Output of the codes

[Neal2020GitHub]

Hi, authors!

May I ask where / how could I check the output of vlfm/run? It seems that there are only .hydra and an empty run.log inside the outputs directory. Are there some files that store evaluation matric, robot path and etc.

Many thanks!

[naokiyokoyama]

The run command should be printing information as the evaluation is progressing. When evaluation is completed, a summarization of the performance statistics should able be printed.

If you do not have access to an interactive session with the command running, the final metrics will also be stored to a log file. The path to this log file can be set using habitat_baselines.log_file.

More config options for habitat_baselines can be found here: https://github.com/facebookresearch/habitat-lab/blob/main/habitat-baselines/habitat_baselines/config/default_structured_configs.py

[Neal2020GitHub]

Thank you.

Found that only setting habitat_baselines.log_file cannot create a log file. Also need to call logger.add_filehandler(config.habitat_baselines.log_file) inside vlfm_trainer.py. Otherwise, logs are just printed in the terminal.

[PEACHTTT]

Thank you. 谢谢。

Found that only setting habitat_baselines.log_file cannot create a log file. Also need to call logger.add_filehandler(config.habitat_baselines.log_file) inside vlfm_trainer.py. Otherwise, logs are just printed in the terminal.发现仅设置 habitat_baselines.log_file 无法创建日志文件。还需要 logger.add_filehandler(config.habitat_baselines.log_file) 打电话进去 vlfm_trainer.py 。否则，日志只会在终端中打印。

Hi,I wonder how you fix the error "FileNotFoundError: Could not find dataset file data/datasets/objectnav/hm3d/v1/val" I'm confused with the measure"debugged" mentioned in another "Issues".