I am training a YOLO Nas s object detection model for a retail store,using 8000 images.

I have alrerady trained this model for 1500 images for 50 epochs. I have generated the checkpoint file, now how should I train this already trained model on another set of 2000 images.

Also how can I check the accuracy of my trained model using the generated checkpoint file(.pth)

I am holding on further training for now. To check the accuracy I tried generating bounding boxes on real time shelf images.

But I am expecting how can we check accuracy rate in number using the best_checkpoint.pth file

from super_gradients.training.dataloaders.dataloaders import coco_detection_yolo_format_train, coco_detection_yolo_format_val

BATCH_SIZE = 1
CLASSES = ['product']
CLASSES += [str(i) for i in range(80 - len(CLASSES))]

dataset_params = {
    'data_dir': r"C:UsersGirirajDocumentsPrernas ML ModelsSKU110KDatasetSKU110K_fixed",
    'train_images_dir':r'C:UsersGirirajDocumentsPrernas ML ModelsSKU110KDatasetSKU110K_fixedimagestrain',
    'train_labels_dir':r'C:UsersGirirajDocumentsPrernas ML ModelsSKU110KDatasetSKU110K_fixedlabelstrain',
    'val_images_dir':r'C:UsersGirirajDocumentsPrernas ML ModelsSKU110KDatasetSKU110K_fixedimagesval',
    'val_labels_dir':r'C:UsersGirirajDocumentsPrernas ML ModelsSKU110KDatasetSKU110K_fixedlabelsval',
    'test_images_dir':r'C:UsersGirirajDocumentsPrernas ML ModelsSKU110KDatasetSKU110K_fixedimagestest',
    'test_labels_dir':r'C:UsersGirirajDocumentsPrernas ML ModelsSKU110KDatasetSKU110K_fixedlabelstest',
    'classes': CLASSES
}

train_data = coco_detection_yolo_format_train(
    dataset_params={
        'data_dir': dataset_params['data_dir'],
        'images_dir': dataset_params['train_images_dir'],
        'labels_dir': dataset_params['train_labels_dir'],
        'classes': dataset_params['classes']
    },
    dataloader_params={
        'batch_size': BATCH_SIZE,
        'num_workers': 2
    }
)

val_data = coco_detection_yolo_format_val(
    dataset_params={
        'data_dir': dataset_params['data_dir'],
        'images_dir': dataset_params['val_images_dir'],
        'labels_dir': dataset_params['val_labels_dir'],
        'classes': dataset_params['classes']
    },
    dataloader_params={
        'batch_size': BATCH_SIZE,
        'num_workers': 2
    }
)

test_data = coco_detection_yolo_format_val(
    dataset_params={
        'data_dir': dataset_params['data_dir'],
        'images_dir': dataset_params['test_images_dir'],
        'labels_dir': dataset_params['test_labels_dir'],
        'classes': dataset_params['classes']
    },
    dataloader_params={
        'batch_size': BATCH_SIZE,
        'num_workers': 2
    }
)

import torch from super_gradients.training import models from super_gradients.training import Trainer

DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu' model = models.get('yolo_nas_s', pretrained_weights="coco").to(DEVICE) trainer = Trainer(experiment_name="SKU110K", ckpt_root_dir="./weights")

from super_gradients.training.losses import PPYoloELoss from super_gradients.training.metrics import DetectionMetrics_050 from super_gradients.training.models.detection_models.pp_yolo_e import PPYoloEPostPredictionCallback

MAX_EPOCHS = 20

train_params = {
    'silent_mode': False,
    "average_best_models":True,
    "warmup_mode": "linear_epoch_step",
    "warmup_initial_lr": 1e-6,
    "lr_warmup_epochs": 3,
    "initial_lr": 5e-4,
    "lr_mode": "cosine",
    "cosine_final_lr_ratio": 0.1,
    "optimizer": "Adam",
    "optimizer_params": {"weight_decay": 0.0001},
    "zero_weight_decay_on_bias_and_bn": True,
    "ema": True,
    "resume":True,
    "ema_params": {"decay": 0.9, "decay_type": "threshold"},
    "max_epochs": MAX_EPOCHS,
    "mixed_precision": True,
    "loss": PPYoloELoss(
        use_static_assigner=False,
        num_classes=len(dataset_params['classes']),
        reg_max=16
    ),
    "valid_metrics_list": [
        DetectionMetrics_050(
            score_thres=0.1,
            top_k_predictions=50,
            num_cls=len(dataset_params['classes']),
            normalize_targets=True,
            post_prediction_callback=PPYoloEPostPredictionCallback(
                score_threshold=0.01,
                nms_top_k=100,
                max_predictions=20,
                nms_threshold=0.7
            )
        )
    ],
    "metric_to_watch": '[email protected]'
}

for epoch in range(START_EPOCH, train_params['max_epochs']): 

trainer.train(model=model, training_params=train_params, train_loader=train_data, valid_loader=val_data)

# Save checkpoint after each epoch
torch.save({
    'epoch': epoch,
    'model_state_dict': model.state_dict(),
    'optimizer_state_dict': optimizer.state_dict(),
    # Add any other relevant information
}, r'C:UsersGirirajsourcereposnew trainingcheckpointJun21.pth')