['--alg', 'sac', '--env', 'Hopper-v3', '--learn', 'brac', '--traj', 'medium', '--seed', '4', '--data', '100000']
1632.7764403783826
episode: 0 training return: tensor(-7.1298e+13, device='cuda:0')
episode: 1 training return: tensor(-1.8229e+12, device='cuda:0')
episode: 2 training return: tensor(-5.8296e+13, device='cuda:0')
episode: 3 training return: tensor(-4.3948e+10, device='cuda:0')
epoch: 1 test_true_pfm: -109.21775369960069
episode: 4 training return: tensor(-2.0169e+13, device='cuda:0')
episode: 5 training return: tensor(-7.2374e+15, device='cuda:0')
episode: 6 training return: tensor(-4.9844e+08, device='cuda:0')
episode: 7 training return: tensor(-2.7832e+13, device='cuda:0')
epoch: 2 test_true_pfm: -105.21734653814234
episode: 8 training return: tensor(-5.6599e+12, device='cuda:0')
episode: 9 training return: tensor(-9.0081e+11, device='cuda:0')
episode: 10 training return: tensor(-2.4998e+10, device='cuda:0')
episode: 11 training return: tensor(-2.5421e+09, device='cuda:0')
epoch: 3 test_true_pfm: 271.890694989634
episode: 12 training return: tensor(-2.0352e+09, device='cuda:0')
episode: 13 training return: tensor(-2.6596e+09, device='cuda:0')
episode: 14 training return: tensor(-3.6090e+09, device='cuda:0')
episode: 15 training return: tensor(-6.4811e+13, device='cuda:0')
epoch: 4 test_true_pfm: 35.743846011234844
episode: 16 training return: tensor(-2.6994e+13, device='cuda:0')
episode: 17 training return: tensor(-4.5300e+15, device='cuda:0')
episode: 18 training return: tensor(-9.1620e+09, device='cuda:0')
episode: 19 training return: tensor(-7.8866e+13, device='cuda:0')
epoch: 5 test_true_pfm: 90.57329714385185
episode: 20 training return: tensor(-1.2367e+10, device='cuda:0')
episode: 21 training return: tensor(-2.6270e+08, device='cuda:0')
episode: 22 training return: tensor(-1.2105e+10, device='cuda:0')
episode: 23 training return: tensor(-6.0183e+08, device='cuda:0')
epoch: 6 test_true_pfm: 91.32370215128883
episode: 24 training return: tensor(-1.2859e+10, device='cuda:0')
episode: 25 training return: tensor(-4.7288e+09, device='cuda:0')
episode: 26 training return: tensor(-1.5386e+09, device='cuda:0')
episode: 27 training return: tensor(-1.8425e+09, device='cuda:0')
epoch: 7 test_true_pfm: 90.98129644371936
episode: 28 training return: tensor(-1.4367e+10, device='cuda:0')
episode: 29 training return: tensor(-3.6970e+09, device='cuda:0')
episode: 30 training return: tensor(-1.4651e+10, device='cuda:0')
episode: 31 training return: tensor(-4.0648e+10, device='cuda:0')
epoch: 8 test_true_pfm: 103.77627365020253
episode: 32 training return: tensor(-3.4476e+10, device='cuda:0')
episode: 33 training return: tensor(-1.5738e+10, device='cuda:0')
episode: 34 training return: tensor(-2.1435e+10, device='cuda:0')
episode: 35 training return: tensor(-1.2102e+10, device='cuda:0')
epoch: 9 test_true_pfm: 91.27856146095542
episode: 36 training return: tensor(-1.6459e+10, device='cuda:0')
episode: 37 training return: tensor(-1.2478e+10, device='cuda:0')
episode: 38 training return: tensor(-4.1344e+09, device='cuda:0')
episode: 39 training return: tensor(-6.4483e+09, device='cuda:0')
epoch: 10 test_true_pfm: 92.31212090142655
episode: 40 training return: tensor(-1.3123e+10, device='cuda:0')
episode: 41 training return: tensor(-2.5976e+09, device='cuda:0')
episode: 42 training return: tensor(-4.2939e+09, device='cuda:0')
episode: 43 training return: tensor(-1.3047e+10, device='cuda:0')
epoch: 11 test_true_pfm: 93.84600067723333
episode: 44 training return: tensor(-4.3861e+09, device='cuda:0')
episode: 45 training return: tensor(-2.5656e+11, device='cuda:0')
episode: 46 training return: tensor(-3.6250e+11, device='cuda:0')
episode: 47 training return: tensor(-9.5097e+11, device='cuda:0')
epoch: 12 test_true_pfm: 124.58493519378304
episode: 48 training return: tensor(-4.4304e+09, device='cuda:0')
episode: 49 training return: tensor(-1.5217e+10, device='cuda:0')
episode: 50 training return: tensor(-1.4097e+10, device='cuda:0')
episode: 51 training return: tensor(-1.3682e+10, device='cuda:0')
epoch: 13 test_true_pfm: 90.522477748262
episode: 52 training return: tensor(-1.3533e+10, device='cuda:0')
episode: 53 training return: tensor(-3.7179e+10, device='cuda:0')
episode: 54 training return: tensor(-1.4409e+10, device='cuda:0')
episode: 55 training return: tensor(-1.3626e+10, device='cuda:0')
epoch: 14 test_true_pfm: 92.46585920651096
episode: 56 training return: tensor(-5.9427e+09, device='cuda:0')
episode: 57 training return: tensor(-1.5821e+10, device='cuda:0')
episode: 58 training return: tensor(-1.3863e+10, device='cuda:0')
episode: 59 training return: tensor(-1.3618e+10, device='cuda:0')
epoch: 15 test_true_pfm: 94.1462807970846
episode: 60 training return: tensor(-7.4437e+09, device='cuda:0')
episode: 61 training return: tensor(-1.2695e+10, device='cuda:0')
episode: 62 training return: tensor(-1.4408e+10, device='cuda:0')
episode: 63 training return: tensor(-1.5975e+10, device='cuda:0')
epoch: 16 test_true_pfm: 90.57737001296783
episode: 64 training return: tensor(-1.4691e+10, device='cuda:0')
episode: 65 training return: tensor(-1.3890e+10, device='cuda:0')
episode: 66 training return: tensor(-1.4832e+10, device='cuda:0')
episode: 67 training return: tensor(-1.5586e+10, device='cuda:0')
epoch: 17 test_true_pfm: 91.89654707333189
episode: 68 training return: tensor(-1.2158e+10, device='cuda:0')
episode: 69 training return: tensor(-1.3329e+10, device='cuda:0')
episode: 70 training return: tensor(-5.6927e+11, device='cuda:0')
episode: 71 training return: tensor(-7.6744e+11, device='cuda:0')
epoch: 18 test_true_pfm: 123.76153382346745
episode: 72 training return: tensor(-2.9434e+11, device='cuda:0')
episode: 73 training return: tensor(-1.5600e+10, device='cuda:0')
episode: 74 training return: tensor(-1.0777e+10, device='cuda:0')
