File size: 16,456 Bytes
fdc673b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
import os
import time
import argparse
import torch
import torchaudio
import torchvision
from torch.utils.data import Dataset, DataLoader
from torch.utils.tensorboard import SummaryWriter
import numpy as np
from efficient_model import MobileNetGRUModel, EfficientNetCNNModel, SqueezeNetTransformerModel

# Print library version information
print(f"\033[92mINFO\033[0m: PyTorch version: {torch.__version__}")
print(f"\033[92mINFO\033[0m: Torchaudio version: {torchaudio.__version__}")
print(f"\033[92mINFO\033[0m: Torchvision version: {torchvision.__version__}")

# Device selection
device = torch.device(
    "cuda"
    if torch.cuda.is_available()
    else "mps" if torch.backends.mps.is_available() else "cpu"
)
print(f"\033[92mINFO\033[0m: Using device: {device}")

# Hyperparameters (using the best configuration from search)
batch_size = 4
epochs = 20
fc_hidden_size = 64
learning_rate = 0.0005
dropout_rate = 0.5

# Model save directory
os.makedirs("./models/", exist_ok=True)


class PreprocessedDataset(Dataset):
    def __init__(self, data_dir):
        self.data_dir = data_dir
        self.samples = [
            os.path.join(data_dir, f) for f in os.listdir(data_dir) if f.endswith(".pt")
        ]

    def __len__(self):
        return len(self.samples)

    def __getitem__(self, idx):
        sample_path = self.samples[idx]
        mfcc, image, label = torch.load(sample_path)
        return mfcc.float(), image.float(), label


def calculate_mae(outputs, labels):
    """Calculate Mean Absolute Error between outputs and labels"""
    return torch.abs(outputs - labels).mean().item()


def evaluate_model(model, test_loader, criterion):
    model.eval()
    test_loss = 0.0
    mae_sum = 0.0
    all_predictions = []
    all_labels = []
    
    # For debugging
    debug_samples = []
    
    with torch.no_grad():
        for mfcc, image, label in test_loader:
            mfcc, image, label = mfcc.to(device), image.to(device), label.to(device)
            output = model(mfcc, image)
            label = label.view(-1, 1).float()
            
            # Store debug samples (handling batch dimension properly)
            if len(debug_samples) < 5:
                # Extract individual samples from the batch
                for i in range(min(len(output), 5 - len(debug_samples))):
                    debug_samples.append((output[i][0].item(), label[i][0].item()))
            
            # Calculate MSE loss
            loss = criterion(output, label)
            test_loss += loss.item()
            
            # Calculate MAE
            mae = torch.abs(output - label).mean()
            mae_sum += mae.item()
            
            # Store predictions and labels for additional analysis
            all_predictions.extend(output.cpu().numpy())
            all_labels.extend(label.cpu().numpy())
    
    avg_loss = test_loss / len(test_loader)
    avg_mae = mae_sum / len(test_loader)
    
    # Convert to numpy arrays for easier analysis
    all_predictions = np.array(all_predictions).flatten()
    all_labels = np.array(all_labels).flatten()
    
    # Print debug samples
    print("\nDEBUG SAMPLES (Prediction, Label):")
    for i, (pred, label) in enumerate(debug_samples):
        print(f"Sample {i+1}: Prediction = {pred:.4f}, Label = {label:.4f}, Difference = {abs(pred-label):.4f}")
    
    return avg_loss, avg_mae, all_predictions, all_labels


def train_model(model_type):
    try:
        # Create model based on type
        if model_type == "mobilenet_gru":
            model = MobileNetGRUModel(
                gru_hidden_size=32, 
                gru_layers=1, 
                fc_hidden_size=fc_hidden_size, 
                dropout_rate=dropout_rate
            ).to(device)
            model_name = "MobileNetGRU"
        elif model_type == "efficientnet_cnn":
            model = EfficientNetCNNModel(
                fc_hidden_size=fc_hidden_size, 
                dropout_rate=dropout_rate
            ).to(device)
            model_name = "EfficientNetCNN"
        elif model_type == "squeezenet_transformer":
            model = SqueezeNetTransformerModel(
                nhead=4, 
                dim_feedforward=128, 
                fc_hidden_size=fc_hidden_size, 
                dropout_rate=dropout_rate
            ).to(device)
            model_name = "SqueezeNetTransformer"
        else:
            raise ValueError(f"Unknown model type: {model_type}")
        
