Zero win rate in SMAC scenario

[George-Chia]

Hello, thanks for your elegant code.
When I directly run:
python dicg_ce_smac_runner.py --map 3s_vs_5z
python dicg_ce_smac_runner.py --map 6h_vs_8z
the win rate is always zero.

When I directly run:
python dicg_ce_smac_runner.py --map 8m_vs_9m
the win rate is lower than 0.1,
which are very different from the results reported in the paper.

Can you give me a hand?

[parachutel]

How many environment steps did your experiments take?

[George-Chia]

More than 800 steps shown in the tensorboard, where I find the EvalAvgReturn is converge to 10.8.
By the way, dicg_de_smac_runner.py --map 8m_vs_9m can achieve 0.9 win rate in the same environment step.