improve image classification readme

[chenhuims]

Please find a few suggestions for improving the readme file and conda yml file below:

• Should we put Frequently asked questions in the end as Notebooks and
  Getting Started may be more important to the user?
• It could be helpful to introduce a few applications of image classification in the beginning.
• "Demonstrates a trained model and inference an image from your computer's webcam." where"inference" should be "inferences".
• The numbering of the notebooks is a bit confusing to me. Should 01 be 0_1 and 11 be 1_1?
• For consistence, maybe change the title "Getting Started" to "Getting started".
• cd ComputerVisionBestPractices/image_classification
  should be
  cd ComputerVisionBestPractices/classification
• The versions of some packages in environment.yml are not specified, e.g., torchvision. It might be better to fix the versions if it's possible.

[vapaunic]

I found additionally a couple of small typos and re-words. Please see below:

Why is “required” in "These notebooks contain introductory "required" material." in classification/README.md in quotes. Is this material not required? Might be nice to follow up with a sentence, explaining this.

“While all notebooks can be executed in Windows, we have found the computer vision tools are often faster and more stable on the Linux operating system” – is this true or is it your preference?

We have also found that some browsers do not render Jupyter widgets correctly. If you have issues, try using an alternative browser, such as Chrome. – maybe also add Edge or Chrome.

Reword notebook descriptions in the table to be consistent. Some descriptions use “[this notebook] Demonstrates how to ….” and some use “[you] Use this notebook to ..”.