episode: 75 training return: tensor(-1.1560e+10, device='cuda:0')
epoch: 19 test_true_pfm: 91.90859970743365
episode: 76 training return: tensor(-1.5976e+10, device='cuda:0')
episode: 77 training return: tensor(-1.3688e+10, device='cuda:0')
episode: 78 training return: tensor(-1.3254e+10, device='cuda:0')
episode: 79 training return: tensor(-1.1731e+10, device='cuda:0')
epoch: 20 test_true_pfm: 124.55134545563074
episode: 80 training return: tensor(-4.8845e+11, device='cuda:0')
episode: 81 training return: tensor(-1.0561e+10, device='cuda:0')
episode: 82 training return: tensor(-1.3239e+10, device='cuda:0')
episode: 83 training return: tensor(-1.1671e+10, device='cuda:0')
epoch: 21 test_true_pfm: 92.20718333736478
episode: 84 training return: tensor(-1.5286e+10, device='cuda:0')
episode: 85 training return: tensor(-1.2792e+10, device='cuda:0')
episode: 86 training return: tensor(-1.3501e+10, device='cuda:0')
episode: 87 training return: tensor(-2.7649e+12, device='cuda:0')
epoch: 22 test_true_pfm: 82.26609550304703
episode: 88 training return: tensor(-6.5310e+13, device='cuda:0')
episode: 89 training return: tensor(-5.1574e+11, device='cuda:0')
episode: 90 training return: tensor(-6.3339e+11, device='cuda:0')
episode: 91 training return: tensor(-3.0239e+11, device='cuda:0')
epoch: 23 test_true_pfm: 128.20443376954518
episode: 92 training return: tensor(-6.3278e+08, device='cuda:0')
episode: 93 training return: tensor(-1.4290e+10, device='cuda:0')
episode: 94 training return: tensor(-1.1853e+10, device='cuda:0')
episode: 95 training return: tensor(-1.3465e+10, device='cuda:0')
epoch: 24 test_true_pfm: 90.98045250240865
episode: 96 training return: tensor(-1.1643e+10, device='cuda:0')
episode: 97 training return: tensor(-1.4742e+10, device='cuda:0')
episode: 98 training return: tensor(-1.3181e+10, device='cuda:0')
episode: 99 training return: tensor(-1.1726e+10, device='cuda:0')
epoch: 25 test_true_pfm: 90.83460112204982
episode: 100 training return: tensor(-1.2534e+10, device='cuda:0')
episode: 101 training return: tensor(-1.5557e+10, device='cuda:0')
episode: 102 training return: tensor(-1.2416e+10, device='cuda:0')
episode: 103 training return: tensor(-1.0984e+10, device='cuda:0')
epoch: 26 test_true_pfm: 91.99603913113229
episode: 104 training return: tensor(-1.3767e+10, device='cuda:0')
episode: 105 training return: tensor(-1.2915e+10, device='cuda:0')
episode: 106 training return: tensor(-1.3520e+10, device='cuda:0')
episode: 107 training return: tensor(-1.3129e+10, device='cuda:0')
epoch: 27 test_true_pfm: 90.4176434595165
episode: 108 training return: tensor(-1.3631e+10, device='cuda:0')
episode: 109 training return: tensor(-1.4355e+10, device='cuda:0')
episode: 110 training return: tensor(-1.4824e+10, device='cuda:0')
episode: 111 training return: tensor(-1.5903e+10, device='cuda:0')
epoch: 28 test_true_pfm: 91.26020243168239
episode: 112 training return: tensor(-1.5396e+10, device='cuda:0')
episode: 113 training return: tensor(-1.4413e+10, device='cuda:0')
episode: 114 training return: tensor(-1.3686e+10, device='cuda:0')
episode: 115 training return: tensor(-4.4671e+13, device='cuda:0')
epoch: 29 test_true_pfm: -3.2372595470451935
episode: 116 training return: tensor(-2.5291e+10, device='cuda:0')
episode: 117 training return: tensor(-6.0833e+11, device='cuda:0')
episode: 118 training return: tensor(-6.0816e+11, device='cuda:0')
episode: 119 training return: tensor(-6.0989e+11, device='cuda:0')
epoch: 30 test_true_pfm: -13.26340058078965
episode: 120 training return: tensor(-1.0694e+10, device='cuda:0')
episode: 121 training return: tensor(-1.0142e+10, device='cuda:0')
episode: 122 training return: tensor(-6.6019e+11, device='cuda:0')
episode: 123 training return: tensor(-6.0530e+11, device='cuda:0')
epoch: 31 test_true_pfm: -9.503791512348801
episode: 124 training return: tensor(-6.0859e+11, device='cuda:0')
episode: 125 training return: tensor(-7.4435e+11, device='cuda:0')
episode: 126 training return: tensor(-6.1348e+11, device='cuda:0')
episode: 127 training return: tensor(-6.0922e+11, device='cuda:0')
epoch: 32 test_true_pfm: -10.459211992362563
episode: 128 training return: tensor(-5.9417e+11, device='cuda:0')
episode: 129 training return: tensor(-5.9105e+11, device='cuda:0')
episode: 130 training return: tensor(-5.9553e+11, device='cuda:0')
episode: 131 training return: tensor(-6.0627e+11, device='cuda:0')
epoch: 33 test_true_pfm: -10.660150068614783
episode: 132 training return: tensor(-6.0449e+11, device='cuda:0')
episode: 133 training return: tensor(-6.0249e+11, device='cuda:0')
episode: 134 training return: tensor(-6.0848e+11, device='cuda:0')
episode: 135 training return: tensor(-3.1267e+10, device='cuda:0')
epoch: 34 test_true_pfm: -11.207563108689442
episode: 136 training return: tensor(-7.7131e+11, device='cuda:0')
episode: 137 training return: tensor(-8.0930e+11, device='cuda:0')
episode: 138 training return: tensor(-1.0426e+10, device='cuda:0')
episode: 139 training return: tensor(-1.4997e+10, device='cuda:0')
epoch: 35 test_true_pfm: -10.771509879286208
episode: 140 training return: tensor(-6.6506e+11, device='cuda:0')
episode: 141 training return: tensor(-5.9854e+11, device='cuda:0')
episode: 142 training return: tensor(-7.8315e+11, device='cuda:0')
episode: 143 training return: tensor(-6.0496e+11, device='cuda:0')
epoch: 36 test_true_pfm: -11.071731008378764
episode: 144 training return: tensor(-6.1845e+11, device='cuda:0')
episode: 145 training return: tensor(-1.1442e+10, device='cuda:0')
episode: 146 training return: tensor(-1.1502e+10, device='cuda:0')
episode: 147 training return: tensor(-1.1404e+10, device='cuda:0')
epoch: 37 test_true_pfm: -23.639706977881463
episode: 148 training return: tensor(-1.3411e+10, device='cuda:0')
episode: 149 training return: tensor(-6.5874e+11, device='cuda:0')
episode: 150 training return: tensor(-6.0122e+11, device='cuda:0')
