| {"global_step": 84167, "epoch": 500, "test/sim_max_reward_100625": 1.0, "test/sim_max_reward_100626": 0.5443378243251776, "test/sim_max_reward_100627": 1.0, "test/sim_max_reward_100628": 1.0, "test/sim_max_reward_100629": 1.0, "test/sim_max_reward_100630": 1.0, "test/sim_max_reward_100631": 1.0, "test/sim_max_reward_100632": 0.986825430828376, "test/sim_max_reward_100633": 0.6514671507085809, "test/sim_max_reward_100634": 0.9943260480811761, "test/sim_max_reward_100635": 0.9811225327872319, "test/sim_max_reward_100636": 1.0, "test/sim_max_reward_100637": 0.4255689331122512, "test/sim_max_reward_100638": 0.9698918305393809, "test/sim_max_reward_100639": 0.9949195705251406, "test/sim_max_reward_100640": 0.9743703477867725, "test/sim_max_reward_100641": 1.0, "test/sim_max_reward_100642": 0.9878246476313166, "test/sim_max_reward_100643": 0.7151200628785718, "test/sim_max_reward_100644": 0.9860415349801745, "test/sim_max_reward_100645": 0.012547685148507874, "test/sim_max_reward_100646": 1.0, "test/sim_max_reward_100647": 1.0, "test/sim_max_reward_100648": 0.7105191206121733, "test/sim_max_reward_100649": 1.0, "test/sim_max_reward_100425": 1.0, "test/sim_max_reward_100426": 1.0, "test/sim_max_reward_100427": 0.005425996669144089, "test/sim_max_reward_100428": 0.7232488282595616, "test/sim_max_reward_100429": 1.0, "test/sim_max_reward_100430": 0.6672046225138566, "test/sim_max_reward_100431": 0.7112155216278788, "test/sim_max_reward_100432": 0.9865116652158371, "test/sim_max_reward_100433": 0.9875898781338502, "test/sim_max_reward_100434": 1.0, "test/sim_max_reward_100435": 0.995477949305561, "test/sim_max_reward_100436": 0.9942311545925934, "test/sim_max_reward_100437": 0.0, "test/sim_max_reward_100438": 1.0, "test/sim_max_reward_100439": 0.6135402611158347, "test/sim_max_reward_100440": 0.6866208906379039, "test/sim_max_reward_100441": 0.7139535577965329, "test/sim_max_reward_100442": 0.9319702193276121, "test/sim_max_reward_100443": 0.9833135684076676, "test/sim_max_reward_100444": 0.997061874769307, "test/sim_max_reward_100445": 0.0, "test/sim_max_reward_100446": 0.6459549403442217, "test/sim_max_reward_100447": 0.9865755784077722, "test/sim_max_reward_100448": 0.0, "test/sim_max_reward_100449": 0.9896852407271837, "test/sim_max_reward_100175": 1.0, "test/sim_max_reward_100176": 0.9737670843268587, "test/sim_max_reward_100177": 0.023066237841451023, "test/sim_max_reward_100178": 0.5758806441285401, "test/sim_max_reward_100179": 1.0, "test/sim_max_reward_100180": 1.0, "test/sim_max_reward_100181": 0.9593020489498411, "test/sim_max_reward_100182": 0.20821634789489044, "test/sim_max_reward_100183": 1.0, "test/sim_max_reward_100184": 1.0, "test/sim_max_reward_100185": 1.0, "test/sim_max_reward_100186": 1.0, "test/sim_max_reward_100187": 0.5856372526441351, "test/sim_max_reward_100188": 1.0, "test/sim_max_reward_100189": 0.7103773802661216, "test/sim_max_reward_100190": 1.0, "test/sim_max_reward_100191": 1.0, "test/sim_max_reward_100192": 0.958043823157261, "test/sim_max_reward_100193": 0.6938666726451833, "test/sim_max_reward_100194": 1.0, "test/sim_max_reward_100195": 0.690836134091043, "test/sim_max_reward_100196": 1.0, "test/sim_max_reward_100197": 1.0, "test/sim_max_reward_100198": 1.0, "test/sim_max_reward_100199": 1.0, "test/sim_max_reward_100575": 1.0, "test/sim_max_reward_100576": 0.9351230949809178, "test/sim_max_reward_100577": 0.61614834450042, "test/sim_max_reward_100578": 0.05745341949891073, "test/sim_max_reward_100579": 1.0, "test/sim_max_reward_100580": 0.9890029203195791, "test/sim_max_reward_100581": 1.0, "test/sim_max_reward_100582": 0.9989304132930541, "test/sim_max_reward_100583": 0.9955134779558222, "test/sim_max_reward_100584": 1.0, "test/sim_max_reward_100585": 1.0, "test/sim_max_reward_100586": 1.0, "test/sim_max_reward_100587": 0.6213559737899764, "test/sim_max_reward_100588": 1.0, "test/sim_max_reward_100589": 0.6785367897565179, "test/sim_max_reward_100590": 0.995965266415375, "test/sim_max_reward_100591": 0.9982348485728728, "test/sim_max_reward_100592": 1.0, "test/sim_max_reward_100593": 1.0, "test/sim_max_reward_100594": 0.41106485989327624, "test/sim_max_reward_100595": 0.716112062774411, "test/sim_max_reward_100596": 1.0, "test/sim_max_reward_100597": 1.0, "test/sim_max_reward_100598": 1.0, "test/sim_max_reward_100599": 1.0, "test/sim_max_reward_100275": 0.9887367185676674, "test/sim_max_reward_100276": 1.0, "test/sim_max_reward_100277": 0.9691376654966989, "test/sim_max_reward_100278": 0.9967802184761105, "test/sim_max_reward_100279": 0.9918216385678721, "test/sim_max_reward_100280": 1.0, "test/sim_max_reward_100281": 1.0, "test/sim_max_reward_100282": 0.0086326329850604, "test/sim_max_reward_100283": 1.0, "test/sim_max_reward_100284": 1.0, "test/sim_max_reward_100285": 1.0, "test/sim_max_reward_100286": 0.5728261920885197, "test/sim_max_reward_100287": 0.3328970908058002, "test/sim_max_reward_100288": 1.0, "test/sim_max_reward_100289": 1.0, "test/sim_max_reward_100290": 0.45429208547504857, "test/sim_max_reward_100291": 1.0, "test/sim_max_reward_100292": 0.9940923845762162, "test/sim_max_reward_100293": 1.0, "test/sim_max_reward_100294": 0.9606134942559803, "test/sim_max_reward_100295": 1.0, "test/sim_max_reward_100296": 0.6528989803808462, "test/sim_max_reward_100297": 0.9496710873731233, "test/sim_max_reward_100298": 1.0, "test/sim_max_reward_100299": 