Add files using upload-large-folder tool

.gitattributes

@@ -963,3 +963,91 @@ ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data
 ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000015/c30_small__simulation_data5__sample_000015_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_25.9_C12_-41.5_phi12_350_C21_404_phi21_54.6_C23_-74.8_phi23_216_C30_4.62e+05_C32_.92e+07_phi56_262/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
 ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000048/c30_small__simulation_data5__sample_000048_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-17.8_C12_36.1_phi12_187_C21_390_phi21_2.56_C23_833_phi23_100.45e+07_phi54_343_C56_2e+07_phi56_117/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
 ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000047/c30_small__simulation_data4__sample_000047_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-276_C12_-45.5_phi12_100_C21_161_phi21_267_C23_211_phi23_261_C30_2.43e+06_C32_-1.14e+04_phi32_13.5/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug2__sample_000030/c30_small__simulation_data_aug2__sample_000030_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-121_C12_-34.9_phi12_330_C21_139_phi21_252_C23_-79.4_phi23_68.4_C30_2.49e+06_C32_-5.58e+03_p.9/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug1__sample_000012/c30_small__simulation_data_aug1__sample_000012_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-133_C12_-38.8_phi12_334_C21_694_phi21_302_C23_96.7_phi23_332_C30_4.51e+06_C32_-8.79e+03_phi.8/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug3__sample_000026/c30_small__simulation_data_aug3__sample_000026_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-528_C12_-40.7_phi12_108_C21_-180_phi21_23.2_C23_-446_phi23_48.9_C30_4.17e+06_C32_-478_phi32.6/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug2__sample_000012/c30_small__simulation_data_aug2__sample_000012_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-170_C12_-8.29_phi12_266_C21_631_phi21_158_C23_308_phi23_227_C30_3.42e+06_C32_.1e+07_phi56_204/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug1__sample_000020/c30_small__simulation_data_aug1__sample_000020_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-10.9_C12_-38.5_phi12_33_C21_-985_phi21_157_C23_740_phi23_336_C30_-1.92e+05_C.88e+06_phi56_188/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug1__sample_000027/c30_small__simulation_data_aug1__sample_000027_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-450_C12_45_phi12_336_C21_-909_phi21_287_C23_927_phi23_286_C30_3.76e+06_C32_-4.52e+03_phi32_.2/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000017/c30_small__simulation_data4__sample_000017_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-320_C12_10.2_phi12_185_C21_505_phi21_140_C23_186_phi23_308_C30_3.99e+05_C32_-4.6.62e+07_phi56_330/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data2__sample_000017/c30_small__simulation_data2__sample_000017_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-281_C12_-33.3_phi12_230_C21_-531_phi21_62.9_C23_-156_phi23_20.4_C30_2.18e+06_C32_1.03e+03_phi32.2/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000018/c30_small__simulation_data1__sample_000018_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-141_C12_42.8_phi12_166_C21_224_phi21_286_C23_-926_phi23_158_C30_3.37e+06_C32_2.5.45e+07_phi56_225/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000026/c30_small__simulation_data1__sample_000026_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-236_C12_17.2_phi12_257_C21_650_phi21_132_C23_-276_phi23_122_C30_4.8e+06_C32_-5.99e+03_phi32_358.9/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000027/c30_small__simulation_data4__sample_000027_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-84.6_C12_17_phi12_221_C21_211_phi21_58.7_C23_395_phi23_124_C30_2.86e+06_C32_-9.7.39e+07_phi56_242/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000005/c30_small__simulation_data3__sample_000005_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-533_C12_36_phi12_292_C21_-578_phi21_9.2_C23_365_phi23_146_C30_1.08e+06_C32_-1.18.86e+07_phi56_340/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug1__sample_000002/c30_small__simulation_data_aug1__sample_000002_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-76_C12_-9.38_phi12_298_C21_-617_phi21_261_C23_150_phi23_270_C30_-9.69e+05_C32.8e+06_phi56_297/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug3__sample_000022/c30_small__simulation_data_aug3__sample_000022_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_13.4_C12_42.8_phi12_331_C21_72.3_phi21_292_C23_901_phi23_82.4_C30_-9.66e+05_C.86e+07_phi56_203/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000008/c30_small__simulation_data3__sample_000008_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-637_C12_9.73_phi12_230_C21_-952_phi21_116_C23_-930_phi23