episode: 151 training return: tensor(-6.0060e+11, device='cuda:0')
epoch: 38 test_true_pfm: -11.044858725539632
episode: 152 training return: tensor(-5.9899e+11, device='cuda:0')
episode: 153 training return: tensor(-5.3314e+11, device='cuda:0')
episode: 154 training return: tensor(-7.5532e+11, device='cuda:0')
episode: 155 training return: tensor(-6.0488e+11, device='cuda:0')
epoch: 39 test_true_pfm: -11.447078375977684
episode: 156 training return: tensor(-6.0601e+11, device='cuda:0')
episode: 157 training return: tensor(-7.8026e+11, device='cuda:0')
episode: 158 training return: tensor(-6.1145e+11, device='cuda:0')
episode: 159 training return: tensor(-6.3510e+11, device='cuda:0')
epoch: 40 test_true_pfm: -11.151795409431438
episode: 160 training return: tensor(-6.6610e+11, device='cuda:0')
episode: 161 training return: tensor(-6.7108e+11, device='cuda:0')
episode: 162 training return: tensor(-7.7676e+11, device='cuda:0')
episode: 163 training return: tensor(-5.9762e+11, device='cuda:0')
epoch: 41 test_true_pfm: -11.466296590783905
episode: 164 training return: tensor(-6.4741e+11, device='cuda:0')
episode: 165 training return: tensor(-6.0167e+11, device='cuda:0')
episode: 166 training return: tensor(-7.4830e+11, device='cuda:0')
episode: 167 training return: tensor(-7.8600e+11, device='cuda:0')
epoch: 42 test_true_pfm: -11.154822789662772
episode: 168 training return: tensor(-6.4753e+11, device='cuda:0')
episode: 169 training return: tensor(-5.9450e+11, device='cuda:0')
episode: 170 training return: tensor(-1.1621e+10, device='cuda:0')
episode: 171 training return: tensor(-7.4546e+11, device='cuda:0')
epoch: 43 test_true_pfm: 37.84035218031849
episode: 172 training return: tensor(-1.2323e+12, device='cuda:0')
episode: 173 training return: tensor(-5.7692e+11, device='cuda:0')
episode: 174 training return: tensor(-5.5052e+11, device='cuda:0')
episode: 175 training return: tensor(-5.9581e+11, device='cuda:0')
epoch: 44 test_true_pfm: -10.787223857363074
episode: 176 training return: tensor(-5.3602e+11, device='cuda:0')
episode: 177 training return: tensor(-6.1010e+11, device='cuda:0')
episode: 178 training return: tensor(-6.5860e+11, device='cuda:0')
episode: 179 training return: tensor(-6.1168e+11, device='cuda:0')
epoch: 45 test_true_pfm: -10.913762290564868
episode: 180 training return: tensor(-7.4104e+11, device='cuda:0')
episode: 181 training return: tensor(-6.1078e+11, device='cuda:0')
episode: 182 training return: tensor(-6.5854e+11, device='cuda:0')
episode: 183 training return: tensor(-1.5067e+10, device='cuda:0')
epoch: 46 test_true_pfm: -16.719442672340268
episode: 184 training return: tensor(-1.5118e+10, device='cuda:0')
episode: 185 training return: tensor(-1.4445e+10, device='cuda:0')
episode: 186 training return: tensor(-5.9747e+11, device='cuda:0')
episode: 187 training return: tensor(-1.0883e+10, device='cuda:0')
epoch: 47 test_true_pfm: -11.029660806159129
episode: 188 training return: tensor(-6.0581e+11, device='cuda:0')
episode: 189 training return: tensor(-5.3406e+11, device='cuda:0')
episode: 190 training return: tensor(-5.3283e+11, device='cuda:0')
episode: 191 training return: tensor(-5.9438e+11, device='cuda:0')
epoch: 48 test_true_pfm: -11.19785455639797
episode: 192 training return: tensor(-6.0851e+11, device='cuda:0')
episode: 193 training return: tensor(-5.9951e+11, device='cuda:0')
episode: 194 training return: tensor(-6.8414e+11, device='cuda:0')
episode: 195 training return: tensor(-5.7408e+11, device='cuda:0')
epoch: 49 test_true_pfm: -11.465504880725836
episode: 196 training return: tensor(-6.0354e+11, device='cuda:0')
episode: 197 training return: tensor(-5.8718e+11, device='cuda:0')
episode: 198 training return: tensor(-6.0953e+11, device='cuda:0')
episode: 199 training return: tensor(-1.4267e+10, device='cuda:0')
epoch: 50 test_true_pfm: -9.557858714581675
episode: 200 training return: tensor(-1.1452e+10, device='cuda:0')
episode: 201 training return: tensor(-5.0442e+09, device='cuda:0')
episode: 202 training return: tensor(-1.5542e+10, device='cuda:0')
episode: 203 training return: tensor(-8.0007e+11, device='cuda:0')
epoch: 51 test_true_pfm: -11.328838086888394
episode: 204 training return: tensor(-6.0227e+11, device='cuda:0')
episode: 205 training return: tensor(-6.1072e+11, device='cuda:0')
episode: 206 training return: tensor(-1.7625e+11, device='cuda:0')
episode: 207 training return: tensor(-9.2799e+15, device='cuda:0')
epoch: 52 test_true_pfm: -3.768066276758432
episode: 208 training return: tensor(-4.2131e+11, device='cuda:0')
episode: 209 training return: tensor(-3.2046e+12, device='cuda:0')
episode: 210 training return: tensor(-1.8798e+12, device='cuda:0')
episode: 211 training return: tensor(-5.7394e+12, device='cuda:0')
epoch: 53 test_true_pfm: 25.61440437575315
episode: 212 training return: tensor(-1.0799e+10, device='cuda:0')
episode: 213 training return: tensor(-1.0890e+10, device='cuda:0')
episode: 214 training return: tensor(-1.0508e+10, device='cuda:0')
episode: 215 training return: tensor(-1.4854e+10, device='cuda:0')
epoch: 54 test_true_pfm: -9.932771358586903
episode: 216 training return: tensor(-1.4504e+10, device='cuda:0')
episode: 217 training return: tensor(-1.0934e+10, device='cuda:0')
episode: 218 training return: tensor(-6.0480e+11, device='cuda:0')
episode: 219 training return: tensor(-5.9576e+11, device='cuda:0')
epoch: 55 test_true_pfm: -11.112103996025946
episode: 220 training return: tensor(-8.0812e+11, device='cuda:0')
episode: 221 training return: tensor(-1.3546e+10, device='cuda:0')
episode: 222 training return: tensor(-1.4032e+11, device='cuda:0')
episode: 223 training return: tensor(-1.1456e+10, device='cuda:0')
epoch: 56 test_true_pfm: -9.593796275507035
episode: 224 training return: tensor(-1.4667e+10, device='cuda:0')
episode: 225 training return: tensor(-1.5470e+10, device='cuda:0')