0.0, "test/sim_max_reward_100075": 0.9867488097386033, "test/sim_max_reward_100076": 1.0, "test/sim_max_reward_100077": 0.9937097109099081, "test/sim_max_reward_100078": 0.8937597473697625, "test/sim_max_reward_100079": 0.9282088340658324, "test/sim_max_reward_100080": 1.0, "test/sim_max_reward_100081": 1.0, "test/sim_max_reward_100082": 0.9887070093632828, "test/sim_max_reward_100083": 0.9371263165436653, "test/sim_max_reward_100084": 0.6126226596558904, "test/sim_max_reward_100085": 0.6735211882218216, "test/sim_max_reward_100086": 1.0, "test/sim_max_reward_100087": 0.966792205154673, "test/sim_max_reward_100088": 1.0, "test/sim_max_reward_100089": 1.0, "test/sim_max_reward_100090": 0.5470574095498856, "test/sim_max_reward_100091": 1.0, "test/sim_max_reward_100092": 1.0, "test/sim_max_reward_100093": 0.0, "test/sim_max_reward_100094": 0.6817988704494945, "test/sim_max_reward_100095": 0.9846912852755506, "test/sim_max_reward_100096": 1.0, "test/sim_max_reward_100097": 1.0, "test/sim_max_reward_100098": 0.9862804475375124, "test/sim_max_reward_100099": 0.6982355976536812, "test/sim_max_reward_100925": 0.5058542856594647, "test/sim_max_reward_100926": 1.0, "test/sim_max_reward_100927": 1.0, "test/sim_max_reward_100928": 1.0, "test/sim_max_reward_100929": 0.6995130317787362, "test/sim_max_reward_100930": 0.708348017018575, "test/sim_max_reward_100931": 1.0, "test/sim_max_reward_100932": 0.8920175219759312, "test/sim_max_reward_100933": 0.820161701517311, "test/sim_max_reward_100934": 1.0, "test/sim_max_reward_100935": 0.9965902539728453, "test/sim_max_reward_100936": 0.6523147211284204, "test/sim_max_reward_100937": 1.0, "test/sim_max_reward_100938": 1.0, "test/sim_max_reward_100939": 0.21401090038310575, "test/sim_max_reward_100940": 0.7226852458686588, "test/sim_max_reward_100941": 0.6494726623890897, "test/sim_max_reward_100942": 0.0, "test/sim_max_reward_100943": 1.0, "test/sim_max_reward_100944": 0.9904827847658062, "test/sim_max_reward_100945": 1.0, "test/sim_max_reward_100946": 0.546114053068784, "test/sim_max_reward_100947": 0.985040312964594, "test/sim_max_reward_100948": 0.9727521548281242, "test/sim_max_reward_100949": 0.9621061199324576, "test/sim_max_reward_100200": 0.9548458572910202, "test/sim_max_reward_100201": 1.0, "test/sim_max_reward_100202": 0.9630898676094249, "test/sim_max_reward_100203": 1.0, "test/sim_max_reward_100204": 1.0, "test/sim_max_reward_100205": 0.9976890332334899, "test/sim_max_reward_100206": 0.6427435413420358, "test/sim_max_reward_100207": 0.9911977941127006, "test/sim_max_reward_100208": 0.7062351579728697, "test/sim_max_reward_100209": 0.991490341645003, "test/sim_max_reward_100210": 1.0, "test/sim_max_reward_100211": 1.0, "test/sim_max_reward_100212": 0.9921052406924438, "test/sim_max_reward_100213": 1.0, "test/sim_max_reward_100214": 0.4812993562092614, "test/sim_max_reward_100215": 0.6680279104114584, "test/sim_max_reward_100216": 0.9916166055243828, "test/sim_max_reward_100217": 0.9920572833890514, "test/sim_max_reward_100218": 0.6047901292901839, "test/sim_max_reward_100219": 1.0, "test/sim_max_reward_100220": 1.0, "test/sim_max_reward_100221": 0.7101385931602662, "test/sim_max_reward_100222": 1.0, "test/sim_max_reward_100223": 1.0, "test/sim_max_reward_100224": 0.963050129452687, "test/sim_max_reward_100800": 0.9857457365053118, "test/sim_max_reward_100801": 0.9971301582797307, "test/sim_max_reward_100802": 1.0, "test/sim_max_reward_100803": 0.6240075880525877, "test/sim_max_reward_100804": 0.9821026998451292, "test/sim_max_reward_100805": 0.6972574246773336, "test/sim_max_reward_100806": 0.0, "test/sim_max_reward_100807": 0.1971483152174601, "test/sim_max_reward_100808": 0.43768387843411977, "test/sim_max_reward_100809": 1.0, "test/sim_max_reward_100810": 1.0, "test/sim_max_reward_100811": 0.9798474574471387, "test/sim_max_reward_100812": 1.0, "test/sim_max_reward_100813": 1.0, "test/sim_max_reward_100814": 1.0, "test/sim_max_reward_100815": 1.0, "test/sim_max_reward_100816": 0.7202606632988292, "test/sim_max_reward_100817": 0.6656786371956755, "test/sim_max_reward_100818": 0.9399309521711129, "test/sim_max_reward_100819": 0.9673506924519998, "test/sim_max_reward_100820": 1.0, "test/sim_max_reward_100821": 0.9529431951264978, "test/sim_max_reward_100822": 0.0, "test/sim_max_reward_100823": 0.8483713350651008, "test/sim_max_reward_100824": 1.0, "test/sim_max_reward_100525": 0.6041509270962931, "test/sim_max_reward_100526": 0.9969746001489811, "test/sim_max_reward_100527": 0.6083313179606783, "test/sim_max_reward_100528": 0.0, "test/sim_max_reward_100529": 0.979079529567076, "test/sim_max_reward_100530": 1.0, "test/sim_max_reward_100531": 0.2782836461064455, "test/sim_max_reward_100532": 1.0, "test/sim_max_reward_100533": 1.0, "test/sim_max_reward_100534": 1.0, "test/sim_max_reward_100535": 1.0, "test/sim_max_reward_100536": 0.578411871693074, "test/sim_max_reward_100537": 0.6511812506322785, "test/sim_max_reward_100538": 1.0, "test/sim_max_reward_100539": 0.6763706751629955, "test/sim_max_reward_100540": 0.4663416407787844, "test/sim_max_reward_100541": 1.0, "test/sim_max_reward_100542": 1.0, "test/sim_max_reward_100543": 1.0, "test/sim_max_reward_100544": 1.0, "test/sim_max_reward_100545": 1.0, "test/sim_max_reward_100546": 0.6990763558737926, "test/sim_max_reward_100547": 1.0, "test/sim_max_reward_100548": 0.7167038023930886, "test/sim_max_reward_100549": 