_350_C30_2.36e+06_C32_6..85e+07_phi56_101/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug1__sample_000003/c30_small__simulation_data_aug1__sample_000003_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-437_C12_-33.9_phi12_37.6_C21_-788_phi21_250_C23_240_phi23_55.9_C30_4.2e+06_C32_5.54e+03_phi.3/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug2__sample_000016/c30_small__simulation_data_aug2__sample_000016_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-533_C12_-33.9_phi12_344_C21_-752_phi21_344_C23_-160_phi23_293_C30_3.86e+06_C32_-6.95e+03_ph.7/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug2__sample_000000/c30_small__simulation_data_aug2__sample_000000_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-565_C12_-3.36_phi12_63.6_C21_-472_phi21_337_C23_165_phi23_259_C30_2.69e+06_C.61e+06_phi56_307/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000020/c30_small__simulation_data3__sample_000020_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-499_C12_-34.2_phi12_62.1_C21_-432_phi21_63.5_C23_-39.2_phi23_128_C30_1.32e+06_C3.38e+06_phi56_337/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug3__sample_000030/c30_small__simulation_data_aug3__sample_000030_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-629_C12_19.5_phi12_302_C21_63_phi21_356_C23_-737_phi23_322_C30_3.76e+06_C32_1.7e+07_phi56_317/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000028/c30_small__simulation_data5__sample_000028_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-37.9_C12_-4.04_phi12_269_C21_890_phi21_107_C23_-228_phi23_154_C30_2.79e+06_C32_-9.21e+03_phi32_.3/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data2__sample_000040/c30_small__simulation_data2__sample_000040_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-482_C12_-12.9_phi12_223_C21_391_phi21_164_C23_-785_phi23_29_C30_1.34e+06_C32_-8..07e+07_phi56_297/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data2__sample_000021/c30_small__simulation_data2__sample_000021_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-217_C12_-17.8_phi12_169_C21_-587_phi21_213_C23_824_phi23_163_C30_6.87e+05_C32_8..62e+07_phi56_238/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data2__sample_000010/c30_small__simulation_data2__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-281_C12_-45_phi12_252_C21_-635_phi21_274_C23_-598_phi23_214_C30_-6.92e+05_C32_5..43e+07_phi56_192/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000043/c30_small__simulation_data5__sample_000043_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_45.5_C12_26.9_phi12_138_C21_403_phi21_339_C23_-846_phi23_254_C30_3.01e+06_C32_6.3.35e+06_phi56_275/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000007/c30_small__simulation_data3__sample_000007_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-235_C12_-32.7_phi12_24.1_C21_554_phi21_351_C23_545_phi23_268_C30_2.56e+06_C32_3..76e+06_phi56_246/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000035/c30_small__simulation_data3__sample_000035_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-22_C12_9.41_phi12_231_C21_313_phi21_77.8_C23_-891_phi23_10.6_C30_2.45e+06_C32_-3.22e+07_phi56_249/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug2__sample_000032/c30_small__simulation_data_aug2__sample_000032_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-658_C12_-20.4_phi12_3.54_C21_980_phi21_64.5_C23_170_phi23_120_C30_4.9e+06_C32_7.78e+03_phi3.4/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000049/c30_small__simulation_data5__sample_000049_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-596_C12_23.6_phi12_26.9_C21_638_phi21_325_C23_-939_phi23_121_C30_2.19e+06_C32_9..99e+05_phi56_325/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug3__sample_000043/c30_small__simulation_data_aug3__sample_000043_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-503_C12_4.46_phi12_292_C21_-942_phi21_251_C23_596_phi23_146_C30_5.03e+05_C32.54e+07_phi56_317/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000004/c30_small__simulation_data1__sample_000004_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-309_C12_0.25_phi12_218_C21_883_phi21_66.5_C23_293_phi23_121_C30_3.42e+06_C32_775_phi32_235_C34_.6/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data2__sample_000034/c30_small__simulation_data2__sample_000034_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-303_C12_46.9_phi12_286_C21_127_phi21_133_C23_144_phi23_138_C30_2.65e+06_C32_5.26e+03_phi32_8.61.8/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug1__sample_000028/c30_small__simulation_data_aug1__sample_000028_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-1.19_C12_3.72_phi12_71.2_C21_-196_phi21_153_C23_-253_phi23_292_C30_4.95e+06_C.9e+07_phi56_342/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000033/c30_small__simulation_data3__sample_000033_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