minigpt4 / train_configs /minigpt4_stage1_laion.yaml
Vision-CAIR's picture
Upload 89 files
181722d
# Copyright (c) 2022, salesforce.com, inc.
# All rights reserved.
# SPDX-License-Identifier: BSD-3-Clause
# For full license text, see the LICENSE file in the repo root or https://opensource.org/licenses/BSD-3-Clause
model:
arch: mini_gpt4
model_type: pretrain_vicuna
freeze_vit: True
freeze_qformer: True
datasets:
laion:
vis_processor:
train:
name: "blip2_image_train"
image_size: 224
text_processor:
train:
name: "blip_caption"
sample_ratio: 115
cc_combine:
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: 3e-5
warmup_lr: 1e-6
weight_decay: 0.05
max_epoch: 4
batch_size_train: 64
batch_size_eval: 64
num_workers: 4
warmup_steps: 5000
iters_per_epoch: 5000
seed: 42
output_dir: "/path/to/save/your/model/"
amp: True
resume_ckpt_path: null
evaluate: False
train_splits: ["train"]
device: "cuda"
world_size: 1
dist_url: "env://"
distributed: True