0.7061812591026145, "test/sim_max_reward_100875": 0.9812643877684724, "test/sim_max_reward_100876": 1.0, "test/sim_max_reward_100877": 0.9863968330222116, "test/sim_max_reward_100878": 1.0, "test/sim_max_reward_100879": 0.9913783397373638, "test/sim_max_reward_100880": 1.0, "test/sim_max_reward_100881": 0.6673584145863498, "test/sim_max_reward_100882": 0.7148762812860178, "test/sim_max_reward_100883": 0.546973839467891, "test/sim_max_reward_100884": 0.9975543131695164, "test/sim_max_reward_100885": 1.0, "test/sim_max_reward_100886": 0.6868360928468094, "test/sim_max_reward_100887": 0.5154464077888132, "test/sim_max_reward_100888": 0.9997440832769982, "test/sim_max_reward_100889": 1.0, "test/sim_max_reward_100890": 0.9859152927131601, "test/sim_max_reward_100891": 1.0, "test/sim_max_reward_100892": 0.0, "test/sim_max_reward_100893": 0.718917568384707, "test/sim_max_reward_100894": 0.6311842021438593, "test/sim_max_reward_100895": 0.9857634883268994, "test/sim_max_reward_100896": 1.0, "test/sim_max_reward_100897": 0.40026311671030645, "test/sim_max_reward_100898": 1.0, "test/sim_max_reward_100899": 1.0, "test/sim_max_reward_100750": 1.0, "test/sim_max_reward_100751": 1.0, "test/sim_max_reward_100752": 1.0, "test/sim_max_reward_100753": 0.9486230037698116, "test/sim_max_reward_100754": 1.0, "test/sim_max_reward_100755": 0.012225269622582323, "test/sim_max_reward_100756": 1.0, "test/sim_max_reward_100757": 0.608552630671357, "test/sim_max_reward_100758": 1.0, "test/sim_max_reward_100759": 1.0, "test/sim_max_reward_100760": 0.9803367297587074, "test/sim_max_reward_100761": 1.0, "test/sim_max_reward_100762": 0.5840871848246562, "test/sim_max_reward_100763": 1.0, "test/sim_max_reward_100764": 0.6840542842278228, "test/sim_max_reward_100765": 1.0, "test/sim_max_reward_100766": 0.9399182717315109, "test/sim_max_reward_100767": 1.0, "test/sim_max_reward_100768": 1.0, "test/sim_max_reward_100769": 0.6593306165691403, "test/sim_max_reward_100770": 1.0, "test/sim_max_reward_100771": 0.9854117071330278, "test/sim_max_reward_100772": 0.9227736830063055, "test/sim_max_reward_100773": 0.9931677992721017, "test/sim_max_reward_100774": 1.0, "test/sim_max_reward_100675": 1.0, "test/sim_max_reward_100676": 0.984393414547096, "test/sim_max_reward_100677": 1.0, "test/sim_max_reward_100678": 0.9526132829292017, "test/sim_max_reward_100679": 0.16136200072119966, "test/sim_max_reward_100680": 1.0, "test/sim_max_reward_100681": 0.9887101986634845, "test/sim_max_reward_100682": 1.0, "test/sim_max_reward_100683": 0.9706976327655654, "test/sim_max_reward_100684": 1.0, "test/sim_max_reward_100685": 0.6835509712773931, "test/sim_max_reward_100686": 0.9890755273635194, "test/sim_max_reward_100687": 0.9309083188120695, "test/sim_max_reward_100688": 1.0, "test/sim_max_reward_100689": 1.0, "test/sim_max_reward_100690": 0.49149131995464285, "test/sim_max_reward_100691": 0.6506986810904036, "test/sim_max_reward_100692": 1.0, "test/sim_max_reward_100693": 0.9906991919226854, "test/sim_max_reward_100694": 0.7499478417315546, "test/sim_max_reward_100695": 1.0, "test/sim_max_reward_100696": 1.0, "test/sim_max_reward_100697": 0.9463181420379614, "test/sim_max_reward_100698": 0.9999718114595827, "test/sim_max_reward_100699": 0.9970877174189147, "test/sim_max_reward_100325": 1.0, "test/sim_max_reward_100326": 0.7136923325599929, "test/sim_max_reward_100327": 1.0, "test/sim_max_reward_100328": 1.0, "test/sim_max_reward_100329": 1.0, "test/sim_max_reward_100330": 0.0035424379596931954, "test/sim_max_reward_100331": 0.9898446672162049, "test/sim_max_reward_100332": 0.9242528838316474, "test/sim_max_reward_100333": 1.0, "test/sim_max_reward_100334": 1.0, "test/sim_max_reward_100335": 0.9939370682475364, "test/sim_max_reward_100336": 1.0, "test/sim_max_reward_100337": 0.7064939719194923, "test/sim_max_reward_100338": 0.6105164334395022, "test/sim_max_reward_100339": 0.5728067720614375, "test/sim_max_reward_100340": 0.7165962953190197, "test/sim_max_reward_100341": 1.0, "test/sim_max_reward_100342": 1.0, "test/sim_max_reward_100343": 0.9859161337227856, "test/sim_max_reward_100344": 0.7142211550901169, "test/sim_max_reward_100345": 1.0, "test/sim_max_reward_100346": 0.998782331224265, "test/sim_max_reward_100347": 0.9965733028372912, "test/sim_max_reward_100348": 0.9293941335914822, "test/sim_max_reward_100349": 0.9660282795157258, "test/sim_max_reward_100225": 0.6972176973124262, "test/sim_max_reward_100226": 1.0, "test/sim_max_reward_100227": 0.688987834335227, "test/sim_max_reward_100228": 0.9907141356555502, "test/sim_max_reward_100229": 1.0, "test/sim_max_reward_100230": 1.0, "test/sim_max_reward_100231": 1.0, "test/sim_max_reward_100232": 0.9801907703451349, "test/sim_max_reward_100233": 1.0, "test/sim_max_reward_100234": 1.0, "test/sim_max_reward_100235": 0.0, "test/sim_max_reward_100236": 1.0, "test/sim_max_reward_100237": 0.5469192056444707, "test/sim_max_reward_100238": 1.0, "test/sim_max_reward_100239": 1.0, "test/sim_max_reward_100240": 1.0, "test/sim_max_reward_100241": 1.0, "test/sim_max_reward_100242": 1.0, "test/sim_max_reward_100243": 1.0, "test/sim_max_reward_100244": 1.0, "test/sim_max_reward_100245": 0.7096261780139952, "test/sim_max_reward_100246": 0.9883486208608893, "test/sim_max_reward_100247": 0.9926477855068874, "test/sim_max_reward_100248": 1.0, "test/sim_max_reward_100249": 0.3645098353870891, "test/sim_max_reward_100450": 