{"batches":[[{"__dict__":[["pw",{"__nd__":[-17.72596340542139],"dtype":"float64","shape":[1]}],["pl",{"__nd__":[-24.25935581481126],"dtype":"float64","shape":[1]}],["rw",{"__nd__":[-18.476414601227845],"dtype":"float64","shape":[1]}],["rl",{"__nd__":[-25.199920531202476],"dtype":"float64","shape":[1]}],["beta",0.1]]},{"__dict__":[["loss",0.7026980348472073],["reward_w",{"__nd__":[0.07504511958064555],"dtype":"float64","shape":[1]}],["grad_w",{"__nd__":[-0.05047526948676755],"dtype":"float64","shape":[1]}],["grad_l",{"__nd__":[0.05047526948676755],"dtype":"float64","shape":[1]}]]}],[{"__dict__":[["pw",{"__nd__":[-29.689145245707966,-25.383656327343555,-18.893288641911525,-22.44050542460115],"dtype":"float64","shape":[4]}],["pl",{"__nd__":[-19.715255003438948,-25.2241022386969,-14.724559824378584,-16.160966233841513],"dtype":"float64","shape":[4]}],["rw",{"__nd__":[-29.755175943269183,-26.51089753431159,-19.360797984163572,-21.58121296171791],"dtype":"float64","shape":[4]}],["rl",{"__nd__":[-20.084005787521445,-24.2652196378679,-15.603010125685858,-16.11104032285526],"dtype":"float64","shape":[4]}],["beta",0.05]]},{"__dict__":[["loss",0.6900391947720743],["reward_w",{"__nd__":[0.0033015348780608637,0.05636206034840168,0.023375467112602345,-0.04296462314416197],"dtype":"float64","shape":[4]}],["grad_w",{"__nd__":[-0.006297299110508959,-0.00592433836212401,-0.006314207265943334,-0.006376446267602915],"dtype":"float64","shape":[4]}],["grad_l",{"__nd__":[0.006297299110508959,0.00592433836212401,0.006314207265943334,0.006376446267602915],"dtype":"float64","shape":[4]}]]}],[{"__dict__":[["pw",{"__nd__":[-21.43655454679784,-24.218420450267583,-13.271313884054353,-19.64367511762312,-22.25558656847041,-22.600824229403674],"dtype":"float64","shape":[6]}],["pl",{"__nd__":[-18.162935287924498,-17.52218689035491,-17.193082770816616,-17.302740716655393,-9.957271702936387,-21.79991375885636],"dtype":"float64","shape":[6]}],["rw",{"__nd__":[-20.924311817726302,-23.404647722019707,-13.887293306629848,-20.77264741034401,-22.141639110815536,-21.760667752441147],"dtype":"float64","shape":[6]}],["rl",{"__nd__":[-17.338454072233258,-18.17277967817961,-17.93633694202006,-17.845894984960587,-9.291761995647693,-22.03207508192308],"dtype":"float64","shape":[6]}],["beta",0.5]]},{"__dict__":[["loss",0.7645429755434551],["reward_w",{"__nd__":[-0.25612136453576895,-0.4068863641239382,0.3079897112877479,0.564486146360446,-0.05697372882743679,-0.42007823848126336],"dtype":"float64","shape":[6]}],["grad_w",{"__nd__":[-0.038420772495049334,-0.05627367341047302,-0.04299199806047931,-0.03560765352856913,-0.035957365962294895,-0.05257653570428326],"dtype":"float64","shape":[6]}],["grad_l",{"__nd__":[0.038420772495049334,0.05627367341047302,0.04299199806047931,0.03560765352856913,0.035957365962294895,0.05257653570428326],"dtype":"float64","shape":[6]}]]}],[{"__dict__":[["pw",{"__nd__":[-18.705344374503504,-18.11320978115501,-15.991581182507488,-19.344374048791156,-15.747456303383373,-19.853408977437155,-19.830089329888,-17.97684608481078,-28.205231385279944,-21.10119533773274],"dtype":"float64","shape":[10]}],["pl",{"__nd__":[-22.209437535393413,-22.50390388714894,-21.80296390246995,-12.366753752751308,-23.70356518449888,-15.467954766762833,-27.95757362052161,-22.336351090995528,-19.549288521040044,-17.45062623720194],"dtype":"float64","shape":[10]}],["rw",{"__nd__":[-19.416570954296358,-18.906557016354935,-15.64285611025905,-18.88202225612659,-16.605432184640527,-19.66210465255554,-18.55440300655008,-16.8435588708073,-27.285779099278333,-21.598356081786505],"dtype":"float64","shape":[10]}],["rl",{"__nd__":[-22.351863271463976,-23.194389241216708,-21.375711256133417,-12.525293443828021,-24.329155578466217,-15.158608227042595,-28.414348858079023,-21.674425149928876,-19.18623467447497,-17.06888834320361],"dtype":"float64","shape":[10]}],["beta",2.0]]},{"__dict__":[["loss",1.0379928114358008],["reward_w",{"__nd__":[1.4224531595857073,1.5866944703998485,-0.6974501444968766,-0.9247035853291337,1.7159517625143081,-0.3826086497632275,-2.551372646675844,-2.2665744280069617,-1.8389045720032229,0.9943214881075306],"dtype":"float64","shape":[10]}],["grad_w",{"__nd__":[-0.04855219322986988,-0.08974993690465088,-0.09216334442703196,-0.15517484261531625,-0.0771709256297994,-0.08825030118382843,-0.19393461544098034,-0.14392985416147486,-0.1505300336681092,-0.029413286020446933],"dtype":"float64","shape":[10]}],["grad_l",{"__nd__":[0.04855219322986988,0.08974993690465088,0.09216334442703196,0.15517484261531625,0.0771709256297994,0.08825030118382843,0.19393461544098034,0.14392985416147486,0.1505300336681092,0.029413286020446933],"dtype":"float64","shape":[10]}]]}]]}