{ "model_args": { "model_name_or_path": "/home/panwen.hu/workspace1/jinxing.zhou/mllm/Crab/pretrained_weights/Llama-2-7b-chat-hf", "freeze_backbone": true, "llm_name": "llama", "vit_ckpt_path": "/home/panwen.hu/workspace1/jinxing.zhou/mllm/Crab/pretrained_weights/clip-vit-large-patch14", "select_feature": "patch", "image_size": 224, "patch_size": 14, "visual_query_token_nums": 32, "BEATs_ckpt_path": "/home/panwen.hu/workspace1/jinxing.zhou/mllm/Crab/pretrained_weights/BEATs_iter3_plus_AS2M_finetuned_on_AS2M_cpt2.pt", "audio_query_token_nums": 32, "prompt_embed_dim": 256, "mask_decoder_transformer_depth": 2, "low_res_mask_size": 112, "image_scale_nums": 2, "token_nums_per_scale": 3, "avs_query_num": 300, "num_classes": 1, "query_generator_num_layers": 2 }, "data_args": { "video_frame_nums": 10, "image_caption_task": false, "video_caption_task": false, "audio_caption_task": false, "segmentation_task": false, "avqa_task": false, "ave_task": false, "avvp_task": false, "arig_task": false, "ms3_task": false, "s4_task": false, "avss_task": false, "avcap_task": false, "ref_avs_task": true, "refavs_meta_csv_path": "../MQ-RAVSBench/train_test_meta_files/metadata.csv", "refavs_test_image_json_path": "../MQ-RAVSBench/train_test_meta_files/test_s_image_filtered.json", "refavs_test_u_image_json_path": "../MQ-RAVSBench/train_test_meta_files/test_u_image_filtered.json", "refavs_test_video_json_path": "../MQ-RAVSBench/train_test_meta_files/test_s_video_filtered.json", "refavs_test_u_video_json_path": "../MQ-RAVSBench/train_test_meta_files/test_u_video_filtered.json", "refavs_cot_json_path": "../MQ-RAVSBench/train_test_meta_files/train_audit_only_filtered.json", "refavs_data_root": "../MQ-RAVSBench", "refavs_eval_mode": "image", "refavs_mask_type_filter": "all", "refavs_mask_rank_filter": -1, "refavs_mask_encode_mode": "mask_and_masked_frame", "refavs_pos_ratio": 0.5, "multi_frames": false, "data_path": null, "model_max_length": 512 }, "training_args": { "output_dir": "results_epoch96/Auditonly_mask_and_masked_frame/epochs96_lr1e-4_bs4_gradacc8_lora_r32alpha64_pos0.5_ioulosswei0", "overwrite_output_dir": false, "do_train": false, "do_eval": false, "do_predict": false, "evaluation_strategy": "no", "prediction_loss_only": false, "per_device_train_batch_size": 4, "per_device_eval_batch_size": 4, "per_gpu_train_batch_size": null, "per_gpu_eval_batch_size": null, "gradient_accumulation_steps": 8, "eval_accumulation_steps": null, "eval_delay": 0, "learning_rate": 0.0001, "weight_decay": 0.0, "adam_beta1": 0.9, "adam_beta2": 0.999, "adam_epsilon": 1e-08, "max_grad_norm": 1.0, "num_train_epochs": 96.0, "max_steps": -1, "lr_scheduler_type": "cosine", "lr_scheduler_kwargs": {}, "warmup_ratio": 0.03, "warmup_steps": 0, "log_level": "passive", "log_level_replica": "warning", "log_on_each_node": true, "logging_dir": "results_epoch96/Auditonly_mask_and_masked_frame/epochs96_lr1e-4_bs4_gradacc8_lora_r32alpha64_pos0.5_ioulosswei0/runs/May18_17-20-46_gpu-17", "logging_strategy": "steps", "logging_first_step": false, "logging_steps": 1.0, "logging_nan_inf_filter": true, "save_strategy": "epoch", "save_steps": -1.0, "save_total_limit": 96, "save_safetensors": true, "save_on_each_node": false, "save_only_model": false, "no_cuda": false, "use_cpu": false, "use_mps_device": false, "seed": 42, "data_seed": null, "jit_mode_eval": false, "use_ipex": false, "bf16": true, "fp16": false, "fp16_opt_level": "O1", "half_precision_backend": "auto", "bf16_full_eval": false, "fp16_full_eval": false, "tf32": false, "local_rank": 0, "ddp_backend": null, "tpu_num_cores": null, "tpu_metrics_debug": false, "debug": [], "dataloader_drop_last": false, "eval_steps": null, "dataloader_num_workers": 4, "past_index": -1, "run_name": "results_epoch96/Auditonly_mask_and_masked_frame/epochs96_lr1e-4_bs4_gradacc8_lora_r32alpha64_pos0.5_ioulosswei0", "disable_tqdm": false, "remove_unused_columns": false, "label_names": null, "load_best_model_at_end": false, "metric_for_best_model": null, "greater_is_better": null, "ignore_data_skip": false, "fsdp": [], "fsdp_min_num_params": 0, "fsdp_config": { "min_num_params": 0, "xla": false, "xla_fsdp_grad_ckpt": false }, "fsdp_transformer_layer_cls_to_wrap": null, "deepspeed": "deepspeed/stage2-offload.json", "label_smoothing_factor": 0.0, "optim": "adamw_torch", "optim_args": null, "adafactor": false, "group_by_length": false, "length_column_name": "length", "report_to": [ "tensorboard", "wandb" ], "ddp_find_unused_parameters": true, "ddp_bucket_cap_mb": null, "ddp_broadcast_buffers": null, "dataloader_pin_memory": true, "dataloader_persistent_workers": false, "skip_memory_metrics": true, "use_legacy_prediction_loop": false, "push_to_hub": false, "resume_from_checkpoint": null, "hub_model_id": null, "hub_strategy": "every_save", "hub_token": null, "hub_private_repo": false, "hub_always_push": false, "gradient_checkpointing": true, "gradient_checkpointing_kwargs": null, "include_inputs_for_metrics": false, "fp16_backend": "auto", "push_to_hub_model_id": null, "push_to_hub_organization": null, "push_to_hub_token": null, "_n_gpu": 1, "mp_parameters": "", "auto_find_batch_size": false, "full_determinism": false, "torchdynamo": null, "ray_scope": "last", "ddp_timeout": 1800, "torch_compile": false, "torch_compile_backend": null, "torch_compile_mode": null, "dispatch_batches": null, "split_batches": false, "include_tokens_per_second": false, "include_num_input_tokens_seen": false, "neftune_noise_alpha": null, "mm_projector_lr": null, "freeze_mm_mlp_adapter": false, "cache_dir": null, "group_by_modality_length": false, "double_quant": true, "quant_type": "nf4", "bits": 32, "lora_enable": true, "lora_r": 32, "lora_alpha": 64, "lora_dropout": 0.05, "lora_weight_path": "", "lora_bias": "none", "ce_loss_weight": 1.0, "dice_loss_weight": 0.5, "bce_loss_weight": 1.0, "iou_loss_weight": 0.0, "audio_branch": true, "visual_branch": true, "seg_branch": false, "pretrain_ckpt_dir": "/home/panwen.hu/workspace1/jinxing.zhou/mllm/Crab/pretrained_weights", "finetune_ckpt_dir": "", "save_modules": "vl_projector,al_projector,lora", "exp_desc": "exp", "use_process": true, "use_hyper_lora": true } }