A dynamic and flexible generalized spatial modulation (GSM) framework is proposed for massive MIMO systems. Our framework is leveraged on the utilization of machine learning methods for GSM in order to improve the error performance in presence of correlated channels and channel estimation errors. Both decision tree and multi-layer perceptrons approaches are adopted for the GSM transmitter. Simulation results indicate that in presence of real-life impairments machine learning based approaches provide a superior performance when compared to the classical Euclidean distance based approach. The observations are validated through measurement results over the designed $16\times 4$ MIMO test-bed using software defined radio nodes.