Unit test test_nonlinear.py generates error

[denson]

I get the following error when I run test_nonlinear.py

.E.

ERROR: testRNN (main.NonLinearTest)

Traceback (most recent call last): File "/Volumes/experiments/GFFS_protein_structure_tensorflow/tutorials/unit_test_original_v2.py", line 79, in testRNN classifier.fit(data, labels) File "/Users/denson/anaconda2/lib/python2.7/site-packages/skflow/estimators/base.py", line 200, in fit self._setup_training() File "/Users/denson/anaconda2/lib/python2.7/site-packages/skflow/estimators/base.py", line 139, in _setup_training self._inp, self._out) File "/Users/denson/anaconda2/lib/python2.7/site-packages/skflow/estimators/rnn.py", line 110, in _model_fn self.initial_state)(X, y) File "/Users/denson/anaconda2/lib/python2.7/site-packages/skflow/models.py", line 251, in rnn_estimator return target_predictor_fn(encoding[-1], y) File "/Users/denson/anaconda2/lib/python2.7/site-packages/tensorflow/python/ops/array_ops.py", line 124, in _SliceHelper raise NotImplementedError("Negative indices are currently unsupported") NotImplementedError: Negative indices are currently unsupported

---

[ilblackdragon]

Interface of rnn op changed - instead of returning list of states it returns last state only now. I'll fix it later today.

[ilblackdragon]

Should be fixed with #119 merge.