pytorch embedding demo code

Here is a simple PyTorch embedding demo code:

import torch
import torch.nn as nn

# Define some input data
input_data = torch.LongTensor([[1, 2, 3], [4, 5, 6]])

# Define the embedding layer
embedding_layer = nn.Embedding(num_embeddings=10, embedding_dim=4)

# Pass the input data through the embedding layer
embedded_data = embedding_layer(input_data)

# Print the output
print(embedded_data)

In this code, we first define some input data as a LongTensor with shape (2, 3). We then define an embedding layer using nn.Embedding, specifying the number of embeddings (num_embeddings) and the dimension of each embedding (embedding_dim). We then pass the input_data through the embedding layer using embedding_layer(input_data), which returns the embedded_data with shape (2, 3, 4), where the last dimension corresponds to the embedding dimension. Finally, we print the output.