-185_C12_-10.1_phi12_352_C21_-667_phi21_324_C23_-698_phi23_158_C30_1.14e+06_C32_-.32e+06_phi56_161/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000029/c30_small__simulation_data3__sample_000029_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-198_C12_-4.8_phi12_306_C21_384_phi21_151_C23_361_phi23_94.1_C30_4.12e+06_C32_8.8.35e+07_phi56_171/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug3__sample_000040/c30_small__simulation_data_aug3__sample_000040_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-354_C12_-22.9_phi12_53.5_C21_-526_phi21_58.8_C23_962_phi23_310_C30_3.77e+06_.36e+07_phi56_306/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000008/c30_small__simulation_data5__sample_000008_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-424_C12_-40.6_phi12_81.7_C21_669_phi21_208_C23_-414_phi23_223_C30_3.08e+05_C32_-8.34e+03_phi32_.4/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug1__sample_000010/c30_small__simulation_data_aug1__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-163_C12_20.4_phi12_117_C21_445_phi21_272_C23_-194_phi23_126_C30_4.33e+06_C32.61e+07_phi56_259/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000009/c30_small__simulation_data1__sample_000009_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-267_C12_18.2_phi12_230_C21_758_phi21_299_C23_882_phi23_3.72_C30_1.33e+05_C32_9.9.88e+05_phi56_333/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000002/c30_small__simulation_data5__sample_000002_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-173_C12_17.9_phi12_159_C21_293_phi21_274_C23_951_phi23_203_C30_2.24e+06_C32_-2.78e+03_phi32_26..4/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000045/c30_small__simulation_data5__sample_000045_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-556_C12_-16.3_phi12_51.9_C21_709_phi21_109_C23_-277_phi23_168_C30_1.8e+06_C32_-5.33e+06_phi56_166/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000019/c30_small__simulation_data4__sample_000019_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-22_C12_27.6_phi12_96.6_C21_-714_phi21_349_C23_-858_phi23_296_C30_3.66e+05_C32_1.9.55e+07_phi56_28/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000032/c30_small__simulation_data3__sample_000032_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-417_C12_4.32_phi12_30.1_C21_-183_phi21_341_C23_968_phi23_181_C30_1.99e+05_C32_2..89e+07_phi56_301/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000024/c30_small__simulation_data4__sample_000024_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-3.75_C12_27.6_phi12_25.6_C21_-779_phi21_359_C23_-671_phi23_122_C30_-3.76e+05_C32.79e+07_phi56_165/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug3__sample_000010/c30_small__simulation_data_aug3__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-326_C12_1.9_phi12_209_C21_981_phi21_162_C23_64.4_phi23_206_C30_2.19e+06_C32_-8.78e+03_phi32.7/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug2__sample_000035/c30_small__simulation_data_aug2__sample_000035_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-470_C12_11.6_phi12_51.1_C21_-463_phi21_111_C23_937_phi23_176_C30_1.03e+06_C3.16e+06_phi56_257/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data2__sample_000033/c30_small__simulation_data2__sample_000033_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_46.7_C12_-4.53_phi12_82.4_C21_553_phi21_66.6_C23_-458_phi23_360_C30_1.98e+06_C32_.02e+07_phi56_237/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug2__sample_000026/c30_small__simulation_data_aug2__sample_000026_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-422_C12_-48.8_phi12_214_C21_363_phi21_165_C23_526_phi23_3.5_C30_8.45e+05_C32_.7e+06_phi56_218/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000034/c30_small__simulation_data3__sample_000034_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-38.3_C12_37.8_phi12_222_C21_-394_phi21_261_C23_-390_phi23_28.3_C30_-8.32e+0.5_C56_2e+07_phi56_342/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000034/c30_small__simulation_data5__sample_000034_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-488_C12_40_phi12_193_C21_120_phi21_53.6_C23_28.1_phi23_148_C30_4.14e+06_C32_1.36.19e+05_phi56_267/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data2__sample_000032/c30_small__simulation_data2__sample_000032_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-460_C12_-11.7_phi12_63.9_C21_-130_phi21_349_C23_-977_phi23_161_C30_1.92e+06_C32_.97e+07_phi56_139/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000012/c30_small__simulation_data3__sample_000012_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_26.7_C12_-19.5_phi12_306_C21_690_phi21_356_C23_60.3_phi23_222_C30_5.67e+04_C32_-7.06e+07_phi56_184/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000037/c30_small__simulation_data4__sample_000037_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