1.0, "test/sim_max_reward_100451": 0.9661396709608172, "test/sim_max_reward_100452": 0.7184082551522004, "test/sim_max_reward_100453": 1.0, "test/sim_max_reward_100454": 0.9671544113686178, "test/sim_max_reward_100455": 1.0, "test/sim_max_reward_100456": 0.9844200931455679, "test/sim_max_reward_100457": 0.5744568137761935, "test/sim_max_reward_100458": 1.0, "test/sim_max_reward_100459": 0.9809429074235169, "test/sim_max_reward_100460": 0.5369389507835944, "test/sim_max_reward_100461": 1.0, "test/sim_max_reward_100462": 0.6869333478375158, "test/sim_max_reward_100463": 1.0, "test/sim_max_reward_100464": 1.0, "test/sim_max_reward_100465": 0.0, "test/sim_max_reward_100466": 0.699179952606589, "test/sim_max_reward_100467": 1.0, "test/sim_max_reward_100468": 1.0, "test/sim_max_reward_100469": 0.9895096478149239, "test/sim_max_reward_100470": 0.9886421336959645, "test/sim_max_reward_100471": 0.0, "test/sim_max_reward_100472": 0.9964547582152733, "test/sim_max_reward_100473": 0.7237943619942014, "test/sim_max_reward_100474": 0.9740148843671416, "test/sim_max_reward_100775": 0.9953558052477879, "test/sim_max_reward_100776": 0.0, "test/sim_max_reward_100777": 0.6092239903442119, "test/sim_max_reward_100778": 1.0, "test/sim_max_reward_100779": 0.9926328201626458, "test/sim_max_reward_100780": 0.9839731676662611, "test/sim_max_reward_100781": 0.7234521262217917, "test/sim_max_reward_100782": 0.7042530116275997, "test/sim_max_reward_100783": 0.992118914127076, "test/sim_max_reward_100784": 0.9325935182028828, "test/sim_max_reward_100785": 1.0, "test/sim_max_reward_100786": 1.0, "test/sim_max_reward_100787": 1.0, "test/sim_max_reward_100788": 0.0, "test/sim_max_reward_100789": 1.0, "test/sim_max_reward_100790": 0.49085267781649167, "test/sim_max_reward_100791": 0.7222039208748195, "test/sim_max_reward_100792": 1.0, "test/sim_max_reward_100793": 1.0, "test/sim_max_reward_100794": 0.6907596511528157, "test/sim_max_reward_100795": 0.9993617869559648, "test/sim_max_reward_100796": 1.0, "test/sim_max_reward_100797": 0.9996965842299173, "test/sim_max_reward_100798": 1.0, "test/sim_max_reward_100799": 0.9908259723105144, "test/sim_max_reward_100250": 0.9611024073997175, "test/sim_max_reward_100251": 0.9939895812271149, "test/sim_max_reward_100252": 1.0, "test/sim_max_reward_100253": 1.0, "test/sim_max_reward_100254": 0.9965741370980331, "test/sim_max_reward_100255": 1.0, "test/sim_max_reward_100256": 0.9821456291763019, "test/sim_max_reward_100257": 0.9993623815523343, "test/sim_max_reward_100258": 0.0, "test/sim_max_reward_100259": 0.9954725875789528, "test/sim_max_reward_100260": 0.9934819186825473, "test/sim_max_reward_100261": 1.0, "test/sim_max_reward_100262": 0.19588706554976798, "test/sim_max_reward_100263": 0.8620602371356237, "test/sim_max_reward_100264": 0.986657081791234, "test/sim_max_reward_100265": 0.9455923695839882, "test/sim_max_reward_100266": 0.7140892690463271, "test/sim_max_reward_100267": 0.701711362841725, "test/sim_max_reward_100268": 1.0, "test/sim_max_reward_100269": 1.0, "test/sim_max_reward_100270": 0.0, "test/sim_max_reward_100271": 1.0, "test/sim_max_reward_100272": 0.9886658338648538, "test/sim_max_reward_100273": 0.7084426354169684, "test/sim_max_reward_100274": 0.9711707711780021, "test/sim_max_reward_100475": 0.6241472328652912, "test/sim_max_reward_100476": 0.7237194941210733, "test/sim_max_reward_100477": 0.9739709601472641, "test/sim_max_reward_100478": 1.0, "test/sim_max_reward_100479": 1.0, "test/sim_max_reward_100480": 0.6986609023410939, "test/sim_max_reward_100481": 1.0, "test/sim_max_reward_100482": 0.722431374890144, "test/sim_max_reward_100483": 0.9676616744719054, "test/sim_max_reward_100484": 0.6070073543970358, "test/sim_max_reward_100485": 1.0, "test/sim_max_reward_100486": 1.0, "test/sim_max_reward_100487": 0.5738645667786856, "test/sim_max_reward_100488": 0.997411323739423, "test/sim_max_reward_100489": 1.0, "test/sim_max_reward_100490": 1.0, "test/sim_max_reward_100491": 1.0, "test/sim_max_reward_100492": 0.36501966017221027, "test/sim_max_reward_100493": 0.17947700903896815, "test/sim_max_reward_100494": 1.0, "test/sim_max_reward_100495": 1.0, "test/sim_max_reward_100496": 1.0, "test/sim_max_reward_100497": 0.16389878608511999, "test/sim_max_reward_100498": 0.6673245110113355, "test/sim_max_reward_100499": 0.9334717802080839, "test/sim_max_reward_100650": 0.9941665677863574, "test/sim_max_reward_100651": 0.9831801582964846, "test/sim_max_reward_100652": 0.9811805414002389, "test/sim_max_reward_100653": 1.0, "test/sim_max_reward_100654": 0.7128874391903809, "test/sim_max_reward_100655": 0.912701440147138, "test/sim_max_reward_100656": 0.8266529175181879, "test/sim_max_reward_100657": 0.03548409867587679, "test/sim_max_reward_100658": 1.0, "test/sim_max_reward_100659": 1.0, "test/sim_max_reward_100660": 0.9838575261672937, "test/sim_max_reward_100661": 1.0, "test/sim_max_reward_100662": 1.0, "test/sim_max_reward_100663": 1.0, "test/sim_max_reward_100664": 1.0, "test/sim_max_reward_100665": 0.13082201814171923, "test/sim_max_reward_100666": 0.7000463804903546, "test/sim_max_reward_100667": 0.679695740915875, "test/sim_max_reward_100668": 0.9779558492909818, "test/sim_max_reward_100669": 0.6664651778073133, "test/sim_max_reward_100670": 1.0, "test/sim_max_reward_100671": 1.0, "test/sim_max_reward_100672": 1.0, "test/sim_max_reward_100673": 