Better example code for the subsection "1. Replaced a variable with a tensor"

# Better example code for the subsection "1. Replaced a variable with a tensor"

**Position:** "1. Replaced a variable with a tensor" subsection

**Link:** https://www.tensorflow.org/guide/autodiff#1_replaced_a_variable_with_a_tensor

**Condition:** The original example is very concise. However, it took me a considerable amount of time to modify the code and test it to understand the example fully. 

I think there are two points that confused me.  
Point 1: The derivative of `y = x + 1` is a constant scalar `1`. This will cause no value change as the number of epoch increases. We can't obviously see the change in the value.  
Point 2: The comment directly gives the correct code `# This should be x.assign_add(1)` without the reason.

**Suggestion:** I have modified the example code as shown below.  
1. The formula has been changed from `y = x + 1` to `y = x**2`. The derivative is now `2x`, and we can obviously see the value change corresponding to epoch. 
2. A comment "The `tf.Variable` has been inadvertently replaced with a `tf.Tensor`." has been added.
3. A epoch indicator is added. `print("epoch:", epoch)`
4. The number of epoch is changed from 2 to 3.

REF: https://github.com/HsienChing/ML_DL_project_State_Estimation_of_Li-ion_Batteries/blob/main/other/Issues_in_TensorFlow_official_doc_Introduction_to_gradients_and_automatic_differentiation.ipynb

Author: HsienChing

site/en/guide/autodiff.ipynb

@@ -799,12 +799,14 @@
       "source": [
         "x = tf.Variable(2.0)\n",
         "\n",
-        "for epoch in range(2):\n",
+        "for epoch in range(3):\n",
         "  with tf.GradientTape() as tape:\n",
-        "    y = x+1\n",
+        "    y = x**2\n",
         "\n",
+        "  print("epoch:", epoch)\n,
         "  print(type(x).__name__, \":\", tape.gradient(y, x))\n",
-        "  x = x + 1   # This should be `x.assign_add(1)`"
+        "  x = x + 1   # The `tf.Variable` has been inadvertently replaced with a `tf.Tensor`."
+        "              # This should be `x.assign_add(1)`."
       ]
     },
     {