Finetune mimic_finetune 25 epochs

llavamed_log1.txt

@@ -1,119 +0,0 @@
-{
-    "run": {
-        "task": "image_text_pretrain",
-        "lr_sched": "linear_warmup_cosine_lr",
-        "init_lr": 3e-05,
-        "min_lr": 1e-05,
-        "warmup_lr": 1e-06,
-        "weight_decay": 0.05,
-        "max_epoch": 20,
-        "iters_per_epoch": 600,
-        "batch_size_train": 15,
-        "batch_size_eval": 1,
-        "num_workers": 4,
-        "warmup_steps": 200,
-        "seed": 42,
-        "output_dir": "output/xraygpt_vqa_finetune",
-        "amp": true,
-        "resume_ckpt_path": null,
-        "evaluate": false,
-        "train_splits": [
-            "train"
-        ],
-        "device": "cuda",
-        "world_size": 2,
-        "dist_url": "env://",
-        "distributed": true,
-        "rank": 0,
-        "gpu": 0,
-        "dist_backend": "nccl"
-    },
-    "model": {
-        "arch": "mini_gpt4",
-        "image_size": 224,
-        "drop_path_rate": 0,
-        "use_grad_checkpoint": false,
-        "vit_precision": "fp16",
-        "freeze_vit": true,
-        "freeze_qformer": true,
-        "num_query_token": 32,
-        "llama_model": "Joycean0301/Llama-3.2-3B-Instruct-Medical-Conversational",
-        "prompt": "",
-        "lora_r": 16,
-        "lora_alpha": 16,
-        "lora_target_modules": [
-            "q_proj",
-            "k_proj",
-            "v_proj",
-            "o_proj",
-            "gate_proj",
-            "up_proj",
-            "down_proj"
-        ],
-        "lora_dropout": 0.1,
-        "lora_bias": "none",
-        "model_type": "pretrain_vicuna",
-        "max_txt_len": 160,
-        "end_sym": "<|eot_id|>",
-        "ckpt": "24_10_29_03_29_41_epoch160.pth"
-    },
-    "preprocess": {
-        "vis_processor": {
-            "train": {
-                "name": "blip2_image_train",
-                "image_size": 224
-            },
-            "eval": {
-                "name": "blip2_image_eval",
-                "image_size": 224
-            }
-        },
-        "text_processor": {
-            "train": {
-                "name": "blip_caption"
-            },
-            "eval": {
-                "name": "blip_caption"
-            }
-        }
-    },
-    "datasets": {
-        "vqa": {
-            "data_type": "images",
-            "build_info": {
-                "storage": "dataset/llava_med_60k_instruct"
-            },
-            "vis_processor": {
-                "train": {
-                    "name": "blip2_image_train",
-                    "image_size": 224
-                }
-            },
-            "text_processor": {
-                "train": {
-                    "name": "blip_caption"
-                }
-            }
-        }
-    }
-}
-{"train_lr": "0.000025", "train_loss": "0.941024"}
-{"train_lr": "0.000030", "train_loss": "0.685024"}
-{"train_lr": "0.000029", "train_loss": "0.622166"}
-{"train_lr": "0.000029", "train_loss": "0.580002"}
-{"train_lr": "0.000028", "train_loss": "0.544959"}
-{"train_lr": "0.000026", "train_loss": "0.512786"}
-{"train_lr": "0.000025", "train_loss": "0.484002"}
-{"train_lr": "0.000024", "train_loss": "0.454916"}
-{"train_lr": "0.000022", "train_loss": "0.427478"}
-{"train_lr": "0.000021", "train_loss": "0.401973"}
-{"train_lr": "0.000019", "train_loss": "0.378341"}
-{"train_lr": "0.000018", "train_loss": "0.353522"}
-{"train_lr": "0.000016", "train_loss": "0.331684"}
-{"train_lr": "0.000015", "train_loss": "0.311488"}
-{"train_lr": "0.000014", "train_loss": "0.291895"}
-{"train_lr": "0.000012", "train_loss": "0.274390"}
-{"train_lr": "0.000011", "train_loss": "0.257786"}
-{"train_lr": "0.000011", "train_loss": "0.243516"}
-{"train_lr": "0.000010", "train_loss": "0.229999"}
-{"train_lr": "0.000010", "train_loss": "0.219022"}