I would clone the first experiment, then in the cloned experiment, I would change the initial weights (assuming there is a parameter storing that) to point to the latest checkpoint, i.e. provide the full path/link. Then enqueue it for execution. The downside is that the iteration counter will start from 0 and not the previous run.
If you have the check point (see output_uri for automatically uploading it) then you can always load it. Do you mean if you can change the iteration/ step counter? Or do you mean with trains-agent?
Is it considered the same experiment? Is it possible to use the trains-agent? Submit a resume from the UI?