episode: 226 training return: tensor(-1.0441e+10, device='cuda:0')
episode: 227 training return: tensor(-9.6328e+10, device='cuda:0')
epoch: 57 test_true_pfm: -10.816986957530595
episode: 228 training return: tensor(-5.9931e+11, device='cuda:0')
episode: 229 training return: tensor(-1.0434e+10, device='cuda:0')
episode: 230 training return: tensor(-1.3585e+10, device='cuda:0')
episode: 231 training return: tensor(-9.4519e+09, device='cuda:0')
epoch: 58 test_true_pfm: -0.7362335854944183
episode: 232 training return: tensor(-1.0195e+10, device='cuda:0')
episode: 233 training return: tensor(-9.6496e+09, device='cuda:0')
episode: 234 training return: tensor(-1.0775e+10, device='cuda:0')
episode: 235 training return: tensor(-1.0729e+10, device='cuda:0')
epoch: 59 test_true_pfm: -8.810297294591459
episode: 236 training return: tensor(-1.0545e+10, device='cuda:0')
episode: 237 training return: tensor(-1.0558e+10, device='cuda:0')
episode: 238 training return: tensor(-1.0737e+10, device='cuda:0')
episode: 239 training return: tensor(-1.0360e+10, device='cuda:0')
epoch: 60 test_true_pfm: -4.184123921519755
episode: 240 training return: tensor(-1.2032e+10, device='cuda:0')
episode: 241 training return: tensor(-1.0358e+10, device='cuda:0')
episode: 242 training return: tensor(-1.0226e+10, device='cuda:0')
episode: 243 training return: tensor(-1.0136e+10, device='cuda:0')
epoch: 61 test_true_pfm: -3.9365519125739232
episode: 244 training return: tensor(-1.2585e+10, device='cuda:0')
episode: 245 training return: tensor(-3.1888e+10, device='cuda:0')
episode: 246 training return: tensor(-1.5753e+10, device='cuda:0')
episode: 247 training return: tensor(-2.0554e+12, device='cuda:0')
epoch: 62 test_true_pfm: -11.334851797064166
episode: 248 training return: tensor(-7.8503e+11, device='cuda:0')
episode: 249 training return: tensor(-5.9733e+11, device='cuda:0')
episode: 250 training return: tensor(-5.7557e+11, device='cuda:0')
episode: 251 training return: tensor(-2.1680e+11, device='cuda:0')
epoch: 63 test_true_pfm: -12.54066790093951
episode: 252 training return: tensor(-1.4035e+11, device='cuda:0')
episode: 253 training return: tensor(-1.1648e+10, device='cuda:0')
episode: 254 training return: tensor(-5.4662e+11, device='cuda:0')
episode: 255 training return: tensor(-5.9779e+11, device='cuda:0')
epoch: 64 test_true_pfm: 61.3563715064189
episode: 256 training return: tensor(-3.3663e+16, device='cuda:0')
episode: 257 training return: tensor(-1.1883e+12, device='cuda:0')
episode: 258 training return: tensor(-1.5436e+13, device='cuda:0')
episode: 259 training return: tensor(-3.4412e+16, device='cuda:0')
epoch: 65 test_true_pfm: 16.636510694577
episode: 260 training return: tensor(-3.7934e+16, device='cuda:0')
episode: 261 training return: tensor(-1.1921e+17, device='cuda:0')
episode: 262 training return: tensor(-9.5231e+16, device='cuda:0')
episode: 263 training return: tensor(-5.4971e+15, device='cuda:0')
epoch: 66 test_true_pfm: 29.451928863026115
episode: 264 training return: tensor(-3.7042e+15, device='cuda:0')
episode: 265 training return: tensor(-6.1132e+11, device='cuda:0')
episode: 266 training return: tensor(-5.9647e+11, device='cuda:0')
episode: 267 training return: tensor(-6.1036e+11, device='cuda:0')
epoch: 67 test_true_pfm: -10.465881814564641
episode: 268 training return: tensor(-6.7169e+11, device='cuda:0')
episode: 269 training return: tensor(-5.9863e+11, device='cuda:0')
episode: 270 training return: tensor(-5.9315e+11, device='cuda:0')
episode: 271 training return: tensor(-6.0132e+11, device='cuda:0')
epoch: 68 test_true_pfm: -10.576759732506543
episode: 272 training return: tensor(-6.2711e+11, device='cuda:0')
episode: 273 training return: tensor(-5.8623e+11, device='cuda:0')
episode: 274 training return: tensor(-7.6854e+11, device='cuda:0')
episode: 275 training return: tensor(-6.3597e+11, device='cuda:0')
epoch: 69 test_true_pfm: -10.89712364772091
episode: 276 training return: tensor(-7.9281e+11, device='cuda:0')
episode: 277 training return: tensor(-6.5641e+11, device='cuda:0')
episode: 278 training return: tensor(-5.8698e+11, device='cuda:0')
episode: 279 training return: tensor(-6.7986e+11, device='cuda:0')
epoch: 70 test_true_pfm: -7.391873487796204
episode: 280 training return: tensor(-6.6765e+11, device='cuda:0')
episode: 281 training return: tensor(-5.9428e+11, device='cuda:0')
episode: 282 training return: tensor(-7.8266e+11, device='cuda:0')
episode: 283 training return: tensor(-6.0076e+11, device='cuda:0')
epoch: 71 test_true_pfm: -10.829488008151307
episode: 284 training return: tensor(-6.1138e+11, device='cuda:0')
episode: 285 training return: tensor(-3.9446e+11, device='cuda:0')
episode: 286 training return: tensor(-6.7435e+11, device='cuda:0')
episode: 287 training return: tensor(-6.6388e+11, device='cuda:0')
epoch: 72 test_true_pfm: -11.036055637434353
episode: 288 training return: tensor(-7.4087e+11, device='cuda:0')
episode: 289 training return: tensor(-5.9779e+11, device='cuda:0')
episode: 290 training return: tensor(-6.0709e+11, device='cuda:0')
episode: 291 training return: tensor(-5.9830e+11, device='cuda:0')
epoch: 73 test_true_pfm: -11.251174613586036
episode: 292 training return: tensor(-6.6155e+11, device='cuda:0')
episode: 293 training return: tensor(-1.0551e+10, device='cuda:0')
episode: 294 training return: tensor(-1.0341e+10, device='cuda:0')
episode: 295 training return: tensor(-6.6246e+11, device='cuda:0')
epoch: 74 test_true_pfm: -11.09736410544975
episode: 296 training return: tensor(-6.5527e+11, device='cuda:0')
episode: 297 training return: tensor(-6.1095e+11, device='cuda:0')
episode: 298 training return: tensor(-6.1016e+11, device='cuda:0')
episode: 299 training return: tensor(-6.6232e+11, device='cuda:0')
epoch: 75 test_true_pfm: -11.052811306581432
episode: 300 training return: tensor(-6.1211e+11, device='cuda:0')
