def cli(experiment_json, run_name, buffer, seed, checkpoint_path):
print('cli')
print('------')
print(experiment_json, run_name, buffer)
print('')
hyp = json_util.load(experiment_json)
hyp['buffer'] = buffer
if run_name:
hyp['run-name'] = run_name
print('params')
print('------')
print(hyp)
print('')
sleep(2)
if checkpoint_path:
print(f'checkpointing from {checkpoint_path}')
print('')
main(**init.init_checkpoint(checkpoint_path))
else:
print(f'starting so fresh, so clean')
print('')
main(**init.init_fresh(hyp))
def test_system():
main(**init_fresh(hyp))
run_path = './experiments/pendulum/test-system'
print(f'deleting {run_path}\n')
rmtree(str(run_path))
return run_path
def test_system_battery():
main(**init_fresh(batt_hyp))
run_path = './experiments/battery/test-system'
print(f'deleting {run_path}\n')
rmtree(str(run_path))
return run_path
"n-episodes": 4,
"test-every": 128,
"n-tests": "all",
"env": {
"name": "battery",
"initial_charge": 0.0,
"episode_length": episode_length,
"n_batteries": n_batteries,
"dataset": dataset,
},
"network": {
"name": "attention",
"size_scale": 8
},
"seed": 42,
}
expt = init_fresh(hyp)
train_one_head_network(
buffer.sample(4),
expt["nets"]["actor"],
[expt["nets"]["online-1"], expt["nets"]["online-2"]],
[expt["nets"]["target-1"], expt["nets"]["target-2"]],
expt["nets"]["alpha"],
expt["writers"]["train"],
expt["optimizers"],
expt["counters"],
hyp,
)