I’m trying to run Tensorflow on a GCP jupyter notebook. The program is running and is generating log output. I’ve tried two approaches.
Approach 1:
LOG_DIR = '/tmp/log'
get_ipython().system_raw(
'tensorboard --logdir {} --host 0.0.0.0 --port 6006 &'
.format(LOG_DIR)
)
! wget https://bin.equinox.io/c/4VmDzA7iaHb/ngrok-stable-linux-amd64.zip
! unzip ngrok-stable-linux-amd64.zip
get_ipython().system_raw('./ngrok http 6006 &')
! curl -s http://localhost:4040/api/tunnels | python3 -c \
"import sys, json; print(json.load(sys.stdin)['tunnels'][0]['public_url'])"
The approach generated a url, and i got an error that says No dashboards are active for the current data set. I’m not sure why the directory didn’t update to the LOG_DIR path.
it still says: Data location: ./runs
Approach two:
run the below on the terminal
jupyter@pytorch-20200422-192725:~$ tensorboard --logdir runs --bind_all
2020-04-23 08:04:18.694994: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libnvinfer.so.6
2020-04-23 08:04:18.696468: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libnvinfer_plugin.so.6
TensorBoard 2.1.1 at http://pytorch-20200422-192725:6007/ (Press CTRL+C to quit)
I tried http://pytorch-20200422-192725:6007/ , http://[GCP internal IP ]:6007/ and http://[GCP external IP ]:6007/
all three gave me the same error: Hmm. We’re having trouble finding that site.
I have allow HTTP requests enabled behind the firewall.
how do I see the ltensorboard log outputs from a virtual machine?