Closed ss2cp closed 5 years ago
Hey mate, let me try to clarify things for you
I only tested in ubuntu, the mac.yml is a contribution from a mac user.
In train_multi_model.py
the following line is missing:
from configs.vars import *
Yeah you need to create a tensorboard
folder and update "local_dir" to this recent create tensorboard folder (u do this if you want to save the checkpoints in the project root folder, if not, use the number 4 idea)
If you don't use the "local_dir"
line your checkpoints will be saved at ~/ray_results/[EXPERIMENT_NAME]
by default.
I know the project is a lil messy, my wish now is to organize it but I'm working in something else and I have no time at the moment, so fell free to contribute if you want and any question let me know.
Thanks :smile:
thanks for the catch
Hi Lucas!
I found your repo from Adam King's project. First of all, great work and respect.
While my mac is training, there is a few issues I ran into.
When installing env with conda on my mac,
matplotlib=2.2.3=py35h0e0179f_0
cannot be found. I removed the last partpy35h ...
and it fixed the problem. So, my question is: did you test this in ubuntu env, or mac?I tried to train multi model as instructed on
readme
,python train_multi_model.py
gave me error not findingSYMBOL_1
(fromvars.py
) I looked into what might caused this, but I could not find where you pass in theSYMBOL
s fromvars.py
any idea how to fix this? or I did something wrongWhile I was looking into
train_multi_model.py
, I noticed inexperiment_spec
, there is a line"local_dir": '/home/lucas/Documents/cryptocurrency_prediction/tensorboard'
. I could not run thistrain_multi_model.py
file, but if I did, I imagine it would give me an error about folder not exist.So instead I tried the second command you provided,
# single pair python train.py \ --algo PPO \ --pair XRP/USDT \
and it is still running as I am writing this post. But I could not find where is the outputtensorflow
folder, so I cannot really track its progress. Again, any idea? or did I do something wrong...?Again, great project! Thx!