das
das copied to clipboard
janclemenslab
AttributeError: module 'keras_tuner.oracles' has no attribute 'BayesianOptimization' when trying to run das.train_tune
Hi Jan,
I ran into an issue while trying to run the automatic architecture tuning using das.train_tune.
First, I got an error that keras-tuner was not installed. If it does not get installed as a requirement when installing/updating das, perhaps the doc can have a line with the install command?
After installing, I tried running it and got the following error. Is it a specific version of keras-tuner that's needed?
import das.train_tune
root_dir = '/data/das/v1'
data_dir = root_dir + '/dataset_v1'
dataset_dir = data_dir + '.npy'
models_dir = data_dir + '_kerastune.res'
model, params, tuner = das.train_tune.train(
data_dir=dataset_dir,
save_dir=models_dir,
tune_config="/code/3_das/tuning.yml",
log_messages=True,
)
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
Cell In[1], [line 8](vscode-notebook-cell:?execution_count=1&line=8)
[5](vscode-notebook-cell:?execution_count=1&line=5) dataset_dir = data_dir + '.npy'
[6](vscode-notebook-cell:?execution_count=1&line=6) models_dir = data_dir + '_kerastuner.res'
----> [8](vscode-notebook-cell:?execution_count=1&line=8) model, params, tuner = das.train_tune.train(
[9](vscode-notebook-cell:?execution_count=1&line=9) data_dir=dataset_dir,
[10](vscode-notebook-cell:?execution_count=1&line=10) save_dir=models_dir,
[11](vscode-notebook-cell:?execution_count=1&line=11) tune_config="/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/tuning.yml",
[12](vscode-notebook-cell:?execution_count=1&line=12) log_messages=True,
[13](vscode-notebook-cell:?execution_count=1&line=13) )
File ~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:432, in train(data_dir, x_suffix, y_suffix, save_dir, save_prefix, save_name, model_name, nb_filters, kernel_size, nb_conv, use_separable, nb_hist, ignore_boundaries, batch_norm, nb_pre_conv, pre_nb_dft, pre_kernel_size, pre_nb_filters, upsample, dilations, nb_lstm_units, verbose, batch_size, nb_epoch, learning_rate, reduce_lr, reduce_lr_patience, fraction_data, seed, batch_level_subsampling, augmentations, tensorboard, wandb_api_token, wandb_project, wandb_entity, log_messages, nb_stacks, with_y_hist, balance, version_data, tune_config, nb_tune_trials, _qt_progress)
[427](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:427) else:
[428](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:428) wandb = None
[430](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:430) tuner = DasTuner(
[431](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:431) params=params,
--> [432](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:432) oracle=kt.oracles.BayesianOptimization(
[433](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:433) objective=kt.Objective("val_loss", "min"),
[434](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:434) max_trials=nb_tune_trials,
[435](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:435) ),
[436](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:436) hypermodel=TunableModel(params, tune_config),
[437](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:437) overwrite=False,
[438](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:438) directory=save_dir,
[439](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:439) project_name=os.path.basename(save_name),
[440](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:440) tracker=wandb,
[441](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:441) )
[443](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:443) logger.info(tuner.search_space_summary())
[444](https://vscode-remote+ssh-002dremote-002besi-002dsvhubx86-002d01-002eesi-002elocal.vscode-resource.vscode-cdn.net/mnt/cs/projects/BWFAFdeactivNpx/code/3_das/~/.conda/envs/das_forge/lib/python3.10/site-packages/das/train_tune.py:444) utils.save_params(params, save_name)
AttributeError: module 'keras_tuner.oracles' has no attribute 'BayesianOptimization'
Also, I noticed the documentation has a lot of new and interesting additions, but I am not sure what the latest/stable version of das is - I seem to recall the installation page previously having this information, but I can't find it there. I did notice keras-tuner is now pyproject file.
INFO:das.cli: Linux-4.18.0-193.28.1.el8_2.x86_64-x86_64-with-glibc2.28
INFO:das.cli: DAS v0.32.4
INFO:das.cli: GUI is available.
INFO:das.cli: xarray-behave v0.35.7
INFO:das.cli: pyqtgraph v0.13.7
INFO:das.cli: PySide6 v6.7.2
INFO:das.cli: Qt v6.7.2
INFO:das.cli:
INFO:das.cli: tensorflow v2.15.0
INFO:das.cli: keras v2.15.0
INFO:das.cli: GPU is not available.
INFO:das.cli:
INFO:das.cli: python v3.10.15 | packaged by conda-forge | (main, Oct 16 2024, 01:24:24) [GCC 13.3.0]
INFO:das.cli: pandas v2.2.3
INFO:das.cli: numpy v1.26.4
INFO:das.cli: h5py v3.12.1
INFO:das.cli: scipy v1.14.1
INFO:das.cli: scikit-learn v1.5.2
INFO:das.cli: xarray v2024.10.0
Additionally, my env has the following keras-related packages:
keras 2.15.0 pyhd8ed1ab_0 conda-forge
keras-tuner 1.4.7 pypi_0 pypi
kernel-headers_linux-64 3.10.0 he073ed8_18 conda-forge
kt-legacy 1.0.5 pypi_0 pypi
Thanks!
