Marcus McCurdy

Type mismatch due to vendored dependency

1

ffjson seems to be getting confused due to the vendor folder in another project. `cannot use event.ID (type "bitbucket.org/50onred/projectA/vendor/github.com/satori/go.uuid".UUID) as type "github.com/satori/go.uuid".UUID in assignment` The code compiles just fine with...

``` python3.5/site-packages/keras/backend/tensorflow_backend.py in in_train_phase(x, alt) 1302 # else: assume learning phase is a placeholder. 1303 x_shape = copy.copy(x.get_shape()) -> 1304 x = tf.python.control_flow_ops.cond(tf.cast(_LEARNING_PHASE, 'bool'), 1305 lambda: x, 1306 lambda: alt)...

Noticed these duplicate `to the` while reading the docs.

The optimizers were using the `weight_decay` key from the `build_info` dict for the learning rates.

LightGBM integration doesn't work properly if you are trying to maximize a custom metric

4

### Expected behavior If the eval metric has is_higher_better set to True then the objective should be maximized. ### Environment - Optuna version: - Optuna Integration version: - Python version:...