| { | |
| "architectures": [ | |
| "MambaVision" | |
| ], | |
| "args": { | |
| "ThreeAugment": false, | |
| "aa": "rand-m9-mstd0.5-inc1", | |
| "activation_tracker": false, | |
| "amp": true, | |
| "ampere_sparsity": false, | |
| "aot_autograd": false, | |
| "apex_amp": false, | |
| "attn_drop_rate": 0, | |
| "aug_repeats": 0, | |
| "aug_splits": 0, | |
| "batch_size": 128, | |
| "bce_loss": false, | |
| "bce_target_thresh": null, | |
| "bn_eps": null, | |
| "bn_momentum": null, | |
| "channels_last": true, | |
| "checkpoint_hist": 1, | |
| "class_map": "", | |
| "clip_grad": 5, | |
| "clip_mode": "norm", | |
| "color_jitter": 0.4, | |
| "cooldown_epochs": 10, | |
| "crop_pct": 1, | |
| "cutmix": 1, | |
| "cutmix_minmax": null, | |
| "data_dir": "/datasets/imagenet_lmdb", | |
| "data_len": 1281167, | |
| "dataset": "", | |
| "dataset_download": false, | |
| "decay_epochs": 100, | |
| "decay_milestones": [ | |
| 30, | |
| 60 | |
| ], | |
| "decay_rate": 0.1, | |
| "dist_bn": "reduce", | |
| "drop_block": null, | |
| "drop_connect": null, | |
| "drop_path": null, | |
| "drop_rate": 0, | |
| "epoch_repeats": 0, | |
| "epochs": 310, | |
| "eval_metric": "top1", | |
| "experiment": "", | |
| "fuser": "", | |
| "gp": null, | |
| "grad_checkpointing": false, | |
| "hflip": 0.5, | |
| "img_size": null, | |
| "initial_checkpoint": "", | |
| "input_size": [ | |
| 3, | |
| 224, | |
| 224 | |
| ], | |
| "interpolation": "", | |
| "jsd_loss": false, | |
| "layer_decay": null, | |
| "checkpoint": "", | |
| "local_rank": 0, | |
| "log_dir": "./log_dir/", | |
| "log_interval": 50, | |
| "log_wandb": false, | |
| "lr": 0.005, | |
| "lr_cycle_decay": 1, | |
| "lr_cycle_limit": 1, | |
| "lr_cycle_mul": 1, | |
| "lr_ep": false, | |
| "lr_k_decay": 1, | |
| "lr_noise": null, | |
| "lr_noise_pct": 0.67, | |
| "lr_noise_std": 1, | |
| "mean": null, | |
| "mesa": 0.5, | |
| "mesa_start_ratio": 0.25, | |
| "min_lr": 0.000005, | |
| "mixup": 0.8, | |
| "mixup_mode": "batch", | |
| "mixup_off_epoch": 0, | |
| "mixup_prob": 1, | |
| "mixup_switch_prob": 0.5, | |
| "model": "mamba_vision_L2", | |
| "model_ema": true, | |
| "model_ema_decay": 0.9998, | |
| "model_ema_force_cpu": false, | |
| "momentum": 0.9, | |
| "native_amp": false, | |
| "no_aug": false, | |
| "no_ddp_bb": false, | |
| "no_prefetcher": false, | |
| "no_resume_opt": false, | |
| "no_saver": false, | |
| "num_classes": null, | |
| "opt": "lamb", | |
| "opt_betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "opt_eps": 1e-8, | |
| "output": "", | |
| "patience_epochs": 10, | |
| "pin_mem": false, | |
| "pretrained": false, | |
| "ratio": [ | |
| 0.75, | |
| 1.3333333333333333 | |
| ], | |
| "recount": 1, | |
| "recovery_interval": 0, | |
| "remode": "pixel", | |
| "reprob": 0.25, | |
| "resplit": false, | |
| "resume": "", | |
| "save_images": false, | |
| "scale": [ | |
| 0.08, | |
| 1 | |
| ], | |
| "sched": "cosine", | |
| "seed": 31, | |
| "smoothing": 0.1, | |
| "split_bn": false, | |
| "start_epoch": null, | |
| "std": null, | |
| "sync_bn": false, | |
| "tag": "mambavision_large2_1k", | |
| "torchscript": false, | |
| "train_interpolation": "random", | |
| "train_split": "train", | |
| "tta": 0, | |
| "use_multi_epochs_loader": false, | |
| "val_split": "validation", | |
| "validate_only": false, | |
| "validation_batch_size": null, | |
| "vflip": 0, | |
| "warmup_epochs": 20, | |
| "warmup_lr": 0.000001, | |
| "weight_decay": 0.05, | |
| "worker_seeding": "all", | |
| "workers": 8 | |
| }, | |
| "auto_map": { | |
| "AutoConfig": "hf_model.MambaVisionConfig", | |
| "AutoModel": "hf_model.MambaVisionModel" | |
| } | |
| } |