| nohup python3 run_image_captioning_flax.py --dataset_name image_caption_dataset.py --dataset_config_name=coco_2017 --data_dir ./coco-data --max_target_length 16 --num_beams 4 --output_dir outputs --do_train --per_device_train_batch_size 4 --do_eval --per_device_eval_batch_size 4 --max_train_samples 16 --max_eval_samples 16 --max_predict_samples 16 --preprocessing_num_workers 1 --num_train_epochs 4 --predict_with_generate --learning_rate 1e-5 --encoder_model_name_or_path ./encoder-decoder/encoder/ --decoder_model_name_or_path ./encoder-decoder/decoder/ --model_type vision-encoder-decoder & |