search

jchengai / forecast-mae

[ICCV'2023] Forecast-MAE: Self-supervised Pre-training for Motion Forecasting with Masked Autoencoders
https://arxiv.org/pdf/2308.09882.pdf
154 stars 16 forks source link

An error when running 'train.py' #9

Closed kelen1 closed 8 months ago

kelen1 commented 8 months ago

Hey! When I try to run 'train.py', an error is reported, the detailed information is below:

Error executing job with overrides: ['data_root=/root/autodl-tmp/data_root', 'model=model_forecast', 'batch_size=16', 'monitor=val_minFDE'] Error locating target 'src.model.trainer_forecast.Trainer', set env var HYDRA_FULL_ERROR=1 to see chained exception. full_key: model.target

I try to find the resaon, but no results. Had anyone had a similar issue? Sincerely for help. Thanks!

jchengai commented 8 months ago

Hi @kelen1, can you paste your command that raises this error?

kelen1 commented 8 months ago

The command is below: root@autodl-container-a1e043b06b-1dc7b9a9:~/autodl-tmp# python3 train.py data_root=/root/autodl-tmp/data_root model=model_forecast batch_size=16 monitor=val_minFDE

The detailed error information is below: /root/miniconda3/lib/python3.8/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'config': Defaults list is missing _self_. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information warnings.warn(msg, UserWarning) Seed set to 2333 Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.rich_model_summary.RichModelSummary'>]. Skipping setting a default ModelSummary callback. GPU available: True (cuda), used: True TPU available: False, using: 0 TPU cores IPU available: False, using: 0 IPUs HPU available: False, using: 0 HPUs conf.model.target {'target': 'src.model.trainer_forecast.Trainer', 'dim': 128, 'historical_steps': 50, 'future_steps': 60, 'encoder_depth': 4, 'num_heads': 8, 'mlp_ratio': 4.0, 'qkv_bias': False, 'drop_path': 0.2, 'pretrained_weights': '${pretrained_weights}', 'lr': '${lr}', 'weight_decay': '${weight_decay}', 'epochs': '${epochs}', 'warmup_epochs': '${warmup_epochs}'} Error executing job with overrides: ['data_root=/root/autodl-tmp/data_root', 'model=model_forecast', 'batch_size=16', 'monitor=val_minFDE'] Error locating target 'src.model.trainer_forecast.Trainer', set env var HYDRA_FULL_ERROR=1 to see chained exception. full_key: model.target

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.

jchengai commented 8 months ago

I see, I'm investing in this issue.

jchengai commented 8 months ago

Hi @kelen1, I think the problem is that there are some other projects in your PYTHONPATH that have similar folder structure project/src/model. Can you check your PYTHONPATH by echo $PYTHONPATH?

kelen1 commented 8 months ago

Hi, @jchengai. It seems to be empty...

kelen1 commented 8 months ago

I downloaded the code and set the environment again, the problem seemed to be solved. Thanks!