train_configs/224_minigpt4_llama2_image.yaml

model:
  arch: mini_gpt4_llama_v2
  model_type: pretrain_vicuna
  llama_model: "meta-llama/Llama-2-7b-chat-hf"
  max_txt_len: 160
  max_context_len: 512


datasets:
  laion:
    batch_size: 64
    vis_processor:
      train:
        name: "blip2_image_train"
        image_size: 224
    text_processor:
      train:
        name: "blip_caption"
    sample_ratio: 115
  cc_sbu:
    batch_size: 64
    vis_processor:
        train:
          name: "blip2_image_train"
          image_size: 224
    text_processor:
        train:
          name: "blip_caption"
    sample_ratio: 14


run:
  task: image_text_pretrain
  # optimizer
  lr_sched: "linear_warmup_cosine_lr"
  init_lr: 1e-4
  min_lr: 8e-5
  warmup_lr: 1e-6

  weight_decay: 0.05
  max_epoch: 4
  num_workers: 4
  warmup_steps: 5000
  iters_per_epoch: 5000

  seed: 42
  output_dir: "output/minigpt4_stage1_pretrain_llama2"

  amp: True
  resume_ckpt_path: null

  evaluate: False 
  train_splits: ["train"]

  device: "cuda"
  world_size: 1
  dist_url: "env://"
  distributed: True

  wandb_log: True
  job_name: minigpt4_llama2_pretrain