Before running the installation script, there are a few values that need to be updated in the .env file.
The values under #Skin can be left blank. They are only necessary if your organization requested a custom skin for the user interface.
Under #Installation please edit the SIMMACHINES_MOUNT_PATH variable with the path/directory you want to use for storage. This is going to be the directory that ML Studio will use as its working directory for data, logs, model serializations, etc. This folder tends to grow rather quickly if you have multiple data sets and models. Please make sure that you have enough space according to your datasets’ size.
Under #API there are two possible values you will be interested in changing: SIMMACHINES_API_HOST is set to the localhost address, but if you intend on using ML Studio remotely, accessing it from a machine that is not where it is installed, you will need to replace this value with the IP address of the machine where ML Studio is installed. The IP address can be public or private, in case it’s going to be accessible within a local network. This variable needs to be set in with a corresponding value for the GUI, as explained below.
SIMMACHINES_API_ACCESS_PORT is set to 9090 as a default value that can be changed at your convenience or need, like the case when the port is already in use.
Next, we have SIMMACHINES_API_MEMORY. This value is set by default to 1 GB in case ML Studio is installed in a small machine for light use. We recommend increasing this value for better performance of ML Studio, especially if you’re going to work with bigger datasets. It is recommended that you add no more than 70 % of your available memory.
Under #GUI, similar to the #API section, you may be interested in changing the value for SIMMACHINES_GUI_HOST in case you will be accessing ML Studio remotely. Depending on your particular installation, the IP address should match the one for the SIMMACHINES_API_HOST. The SIMMACHINES_GUI_ACCESS_PORT can be used to change the port used to access the graphical user interface via the web. The default port is 8080.
The SIMMACHINES_GUI_MEMORY parameter works in the same way as the API memory. In contrast to the values needed for the API, this parameter is set to 1 GB, which is more than enough for most use cases. Depending on your individual needs and the load it’s put under, we would recommend increasing it up to 2 GB.
Once you update the parameters mentioned above, you have to save the changes and move to the final step.
Comments
0 comments
Please sign in to leave a comment.