Question: Evaluation continues beyond specified number of generations.

First off, I've been working with tpot recently and it's pretty awesome, so thanks for that! I've been using it with a small dataset of 1 GB and noticed that the evaluation doesn't always stop when the number of generations has been reached.

For example, printing the classifier before training:
```python
TPOTClassifier(config_dict='TPOT light', crossover_rate=0.1, cv=5,
disable_update_check=False, early_stop=None, generations=5,
max_eval_time_mins=5, max_time_mins=120, memory='auto',
mutation_rate=0.9, n_jobs=8, offspring_size=None,
periodic_checkpoint_folder='tpot', population_size=10,
random_state=None, scoring=None, subsample=1.0,
template='RandomTree', use_dask=False, verbosity=2,
warm_start=False)

```

The logs are like:
```
Starting training at: 2019-05-03 13:35:13.165409

Generation 1 - Current best internal CV score: 0.9391366044449356
Generation 2 - Current best internal CV score: 0.9391379502561822
Generation 3 - Current best internal CV score: 0.9445918220779804
Generation 4 - Current best internal CV score: 0.9445918220779804
Generation 5 - Current best internal CV score: 0.9495054028474923
Generation 6 - Current best internal CV score: 0.9495054028474923
Generation 7 - Current best internal CV score: 0.9495054028474923
Generation 8 - Current best internal CV score: 0.9495054028474923

165.17067630000003 minutes have elapsed. TPOT will close down.
TPOT closed during evaluation in one generation.
WARNING: TPOT may not provide a good pipeline if TPOT is stopped/interrupted in a early generation.

TPOT closed prematurely. Will use the current best pipeline.

Best pipeline: GaussianNB(Normalizer(SelectPercentile(input_matrix, percentile=90), norm=l1))

Finished training at: 2019-05-03 16:23:12.062465
```

It takes longer in time 165 minutes vs 120 as specified before termination, and continuous for 8 generations, even though I passed in 5. Any idea why this is happening?