        # Data loading
        data_dir = "./processed/"
        dataset = PreprocessedDataset(data_dir)
        n_samples = len(dataset)
        
        # Check label range
        all_labels = []
        for i in range(min(10, len(dataset))):
            _, _, label = dataset[i]
            all_labels.append(label)
        
        print("\nLABEL RANGE CHECK:")
        print(f"Sample labels: {all_labels}")
        print(f"Min label: {min(all_labels)}, Max label: {max(all_labels)}")
        
        train_size = int(0.7 * n_samples)
        val_size = int(0.2 * n_samples)
        test_size = n_samples - train_size - val_size

        train_dataset, val_dataset, test_dataset = torch.utils.data.random_split(
            dataset, [train_size, val_size, test_size]
        )

        train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
        val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)
        test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False)

        # Loss function and optimizer
        criterion = torch.nn.MSELoss()
        optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

        # TensorBoard
        writer = SummaryWriter(f"runs/{model_name}/")
        global_step = 0

        print(f"\033[92mINFO\033[0m: Training {model_name} model for {epochs} epochs")
        print(f"\033[92mINFO\033[0m: Training samples: {len(train_dataset)}")
        print(f"\033[92mINFO\033[0m: Validation samples: {len(val_dataset)}")
        print(f"\033[92mINFO\033[0m: Test samples: {len(test_dataset)}")
        print(f"\033[92mINFO\033[0m: Batch size: {batch_size}")
        print(f"\033[92mINFO\033[0m: Learning rate: {learning_rate}")
        print(f"\033[92mINFO\033[0m: Dropout rate: {dropout_rate}")

        best_val_loss = float('inf')
        best_model_path = None
        
        # Calculate model size
        model_size = sum(p.numel() for p in model.parameters()) / 1e6  # in millions
        print(f"\033[92mINFO\033[0m: Model parameters: {model_size:.2f}M")

        # Training loop
        for epoch in range(epochs):
            print(f"\033[92mINFO\033[0m: Training epoch ({epoch+1}/{epochs})")

            model.train()
            running_loss = 0.0
            running_mae = 0.0
            n_batches = 0
            
            start_time = time.time()
            
            try:
                for mfcc, image, label in train_loader:
                    mfcc, image, label = mfcc.to(device), image.to(device), label.to(device)

                    optimizer.zero_grad()
                    output = model(mfcc, image)
                    label = label.view(-1, 1).float()
                    loss = criterion(output, label)
                    loss.backward()
                    optimizer.step()

                    running_loss += loss.item()
                    running_mae += calculate_mae(output, label)
                    n_batches += 1
                    
                    writer.add_scalar("Training/Loss", loss.item(), global_step)
                    writer.add_scalar("Training/MAE", calculate_mae(output, label), global_step)
                    global_step += 1
            except Exception as e:
                print(f"\033[91mERR!\033[0m: {e}")
            
            epoch_time = time.time() - start_time

            # Validation phase
            model.eval()
            val_loss = 0.0
            val_mae = 0.0
            val_batches = 0
            
            with torch.no_grad():
                try:
                    for mfcc, image, label in val_loader:
                        mfcc, image, label = (
                            mfcc.to(device),
                            image.to(device),
                            label.to(device),
                        )
                        output = model(mfcc, image)
                        label = label.view(-1, 1).float()
                        