1.0, "test/sim_max_reward_100674": 0.9849608786417285, "test/sim_max_reward_100725": 0.9888381854035336, "test/sim_max_reward_100726": 0.6739146183531095, "test/sim_max_reward_100727": 0.9777718508877091, "test/sim_max_reward_100728": 0.9871381337146221, "test/sim_max_reward_100729": 0.6942730767916736, "test/sim_max_reward_100730": 0.7187509344747205, "test/sim_max_reward_100731": 1.0, "test/sim_max_reward_100732": 1.0, "test/sim_max_reward_100733": 0.7132701607999523, "test/sim_max_reward_100734": 1.0, "test/sim_max_reward_100735": 1.0, "test/sim_max_reward_100736": 1.0, "test/sim_max_reward_100737": 0.0, "test/sim_max_reward_100738": 0.9984713728112946, "test/sim_max_reward_100739": 0.97992803676299, "test/sim_max_reward_100740": 1.0, "test/sim_max_reward_100741": 0.9634337887944567, "test/sim_max_reward_100742": 0.9945363832907435, "test/sim_max_reward_100743": 1.0, "test/sim_max_reward_100744": 1.0, "test/sim_max_reward_100745": 1.0, "test/sim_max_reward_100746": 1.0, "test/sim_max_reward_100747": 0.7120064977330408, "test/sim_max_reward_100748": 1.0, "test/sim_max_reward_100749": 0.47051053756967565, "test/sim_max_reward_100350": 1.0, "test/sim_max_reward_100351": 0.9952983281904603, "test/sim_max_reward_100352": 0.9977152409089979, "test/sim_max_reward_100353": 1.0, "test/sim_max_reward_100354": 0.6968503767218086, "test/sim_max_reward_100355": 1.0, "test/sim_max_reward_100356": 0.6797440794089685, "test/sim_max_reward_100357": 0.6945927658511764, "test/sim_max_reward_100358": 0.44365140699853073, "test/sim_max_reward_100359": 1.0, "test/sim_max_reward_100360": 1.0, "test/sim_max_reward_100361": 1.0, "test/sim_max_reward_100362": 0.7162558647607562, "test/sim_max_reward_100363": 0.9977194775976106, "test/sim_max_reward_100364": 0.7035446630687576, "test/sim_max_reward_100365": 1.0, "test/sim_max_reward_100366": 1.0, "test/sim_max_reward_100367": 0.0, "test/sim_max_reward_100368": 1.0, "test/sim_max_reward_100369": 0.6798561304931509, "test/sim_max_reward_100370": 1.0, "test/sim_max_reward_100371": 0.9986233024592956, "test/sim_max_reward_100372": 0.9703194990446938, "test/sim_max_reward_100373": 0.9068761189623594, "test/sim_max_reward_100374": 0.6690504411015888, "test/sim_max_reward_100050": 0.5831408753252301, "test/sim_max_reward_100051": 1.0, "test/sim_max_reward_100052": 0.977483844351538, "test/sim_max_reward_100053": 0.9926701294625127, "test/sim_max_reward_100054": 1.0, "test/sim_max_reward_100055": 1.0, "test/sim_max_reward_100056": 1.0, "test/sim_max_reward_100057": 0.4328404145384277, "test/sim_max_reward_100058": 1.0, "test/sim_max_reward_100059": 1.0, "test/sim_max_reward_100060": 1.0, "test/sim_max_reward_100061": 0.6167149289940747, "test/sim_max_reward_100062": 1.0, "test/sim_max_reward_100063": 1.0, "test/sim_max_reward_100064": 1.0, "test/sim_max_reward_100065": 0.9103442837029637, "test/sim_max_reward_100066": 0.9737201319599423, "test/sim_max_reward_100067": 1.0, "test/sim_max_reward_100068": 0.6867658098738922, "test/sim_max_reward_100069": 0.6931581266289254, "test/sim_max_reward_100070": 0.7063576746188082, "test/sim_max_reward_100071": 0.6772117204093292, "test/sim_max_reward_100072": 1.0, "test/sim_max_reward_100073": 0.6603870731038683, "test/sim_max_reward_100074": 0.9999961584655421, "test/sim_max_reward_100150": 0.6651423865570469, "test/sim_max_reward_100151": 1.0, "test/sim_max_reward_100152": 0.511631843922063, "test/sim_max_reward_100153": 0.5254341046400028, "test/sim_max_reward_100154": 0.09498956349821815, "test/sim_max_reward_100155": 0.69724566541936, "test/sim_max_reward_100156": 0.4968647046963382, "test/sim_max_reward_100157": 0.7235419006564273, "test/sim_max_reward_100158": 1.0, "test/sim_max_reward_100159": 1.0, "test/sim_max_reward_100160": 1.0, "test/sim_max_reward_100161": 1.0, "test/sim_max_reward_100162": 1.0, "test/sim_max_reward_100163": 0.6738432582110455, "test/sim_max_reward_100164": 1.0, "test/sim_max_reward_100165": 0.5681223676763366, "test/sim_max_reward_100166": 1.0, "test/sim_max_reward_100167": 0.2963780721839753, "test/sim_max_reward_100168": 1.0, "test/sim_max_reward_100169": 0.9347457182759928, "test/sim_max_reward_100170": 0.9909382383853294, "test/sim_max_reward_100171": 0.6890691021179148, "test/sim_max_reward_100172": 0.9313466435248519, "test/sim_max_reward_100173": 0.6379280167382309, "test/sim_max_reward_100174": 0.9933787027781253, "test/sim_max_reward_100025": 1.0, "test/sim_max_reward_100026": 0.0, "test/sim_max_reward_100027": 0.9773264741576219, "test/sim_max_reward_100028": 0.9936461612105462, "test/sim_max_reward_100029": 1.0, "test/sim_max_reward_100030": 0.3486406872716362, "test/sim_max_reward_100031": 1.0, "test/sim_max_reward_100032": 0.6062189761945357, "test/sim_max_reward_100033": 1.0, "test/sim_max_reward_100034": 0.6823445997329065, "test/sim_max_reward_100035": 0.7203352963580117, "test/sim_max_reward_100036": 1.0, "test/sim_max_reward_100037": 0.9949777085910313, "test/sim_max_reward_100038": 0.3682597420013055, "test/sim_max_reward_100039": 1.0, "test/sim_max_reward_100040": 1.0, "test/sim_max_reward_100041": 0.9958465714009604, "test/sim_max_reward_100042": 1.0, "test/sim_max_reward_100043": 0.9377364606839046, "test/sim_max_reward_100044": 0.9880485775665266, "test/sim_max_reward_100045": 1.0, "test/sim_max_reward_100046": 0.9791291535272828, "test/sim_max_reward_100047": 0.9344493184828752, "test/sim_max_reward_100048": 