Plot of predictions with test data in Colab notebook doesn't make much sense

In the Colab notebook, after the training we plot some graphs.
The last one, which is "Plot the predictions along with to the test data", is literally taken from a standard TensorFlow Lite example notebook (see https://colab.research.google.com/github/tensorflow/tensorflow/blob/master/tensorflow/lite/experimental/micro/examples/hello_world/create_sine_model.ipynb#scrollTo=L_vE-ZDkHVxe )
but does not make much sense in this context.

Besides that, the code for that plot currently (ab)uses a deprecated matplotlib behaviour
```
/usr/local/lib/python3.6/dist-packages/matplotlib/axes/_base.py:380: MatplotlibDeprecationWarning:
cycling among columns of inputs with non-matching shapes is deprecated.
```
which also suggests that the current plot is a bit nonsensical.

Abra o notebook do Colab e inspecione a seção final "Plot the predictions along with to the test data", comparando-a com o exemplo vinculado do TensorFlow Lite. Execute o notebook até o treinamento e revise o plot para que ele represente os dados de teste e as previsões de forma significativa, sem o aviso de depreciação do Matplotlib relatado.

Escrita pelo modelo de indexação a partir do texto da issue.