import torch
import torch.nn as nn
import numpy as np

# Positional encoding
class PositionalEncoding(nn.Module):
    def __init__(self, d_model, max_len=32):
        super().__init__()
        pe = torch.zeros(max_len, d_model)
        pos = torch.arange(0, max_len).unsqueeze(1)
        div = torch.exp(torch.arange(0, d_model, 2) * (-np.log(10000.0) / d_model))
        pe[:, 0::2] = torch.sin(pos * div)
        pe[:, 1::2] = torch.cos(pos * div)
        self.pe = pe.unsqueeze(0)

    def forward(self, x):
        return x + self.pe[:, :x.size(1)].to(x.device)

# Transformer emotion classifier
class EmotionTransformer(nn.Module):
    def __init__(self, vocab_size, embed_dim=64, num_heads=4, num_classes=5, dropout=0.3):
        super().__init__()
        self.embedding   = nn.Embedding(vocab_size, embed_dim, padding_idx=0)
        self.pos_encoder = PositionalEncoding(embed_dim)
        encoder_layer    = nn.TransformerEncoderLayer(
            d_model=embed_dim, nhead=num_heads, batch_first=True
        )
        self.transformer = nn.TransformerEncoder(encoder_layer, num_layers=2)
        self.dropout     = nn.Dropout(dropout)
        self.fc          = nn.Linear(embed_dim, num_classes)

    def forward(self, x):
        mask = (x == 0)
        x = self.embedding(x)
        x = self.pos_encoder(x)
        x = self.transformer(x, src_key_padding_mask=mask)
        x = self.dropout(x.mean(dim=1))
        return self.fc(x)