0.947367279853142, "test/sim_max_reward_100049": 0.0, "test/sim_max_reward_100825": 0.6749381313566784, "test/sim_max_reward_100826": 0.5822282217598244, "test/sim_max_reward_100827": 0.9674141426761877, "test/sim_max_reward_100828": 0.658967117397995, "test/sim_max_reward_100829": 0.9737130795997487, "test/sim_max_reward_100830": 1.0, "test/sim_max_reward_100831": 1.0, "test/sim_max_reward_100832": 0.9529219599488082, "test/sim_max_reward_100833": 1.0, "test/sim_max_reward_100834": 1.0, "test/sim_max_reward_100835": 0.9678976734736218, "test/sim_max_reward_100836": 0.9877585562834147, "test/sim_max_reward_100837": 1.0, "test/sim_max_reward_100838": 0.7048611652422763, "test/sim_max_reward_100839": 0.9803630136209346, "test/sim_max_reward_100840": 1.0, "test/sim_max_reward_100841": 0.713573015479122, "test/sim_max_reward_100842": 1.0, "test/sim_max_reward_100843": 1.0, "test/sim_max_reward_100844": 0.7087852243937106, "test/sim_max_reward_100845": 1.0, "test/sim_max_reward_100846": 1.0, "test/sim_max_reward_100847": 1.0, "test/sim_max_reward_100848": 1.0, "test/sim_max_reward_100849": 0.9255567376702386, "test/sim_max_reward_100300": 0.9690169730099315, "test/sim_max_reward_100301": 0.9964421062726753, "test/sim_max_reward_100302": 0.6828526337108256, "test/sim_max_reward_100303": 0.5381770737432809, "test/sim_max_reward_100304": 1.0, "test/sim_max_reward_100305": 1.0, "test/sim_max_reward_100306": 0.9899955629452704, "test/sim_max_reward_100307": 0.5054961495892046, "test/sim_max_reward_100308": 0.9953144190984448, "test/sim_max_reward_100309": 0.9999392295522787, "test/sim_max_reward_100310": 0.4939443092230453, "test/sim_max_reward_100311": 0.7141915109761787, "test/sim_max_reward_100312": 0.9881005168523488, "test/sim_max_reward_100313": 0.702786933247557, "test/sim_max_reward_100314": 1.0, "test/sim_max_reward_100315": 1.0, "test/sim_max_reward_100316": 1.0, "test/sim_max_reward_100317": 0.9795210149404869, "test/sim_max_reward_100318": 0.9551237905786226, "test/sim_max_reward_100319": 1.0, "test/sim_max_reward_100320": 0.986058563416332, "test/sim_max_reward_100321": 1.0, "test/sim_max_reward_100322": 1.0, "test/sim_max_reward_100323": 0.45198679266913083, "test/sim_max_reward_100324": 0.6699202620157048, "test/sim_max_reward_100400": 1.0, "test/sim_max_reward_100401": 0.5986978285250345, "test/sim_max_reward_100402": 1.0, "test/sim_max_reward_100403": 1.0, "test/sim_max_reward_100404": 1.0, "test/sim_max_reward_100405": 1.0, "test/sim_max_reward_100406": 0.7253199122816373, "test/sim_max_reward_100407": 0.9999133886068183, "test/sim_max_reward_100408": 1.0, "test/sim_max_reward_100409": 0.9730584889010215, "test/sim_max_reward_100410": 0.9765589094099539, "test/sim_max_reward_100411": 0.2629160928858833, "test/sim_max_reward_100412": 1.0, "test/sim_max_reward_100413": 0.6510755128493579, "test/sim_max_reward_100414": 0.927352175375545, "test/sim_max_reward_100415": 1.0, "test/sim_max_reward_100416": 1.0, "test/sim_max_reward_100417": 0.7149284677979607, "test/sim_max_reward_100418": 1.0, "test/sim_max_reward_100419": 1.0, "test/sim_max_reward_100420": 0.98258749884071, "test/sim_max_reward_100421": 1.0, "test/sim_max_reward_100422": 0.541953567325924, "test/sim_max_reward_100423": 1.0, "test/sim_max_reward_100424": 0.6737352890751213, "test/sim_max_reward_100100": 1.0, "test/sim_max_reward_100101": 0.6519856320041221, "test/sim_max_reward_100102": 0.9851074274464794, "test/sim_max_reward_100103": 1.0, "test/sim_max_reward_100104": 0.9984603680472057, "test/sim_max_reward_100105": 1.0, "test/sim_max_reward_100106": 0.6993923920630645, "test/sim_max_reward_100107": 0.8782395861964776, "test/sim_max_reward_100108": 1.0, "test/sim_max_reward_100109": 0.9783914997945602, "test/sim_max_reward_100110": 1.0, "test/sim_max_reward_100111": 1.0, "test/sim_max_reward_100112": 1.0, "test/sim_max_reward_100113": 1.0, "test/sim_max_reward_100114": 0.9948814951887113, "test/sim_max_reward_100115": 0.9713133935455887, "test/sim_max_reward_100116": 0.9835060820114603, "test/sim_max_reward_100117": 0.9917231593463588, "test/sim_max_reward_100118": 0.6522012767484302, "test/sim_max_reward_100119": 1.0, "test/sim_max_reward_100120": 0.5691984917660458, "test/sim_max_reward_100121": 0.6634782151786813, "test/sim_max_reward_100122": 1.0, "test/sim_max_reward_100123": 0.9813747001818435, "test/sim_max_reward_100124": 1.0, "test/sim_max_reward_100900": 1.0, "test/sim_max_reward_100901": 0.7086727996222016, "test/sim_max_reward_100902": 1.0, "test/sim_max_reward_100903": 1.0, "test/sim_max_reward_100904": 0.542143324829692, "test/sim_max_reward_100905": 1.0, "test/sim_max_reward_100906": 0.9372526605528433, "test/sim_max_reward_100907": 1.0, "test/sim_max_reward_100908": 0.9923134713609583, "test/sim_max_reward_100909": 0.6443748579749907, "test/sim_max_reward_100910": 1.0, "test/sim_max_reward_100911": 0.9911666571276517, "test/sim_max_reward_100912": 0.9958170263788438, "test/sim_max_reward_100913": 1.0, "test/sim_max_reward_100914": 0.18730930135508023, "test/sim_max_reward_100915": 0.9527745426215349, "test/sim_max_reward_100916": 0.7140200711687102, "test/sim_max_reward_100917": 0.9437093241185952, "test/sim_max_reward_100918": 0.6757578239915515, "test/sim_max_reward_100919": 0.9957205274900777, "test/sim_max_reward_100920": 1.0, "test/sim_max_reward_100921": 0.7632447070698789, "test/sim_max_reward_100922": 