run_att.sh

#! /bin/sh

id="fc"
ckpt_path="log_"$id
if [ ! -d $ckpt_path ]; then
  mkdir $ckpt_path
fi
if [ ! -f $ckpt_path"/infos_"$id".pkl" ]; then
start_from=""
else
start_from="--start_from "$ckpt_path
fi

python train.py \
--id $id \
--caption_model fc \
--vse_model fc \
--share_embed 0 \
--input_json data/cocotalk.json \
--input_label_h5 data/cocotalk_label.h5 \
--input_fc_dir ../data/cocotalk_fc \
--input_att_dir ../data/cocobu_att \
--batch_size 128 \
--beam_size 3 \
--learning_rate 5e-4 \
--learning_rate_decay_start 0 \
--learning_rate_decay_every 15 \
--scheduled_sampling_start 0 \
--checkpoint_path $ckpt_path $start_from \
--save_checkpoint_every 3000 \
--language_eval 1 \
--val_images_use 5000 \
--max_epochs 100 \
--vse_loss_weight 0 \
--retrieval_reward_weight 0 \
--initialize_retrieval log_fc_con/model_vse-best.pth \
--cached_tokens coco-train2014-df-new