After reviewing some online tutorials regarding Tensorboard, I've tried implementing it in my research notebook but I can't seem to get this to work.
I've set
%load_ext tensorboard
near the top of my notebook, but the callback in my model doesn't seem to trigger anything. I would expect a web page to pop up like what running the CLI command does:
lean research “My Notebook"
From what I've read, Tensorboard runs typically on port 6006. I tried installing a separate docker image following the instructions - it failed due to “ports already in use”, so I converted the command to:
docker run -it -p 6006:8888 tensorflow/tensorflow:nightly-py3-jupyter
Given the docker image script seems to want to expose 8888 internally, I mapped 6006 to it and it loads some sort of gateway UI that is asking for login credentials for some reason. I'm not at all familiar with Tensorboard or even if I've set things up correctly. Has anyone else successfully used Tensorboard with Lean CLI yet?
Jake Mitchell
I removed the additional docker image as I found out that tensorboard is installed with tensorflow. I verified this by putting the following into the notebook:
Known TensorBoard instances:
- port 6006: logdir logs/fit (started 0:07:10 ago; pid 19999)
Which indicates that tensorboard is running. However when I use:
I get the HTTP error “localhost refused to connect”. I think this might be due to the Docker port 6006 not being forwarded to the host system, but I'm unsure of how to configure that using the Lean CLI.
Varad Kabade
Hi Jake Mitchell
Unfortunately, the tensorboard GUI isn't supported within QC's cloud environment. However, it is possible to use the tensorboard GUI from a notebook deployed with Lean-CLI because tensorboard is a supported library. We recommend the following video explaining the installation of LEAN and this link describing how to run a research notebook. One work around for visualizing your model training within QC is to plot your training data using either the Plotting method or a plotting library like matplotlib.
Best,
Varad Kabade
Jake Mitchell
The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.
To unlock posting to the community forums please complete at least 30% of Boot Camp.
You can continue your Boot Camp training progress from the terminal. We hope to see you in the community soon!