episode: 301 training return: tensor(-6.0326e+11, device='cuda:0')
episode: 302 training return: tensor(-5.9669e+11, device='cuda:0')
episode: 303 training return: tensor(-6.1227e+11, device='cuda:0')
epoch: 76 test_true_pfm: -8.696909780976966
episode: 304 training return: tensor(-5.9651e+11, device='cuda:0')
episode: 305 training return: tensor(-6.1457e+11, device='cuda:0')
episode: 306 training return: tensor(-7.7815e+11, device='cuda:0')
episode: 307 training return: tensor(-5.9564e+11, device='cuda:0')
epoch: 77 test_true_pfm: -10.616235402590105
episode: 308 training return: tensor(-6.6405e+11, device='cuda:0')
episode: 309 training return: tensor(-1.4678e+12, device='cuda:0')
episode: 310 training return: tensor(-4.4626e+11, device='cuda:0')
episode: 311 training return: tensor(-1.1034e+17, device='cuda:0')
epoch: 78 test_true_pfm: -11.308975134745545
episode: 312 training return: tensor(-5.8900e+11, device='cuda:0')
episode: 313 training return: tensor(-3.0884e+16, device='cuda:0')
episode: 314 training return: tensor(-7.1451e+13, device='cuda:0')
episode: 315 training return: tensor(-5.1342e+15, device='cuda:0')
epoch: 79 test_true_pfm: 81.1061167213602
episode: 316 training return: tensor(-2.4003e+12, device='cuda:0')
episode: 317 training return: tensor(-4.1926e+13, device='cuda:0')
episode: 318 training return: tensor(-1.5391e+10, device='cuda:0')
episode: 319 training return: tensor(-1.4266e+10, device='cuda:0')
epoch: 80 test_true_pfm: -4.934832116487345
episode: 320 training return: tensor(-1.2099e+10, device='cuda:0')
episode: 321 training return: tensor(-6.3163e+11, device='cuda:0')
episode: 322 training return: tensor(-6.6740e+11, device='cuda:0')
episode: 323 training return: tensor(-6.7943e+11, device='cuda:0')
epoch: 81 test_true_pfm: -11.073500694646887
episode: 324 training return: tensor(-5.9967e+11, device='cuda:0')
episode: 325 training return: tensor(-8.0181e+11, device='cuda:0')
episode: 326 training return: tensor(-7.9452e+11, device='cuda:0')
episode: 327 training return: tensor(-6.0839e+11, device='cuda:0')
epoch: 82 test_true_pfm: -11.185196582159216
episode: 328 training return: tensor(-7.9365e+11, device='cuda:0')
episode: 329 training return: tensor(-5.8661e+11, device='cuda:0')
episode: 330 training return: tensor(-6.0760e+11, device='cuda:0')
episode: 331 training return: tensor(-1.0860e+17, device='cuda:0')
epoch: 83 test_true_pfm: -10.820994489667846
episode: 332 training return: tensor(-5.3250e+11, device='cuda:0')
episode: 333 training return: tensor(-1.4865e+12, device='cuda:0')
episode: 334 training return: tensor(-2.3074e+13, device='cuda:0')
episode: 335 training return: tensor(-1.0416e+12, device='cuda:0')
epoch: 84 test_true_pfm: -10.635252918271997
episode: 336 training return: tensor(-5.9754e+11, device='cuda:0')
episode: 337 training return: tensor(-5.9923e+11, device='cuda:0')
episode: 338 training return: tensor(-6.1478e+11, device='cuda:0')
episode: 339 training return: tensor(-5.9257e+11, device='cuda:0')
epoch: 85 test_true_pfm: -11.502488102389194
episode: 340 training return: tensor(-6.0427e+11, device='cuda:0')
episode: 341 training return: tensor(-7.4103e+11, device='cuda:0')
episode: 342 training return: tensor(-1.0383e+10, device='cuda:0')
episode: 343 training return: tensor(-2.3745e+10, device='cuda:0')
epoch: 86 test_true_pfm: -10.770376252231523
episode: 344 training return: tensor(-6.6724e+11, device='cuda:0')
episode: 345 training return: tensor(-6.1568e+11, device='cuda:0')
episode: 346 training return: tensor(-5.9828e+11, device='cuda:0')
episode: 347 training return: tensor(-5.8610e+11, device='cuda:0')
epoch: 87 test_true_pfm: -3.4573327258612916
episode: 348 training return: tensor(-1.2353e+10, device='cuda:0')
episode: 349 training return: tensor(-1.4432e+10, device='cuda:0')
episode: 350 training return: tensor(-1.4247e+10, device='cuda:0')
episode: 351 training return: tensor(-7.4188e+11, device='cuda:0')
epoch: 88 test_true_pfm: -11.274905150729984
episode: 352 training return: tensor(-6.4307e+11, device='cuda:0')
episode: 353 training return: tensor(-6.1043e+11, device='cuda:0')
episode: 354 training return: tensor(-6.0085e+11, device='cuda:0')
episode: 355 training return: tensor(-6.5040e+11, device='cuda:0')
epoch: 89 test_true_pfm: -11.844190084142712
episode: 356 training return: tensor(-5.8663e+11, device='cuda:0')
episode: 357 training return: tensor(-5.8393e+11, device='cuda:0')
episode: 358 training return: tensor(-6.0487e+11, device='cuda:0')
episode: 359 training return: tensor(-1.8545e+10, device='cuda:0')
epoch: 90 test_true_pfm: -7.467551704797093
episode: 360 training return: tensor(-1.4076e+10, device='cuda:0')
episode: 361 training return: tensor(-4.1492e+16, device='cuda:0')
episode: 362 training return: tensor(-1.0915e+10, device='cuda:0')
episode: 363 training return: tensor(-1.0481e+10, device='cuda:0')
epoch: 91 test_true_pfm: -23.009420530679392
episode: 364 training return: tensor(-1.0406e+10, device='cuda:0')
episode: 365 training return: tensor(-5.5783e+14, device='cuda:0')
episode: 366 training return: tensor(-1.0939e+10, device='cuda:0')
episode: 367 training return: tensor(-1.0158e+10, device='cuda:0')
epoch: 92 test_true_pfm: -9.854414789577676
episode: 368 training return: tensor(-6.9706e+12, device='cuda:0')
episode: 369 training return: tensor(-4.0025e+11, device='cuda:0')
episode: 370 training return: tensor(-2.3603e+16, device='cuda:0')
episode: 371 training return: tensor(-2.4958e+16, device='cuda:0')
epoch: 93 test_true_pfm: 15.738349224377918
episode: 372 training return: tensor(-3.2166e+16, device='cuda:0')
episode: 373 training return: tensor(-2.1926e+14, device='cuda:0')
episode: 374 training return: tensor(-1.2327e+17, device='cuda:0')
episode: 375 training return: tensor(-1.1497e+16, device='cuda:0')
