By Weijie Su, Xizhou Zhu, Yue Cao, Bin Li, Lewei Lu, Furu Wei, Jifeng Dai.
This repository is an official implementation of the paper VL-BERT: Pre-training of Generic Visual-Linguistic Representations.
Update on 2020/01/16 Add code of visualization.
Update on 2019/12/20 Our VL-BERT got accepted by ICLR 2020.
VL-BERT is a simple yet powerful pre-trainable generic representation for visual-linguistic tasks. It is pre-trained on the massive-scale caption dataset and text-only corpus, and can be fine-tuned for various down-stream visual-linguistic tasks, such as Visual Commonsense Reasoning, Visual Question Answering and Referring Expression Comprehension.
Thanks to PyTorch and its 3rd-party libraries, this codebase also contains following features:
@inproceedings{
Su2020VL-BERT:,
title={VL-BERT: Pre-training of Generic Visual-Linguistic Representations},
author={Weijie Su and Xizhou Zhu and Yue Cao and Bin Li and Lewei Lu and Furu Wei and Jifeng Dai},
booktitle={International Conference on Learning Representations},
year={2020},
url={https://openreview.net/forum?id=SygXPaEYvH}
}
# We recommend you to use Anaconda/Miniconda to create a conda environment
conda create -n vl-bert python=3.6 pip
conda activate vl-bert
conda install pytorch=1.1.0 cudatoolkit=9.0 -c pytorch
git clone https://github.com/jackroos/apex
cd ./apex
pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./
pip install Cython
pip install -r requirements.txt
./scripts/init.sh
See PREPARE_DATA.md.
See PREPARE_PRETRAINED_MODELS.md.
./scripts/dist_run_single.sh <num_gpus> <task>/train_end2end.py <path_to_cfg> <dir_to_store_checkpoint>
<num_gpus>
: number of gpus to use.<task>
: pretrain/vcr/vqa/refcoco.<path_to_cfg>
: config yaml file under ./cfgs/<task>
.<dir_to_store_checkpoint>
: root directory to store checkpoints.Following is a more concrete example:
./scripts/dist_run_single.sh 4 vcr/train_end2end.py ./cfgs/vcr/base_q2a_4x16G_fp32.yaml ./
For example, on 2 machines (A and B), each with 4 GPUs,
run following command on machine A:
./scripts/dist_run_multi.sh 2 0 <ip_addr_of_A> 4 <task>/train_end2end.py <path_to_cfg> <dir_to_store_checkpoint>
run following command on machine B:
./scripts/dist_run_multi.sh 2 1 <ip_addr_of_A> 4 <task>/train_end2end.py <path_to_cfg> <dir_to_store_checkpoint>
./scripts/nondist_run.sh <task>/train_end2end.py <path_to_cfg> <dir_to_store_checkpoint>
Note:
In yaml files under ./cfgs
, we set batch size for GPUs with at least 16G memory, you may need to adapt the batch size and
gradient accumulation steps according to your actual case, e.g., if you decrease the batch size, you should also
increase the gradient accumulation steps accordingly to keep 'actual' batch size for SGD unchanged.
For efficiency, we recommend you to use distributed training even on single-machine. But for RefCOCO+, you may meet deadlock using distributed training due to unknown reason (it may be related to PyTorch dataloader deadloack), you can simply use non-distributed training to solve this problem.
Local evaluation on val set:
python vcr/val.py \
--a-cfg <cfg_of_q2a> --r-cfg <cfg_of_qa2r> \
--a-ckpt <checkpoint_of_q2a> --r-ckpt <checkpoint_of_qa2r> \
--gpus <indexes_of_gpus_to_use> \
--result-path <dir_to_save_result> --result-name <result_file_name>
Note: <indexes_of_gpus_to_use>
is gpu indexes, e.g., 0 1 2 3
.
Generate prediction results on test set for leaderboard submission:
python vcr/test.py \
--a-cfg <cfg_of_q2a> --r-cfg <cfg_of_qa2r> \
--a-ckpt <checkpoint_of_q2a> --r-ckpt <checkpoint_of_qa2r> \
--gpus <indexes_of_gpus_to_use> \
--result-path <dir_to_save_result> --result-name <result_file_name>
python vqa/test.py \
--cfg <cfg_file> \
--ckpt <checkpoint> \
--gpus <indexes_of_gpus_to_use> \
--result-path <dir_to_save_result> --result-name <result_file_name>
python refcoco/test.py \
--split <val|testA|testB> \
--cfg <cfg_file> \
--ckpt <checkpoint> \
--gpus <indexes_of_gpus_to_use> \
--result-path <dir_to_save_result> --result-name <result_file_name>
See VISUALIZATION.md.
Many thanks to following codes that help us a lot in building this codebase: