Extended discritions of usage of tensor classes

Author: IlyaKisil

1_N-dimensional_arrays_and_Tensor_class.ipynb

@@ -5,7 +5,7 @@
  "metadata": {},
  "source": [
  "# Multi-dimensional arrays and **Tensor** class\n",
- "### Last modification (05.04.2018).\n",
+ "### Last modification (14.04.2018).\n",
  "\n",
  "In this tutorial we will show the core data structures of multidimenaional arrays within tensor algebra and illustrate how they are integrated into [hottbox](https://github.com/hottbox/hottbox). For more details visit the [documentation page](https://hottbox.github.io/stable/api/hottbox.core.html#module-hottbox.core).\n",
  "\n",
@@ -40,8 +40,6 @@
  "\n",
  "In order to create tensor using ``hottbox``, you simply need to pass numpy ndarray to the constructor of the ``Tensor`` class. \n",
  "\n",
- "> **Note:** the ND-array is stored in the attribute `data`.\n",
- "\n",
  "This will allow you to use top level API for the most common properties and operations on the tensor itself that correspond to the conventional definitions. \n",
  "\n",
  "> **Note:** In order to be consistent with python indexing, count of modes starts from zeros."
@@ -548,7 +546,7 @@
  "cell_type": "markdown",
  "metadata": {},
  "source": [
- "Here, despite the **`Y`** and **`Z`** are being different objects, their attribute **`data`** will have the same values."
+ "Here, despite the **`Y`** and **`Z`** are being different objects, their data values will remaain the same."
  ]
  },
  {
@@ -647,6 +645,19 @@
  "source": [
  "As expected, the result remains the same."
  ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Additional notes on API of **Tensor** class\n",
+ "\n",
+ "1. When object of **`Tensor`** class is created, the data values are stored in **`_data`** attribute and should only be accessed by calling the corresponding property **`data`**, as has been shown previously. \n",
+ "\n",
+ "- If you want to modify these values, then call the corresponding transformation method available for the **`Tensor`** class.\n",
+ "\n",
+ "- In order to create a duplicate of an object of **`Tensor`** class then use method **`copy()`**"
+ ]
  }
  ],
  "metadata": {
@@ -665,7 +676,7 @@
  "name": "python",
  "nbconvert_exporter": "python",
  "pygments_lexer": "ipython3",
- "version": "3.6.5"
+ "version": "3.6.2"
  }
  },
  "nbformat": 4,