I want to try something similar to the example in coach/tutorials/4. Batch Reinforcement Learning.ipynb with a dataset similar to the _acrobotdataset.csv. However, when I try to make a small adjustment to the dataset (e.g. change the episode_name, I assumed it should not make a difference) and re-run the example code, I get the error ValueError: cannot convert float NaN to integer.
Does anyone have an idea why this occurs? How am I supposed to use another dataset than the example dataset _acrobatdataset.csv?
Hi all,
I want to try something similar to the example in coach/tutorials/4. Batch Reinforcement Learning.ipynb with a dataset similar to the _acrobotdataset.csv. However, when I try to make a small adjustment to the dataset (e.g. change the episode_name, I assumed it should not make a difference) and re-run the example code, I get the error ValueError: cannot convert float NaN to integer.
Does anyone have an idea why this occurs? How am I supposed to use another dataset than the example dataset _acrobatdataset.csv?