-223_C12_-0.047_phi12_206_C21_226_phi21_12.9_C23_-287_phi23_142_C30_-3.17e+05_C32.76e+06_phi56_341/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000033/c30_small__simulation_data4__sample_000033_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-290_C12_-29_phi12_134_C21_-67.7_phi21_199_C23_435_phi23_343_C30_1.47e+05_C32_-8..31e+07_phi56_348/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000021/c30_small__simulation_data3__sample_000021_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-173_C12_42.1_phi12_130_C21_976_phi21_344_C23_-650_phi23_78.2_C30_2.62e+06_C32_2..87e+07_phi56_195/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000028/c30_small__simulation_data3__sample_000028_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-430_C12_47.3_phi12_226_C21_899_phi21_86.6_C23_-716_phi23_180_C30_1.18e+06_C32_6..85e+07_phi56_139/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000020/c30_small__simulation_data5__sample_000020_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-13.3_C12_-19_phi12_197_C21_-186_phi21_246_C23_-23.8_phi23_270_C30_1.08e+06_C32_-.14e+07_phi56_203/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000016/c30_small__simulation_data1__sample_000016_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-644_C12_-38.8_phi12_109_C21_-35.9_phi21_346_C23_643_phi23_271_C30_1.46e+06_C32_4.66e+07_phi56_303/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug2__sample_000036/c30_small__simulation_data_aug2__sample_000036_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-180_C12_22.1_phi12_212_C21_-933_phi21_359_C23_312_phi23_295_C30_4.93e+06_C32.54e+07_phi56_340/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug2__sample_000017/c30_small__simulation_data_aug2__sample_000017_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-592_C12_11.7_phi12_89_C21_697_phi21_194_C23_-146_phi23_126_C30_1.13e+06_C32_-.2e+07_phi56_210/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data3__sample_000009/c30_small__simulation_data3__sample_000009_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-166_C12_30_phi12_343_C21_-452_phi21_165_C23_-219_phi23_177_C30_1.02e+06_C32_8.99e+03_phi32_66.1.9/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug1__sample_000019/c30_small__simulation_data_aug1__sample_000019_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-122_C12_-0.499_phi12_236_C21_-698_phi21_251_C23_-411_phi23_142_C30_-6.97e+05.12e+06_phi56_201/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug3__sample_000035/c30_small__simulation_data_aug3__sample_000035_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-677_C12_21.9_phi12_173_C21_-394_phi21_72.5_C23_-866_phi23_219_C30_2.55e+06_C3.8e+06_phi56_184/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug1__sample_000015/c30_small__simulation_data_aug1__sample_000015_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-311_C12_-46.5_phi12_111_C21_-982_phi21_155_C23_980_phi23_280_C30_3.88e+06_C3.92e+05_phi56_265/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000019/c30_small__simulation_data1__sample_000019_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_69.8_C12_0.828_phi12_138_C21_806_phi21_72.5_C23_640_phi23_275_C30_1.32e+06_C32_5.13e+03_phi32_23.8/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000025/c30_small__simulation_data4__sample_000025_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-686_C12_46.6_phi12_62.9_C21_-56.1_phi21_39.9_C23_-490_phi23_160_C30_3.24e+06_C32.49e+07_phi56_344/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug2__sample_000037/c30_small__simulation_data_aug2__sample_000037_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-369_C12_-1.7_phi12_63.5_C21_-999_phi21_159_C23_442_phi23_241_C30_5.06e+04_C3.09e+06_phi56_177/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000013/c30_small__simulation_data5__sample_000013_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_37.3_C12_10.5_phi12_229_C21_-81.6_phi21_312_C23_-805_phi23_343_C30_4.29e+06_C32_-8.72e+03_phi32_.6/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000021/c30_small__simulation_data1__sample_000021_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-260_C12_20.2_phi12_355_C21_29.5_phi21_352_C23_90.8_phi23_20.1_C30_2.86e+06_C32_-.92e+07_phi56_322/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug3__sample_000018/c30_small__simulation_data_aug3__sample_000018_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_19.3_C12_38.1_phi12_134_C21_-790_phi21_68.4_C23_-563_phi23_226_C30_2.94e+06_C.97e+07_phi56_230/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000033/c30_small__simulation_data5__sample_000033_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-371_C12_7.54_phi12_302_C21_-954_phi21_116_C23_-148_phi23_215_C30_1.07e+06_C32_-1.19e+07_phi56_160/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000025/c30_small__simulation_data1__sample_000025_