Model predicting NaN choosed as best model

**System Information (please complete the following information):**
- Model Builder or CLI Version: 16.13.9.2235601
- Visual Studio Version (if applicable): 17.5.2

**Describe the bug**
- On which page of the tutorial did you run into an issue (URL): https://dotnet.microsoft.com/en-us/learn/ml-dotnet/get-started-tutorial/evaluate
- Clear description of the problem:

1. Model's predictions are always NaN
2. RSquared of all Models is either 0.0 either NaN

**To Reproduce**
Steps to reproduce the behavior:

1. Go to Model Builder
2. Click on Predicting values (2nd scenario)
3. Download the [train.csv file](https://github.com/dotnet/machinelearning-modelbuilder/files/11011847/train.csv) and chose it in MB
4. Choose first column as Feature, second as Label
5. Train and evaluate

**Expected behavior**

1. Not getting the NaN from any Model
2. Not considering NaN as 0.0 loss (FastForestRegression)
3. Show an warning if something wrong with input data