata
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/data/preprocessing.py", line 62, in build_dataset
    **kwargs
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/data/preprocessing.py", line 90, in build_dataset_df
    global_preprocessing_parameters
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/data/preprocessing.py", line 153, in build_data
    preprocessing_parameters
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/ludwig/data/preprocessing.py", line 174, in handle_missing_values
    dataset_df[feature['name']] = dataset_df[feature['name']].fillna(
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/pandas/core/frame.py", line 2927, in __getitem__
    indexer = self.columns.get_loc(key)
  File "/Users/tommylees/miniconda3/envs/ludwig/lib/python3.6/site-packages/pandas/core/indexes/base.py", line 2658, in get_loc
    return self._engine.get_loc(self._maybe_cast_indexer(key))
  File "pandas/_libs/index.pyx", line 108, in pandas._libs.index.IndexEngine.get_loc
  File "pandas/_libs/index.pyx", line 132, in pandas._libs.index.IndexEngine.get_loc
  File "pandas/_libs/hashtable_class_helper.pxi", line 1601, in pandas._libs.hashtable.PyObjectHashTable.get_item
  File "pandas/_libs/hashtable_class_helper.pxi", line 1608, in pandas._libs.hashtable.PyObjectHashTable.get_item
KeyError: 'Survived'

train.csv

PassengerId,Survived,Pclass,Name,Sex,Age,SibSp,Parch,Ticket,Fare,Cabin,Embarked
1,0,3,"Braund, Mr. Owen Harris",male,22,1,0,A/5 21171,7.25,,S
2,1,1,"Cumings, Mrs. John Bradley (Florence Briggs Thayer)",female,38,1,0,PC 17599,71.2833,C85,C

test.csv

PassengerId,Pclass,Name,Sex,Age,SibSp,Parch,Ticket,Fare,Cabin,Embarked
892,3,"Kelly, Mr. James",male,34.5,0,0,330911,7.8292,,Q
893,3,"Wilkes, Mrs. James (Ellen Needs)",female,47,1,0,363272,7,,S

[maacl]

What happens if you just add the "Survived" column to the header row in test.csv?

[w4nderlust]

Please use the --only_predictions option as documented in the docs and asked in many other questions already. The next release of Ludwig will default to it.

[tommylees112]

Sorry! I am reading the docs now. They are extremely detailed. Thank you very much for your hard work!

[w4nderlust]

Thank you for your kind words, you're welcome. I love the noble causes you use machine learning for, and if Ludwig can help you with that, I'm even more glad.