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-466_C12_-4.33_phi12_247_C21_885_phi21_331_C23_177_phi23_50.9_C30_1.72e+06_C32_41.56e+07_phi56_242/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000043/c30_small__simulation_data4__sample_000043_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-640_C12_-17.7_phi12_49.5_C21_-120_phi21_108_C23_-32.8_phi23_91.8_C30_2.91e+06_C3.03e+07_phi56_130/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000026/c30_small__simulation_data5__sample_000026_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-587_C12_-36.8_phi12_109_C21_440_phi21_255_C23_-480_phi23_204_C30_2.4e+06_C32_2.42e+03_phi32_143.7/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000040/c30_small__simulation_data5__sample_000040_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-291_C12_-0.844_phi12_321_C21_-410_phi21_61.8_C23_822_phi23_345_C30_6.56e+05_C32_.88e+07_phi56_139/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000010/c30_small__simulation_data1__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_6.82_C12_-46.2_phi12_189_C21_698_phi21_28_C23_-991_phi23_186_C30_2.87e+06_C32_-7..72e+07_phi56_117/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000002/c30_small__simulation_data1__sample_000002_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-4.82_C12_0.327_phi12_59.7_C21_135_phi21_127_C23_251_phi23_21.1_C30_2.39e+06_C32_-5.99e+03_phi32.3/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000052/c30_small__simulation_data4__sample_000052_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-238_C12_35.6_phi12_31.7_C21_-515_phi21_204_C23_-297_phi23_238_C30_4.43e+06_C32_-3.02e+03_phi32_.9/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data2__sample_000024/c30_small__simulation_data2__sample_000024_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-192_C12_28_phi12_215_C21_104_phi21_32.8_C23_-809_phi23_223_C30_-2.88e+05_C32_-1..73e+05_phi56_333/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug3__sample_000015/c30_small__simulation_data_aug3__sample_000015_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_89.3_C12_35.1_phi12_207_C21_187_phi21_80_C23_-528_phi23_100_C30_-5.47e+05_C32.17e+07_phi56_198/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000000/c30_small__simulation_data4__sample_000000_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-610_C12_21.4_phi12_12.7_C21_-933_phi21_98.9_C23_-94.5_phi23_244_C30_3.95e+06_C32_-5.2e+03_phi32.9/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000005/c30_small__simulation_data4__sample_000005_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-413_C12_-9.57_phi12_11.3_C21_379_phi21_16.6_C23_161_phi23_287_C30_-8.96e+05_C32_.94e+05_phi56_204/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000039/c30_small__simulation_data4__sample_000039_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-266_C12_-38.6_phi12_131_C21_-18.2_phi21_65.8_C23_-823_phi23_75_C30_3.53e+06_C32_.65e+07_phi56_298/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data_aug3__sample_000042/c30_small__simulation_data_aug3__sample_000042_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-677_C12_41.1_phi12_38.4_C21_-400_phi21_307_C23_-728_phi23_217_C30_3.63e+06_C.47e+07_phi56_102/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000003/c30_small__simulation_data4__sample_000003_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-15.6_C12_20.3_phi12_316_C21_-830_phi21_217_C23_388_phi23_178_C30_3.67e+06_C32_-1.02e+04_phi32_1.5/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data4__sample_000021/c30_small__simulation_data4__sample_000021_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_93_C12_4.67_phi12_75.1_C21_175_phi21_222_C23_-579_phi23_343_C30_-8.29e+05_C32_7.5.33e+07_phi56_357/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data5__sample_000019/c30_small__simulation_data5__sample_000019_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-596_C12_-28.4_phi12_99.5_C21_-66.6_phi21_323_C23_-922_phi23_165_C30_2.17e+06_C32_8.5e+03_phi32_.8/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text
+ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data2__sample_000044/c30_small__simulation_data2__sample_000044_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-579_C12_-41.3_phi12_96.3_C21_851_phi21_37.7_C23_563_phi23_0.171_C30_1.05e+06_C32_.5e+06_phi56_236/model_iter0100.hdf5 filter=lfs diff=lfs merge=lfs -text

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000002/c30_small__simulation_data1__sample_000002_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-4.82_C12_0.327_phi12_59.7_C21_135_phi21_127_C23_251_phi23_21.1_C30_2.39e+06_C32_-5.99e+03_phi32.3/20260624_053439_ptyrad_log.txt

@@ -0,0 +1,417 @@
+2026-06-24 05:34:37,885 - ### PtyRAD LoggingManager configuration ###
+2026-06-24 05:34:37,886 - log_file       = 'ptyrad_log.txt'. If log_file = None, no log file will be created.
