Dashboard not supported on Windows. Disable the dashboard and save the finalized study instead.How do I tackle this prblem
michaelGK92 opened this issue · 5 comments
the code is from your wiki.I test it on my python.when I use it in windows, it does not work.
the random forest.py
`from sklearn.datasets import load_breast_cancer
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import cross_val_score
import time
import sherpa
import sherpa.algorithms.bayesian_optimization as bayesian_optimization
parameters = [sherpa.Discrete('n_estimators', [2, 50]),
sherpa.Choice('criterion', ['gini', 'entropy']),
sherpa.Continuous('max_features', [0.1, 0.9])]
algorithm = bayesian_optimization.GPyOpt(max_concurrent=1,
model_type='GP_MCMC',
acquisition_type='EI_MCMC',
max_num_trials=100)
X, y = load_breast_cancer(return_X_y=True)
study = sherpa.Study(parameters=parameters,
algorithm=algorithm,
lower_is_better=False)
for trial in study:
print("Trial ", trial.id, " with parameters ", trial.parameters)
clf = RandomForestClassifier(criterion=trial.parameters['criterion'],
max_features=trial.parameters['max_features'],
n_estimators=trial.parameters['n_estimators'],
random_state=0)
scores = cross_val_score(clf, X, y, cv=5)
print("Score: ", scores.mean())
study.add_observation(trial, iteration=1, objective=scores.mean())
study.finalize(trial)
print(study.get_best_result())
Hi @michaelGK92 ,
Try this code (inside sherpa.Study I added disable_dashboard=True).
from sklearn.datasets import load_breast_cancer
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import cross_val_score
import time
import sherpa
import sherpa.algorithms.bayesian_optimization as bayesian_optimization
parameters = [sherpa.Discrete('n_estimators', [2, 50]),
sherpa.Choice('criterion', ['gini', 'entropy']),
sherpa.Continuous('max_features', [0.1, 0.9])]
algorithm = bayesian_optimization.GPyOpt(max_concurrent=1,
model_type='GP_MCMC',
acquisition_type='EI_MCMC',
max_num_trials=100)
X, y = load_breast_cancer(return_X_y=True)
study = sherpa.Study(parameters=parameters,
algorithm=algorithm,
disable_dashboard=True,
lower_is_better=False)
for trial in study:
print("Trial ", trial.id, " with parameters ", trial.parameters)
clf = RandomForestClassifier(criterion=trial.parameters['criterion'],
max_features=trial.parameters['max_features'],
n_estimators=trial.parameters['n_estimators'],
random_state=0)
scores = cross_val_score(clf, X, y, cv=5)
print("Score: ", scores.mean())
study.add_observation(trial, iteration=1, objective=scores.mean())
study.finalize(trial)
print(study.get_best_result())
study.save(".")
Then later you can do
import sherpa
sherpa.Study.load_dashboard(".")
from the same directory.
I installed ubuntu app on my windows, installed python and copied the file:
sudo cp -i /mnt/c/Users/.... /xyz.py
I run the code from ubuntu.
python3 xyz.py
my file includes:
import tempfile
model_dir = tempfile.mkdtemp()
...
study.save(model_dir)
print (model_dir)
opening the file e.g.:
import sherpa
sherpa.Study.load_dashboard("/tmp/tmpja_m1w61")
removing all tmp files:
cd /tmp
find . -type d -name 'tmp*' -exec rm -r {} ; -prune