                        # Calculate loss
                        loss = criterion(output, label)
                        val_loss += loss.item()
                        
                        # Calculate MAE
                        val_mae += calculate_mae(output, label)
                        val_batches += 1
                except Exception as e:
                    print(f"\033[91mERR!\033[0m: {e}")

            avg_train_loss = running_loss / n_batches
            avg_train_mae = running_mae / n_batches
            avg_val_loss = val_loss / val_batches
            avg_val_mae = val_mae / val_batches
            
            # Record validation metrics
            writer.add_scalar("Validation/Loss", avg_val_loss, epoch)
            writer.add_scalar("Validation/MAE", avg_val_mae, epoch)

            print(
                f"Epoch [{epoch+1}/{epochs}], Time: {epoch_time:.2f}s, "
                f"Train Loss: {avg_train_loss:.4f}, Train MAE: {avg_train_mae:.4f}, "
                f"Val Loss: {avg_val_loss:.4f}, Val MAE: {avg_val_mae:.4f}"
            )

            # Save model checkpoint
            timestamp = time.strftime("%Y%m%d-%H%M%S")
            model_path = f"models/{model_name}_model_{epoch+1}_{timestamp}.pt"
            torch.save(model.state_dict(), model_path)
            
            # Save the best model based on validation loss
            if avg_val_loss < best_val_loss:
                best_val_loss = avg_val_loss
                best_model_path = model_path
                print(f"\033[92mINFO\033[0m: New best model saved with validation loss: {best_val_loss:.4f}")

            print(
                f"\033[92mINFO\033[0m: Model checkpoint epoch [{epoch+1}/{epochs}] saved: {model_path}"
            )

        print(f"\033[92mINFO\033[0m: Training complete")
        
        # Load the best model for testing
        print(f"\033[92mINFO\033[0m: Loading best model from {best_model_path} for testing")
        model.load_state_dict(torch.load(best_model_path))
        
        # Evaluate on test set
        test_loss, test_mae, predictions, labels = evaluate_model(model, test_loader, criterion)
        
        # Calculate additional metrics
        max_error = np.max(np.abs(predictions - labels))
        min_error = np.min(np.abs(predictions - labels))
        
        print("\n" + "="*50)
        print(f"TEST RESULTS FOR {model_name}:")
        print(f"Test Loss (MSE): {test_loss:.4f}")
        print(f"Mean Absolute Error: {test_mae:.4f}")
        print(f"Maximum Absolute Error: {max_error:.4f}")
        print(f"Minimum Absolute Error: {min_error:.4f}")
        
        # Add test results to TensorBoard
        writer.add_scalar("Test/MSE", test_loss, 0)
        writer.add_scalar("Test/MAE", test_mae, 0)
        writer.add_scalar("Test/Max_Error", max_error, 0)
        writer.add_scalar("Test/Min_Error", min_error, 0)
        
        # Create a histogram of absolute errors
        abs_errors = np.abs(predictions - labels)
        writer.add_histogram("Test/Absolute_Errors", abs_errors, 0)
        
        print("="*50)
        
        # Final summary
        print("\nTRAINING SUMMARY:")
        print(f"Model: {model_name}")
        print(f"Model Size: {model_size:.2f}M parameters")
        print(f"Best Validation Loss: {best_val_loss:.4f}")
        print(f"Final Test Loss: {test_loss:.4f}")
        print(f"Final Test MAE: {test_mae:.4f}")
        print(f"Best model saved at: {best_model_path}")
        
        writer.close()
        
