python3.8 read_models.py --path ./trained_models/[YYY]/square/FILE.pt

where

YYY is a train-variant with possible values in [fedavg, fedprox]

FILE contains trained models, these files are stores in  ./trained_models/[YYY]/square/

        for example results_SimpleFC_square_sds_8_8_pr_[XXX]_mu_0.001_cts_1_data_default.fedavg.2000.pt in ./trained_models/fedavg/square/

        where XXX is personalization round with possible values in [1,2,...,100]

Ex1: personalization round = 0, training = fedavg (if personalization round = 0 then there is no personalization)

python3.8 read_models.py --path ./trained_models/fedavg/square/results_SimpleFC_square_sds_8_8_pr_0_mu_0.001_cts_1_data_default.fedavg.2000.pt

Ex2: personalization round = 100, training = fedavg

python3.8 read_models.py --path ./trained_models/fedavg/square/results_SimpleFC_square_sds_8_8_pr_100_mu_0.001_cts_1_data_default.fedavg.2000.pt

Ex3: personalization round = 0, training = fedprox (if personalization round = 0 then there is no personalization)

python3.8 read_models.py --path ./trained_models/fedprox/square/results_SimpleFC_square_sds_8_8_pr_0_mu_0.5_cts_2_data_default.fedprox.2000.pt

Ex4: personalization round = 100, training = fedprox

python3.8 read_models.py --path ./trained_models/fedprox/square/results_SimpleFC_square_sds_8_8_pr_100_mu_0.5_cts_2_data_default.fedprox.2000.pt