mirror of
https://github.com/Vision-CAIR/MiniGPT-4.git
synced 2025-04-24 22:50:46 +00:00
56 lines
1.3 KiB
YAML
56 lines
1.3 KiB
YAML
|
# 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: blip2_vicuna_instruct
|
||
|
|
load_finetuned: False
|
||
|
|
load_pretrained: True
|
||
|
|
|
||
|
|
pretrained: "/mnt/pfs-guan-ssai/nlu/wanghanzi/models/blip2/blip2_vicuna7b/blip2_pretrained_vicuna7b.pth"
|
||
|
|
finetuned: ""
|
||
|
|
q_former_model: "/mnt/pfs-guan-ssai/nlu/wanghanzi/models/blip2/blip2_vicuna7b/blip2_pretrained_vicuna7b.pth"
|
||
|
|
|
||
|
|
# vit encoder
|
||
|
|
image_size: 224
|
||
|
|
drop_path_rate: 0
|
||
|
|
use_grad_checkpoint: False
|
||
|
|
vit_precision: "fp16"
|
||
|
|
|
||
|
|
# Q-Former
|
||
|
|
num_query_token: 32
|
||
|
|
qformer_text_input: True
|
||
|
|
|
||
|
|
# path to Vicuna checkpoint
|
||
|
|
llm_model: "/mnt/pfs-guan-ssai/nlu/wanghanzi/models/vicuna-7b-v1.1"
|
||
|
|
prompt: ""
|
||
|
|
max_txt_len: 256
|
||
|
|
max_output_txt_len: 256
|
||
|
|
|
||
|
|
# freeze
|
||
|
|
freeze_vit: True
|
||
|
|
freeze_llm: True
|
||
|
|
freeze_qformer: False
|
||
|
|
freeze_t5_proj: False
|
||
|
|
|
||
|
|
# moe
|
||
|
|
use_moeqformer: True
|
||
|
|
use_route_moe: True
|
||
|
|
moebert_expert_num: 5
|
||
|
|
moebert_num_beams: 2
|
||
|
|
|
||
|
|
preprocess:
|
||
|
|
vis_processor:
|
||
|
|
train:
|
||
|
|
name: "blip2_image_train"
|
||
|
|
image_size: 224
|
||
|
|
eval:
|
||
|
|
name: "blip_image_eval"
|
||
|
|
image_size: 224
|
||
|
|
text_processor:
|
||
|
|
train:
|
||
|
|
name: "blip_caption"
|
||
|
|
eval:
|
||
|
|
name: "blip_caption"
|