deepethogram

deepethogram copied to clipboard

Prediction on new videos

6

Dear deepethogram devs, I was wondering what would be the minimum steps required for, given a set of flow generator + feature extractor + sequence models already trained on my...

antortjim

TPUAccelerator can not run on your system since the accelerator is not available. The following accelerator(s) is available and can be passed into `accelerator` argument of `Trainer`: ['cpu'].

1

We get the following error when we start the training process. Has anyone experienced this or found a solution `pytorch_lightning.utilities.exceptions.MisconfigurationException: TPUAccelerator can not run on your system since the accelerator...

NaohikoOkabe

I'm beginning to train by running the feature extractor. I then receive the following error: RuntimeError: The expanded size of the tensor (5) must match the existing size (4) at...

jbwallace123

AssertionError: model path does not exist! during FlowGenerator [GOOGLE COLAB]

1

[2022-10-01 11:47:56,139] INFO [deepethogram.projects.convert_config_paths_to_absolute:1135] cwd in absolute: /content --------------------------------------------------------------------------- AssertionError Traceback (most recent call last) [](https://localhost:8080/#) in ----> 1 flow_generator = flow_generator_train(cfg) 2 frames [/usr/local/lib/python3.7/dist-packages/deepethogram/projects.py](https://localhost:8080/#) in convert_config_paths_to_absolute(project_cfg, raise_error_if_pretrained_missing) 1137 if...

AnanyaPassi-prog

PossibleUserWarning: Shuffling enabled in 'val_dataloader' during FlowGenerator Training

1

I started training the FlowGenerator on my videos. The training seems to be going fine - since several epochs are being completed. However, before each epoch I get the following...

AnanyaPassi-prog

``` [2022-09-29 23:45:59,066] INFO [deepethogram.gui.main.project_loaded_buttons:175] Number finalized labels: 3 [2022-09-29 23:46:19,448] INFO [deepethogram.gui.main.get_selected_models:1136] {'flow_generator': {'no pretrained weights': None, '200221_115158_TinyMotionNet': '/home/anuj/Desktop/ananya/test_deepethogram/models/pretrained_models/200221_115158_TinyMotionNet/checkpoint.pt', '220929_172218_flow_generator_train': '/home/anuj/Desktop/ananya/test_deepethogram/models/220929_172218_flow_generator_train/lightning_checkpoints/epoch=2-step=3000.ckpt'}, 'feature_extractor': {'no pretrained weights': None, '200415_125824_hidden_two_stream_kinetics_degf': '/home/anuj/Desktop/ananya/test_deepethogram/models/pretrained_models/200415_125824_hidden_two_stream_kinetics_degf/checkpoint.pt', '220929_194952_feature_extractor_train':...

sh00kti

The function projects.add_label_to_project() is broken. with a label file that does not have the 'background' column, it deletes the original first column (i.e., first behavior) and added a background column.

yihan0512

specify pytorch_lightning==1.6.5 https://github.com/jbohnslav/deepethogram/issues/111#issuecomment-1237118674

EvansKathryn

Lightning Trainer arguments 'reload_dataloaders_every_epoch' and 'progress_bar_refresh_rate' deprecated

3

[2022-08-10 14:46:36,281] INFO [__main__.flow_generator_train:63] split: reload: true file: null train_val_test: - 0.8 - 0.2 - 0.0 compute: fp16: false num_workers: 8 batch_size: 32 min_batch_size: 8 max_batch_size: 512 distributed: false gpu_id:...

drbcary

hi, i'm using GUI, but some functions aren't available for me (see the short cut). any suggestions? i already ran the flow generator and feature extractor. thank you very much!

wenyu-ding