NLP Concepts Everyone Needs to Know (Part 3/5) | by Souradip Pal | Mar, 2025

Welcome back! In Part 2, we explored classical NLP models. Now, let’s dive into modern NLP techniques that have revolutionized language understanding. These include Recurrent Neural Networks, Transformers, and advanced architectures like BERT and GPT.

RNNs are designed to handle sequential data like text, speech, and time series by maintaining a hidden state that carries information from previous steps.

🔹 Example: Predicting the next word in “I love deep ___” using previous words.

import torch
import torch.nn as nn
class SimpleRNN(nn.Module):
def __init__(self, input_size, hidden_size, output_size):
super(SimpleRNN, self).__init__()
self.rnn = nn.RNN(input_size, hidden_size, batch_first=True)
self.fc = nn.Linear(hidden_size, output_size)
def forward(self, x):
out, _ = self.rnn(x)
out = self.fc(out[:, -1, :])
return out

LSTMs improve upon RNNs by avoiding the vanishing gradient problem and retaining information over longer sequences.