readme fix

README.md

@@ -16,25 +16,26 @@ pinned: false
 
 ## Metric Description
 
-This metrics computes the expected calibration error (ECE). ECE evaluates how well a model is calibrated, i.e. how well its output probabilities match the actual ground truth distribution. It measures the $$L^p$$ norm difference between a model’s posterior and the true likelihood of being correct.
+This metrics computes the expected calibration error (ECE). ECE evaluates how well a model is calibrated, i.e. how well its output probabilities match the actual ground truth distribution. It measures the $L^p$ norm difference between a model’s posterior and the true likelihood of being correct.
 This module directly calls the [torchmetrics package implementation](https://torchmetrics.readthedocs.io/en/stable/classification/calibration_error.html), allowing to use its flexible arguments.
 
 ## How to Use
 
 ### Inputs
+
 *List all input arguments in the format below*
 - **predictions** *(float32): predictions (after softmax). They must have a shape (N,C) if multiclass, or (N,...) if binary;*
 - **references** *(int64): reference for each prediction, with a shape (N,...);*
-- **kwargs** *arguments to pass to the [ece](https://torchmetrics.readthedocs.io/en/stable/classification/calibration_error.html) methods.*
+- **kwargs** *arguments to pass to the [calibration error](https://torchmetrics.readthedocs.io/en/stable/classification/calibration_error.html) method.*
 
 ### Output Values
 
-ECE as float.
+ECE as a float number.
 
 ### Examples
 
 ```Python
-ce = evaluate.load("Natooz/ece")
+ece = evaluate.load("Natooz/ece")
 results = ece.compute(
     references=np.array([[0.25, 0.20, 0.55],
                          [0.55, 0.05, 0.40],