+2026-06-24 05:34:37,886 - log_dir        = 'auto'. If log_dir = 'auto', then log will be saved to `output_path` or 'logs/'.
+2026-06-24 05:34:37,886 - flush_file     = True. Automatically set to True if `log_file is not None`
+2026-06-24 05:34:37,886 - prefix_time    = datetime. If true, preset strings ('date', 'time', 'datetime'), or a string of time format, a datetime str is prefixed to the `log_file`.
+2026-06-24 05:34:37,886 - prefix_jobid   = '0'. If not 0, it'll be prefixed to the log file. This is used for hypertune mode with multiple GPUs.
+2026-06-24 05:34:37,886 - append_to_file = True. If true, logs will be appended to the existing file. If false, the log file will be overwritten.
+2026-06-24 05:34:37,886 - show_timestamp = True. If true, the printed information will contain a timestamp.
+2026-06-24 05:34:37,886 -  
+2026-06-24 05:34:37,970 - ### Initializing HuggingFace accelerator ###
+2026-06-24 05:34:37,970 - Accelerator.distributed_type = DistributedType.NO
+2026-06-24 05:34:37,970 - Accelerator.num_process      = 1
+2026-06-24 05:34:37,970 - Accelerator.mixed_precision  = no
+2026-06-24 05:34:37,980 - 'accelerate' is available but NOT using distributed mode or mixed precision
+2026-06-24 05:34:37,980 - If you want to utilize 'accelerate' for multiGPU or mixed precision, 
+2026-06-24 05:34:37,980 - Run `accelerate launch --multi_gpu --num_processes=2 --mixed_precision='no' -m ptyrad run <PTYRAD_ARGUMENTS> --gpuid 'acc'` in your terminal
+2026-06-24 05:34:37,980 -  
+2026-06-24 05:34:37,980 - ### System information ###
+2026-06-24 05:34:37,980 - Platform: Linux-4.18.0-553.69.1.el8_10.x86_64-x86_64-with-glibc2.28
+2026-06-24 05:34:37,980 - Operating System: Linux 4.18.0-553.69.1.el8_10.x86_64
+2026-06-24 05:34:37,980 - OS Version: #1 SMP Thu Aug 7 18:10:00 EDT 2025
+2026-06-24 05:34:37,980 - Machine: x86_64
+2026-06-24 05:34:37,980 - Processor: x86_64
+2026-06-24 05:34:37,980 - Available CPU cores: 8
+2026-06-24 05:34:37,980 - SLURM-Allocated Total Memory: 78.12 GB
+2026-06-24 05:34:37,980 -  
+2026-06-24 05:34:37,980 - ### GPU information ###
+2026-06-24 05:34:37,980 - CUDA Available: True
+2026-06-24 05:34:37,980 - CUDA Version: 13.0
+2026-06-24 05:34:37,981 - Available CUDA GPUs: ['NVIDIA A100 80GB PCIe']
+2026-06-24 05:34:37,981 - CUDA Compute Capability: ['8.0']
+2026-06-24 05:34:37,981 -   INFO: For torch.compile with Triton, you'll need CUDA GPU with Compute Capability >= 7.0.
+2026-06-24 05:34:37,981 -         In addition, Triton does not directly support Windows.
+2026-06-24 05:34:37,981 -         For Windows users, please follow the instruction and download `triton-windows` from https://github.com/woct0rdho/triton-windows.
+2026-06-24 05:34:37,996 - MIG (Multi-Instance GPU) mode = False
+2026-06-24 05:34:37,996 -   INFO: MIG splits a physical GPU into multiple GPU slices, but multiGPU does not support these MIG slices.
