By default Tensorflow preallocates GPU memory eagerly. Background for that is that it wants to prevent memory fragmentation. The amount of memory it allocates is around 80% of the memory available. Although running a rather small model with less than 45k of parameters, the monitoring tool nvidia-smi shows this:
This site wants your consent to use cookies. It uses cookies in two ways: Cookies are used for technical reasons to allow proper navigation and handling of requests sent to the server infrastructure. Moreover, also statistical analysis may be performed to provide insight on who is using the webpage. Click here to show our data protection policy.
The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.