        # Return metrics for comparison
        return {
            "model_name": model_name,
            "model_size": model_size,
            "val_loss": best_val_loss,
            "test_loss": test_loss,
            "test_mae": test_mae,
            "model_path": best_model_path
        }

    except Exception as e:
        print(f"\033[91mERR!\033[0m: Error training {model_type}: {e}")
        # Return a placeholder result
        return {
            "model_name": model_type,
            "model_size": 0,
            "val_loss": float('inf'),
            "test_loss": float('inf'),
            "test_mae": float('inf'),
            "model_path": None,
            "error": str(e)
        }


def test_cpu_inference(model_path, model_type):
    """Test CPU inference speed for the given model"""
    # Create model based on type
    if model_type == "mobilenet_gru":
        model = MobileNetGRUModel(
            gru_hidden_size=32, 
            gru_layers=1, 
            fc_hidden_size=fc_hidden_size, 
            dropout_rate=dropout_rate
        )
        model_name = "MobileNetGRU"
    elif model_type == "efficientnet_cnn":
        model = EfficientNetCNNModel(
            fc_hidden_size=fc_hidden_size, 
            dropout_rate=dropout_rate
        )
        model_name = "EfficientNetCNN"
    elif model_type == "squeezenet_transformer":
        model = SqueezeNetTransformerModel(
            nhead=4, 
            dim_feedforward=128, 
            fc_hidden_size=fc_hidden_size, 
            dropout_rate=dropout_rate
        )
        model_name = "SqueezeNetTransformer"
    else:
        raise ValueError(f"Unknown model type: {model_type}")
    
    # Load model weights
    model.load_state_dict(torch.load(model_path, map_location=torch.device('cpu')))
    model.eval()
    
    # Create dummy input
    dummy_mfcc = torch.randn(1, 10, 376)  # Batch size 1, 10 time steps, 376 features
    dummy_image = torch.randn(1, 3, 224, 224)  # Batch size 1, 3 channels, 224x224 image
    
    # Warm-up
    for _ in range(10):
        _ = model(dummy_mfcc, dummy_image)
    
    # Measure inference time
    num_runs = 100
    start_time = time.time()
    for _ in range(num_runs):
        _ = model(dummy_mfcc, dummy_image)
    end_time = time.time()
    
    avg_time = (end_time - start_time) / num_runs
    
    print(f"\n{model_name} CPU Inference Time:")
    print(f"Average over {num_runs} runs: {avg_time*1000:.2f} ms")
    
    return avg_time


if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="Train and evaluate efficient models")
    parser.add_argument(
        "--model", 
        type=str, 
        choices=["mobilenet_gru", "efficientnet_cnn", "squeezenet_transformer", "all"],
        default="all",
        help="Model architecture to train"
    )
    args = parser.parse_args()
    
    results = []
    
    if args.model == "all":
        # Train all models
        for model_type in ["mobilenet_gru", "efficientnet_cnn", "squeezenet_transformer"]:
            print(f"\n\n{'='*50}")
            print(f"TRAINING {model_type.upper()}")
            print(f"{'='*50}\n")
            result = train_model(model_type)
            results.append(result)
            
            # Test CPU inference
            inference_time = test_cpu_inference(result["model_path"], model_type)
            result["inference_time"] = inference_time
    else:
        # Train specific model
        result = train_model(args.model)
        results.append(result)
        
        # Test CPU inference
        inference_time = test_cpu_inference(result["model_path"], args.model)
        result["inference_time"] = inference_time
    
    # Compare results
    print("\n\n" + "="*80)
    print("MODEL COMPARISON")
    print("="*80)
    print(f"{'Model':<25} {'Size (M)':<10} {'Val Loss':<10} {'Test Loss':<10} {'Test MAE':<10} {'CPU Time (ms)':<15}")
    print("-"*80)
    
    for result in results:
        print(f"{result['model_name']:<25} {result['model_size']:<10.2f} {result['val_loss']:<10.4f} "
              f"{result['test_loss']:<10.4f} {result['test_mae']:<10.4f} {result['inference_time']*1000:<15.2f}")
    
    print("="*80)
    
    # Find best model
    best_model = min(results, key=lambda x: x["test_mae"])
    print(f"\nBEST MODEL: {best_model['model_name']}")
    print(f"Test MAE: {best_model['test_mae']:.4f}")
    print(f"CPU Inference Time: {best_model['inference_time']*1000:.2f} ms")
    print(f"Model Path: {best_model['model_path']}")