+2026-06-24 05:34:37,996 -         In addition, multiGPU is currently only available on Linux due to the limited NCCL support.
+2026-06-24 05:34:37,996 -       -> If you're doing normal reconstruction/hypertune, you can safely ignore this.
+2026-06-24 05:34:37,996 -       -> If you want to do multiGPU, you must provide multiple 'full' GPUs that are not in MIG mode.
+2026-06-24 05:34:37,996 -  
+2026-06-24 05:34:37,996 - ### Python information ###
+2026-06-24 05:34:37,998 - Python Executable: /home/tnguye11/anaconda3/envs/ptyrad/bin/python3.12
+2026-06-24 05:34:37,998 - Python Version: 3.12.13 | packaged by Anaconda, Inc. | (main, Mar 19 2026, 20:20:58) [GCC 14.3.0]
+2026-06-24 05:34:37,998 -  
+2026-06-24 05:34:37,998 - ### Packages information ###
+2026-06-24 05:34:37,999 - Numpy Version (metadata): 2.4.6
+2026-06-24 05:34:38,000 - PyTorch Version (metadata): 2.12.0
+2026-06-24 05:34:38,000 - Optuna Version (metadata): 4.9.0
+2026-06-24 05:34:38,001 - Accelerate Version (metadata): 1.13.0
+2026-06-24 05:34:38,001 - PtyRAD Version (ptyrad/__init__.py): 1.0.0
+2026-06-24 05:34:38,002 - PtyRAD is located at: /home/tnguye11/anaconda3/envs/ptyrad/lib/python3.12/site-packages/ptyrad/__init__.py
+2026-06-24 05:34:38,002 -  
+2026-06-24 05:34:38,002 - ### Loading params file ###
+2026-06-24 05:34:38,002 - params_path = /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/params/c30_small/simulation_data1__sample_000002.yaml
+2026-06-24 05:34:38,008 - validate = True: Filling defaults and validating the params file...
+2026-06-24 05:34:38,009 - Success! Params file validated and defaults applied.
+2026-06-24 05:34:38,009 -  
+2026-06-24 05:34:38,009 - ### Setting GPU Device ###
+2026-06-24 05:34:38,010 - Selected GPU device: cuda:0 (NVIDIA A100 80GB PCIe)
+2026-06-24 05:34:38,010 -  
+2026-06-24 05:34:38,010 - Random seed: 20261846 provided by params file
+2026-06-24 05:34:38,010 - ### Initializing Initializer ###
+2026-06-24 05:34:38,010 - init_params are displayed below:
+2026-06-24 05:34:38,010 -   random_seed: 20261846
+2026-06-24 05:34:38,010 -   probe_illum_type: electron
+2026-06-24 05:34:38,010 -   probe_kv: 200.0
+2026-06-24 05:34:38,010 -   probe_conv_angle: 12.791192054748535
+2026-06-24 05:34:38,010 -   probe_aberrations: {'C10': -4.824, 'C12': 0.327, 'phi12': 59.698, 'C21': 134.821, 'phi21': 127.139, 'C23': 251.001, 'phi23': 21.108, 'C30': 2393991.0, 'C32': -5988.508, 'phi32': 313.77, 'C34': -4476.031, 'phi34': 67.073, 'C41': -304813.312, 'phi41': 327.112, 'C43': 364127.75, 'phi43': 77.758, 'C45': -97685.359, 'phi45': 184.903, 'C50': -410699616.0, 'C52': -18312922.0, 'phi52': 64.88, 'C54': -17729770.0, 'phi54': 177.945, 'C56': -5185558.0, 'phi56': 49.278}
+2026-06-24 05:34:38,010 -   beam_kev: None
+2026-06-24 05:34:38,010 -   probe_dRn: None
+2026-06-24 05:34:38,010 -   probe_Rn: None
+2026-06-24 05:34:38,010 -   probe_D_H: None
+2026-06-24 05:34:38,010 -   probe_D_FZP: None
+2026-06-24 05:34:38,010 -   probe_Ls: None
+2026-06-24 05:34:38,010 -   meas_Npix: 128
+2026-06-24 05:34:38,010 -   pos_N_scans: 256
+2026-06-24 05:34:38,010 -   pos_N_scan_slow: 16
+2026-06-24 05:34:38,011 -   pos_N_scan_fast: 16
+2026-06-24 05:34:38,011 -   pos_scan_step_size: 0.4755619168281555
