Category : ray

I have start ray as a head node in one machine: ray start –head –port=6379 And when I tried to connect another node to the head, I get this error message: :~/dev_root$ ray start –address=’10.50.131.250:6379′ –redis-password=’5241590000000000′ Local node IP: 10.50.131.67 [2021-10-20 21:28:14,052 C 18965 18965] global_state_accessor.cc:342: Failed to get system config within the timeout setting. ..

Read more

I am installing ray on a local server without internet connection. I downloaded the ray source from the github and run the command python setup.py install in the folder /ray/python, however, I get an error No such file of directory: ‘ray/_raylet.so’ I cannot find any files of this name in github. How should I solve ..

Read more

I have a big python project with many folders -model -utils -compute My ray remote code is some function in compute folder and I need to run in remote task code from model and util Currently, I’m getting errors no such module for different project folders from utils.osops import run_command from model.model_desc import ModelInsance @ray.remote ..

Read more

I’m trying to follow this tutorial to understand the basics of RLlib. I’m using pipenv on OS X to setup my environment with the following Pipfile: [[source]] name = "pypi" url = "https://pypi.python.org/simple" verify_ssl = true [dev-packages] [packages] ray = {extras = ["default", "rllib"], version = "*"} torch = "*" jupyterlab = "*" ipywidgets = ..

Read more

Can anyone help me speeding up this this loop with Ray or Multiprocessing. Trying to get as much help as possible so any tip or advice is welcome. Thanks, M. n_train = 180 n_records = len(X) pred = [] for i in range(n_train, n_records): train_model = MultiOutputClassifier(DecisionTreeClassifier(), n_jobs=-1).fit(X[1:i], y[1:i]) predict_model = train_model.predict(X[i:i+1]) pred.append(predict_model) print(predict_model) Source: ..

Read more

In an effort to make fitting multiple models more efficient, I have been trying to use all available CPU’s and/or parallelize the process. I found out that quite some sklearn functions support the n_jobs argument which allows for the use of all CPU cores. This is not available for all models and functions, especially when ..

Read more