epoch: 94 test_true_pfm: -11.220260493717591
episode: 376 training return: tensor(-7.2482e+11, device='cuda:0')
episode: 377 training return: tensor(-6.0254e+11, device='cuda:0')
episode: 378 training return: tensor(-1.0423e+10, device='cuda:0')
episode: 379 training return: tensor(-6.5483e+10, device='cuda:0')
epoch: 95 test_true_pfm: -14.06758650825268
episode: 380 training return: tensor(-1.3140e+10, device='cuda:0')
episode: 381 training return: tensor(-1.0389e+10, device='cuda:0')
episode: 382 training return: tensor(-1.0696e+10, device='cuda:0')
episode: 383 training return: tensor(-5.9584e+11, device='cuda:0')
epoch: 96 test_true_pfm: 86.28819624849741
episode: 384 training return: tensor(-1.1050e+17, device='cuda:0')
episode: 385 training return: tensor(-2.0083e+15, device='cuda:0')
episode: 386 training return: tensor(-6.1399e+11, device='cuda:0')
episode: 387 training return: tensor(-3.4892e+11, device='cuda:0')
epoch: 97 test_true_pfm: -10.53408093416733
episode: 388 training return: tensor(-6.0828e+11, device='cuda:0')
episode: 389 training return: tensor(-5.5689e+11, device='cuda:0')
episode: 390 training return: tensor(-5.3482e+11, device='cuda:0')
episode: 391 training return: tensor(-5.3281e+11, device='cuda:0')
epoch: 98 test_true_pfm: 23.896040316653266
episode: 392 training return: tensor(-4.0784e+12, device='cuda:0')
episode: 393 training return: tensor(-1.1108e+12, device='cuda:0')
episode: 394 training return: tensor(-1.4478e+12, device='cuda:0')
episode: 395 training return: tensor(-5.6654e+11, device='cuda:0')
epoch: 99 test_true_pfm: -10.7871396504758
episode: 396 training return: tensor(-5.3641e+11, device='cuda:0')
episode: 397 training return: tensor(-5.3511e+11, device='cuda:0')
episode: 398 training return: tensor(-5.4119e+11, device='cuda:0')
episode: 399 training return: tensor(-5.3301e+11, device='cuda:0')
epoch: 100 test_true_pfm: -10.689681946616629
episode: 400 training return: tensor(-5.3161e+11, device='cuda:0')
episode: 401 training return: tensor(-3.4721e+15, device='cuda:0')
episode: 402 training return: tensor(-1.1708e+12, device='cuda:0')
episode: 403 training return: tensor(-1.6693e+12, device='cuda:0')
epoch: 101 test_true_pfm: 13.195350548322226
episode: 404 training return: tensor(-5.0677e+11, device='cuda:0')
episode: 405 training return: tensor(-6.6724e+12, device='cuda:0')
episode: 406 training return: tensor(-2.0820e+10, device='cuda:0')
episode: 407 training return: tensor(-2.1112e+11, device='cuda:0')
epoch: 102 test_true_pfm: -11.187875234369477
episode: 408 training return: tensor(-6.0892e+11, device='cuda:0')
episode: 409 training return: tensor(-6.0201e+11, device='cuda:0')
episode: 410 training return: tensor(-7.9057e+11, device='cuda:0')
episode: 411 training return: tensor(-7.8789e+12, device='cuda:0')
epoch: 103 test_true_pfm: -20.09415152359984
episode: 412 training return: tensor(-1.1260e+10, device='cuda:0')
episode: 413 training return: tensor(-6.0522e+11, device='cuda:0')
episode: 414 training return: tensor(-5.8580e+11, device='cuda:0')
episode: 415 training return: tensor(-6.1003e+11, device='cuda:0')
epoch: 104 test_true_pfm: -11.411988518015667
episode: 416 training return: tensor(-6.1386e+11, device='cuda:0')
episode: 417 training return: tensor(-6.0952e+11, device='cuda:0')
episode: 418 training return: tensor(-7.7723e+11, device='cuda:0')
episode: 419 training return: tensor(-6.0514e+11, device='cuda:0')
epoch: 105 test_true_pfm: -10.754474449426937
episode: 420 training return: tensor(-5.8374e+11, device='cuda:0')
episode: 421 training return: tensor(-6.0077e+11, device='cuda:0')
episode: 422 training return: tensor(-6.1140e+11, device='cuda:0')
episode: 423 training return: tensor(-6.0907e+11, device='cuda:0')
epoch: 106 test_true_pfm: 2.821952173913914
episode: 424 training return: tensor(-2.9521e+10, device='cuda:0')
episode: 425 training return: tensor(-7.6850e+11, device='cuda:0')
episode: 426 training return: tensor(-6.0004e+11, device='cuda:0')
episode: 427 training return: tensor(-5.8378e+11, device='cuda:0')
epoch: 107 test_true_pfm: -7.849315576948617
episode: 428 training return: tensor(-6.5625e+11, device='cuda:0')
episode: 429 training return: tensor(-6.1297e+11, device='cuda:0')
episode: 430 training return: tensor(-7.8122e+11, device='cuda:0')
episode: 431 training return: tensor(-1.7454e+12, device='cuda:0')
epoch: 108 test_true_pfm: -11.221642538401062
episode: 432 training return: tensor(-6.0760e+11, device='cuda:0')
episode: 433 training return: tensor(-7.4200e+11, device='cuda:0')
episode: 434 training return: tensor(-6.0623e+11, device='cuda:0')
episode: 435 training return: tensor(-1.2414e+14, device='cuda:0')
epoch: 109 test_true_pfm: -10.851452700546702
episode: 436 training return: tensor(-1.0799e+10, device='cuda:0')
episode: 437 training return: tensor(-7.7480e+11, device='cuda:0')
episode: 438 training return: tensor(-6.6965e+11, device='cuda:0')
episode: 439 training return: tensor(-6.2531e+11, device='cuda:0')
epoch: 110 test_true_pfm: -11.307395639874981
episode: 440 training return: tensor(-6.0049e+11, device='cuda:0')
episode: 441 training return: tensor(-1.0705e+14, device='cuda:0')
episode: 442 training return: tensor(-1.0954e+10, device='cuda:0')
episode: 443 training return: tensor(-1.0816e+10, device='cuda:0')
epoch: 111 test_true_pfm: -10.288171478511009
episode: 444 training return: tensor(-1.0849e+10, device='cuda:0')
episode: 445 training return: tensor(-1.0820e+10, device='cuda:0')
episode: 446 training return: tensor(-1.0792e+10, device='cuda:0')
episode: 447 training return: tensor(-6.1030e+11, device='cuda:0')
epoch: 112 test_true_pfm: -11.253301124933794
episode: 448 training return: tensor(-1.3354e+10, device='cuda:0')
episode: 449 training return: tensor(-1.5675e+10, device='cuda:0')
episode: 450 training return: tensor(-1.0617e+10, device='cuda:0')