0.0, "test/sim_max_reward_100923": 0.3521353833030189, "test/sim_max_reward_100924": 0.7028976597155687, "test/sim_max_reward_100700": 1.0, "test/sim_max_reward_100701": 0.6771554855576192, "test/sim_max_reward_100702": 1.0, "test/sim_max_reward_100703": 0.7743032289009057, "test/sim_max_reward_100704": 1.0, "test/sim_max_reward_100705": 0.6895959430813166, "test/sim_max_reward_100706": 0.0, "test/sim_max_reward_100707": 1.0, "test/sim_max_reward_100708": 0.9568845839040172, "test/sim_max_reward_100709": 1.0, "test/sim_max_reward_100710": 1.0, "test/sim_max_reward_100711": 0.9820158224481877, "test/sim_max_reward_100712": 1.0, "test/sim_max_reward_100713": 0.9632527452415744, "test/sim_max_reward_100714": 0.6952275595466184, "test/sim_max_reward_100715": 1.0, "test/sim_max_reward_100716": 0.999781803401674, "test/sim_max_reward_100717": 0.9333845175535912, "test/sim_max_reward_100718": 0.9623378631528209, "test/sim_max_reward_100719": 0.006667550673181165, "test/sim_max_reward_100720": 1.0, "test/sim_max_reward_100721": 1.0, "test/sim_max_reward_100722": 1.0, "test/sim_max_reward_100723": 1.0, "test/sim_max_reward_100724": 0.9897668217374606, "test/sim_max_reward_100600": 0.9722925836655789, "test/sim_max_reward_100601": 0.9976432126241893, "test/sim_max_reward_100602": 0.9976791781559015, "test/sim_max_reward_100603": 1.0, "test/sim_max_reward_100604": 0.99567586363032, "test/sim_max_reward_100605": 0.6435722893755907, "test/sim_max_reward_100606": 0.028508234201441794, "test/sim_max_reward_100607": 0.9775865578124275, "test/sim_max_reward_100608": 0.9759524854716896, "test/sim_max_reward_100609": 1.0, "test/sim_max_reward_100610": 1.0, "test/sim_max_reward_100611": 1.0, "test/sim_max_reward_100612": 0.6208817158982506, "test/sim_max_reward_100613": 0.7999829917614806, "test/sim_max_reward_100614": 1.0, "test/sim_max_reward_100615": 0.6430285990948628, "test/sim_max_reward_100616": 0.971380256729085, "test/sim_max_reward_100617": 1.0, "test/sim_max_reward_100618": 0.9720453805656807, "test/sim_max_reward_100619": 0.9932850723293462, "test/sim_max_reward_100620": 0.9803210739693895, "test/sim_max_reward_100621": 0.5596186296202673, "test/sim_max_reward_100622": 1.0, "test/sim_max_reward_100623": 1.0, "test/sim_max_reward_100624": 1.0, "test/sim_max_reward_100950": 0.5101201522232115, "test/sim_max_reward_100951": 0.3350605513374616, "test/sim_max_reward_100952": 0.6594272416365613, "test/sim_max_reward_100953": 1.0, "test/sim_max_reward_100954": 1.0, "test/sim_max_reward_100955": 0.9741036702328726, "test/sim_max_reward_100956": 0.19047255401268165, "test/sim_max_reward_100957": 0.8237166405249583, "test/sim_max_reward_100958": 1.0, "test/sim_max_reward_100959": 1.0, "test/sim_max_reward_100960": 1.0, "test/sim_max_reward_100961": 0.6673238915341113, "test/sim_max_reward_100962": 1.0, "test/sim_max_reward_100963": 1.0, "test/sim_max_reward_100964": 0.3074255331601559, "test/sim_max_reward_100965": 1.0, "test/sim_max_reward_100966": 0.9836250086997934, "test/sim_max_reward_100967": 0.6018243307547567, "test/sim_max_reward_100968": 1.0, "test/sim_max_reward_100969": 1.0, "test/sim_max_reward_100970": 1.0, "test/sim_max_reward_100971": 1.0, "test/sim_max_reward_100972": 1.0, "test/sim_max_reward_100973": 0.5277465588540818, "test/sim_max_reward_100974": 1.0, "test/sim_max_reward_100850": 0.9926689656441282, "test/sim_max_reward_100851": 1.0, "test/sim_max_reward_100852": 0.0, "test/sim_max_reward_100853": 1.0, "test/sim_max_reward_100854": 1.0, "test/sim_max_reward_100855": 0.9691966352639182, "test/sim_max_reward_100856": 1.0, "test/sim_max_reward_100857": 0.7083319781859274, "test/sim_max_reward_100858": 1.0, "test/sim_max_reward_100859": 0.9793427719555234, "test/sim_max_reward_100860": 0.6709976078776083, "test/sim_max_reward_100861": 1.0, "test/sim_max_reward_100862": 1.0, "test/sim_max_reward_100863": 0.9913266220263562, "test/sim_max_reward_100864": 1.0, "test/sim_max_reward_100865": 1.0, "test/sim_max_reward_100866": 0.5585864707032023, "test/sim_max_reward_100867": 1.0, "test/sim_max_reward_100868": 1.0, "test/sim_max_reward_100869": 1.0, "test/sim_max_reward_100870": 1.0, "test/sim_max_reward_100871": 0.9312769663293282, "test/sim_max_reward_100872": 0.48257015758899285, "test/sim_max_reward_100873": 0.9826668533822678, "test/sim_max_reward_100874": 0.0, "test/sim_max_reward_100550": 0.9886056620027138, "test/sim_max_reward_100551": 1.0, "test/sim_max_reward_100552": 0.9712700638480039, "test/sim_max_reward_100553": 0.5432817130325043, "test/sim_max_reward_100554": 1.0, "test/sim_max_reward_100555": 1.0, "test/sim_max_reward_100556": 1.0, "test/sim_max_reward_100557": 0.5410785816946395, "test/sim_max_reward_100558": 0.6551508504629521, "test/sim_max_reward_100559": 1.0, "test/sim_max_reward_100560": 1.0, "test/sim_max_reward_100561": 0.9827108470179285, "test/sim_max_reward_100562": 0.6891337273012316, "test/sim_max_reward_100563": 1.0, "test/sim_max_reward_100564": 0.9744153480818621, "test/sim_max_reward_100565": 1.0, "test/sim_max_reward_100566": 0.6830399186513466, "test/sim_max_reward_100567": 0.7506150949837664, "test/sim_max_reward_100568": 0.9804205978472699, "test/sim_max_reward_100569": 1.0, "test/sim_max_reward_100570": 0.7052192565672345, "test/sim_max_reward_100571": 0.9721556304679906, "test/sim_max_reward_100572": 0.7188465190066166, "test/sim_max_reward_100573": 