File size: 6,729 Bytes
ab80e91 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 |
"""
Universal Image Classification Models
Flexible architectures for any image classification task
"""
import torch
import torch.nn as nn
import torch.nn.functional as F
from typing import Optional
from config import ModelConfig
class ClassificationLoss(nn.Module):
"""Multi-class classification loss with label smoothing option"""
def __init__(self, label_smoothing: float = 0.0):
super().__init__()
self.label_smoothing = label_smoothing
def forward(self, logits: torch.Tensor, target: torch.LongTensor) -> torch.Tensor:
if self.label_smoothing > 0:
return F.cross_entropy(logits, target, label_smoothing=self.label_smoothing)
return F.cross_entropy(logits, target)
class LinearClassifier(nn.Module):
"""Simple linear baseline classifier"""
def __init__(self, config: Optional[ModelConfig] = None):
super().__init__()
if config is None:
config = ModelConfig()
self.config = config
in_features = 3 * config.input_height * config.input_width
self.linear = nn.Linear(in_features, config.num_classes)
def forward(self, x: torch.Tensor) -> torch.Tensor:
x = x.view(x.size(0), -1)
return self.linear(x)
class MLPClassifier(nn.Module):
"""MLP classifier with configurable architecture"""
def __init__(self, config: Optional[ModelConfig] = None):
super().__init__()
if config is None:
config = ModelConfig()
self.config = config
in_features = 3 * config.input_height * config.input_width
layers = []
layers.append(nn.Linear(in_features, config.hidden_dim))
if config.use_batch_norm:
layers.append(nn.BatchNorm1d(config.hidden_dim))
layers.append(nn.ReLU())
if config.dropout_rate > 0:
layers.append(nn.Dropout(config.dropout_rate))
layers.append(nn.Linear(config.hidden_dim, config.num_classes))
self.net = nn.Sequential(*layers)
def forward(self, x: torch.Tensor) -> torch.Tensor:
x = x.view(x.size(0), -1)
return self.net(x)
class MLPClassifierDeep(nn.Module):
"""Deep MLP classifier with multiple hidden layers"""
def __init__(self, config: Optional[ModelConfig] = None):
super().__init__()
if config is None:
config = ModelConfig()
self.config = config
in_features = 3 * config.input_height * config.input_width
layers = []
# Input layer
layers.append(nn.Linear(in_features, config.hidden_dim))
if config.use_batch_norm:
layers.append(nn.BatchNorm1d(config.hidden_dim))
layers.append(nn.ReLU())
if config.dropout_rate > 0:
layers.append(nn.Dropout(config.dropout_rate))
# Hidden layers
for _ in range(config.num_layers - 2):
layers.append(nn.Linear(config.hidden_dim, config.hidden_dim))
if config.use_batch_norm:
layers.append(nn.BatchNorm1d(config.hidden_dim))
layers.append(nn.ReLU())
if config.dropout_rate > 0:
layers.append(nn.Dropout(config.dropout_rate))
# Output layer
layers.append(nn.Linear(config.hidden_dim, config.num_classes))
self.net = nn.Sequential(*layers)
def forward(self, x: torch.Tensor) -> torch.Tensor:
x = x.view(x.size(0), -1)
return self.net(x)
class ResidualBlock(nn.Module):
"""Residual block for deep networks"""
def __init__(self, hidden_dim: int, use_batch_norm: bool = True, dropout_rate: float = 0.0):
super().__init__()
layers = []
layers.append(nn.Linear(hidden_dim, hidden_dim))
if use_batch_norm:
layers.append(nn.BatchNorm1d(hidden_dim))
layers.append(nn.ReLU())
if dropout_rate > 0:
layers.append(nn.Dropout(dropout_rate))
layers.append(nn.Linear(hidden_dim, hidden_dim))
if use_batch_norm:
layers.append(nn.BatchNorm1d(hidden_dim))
self.block = nn.Sequential(*layers)
self.relu = nn.ReLU()
def forward(self, x: torch.Tensor) -> torch.Tensor:
residual = x
out = self.block(x)
out = out + residual # Residual connection
return self.relu(out)
class MLPClassifierDeepResidual(nn.Module):
"""Deep MLP with residual connections"""
def __init__(self, config: Optional[ModelConfig] = None):
super().__init__()
if config is None:
config = ModelConfig()
self.config = config
in_features = 3 * config.input_height * config.input_width
# Input projection
self.input_layer = nn.Linear(in_features, config.hidden_dim)
self.input_bn = nn.BatchNorm1d(config.hidden_dim) if config.use_batch_norm else nn.Identity()
self.input_relu = nn.ReLU()
self.input_dropout = nn.Dropout(config.dropout_rate) if config.dropout_rate > 0 else nn.Identity()
# Residual blocks
self.residual_blocks = nn.ModuleList([
ResidualBlock(config.hidden_dim, config.use_batch_norm, config.dropout_rate)
for _ in range(config.num_layers - 2)
])
# Output layer
self.output_layer = nn.Linear(config.hidden_dim, config.num_classes)
def forward(self, x: torch.Tensor) -> torch.Tensor:
x = x.view(x.size(0), -1)
# Input projection
x = self.input_layer(x)
x = self.input_bn(x)
x = self.input_relu(x)
x = self.input_dropout(x)
# Residual blocks
for block in self.residual_blocks:
x = block(x)
# Output
return self.output_layer(x)
# Model factory for easy instantiation
MODEL_REGISTRY = {
"linear": LinearClassifier,
"mlp": MLPClassifier,
"mlp_deep": MLPClassifierDeep,
"mlp_deep_residual": MLPClassifierDeepResidual,
}
def create_model(model_name: str, config: Optional[ModelConfig] = None) -> nn.Module:
"""Factory function to create models"""
if model_name not in MODEL_REGISTRY:
raise ValueError(f"Unknown model: {model_name}. Available: {list(MODEL_REGISTRY.keys())}")
return MODEL_REGISTRY[model_name](config)
def count_parameters(model: nn.Module) -> int:
"""Count trainable parameters in a model"""
return sum(p.numel() for p in model.parameters() if p.requires_grad)
def calculate_model_size_mb(model: nn.Module) -> float:
"""Calculate model size in MB"""
return count_parameters(model) * 4 / 1024 / 1024 # 4 bytes per float32
|