Table 3.

Overall performance of GBM in training using the combined dataset of all states of India

S. No	Distribution	Performance measures of GBM
		Active cases modeling					Recovered cases modeling
		R2	MSE	RMSE	MAE	MRD	R2	MSE	RMSE	MAE	MRD
1	Gaussian	0.97	2.13	86.07	61.29	0.97	0.94	535.41	23.14	15.92	535.41
2	Tweedie	0.98	4805.18	69.32	38.73	2.13	0.97	232.60	15.25	8.61	1.87
3	Huber	0.85	20,673.86	143.78	60.49	0.91	0.68	26.53	57.99	19.37	642.39
4	Laplace	0.72	74,248.96	272.49	104.08	0.67	0.48	6228.91	78.92	26.53	26.53
5	Poisson	0.99	3075.37	55.46	35.89	−2846.53	0.99	94.35	9.71	6.63	−357.78
6	Quantile	0.70	69,855.46	264.30	113.82	56.91	0.50	6614.20	81.33	26.37	13.18
7	Gamma	0.85	27,019.53	164.38	70.78	10.21	0.83	1682.18	41.01	16.92	6.05

Note: R² (coefficient of determination), MSE (mean square error), RMSE (root mean square error), MAE (mean average error), and MRD (mean residual deviance)