File size: 2,762 Bytes
26fd00c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 |
#!/usr/bin/env bash
# The port for communication. Note that if you want to run multiple tasks on the same machine,
# you need to specify different port numbers.
# Number of GPUs per GPU worker
export GPUS_PER_NODE=8
# Number of GPU workers, for single-worker training, please set to 1
export NUM_NODES=$SLURM_NNODES
# The ip address of the rank-0 worker, for single-worker training, please set to localhost
master_addr=$(scontrol show hostnames "$SLURM_JOB_NODELIST" | head -n 1)
export MASTER_ADDR=$master_addr
# The port for communication
export MASTER_PORT=12350
# The rank of this worker, should be in {0, ..., WORKER_CNT-1}, for single-worker training, please set to 0
export RANK=$SLURM_NODEID
echo "MASTER_ADDR: $MASTER_ADDR"
echo "RANK :$RANK"
echo "NUM_NODES :$NUM_NODES"
echo "GPUS_PER_NODE :$GPUS_PER_NODE"
export MIOPEN_USER_DB_PATH=/lus/home/NAT/gda2204/mshukor/.config/miopen_${MASTER_ADDR}_${SLURM_PROCID}/
echo "MIOPEN_USER_DB_PATH :$MIOPEN_USER_DB_PATH"
num_workers=0
ofa_dir=/lus/home/NAT/gda2204/mshukor/code/unival
base_data_dir=/lus/scratch/NAT/gda2204/SHARED/data
base_log_dir=/work/NAT/gda2204/mshukor/logs
bpe_dir=${ofa_dir}/utils/BPE
user_dir=${ofa_dir}/ofa_module
selected_cols=0,4,2,3
image_encoder_name=resnet #vit_base_patch16_224
acc_thresh='0.4,0.5,0.6,0.7,0.8,0.9'
metric=map
min_area_size=100000 # max 1000000
max_area_size=30000
for l in {0.00,0.20,0.40,0.60,0.80,1.00};do
new_base_log_dir=/lus/scratch/NAT/gda2204/SHARED/logs
exp_name=eval_refcocoplus_base_best_avg_postfuse_refvqa${l}
path=${new_base_log_dir}/ofa/pretrained_models/average_models/avg_postfuse_refvqa_l${l}.pt
echo ${path}
result_path=${new_base_log_dir}/ofa/results/refcocoplus/${exp_name}
mkdir ${result_path}
data=${base_data_dir}/ofa/refcocoplus_data/refcocoplus_val.tsv
split='refcocoplus_val'
python3 -m torch.distributed.launch \
--nnodes=${NUM_NODES} \
--nproc_per_node=${GPUS_PER_NODE} \
--master_port=${MASTER_PORT} \
--node_rank=${RANK} \
--master_addr=${MASTER_ADDR} \
--use_env ${ofa_dir}/evaluate.py \
${data} \
--path=${path} \
--user-dir=${user_dir} \
--task=refcoco \
--batch-size=16 \
--log-format=simple --log-interval=10 \
--seed=7 \
--gen-subset=${split} \
--results-path=${result_path} \
--beam=5 \
--min-len=4 \
--max-len-a=0 \
--max-len-b=4 \
--no-repeat-ngram-size=3 \
--fp16 \
--num-workers=0 \
--model-overrides="{\"data\":\"${data}\",\"bpe_dir\":\"${bpe_dir}\",\"selected_cols\":\"${selected_cols}\"}" \
--acc-thresh=${acc_thresh} \
--metric=${metric}
done |