episode: 451 training return: tensor(-1.0697e+10, device='cuda:0')
epoch: 113 test_true_pfm: -8.024264659092758
episode: 452 training return: tensor(-7.8258e+11, device='cuda:0')
episode: 453 training return: tensor(-7.2939e+11, device='cuda:0')
episode: 454 training return: tensor(-7.9105e+11, device='cuda:0')
episode: 455 training return: tensor(-7.7073e+11, device='cuda:0')
epoch: 114 test_true_pfm: 7.504487728127022
episode: 456 training return: tensor(-1.5194e+10, device='cuda:0')
episode: 457 training return: tensor(-1.4592e+08, device='cuda:0')
episode: 458 training return: tensor(-2.7553e+09, device='cuda:0')
episode: 459 training return: tensor(-1.2515e+12, device='cuda:0')
epoch: 115 test_true_pfm: -9.428452745970517
episode: 460 training return: tensor(-1.7804e+12, device='cuda:0')
episode: 461 training return: tensor(-1.7002e+10, device='cuda:0')
episode: 462 training return: tensor(-6.2643e+11, device='cuda:0')
episode: 463 training return: tensor(-5.7067e+11, device='cuda:0')
epoch: 116 test_true_pfm: -10.834020038498318
episode: 464 training return: tensor(-6.1226e+11, device='cuda:0')
episode: 465 training return: tensor(-6.1035e+11, device='cuda:0')
episode: 466 training return: tensor(-6.0725e+11, device='cuda:0')
episode: 467 training return: tensor(-1.0169e+10, device='cuda:0')
epoch: 117 test_true_pfm: -11.281720443726892
episode: 468 training return: tensor(-7.3181e+11, device='cuda:0')
episode: 469 training return: tensor(-2.5705e+15, device='cuda:0')
episode: 470 training return: tensor(-6.3859e+11, device='cuda:0')
episode: 471 training return: tensor(-6.1084e+11, device='cuda:0')
epoch: 118 test_true_pfm: -10.959291534351161
episode: 472 training return: tensor(-5.9788e+11, device='cuda:0')
episode: 473 training return: tensor(-2.6493e+08, device='cuda:0')
episode: 474 training return: tensor(-1.2262e+10, device='cuda:0')
episode: 475 training return: tensor(-6.0658e+11, device='cuda:0')
epoch: 119 test_true_pfm: -10.711043695425593
episode: 476 training return: tensor(-5.9936e+11, device='cuda:0')
episode: 477 training return: tensor(-1.3227e+12, device='cuda:0')
episode: 478 training return: tensor(-7.5734e+10, device='cuda:0')
episode: 479 training return: tensor(-2.7203e+10, device='cuda:0')
epoch: 120 test_true_pfm: -8.853867063881266
episode: 480 training return: tensor(-1.7042e+10, device='cuda:0')
episode: 481 training return: tensor(-1.0659e+10, device='cuda:0')
episode: 482 training return: tensor(-5.2986e+11, device='cuda:0')
episode: 483 training return: tensor(-1.0282e+10, device='cuda:0')
epoch: 121 test_true_pfm: -13.444698299120462
episode: 484 training return: tensor(-2.8153e+10, device='cuda:0')
episode: 485 training return: tensor(-6.4307e+11, device='cuda:0')
episode: 486 training return: tensor(-6.1209e+11, device='cuda:0')
episode: 487 training return: tensor(-5.9872e+11, device='cuda:0')
epoch: 122 test_true_pfm: -11.528241055814695
episode: 488 training return: tensor(-5.9971e+11, device='cuda:0')
episode: 489 training return: tensor(-6.1325e+11, device='cuda:0')
episode: 490 training return: tensor(-7.4574e+11, device='cuda:0')
episode: 491 training return: tensor(-6.0480e+11, device='cuda:0')
epoch: 123 test_true_pfm: -11.305734546805995
episode: 492 training return: tensor(-2.0627e+10, device='cuda:0')
episode: 493 training return: tensor(-3.7460e+15, device='cuda:0')
episode: 494 training return: tensor(-7.3772e+11, device='cuda:0')
episode: 495 training return: tensor(-6.7884e+11, device='cuda:0')
epoch: 124 test_true_pfm: -11.022334123449893
episode: 496 training return: tensor(-5.9450e+11, device='cuda:0')
episode: 497 training return: tensor(-5.9413e+11, device='cuda:0')
episode: 498 training return: tensor(-6.1313e+11, device='cuda:0')
episode: 499 training return: tensor(-2.9536e+10, device='cuda:0')
epoch: 125 test_true_pfm: -11.595259729529507
episode: 500 training return: tensor(-6.0868e+11, device='cuda:0')
episode: 501 training return: tensor(-6.0485e+11, device='cuda:0')
episode: 502 training return: tensor(-1.3262e+10, device='cuda:0')
episode: 503 training return: tensor(-1.0856e+10, device='cuda:0')
epoch: 126 test_true_pfm: -11.245599587225238
episode: 504 training return: tensor(-7.4644e+11, device='cuda:0')
episode: 505 training return: tensor(-5.9321e+11, device='cuda:0')
episode: 506 training return: tensor(-5.8941e+11, device='cuda:0')
episode: 507 training return: tensor(-6.6517e+11, device='cuda:0')
epoch: 127 test_true_pfm: -10.85854926672313
episode: 508 training return: tensor(-7.7517e+11, device='cuda:0')
episode: 509 training return: tensor(-5.7515e+11, device='cuda:0')
episode: 510 training return: tensor(-2.4887e+13, device='cuda:0')
episode: 511 training return: tensor(-8.8776e+10, device='cuda:0')
epoch: 128 test_true_pfm: -0.5979745349649702
episode: 512 training return: tensor(-1.7256e+11, device='cuda:0')
episode: 513 training return: tensor(-6.4026e+11, device='cuda:0')
episode: 514 training return: tensor(-1.6094e+10, device='cuda:0')
episode: 515 training return: tensor(-4.4371e+14, device='cuda:0')
epoch: 129 test_true_pfm: -14.142315833572768
episode: 516 training return: tensor(-1.0516e+10, device='cuda:0')
episode: 517 training return: tensor(-2.0492e+10, device='cuda:0')
episode: 518 training return: tensor(-3.8787e+11, device='cuda:0')
episode: 519 training return: tensor(-3.0631e+09, device='cuda:0')
epoch: 130 test_true_pfm: -14.56033473346339
episode: 520 training return: tensor(-1.0486e+10, device='cuda:0')
episode: 521 training return: tensor(-6.7553e+11, device='cuda:0')
episode: 522 training return: tensor(-1.1666e+12, device='cuda:0')
episode: 523 training return: tensor(-1.0056e+14, device='cuda:0')
epoch: 131 test_true_pfm: -19.033797456182487
episode: 524 training return: tensor(-1.6885e+10, device='cuda:0')
