neural networks

welcome to the neural networks section

The first place that you will be directed is here: https://cs.stanford.edu/people/karpathy/convnetjs

but there will be much more too.

natural language

For the purpose of natural language, we will look at this module: NLTK

And maybe start with this youtube tutorial: makemore

The toolkit is on GitHub: https://github.com/karpathy/makemore

The original dataset (around 32,000 names) is here: https://www.ssa.gov/oact/babynames

sorting

a quick note on sorting in python (a few methods):

This method avoids lambda functions:

def get_value(item): return item[1]

my_dict = {'a': 3, 'b': 1, 'c': 2}

sorted_dict = dict(sorted(my_dict.items(), key=get_value, reverse=True))

print(sorted_dict)

This method uses lambda functions:

my_dict = {'a': 3, 'b': 1, 'c': 2}

# Sort the dictionary by value using the items() method

sorted_dict = dict(sorted(my_dict.items(), key=lambda item: item[1], reverse=True))

print(sorted_dict)

torch

some requirements

• • pip install torch

  • pip install ipywidgets

test code:

import torch

# 5 rows, 3 columns

x = torch.rand(5, 3)

print(x)

The result is this:

tensor([[0.2297, 0.4190, 0.2671],

[0.5142, 0.1679, 0.4740],

[0.9949, 0.1533, 0.7456],

[0.3277, 0.4973, 0.2362],

[0.5808, 0.7602, 0.2213]])