0.973915559060094, "test/sim_max_reward_100574": 1.0, "test/sim_max_reward_100975": 1.0, "test/sim_max_reward_100976": 1.0, "test/sim_max_reward_100977": 1.0, "test/sim_max_reward_100978": 1.0, "test/sim_max_reward_100979": 0.9512151558375449, "test/sim_max_reward_100980": 1.0, "test/sim_max_reward_100981": 1.0, "test/sim_max_reward_100982": 1.0, "test/sim_max_reward_100983": 0.9967004519306125, "test/sim_max_reward_100984": 1.0, "test/sim_max_reward_100985": 1.0, "test/sim_max_reward_100986": 0.26687243394217347, "test/sim_max_reward_100987": 1.0, "test/sim_max_reward_100988": 1.0, "test/sim_max_reward_100989": 0.6148575421735782, "test/sim_max_reward_100990": 0.371723750050597, "test/sim_max_reward_100991": 1.0, "test/sim_max_reward_100992": 1.0, "test/sim_max_reward_100993": 1.0, "test/sim_max_reward_100994": 1.0, "test/sim_max_reward_100995": 0.9881617547085838, "test/sim_max_reward_100996": 0.6675079398467627, "test/sim_max_reward_100997": 1.0, "test/sim_max_reward_100998": 0.6266655936177918, "test/sim_max_reward_100999": 1.0, "test/sim_max_reward_100500": 0.7178140803801009, "test/sim_max_reward_100501": 0.6863817880111507, "test/sim_max_reward_100502": 1.0, "test/sim_max_reward_100503": 0.714693082224763, "test/sim_max_reward_100504": 1.0, "test/sim_max_reward_100505": 1.0, "test/sim_max_reward_100506": 1.0, "test/sim_max_reward_100507": 0.26293381348110123, "test/sim_max_reward_100508": 0.997308354085337, "test/sim_max_reward_100509": 1.0, "test/sim_max_reward_100510": 1.0, "test/sim_max_reward_100511": 0.7086972494189119, "test/sim_max_reward_100512": 0.9993846290644778, "test/sim_max_reward_100513": 0.9980404881280771, "test/sim_max_reward_100514": 1.0, "test/sim_max_reward_100515": 1.0, "test/sim_max_reward_100516": 0.5319212841763832, "test/sim_max_reward_100517": 0.6019166514996077, "test/sim_max_reward_100518": 0.6129336930822612, "test/sim_max_reward_100519": 0.9581225178927966, "test/sim_max_reward_100520": 1.0, "test/sim_max_reward_100521": 1.0, "test/sim_max_reward_100522": 0.0, "test/sim_max_reward_100523": 0.5899872890059417, "test/sim_max_reward_100524": 0.3299185960826198, "test/sim_max_reward_100125": 0.5930847291829033, "test/sim_max_reward_100126": 1.0, "test/sim_max_reward_100127": 1.0, "test/sim_max_reward_100128": 1.0, "test/sim_max_reward_100129": 0.993741637533495, "test/sim_max_reward_100130": 0.0046859812481875, "test/sim_max_reward_100131": 0.9989676248142081, "test/sim_max_reward_100132": 0.6622271343488267, "test/sim_max_reward_100133": 1.0, "test/sim_max_reward_100134": 1.0, "test/sim_max_reward_100135": 0.961362933587572, "test/sim_max_reward_100136": 1.0, "test/sim_max_reward_100137": 0.9892697924756663, "test/sim_max_reward_100138": 1.0, "test/sim_max_reward_100139": 1.0, "test/sim_max_reward_100140": 1.0, "test/sim_max_reward_100141": 1.0, "test/sim_max_reward_100142": 1.0, "test/sim_max_reward_100143": 0.9927387495819074, "test/sim_max_reward_100144": 0.7233271613021788, "test/sim_max_reward_100145": 0.6452089155731757, "test/sim_max_reward_100146": 1.0, "test/sim_max_reward_100147": 1.0, "test/sim_max_reward_100148": 0.0, "test/sim_max_reward_100149": 0.9995282034378764, "test/sim_max_reward_100375": 0.5297083384290737, "test/sim_max_reward_100376": 0.9013771258070151, "test/sim_max_reward_100377": 1.0, "test/sim_max_reward_100378": 0.9949131164869597, "test/sim_max_reward_100379": 0.9041383623755338, "test/sim_max_reward_100380": 0.9764913894384194, "test/sim_max_reward_100381": 1.0, "test/sim_max_reward_100382": 0.9662144190417742, "test/sim_max_reward_100383": 0.6912441449158356, "test/sim_max_reward_100384": 1.0, "test/sim_max_reward_100385": 0.9256352680869185, "test/sim_max_reward_100386": 0.9959309164581169, "test/sim_max_reward_100387": 0.9854050627102866, "test/sim_max_reward_100388": 0.990142669372099, "test/sim_max_reward_100389": 0.9612483707537997, "test/sim_max_reward_100390": 1.0, "test/sim_max_reward_100391": 0.7138457149672517, "test/sim_max_reward_100392": 0.9507389705599374, "test/sim_max_reward_100393": 1.0, "test/sim_max_reward_100394": 1.0, "test/sim_max_reward_100395": 0.6978869266854597, "test/sim_max_reward_100396": 0.6997337636767474, "test/sim_max_reward_100397": 1.0, "test/sim_max_reward_100398": 1.0, "test/sim_max_reward_100399": 1.0, "test/sim_max_reward_100000": 1.0, "test/sim_max_reward_100001": 0.9350465750593948, "test/sim_max_reward_100002": 1.0, "test/sim_max_reward_100003": 0.6766595239118919, "test/sim_max_reward_100004": 1.0, "test/sim_max_reward_100005": 1.0, "test/sim_max_reward_100006": 1.0, "test/sim_max_reward_100007": 0.9733575026367609, "test/sim_max_reward_100008": 1.0, "test/sim_max_reward_100009": 0.9909548426373472, "test/sim_max_reward_100010": 0.9992908036870163, "test/sim_max_reward_100011": 1.0, "test/sim_max_reward_100012": 0.9057649656198549, "test/sim_max_reward_100013": 1.0, "test/sim_max_reward_100014": 1.0, "test/sim_max_reward_100015": 0.6777427340915948, "test/sim_max_reward_100016": 1.0, "test/sim_max_reward_100017": 0.9856531594971153, "test/sim_max_reward_100018": 1.0, "test/sim_max_reward_100019": 0.7078112058459364, "test/sim_max_reward_100020": 1.0, "test/sim_max_reward_100021": 0.9827487964608749, "test/sim_max_reward_100022": 0.7027940168764812, "test/sim_max_reward_100023": 0.9938890034762538, "test/sim_max_reward_100024": 1.0, "test/mean_score": 0.8556943428758327} |