episode: 525 training return: tensor(-3.8381e+10, device='cuda:0')
episode: 526 training return: tensor(-1.9082e+10, device='cuda:0')
episode: 527 training return: tensor(-1.6570e+10, device='cuda:0')
epoch: 132 test_true_pfm: -10.08796534545834
episode: 528 training return: tensor(-1.1083e+10, device='cuda:0')
episode: 529 training return: tensor(-1.1054e+10, device='cuda:0')
episode: 530 training return: tensor(-6.0472e+11, device='cuda:0')
episode: 531 training return: tensor(-7.3050e+11, device='cuda:0')
epoch: 133 test_true_pfm: -10.649891245972189
episode: 532 training return: tensor(-8.7337e+11, device='cuda:0')
episode: 533 training return: tensor(-1.3505e+10, device='cuda:0')
episode: 534 training return: tensor(-1.1875e+12, device='cuda:0')
episode: 535 training return: tensor(-1.0142e+13, device='cuda:0')
epoch: 134 test_true_pfm: -11.016112454449285
episode: 536 training return: tensor(-6.0527e+11, device='cuda:0')
episode: 537 training return: tensor(-7.9671e+11, device='cuda:0')
episode: 538 training return: tensor(-5.9963e+11, device='cuda:0')
episode: 539 training return: tensor(-6.0130e+11, device='cuda:0')
epoch: 135 test_true_pfm: -11.588968576372139
episode: 540 training return: tensor(-6.0778e+11, device='cuda:0')
episode: 541 training return: tensor(-6.5919e+11, device='cuda:0')
episode: 542 training return: tensor(-7.8598e+11, device='cuda:0')
episode: 543 training return: tensor(-6.1380e+11, device='cuda:0')
epoch: 136 test_true_pfm: -11.518521214521314
episode: 544 training return: tensor(-7.6586e+11, device='cuda:0')
episode: 545 training return: tensor(-7.9365e+11, device='cuda:0')
episode: 546 training return: tensor(-6.1113e+11, device='cuda:0')
episode: 547 training return: tensor(-6.0322e+11, device='cuda:0')
epoch: 137 test_true_pfm: -11.199686030595968
episode: 548 training return: tensor(-6.0075e+11, device='cuda:0')
episode: 549 training return: tensor(-6.2987e+11, device='cuda:0')
episode: 550 training return: tensor(-6.0731e+11, device='cuda:0')
episode: 551 training return: tensor(-5.9696e+11, device='cuda:0')
epoch: 138 test_true_pfm: -11.346851389998042
episode: 552 training return: tensor(-6.4041e+11, device='cuda:0')
episode: 553 training return: tensor(-5.9529e+11, device='cuda:0')
episode: 554 training return: tensor(-6.1182e+11, device='cuda:0')
episode: 555 training return: tensor(-4.8525e+09, device='cuda:0')
epoch: 139 test_true_pfm: -11.196390462367853
episode: 556 training return: tensor(-1.1155e+10, device='cuda:0')
episode: 557 training return: tensor(-2.0980e+12, device='cuda:0')
episode: 558 training return: tensor(-6.1111e+11, device='cuda:0')
episode: 559 training return: tensor(-6.0759e+11, device='cuda:0')
epoch: 140 test_true_pfm: -11.04102321349562
episode: 560 training return: tensor(-6.0035e+11, device='cuda:0')
episode: 561 training return: tensor(-6.3881e+11, device='cuda:0')
episode: 562 training return: tensor(-1.1484e+10, device='cuda:0')
episode: 563 training return: tensor(-2.3616e+10, device='cuda:0')
epoch: 141 test_true_pfm: -10.66744332898149
episode: 564 training return: tensor(-1.0539e+10, device='cuda:0')
episode: 565 training return: tensor(-6.5291e+11, device='cuda:0')
episode: 566 training return: tensor(-1.4187e+12, device='cuda:0')
episode: 567 training return: tensor(-3.3207e+14, device='cuda:0')
epoch: 142 test_true_pfm: -9.114416757971869
episode: 568 training return: tensor(-1.0640e+10, device='cuda:0')
episode: 569 training return: tensor(-1.1403e+10, device='cuda:0')
episode: 570 training return: tensor(-1.0600e+10, device='cuda:0')
episode: 571 training return: tensor(-1.0705e+10, device='cuda:0')
epoch: 143 test_true_pfm: -25.157352794564755
episode: 572 training return: tensor(-2.9512e+10, device='cuda:0')
episode: 573 training return: tensor(-6.2824e+09, device='cuda:0')
episode: 574 training return: tensor(-1.1448e+10, device='cuda:0')
episode: 575 training return: tensor(-6.0087e+11, device='cuda:0')
epoch: 144 test_true_pfm: -11.376331151139809
episode: 576 training return: tensor(-1.0243e+10, device='cuda:0')
episode: 577 training return: tensor(-2.2535e+10, device='cuda:0')
episode: 578 training return: tensor(-2.8277e+11, device='cuda:0')
episode: 579 training return: tensor(-2.6857e+10, device='cuda:0')
epoch: 145 test_true_pfm: -10.075744760817747
episode: 580 training return: tensor(-1.1334e+10, device='cuda:0')
episode: 581 training return: tensor(-1.0330e+10, device='cuda:0')
episode: 582 training return: tensor(-1.1224e+10, device='cuda:0')
episode: 583 training return: tensor(-6.1295e+11, device='cuda:0')
epoch: 146 test_true_pfm: -13.06544846125031
episode: 584 training return: tensor(-1.0500e+10, device='cuda:0')
episode: 585 training return: tensor(-2.2515e+10, device='cuda:0')
episode: 586 training return: tensor(-2.3291e+10, device='cuda:0')
episode: 587 training return: tensor(-2.3253e+10, device='cuda:0')
epoch: 147 test_true_pfm: -10.999967155791
episode: 588 training return: tensor(-6.5854e+11, device='cuda:0')
episode: 589 training return: tensor(-1.8966e+10, device='cuda:0')
episode: 590 training return: tensor(-1.1531e+10, device='cuda:0')
episode: 591 training return: tensor(-1.1144e+10, device='cuda:0')
epoch: 148 test_true_pfm: -4.459046476219872
episode: 592 training return: tensor(-1.0467e+10, device='cuda:0')
episode: 593 training return: tensor(-1.0904e+10, device='cuda:0')
episode: 594 training return: tensor(-1.8803e+10, device='cuda:0')
episode: 595 training return: tensor(-5.3237e+11, device='cuda:0')
epoch: 149 test_true_pfm: -11.354268829702514
episode: 596 training return: tensor(-6.0067e+11, device='cuda:0')
episode: 597 training return: tensor(-6.7234e+11, device='cuda:0')
episode: 598 training return: tensor(-6.0203e+11, device='cuda:0')
episode: 599 training return: tensor(-5.8427e+11, device='cuda:0')
epoch: 150 test_true_pfm: -11.145002489544998
