Fig. 6From: Conditional reduction of the loss value versus reinforcement learning for biassing a de-novo drug design generatorThe steps followed for data preparationBack to article page