| echo "Running evaluation for RVE" | |
| cd evaluation | |
| revisit_eval_out_base="outputs_rve" | |
| all_setting_names=( | |
| "GeometryForcingREPA" | |
| ) | |
| all_gif_dirs=( | |
| "output/evaluations/{put_your_run_name_here}/wandb/latest-run/files/media/videos/prediction_vis" | |
| ) | |
| for i in "${!all_setting_names[@]}"; do | |
| setting_name=${all_setting_names[i]} | |
| gif_dir=${all_gif_dirs[i]} | |
| # Create log directory if it doesn't exist | |
| log_dir=${revisit_eval_out_base}/${setting_name} | |
| mkdir -p ${log_dir} | |
| temp_dir=${log_dir}/"video_vis" | |
| output_dir=${log_dir}/"eval" | |
| python eval_revisiting.py \ | |
| --temp_dir ${temp_dir} \ | |
| --output_dir ${output_dir} \ | |
| --gif_dir ${gif_dir} \ | |
| > ${log_dir}/log.txt 2>&1 | |
| done |