Tensorboard: TensorBoard Error: Your CPU supports instructions that this TensorFlow binary was not compiled to use:AVX2

[vipulroxx]

I noticed that there is an earlier issue similar to this, but the issue was closed so I decided to make a new one since the suggestions weren't working.

...[previous code]
model.fit(data, labels, batch_size=32, epochs=5)

callbacks = [
    tf.keras.callbacks.EarlyStopping(patience=2, monitor='val_loss'),
    tf.keras.callbacks.TensorBoard(log_dir='./logs')
]
...[latter code]

My code did compile and it did create a log folder with profile directory and other files in the directory.

I ran this on the terminal(making sure that I am in the right path):

tensorboard --logdir=./logs

This is what I get as output

Vipuls-MacBook-Air:Keras vipulsharma$ tensorboard --logdir=./logs
2019-06-24 14:31:20.436096: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
TensorBoard 1.7.0 at http://Vipuls-Air:6006 (Press CTRL+C to quit)

When I run the address on the browser, it dosen't show anything.

[stephanwlee]

Hi @vipulroxx, the TensorFlow log (info) on CPU instruction can be ignored. What does your code look like and what is populated in the log_dir? How large is an event file in it? Also, when you say

it dosen't show anything

Do you mean it is totally blank?

[nfelt]

Can you just try going to http://localhost:6006 instead? It looks like your MacBook Air's hostname is not one that actually resolves in DNS.

[vipulroxx]

Thanks @stephanwlee for the help.

Thanks alot @nfelt . It wasn't resolving the DNS, and it worked with http://localhost:6006. You guys are great! Cheers :)