Update machine-learning/03-Training and AutoML.ipynb

Co-authored-by: Luis Quintanilla <46974588+luisquintanilla@users.noreply.github.com>

Author: LittleLittleCloud

machine-learning/03-Training and AutoML.ipynb

@@ -31,7 +31,7 @@
         "\n",
         "In machine learning, the difference or distance between predicted and actual label is usually called **loss**. Similar to training algorithms, you use different loss measures based on the task. For classification softmax is a common loss measure. For regression, Root Mean Squared Error (RMSE) is a common loss measure. In general though, they are all metrics to quantify the distance between the predicted and actual value. In most of cases, a **lower loss means a better model**. For more information, see the [ML.NET evaluation metrics guide](https://docs.microsoft.com/dotnet/machine-learning/resources/metrics).\n"
         "\n",
-        "So what `Fit(x)` does is to lower the `loss`. A good model means a model with small `loss`, when you train a model, you want to decrease its loss so as to make the prediction of that model closer to actual value. That's what __Train__ actually mean."
+        "So what `Fit` does is apply an algorithm to your data to identify patterns and estimate a function that lowers the loss. When you train a model, you want to decrease its loss to make the prediction of that model closer to the actual value."
       ]
     },
     {