44e8b6ae7d
Implement contrastive learning with LoRA to fine-tune CLIP's vision encoder on LogoDet-3K dataset for improved logo embedding similarity. New training module (training/): - config.py: TrainingConfig dataclass with all hyperparameters - dataset.py: LogoContrastiveDataset with logo-level splits - model.py: LogoFineTunedCLIP wrapper with LoRA support - losses.py: InfoNCE, TripletLoss, SupConLoss implementations - trainer.py: Training loop with mixed precision and checkpointing - evaluation.py: EmbeddingEvaluator for validation metrics New scripts: - train_clip_logo.py: Main training entry point - export_model.py: Export to HuggingFace-compatible format Configurations: - configs/jetson_orin.yaml: Optimized for Jetson Orin AGX - configs/cloud_rtx4090.yaml: Optimized for 24GB cloud GPUs - configs/cloud_a100.yaml: Optimized for 80GB cloud GPUs Documentation: - CLIP_FINETUNING.md: Training guide and usage instructions - CLOUD_TRAINING.md: Cloud GPU recommendations and cost estimates Modified: - logo_detection_detr.py: Add fine-tuned model loading support - pyproject.toml: Add peft, pyyaml, torchvision dependencies
25 lines
625 B
Python
25 lines
625 B
Python
"""
|
|
CLIP fine-tuning module for logo recognition.
|
|
|
|
This module provides tools for fine-tuning CLIP's vision encoder using
|
|
contrastive learning on the LogoDet-3K dataset.
|
|
"""
|
|
|
|
from .config import TrainingConfig
|
|
from .dataset import LogoContrastiveDataset, create_dataloaders
|
|
from .model import LogoFineTunedCLIP
|
|
from .losses import InfoNCELoss, TripletLoss
|
|
from .trainer import Trainer
|
|
from .evaluation import EmbeddingEvaluator
|
|
|
|
__all__ = [
|
|
"TrainingConfig",
|
|
"LogoContrastiveDataset",
|
|
"create_dataloaders",
|
|
"LogoFineTunedCLIP",
|
|
"InfoNCELoss",
|
|
"TripletLoss",
|
|
"Trainer",
|
|
"EmbeddingEvaluator",
|
|
]
|