+2026-06-24 05:34:38,011 -   meas_calibration: {'mode': 'kMax', 'value': 2.5}
+2026-06-24 05:34:38,011 -   probe_pmode_max: 6
+2026-06-24 05:34:38,011 -   probe_pmode_init_pows: [0.02]
+2026-06-24 05:34:38,011 -   obj_omode_max: 1
+2026-06-24 05:34:38,011 -   obj_omode_init_occu: {'occu_type': 'uniform', 'init_occu': None}
+2026-06-24 05:34:38,011 -   obj_Nlayer: 1
+2026-06-24 05:34:38,011 -   obj_slice_thickness: 20.0
+2026-06-24 05:34:38,011 -   simu_Npix: None
+2026-06-24 05:34:38,011 -   simu_match_mode: None
+2026-06-24 05:34:38,011 -   meas_permute: None
+2026-06-24 05:34:38,011 -   meas_reshape: [256, 128, 128]
+2026-06-24 05:34:38,011 -   meas_flipT: [0, 0, 0]
+2026-06-24 05:34:38,011 -   meas_crop: None
+2026-06-24 05:34:38,011 -   meas_pad: None
+2026-06-24 05:34:38,011 -   meas_resample: None
+2026-06-24 05:34:38,011 -   meas_add_source_size: None
+2026-06-24 05:34:38,011 -   meas_add_detector_blur: None
+2026-06-24 05:34:38,011 -   meas_remove_neg_values: {'mode': 'clip_neg', 'value': None, 'force': False}
+2026-06-24 05:34:38,011 -   meas_normalization: {'mode': 'max_at_one', 'value': None}
+2026-06-24 05:34:38,011 -   meas_add_poisson_noise: None
+2026-06-24 05:34:38,011 -   meas_export: None
+2026-06-24 05:34:38,011 -   probe_permute: None
+2026-06-24 05:34:38,011 -   probe_z_shift: None
+2026-06-24 05:34:38,011 -   probe_normalization: {'mode': 'mean_total_ints', 'value': None}
+2026-06-24 05:34:38,011 -   pos_scan_flipT: None
+2026-06-24 05:34:38,011 -   pos_scan_affine: None
+2026-06-24 05:34:38,011 -   pos_scan_rand_std: 0.15
+2026-06-24 05:34:38,011 -   obj_z_crop: None
+2026-06-24 05:34:38,011 -   obj_z_pad: None
+2026-06-24 05:34:38,011 -   obj_z_resample: None
+2026-06-24 05:34:38,011 -   meas_source: file
+2026-06-24 05:34:38,011 -   meas_params: {'path': '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000002_measurement.h5', 'key': 'measurement', 'shape': None, 'offset': None, 'gap': None, 'selection': None, 'zarr_kwargs': None}
+2026-06-24 05:34:38,011 -   probe_source: simu
+2026-06-24 05:34:38,011 -   probe_params: None
+2026-06-24 05:34:38,011 -   pos_source: simu
+2026-06-24 05:34:38,011 -   pos_params: None
+2026-06-24 05:34:38,011 -   obj_source: simu
+2026-06-24 05:34:38,011 -   obj_params: None
+2026-06-24 05:34:38,011 -   tilt_source: simu
+2026-06-24 05:34:38,012 -   tilt_params: {'tilt_type': 'all', 'init_tilts': [[0, 0]]}
+2026-06-24 05:34:38,012 -  
+2026-06-24 05:34:38,012 - ### Initializing cache ###
+2026-06-24 05:34:38,012 - use_cached_obj   = False
+2026-06-24 05:34:38,012 - use_cached_probe = False
+2026-06-24 05:34:38,012 - use_cached_pos   = False
+2026-06-24 05:34:38,012 -  
+2026-06-24 05:34:38,012 - ### Initializing measurements ###
+2026-06-24 05:34:38,012 - Loading measurements from source = 'file'
+2026-06-24 05:34:38,012 - Detected measurement file type = '.h5'
+2026-06-24 05:34:38,089 - Original measurements dtype is float32, casting to float32 (single precision) for computational efficiency.
+2026-06-24 05:34:38,089 - Imported meausrements shape / dtype = (16, 16, 128, 128), dtype = float32
+2026-06-24 05:34:38,091 - Imported meausrements int. statistics (min, mean, max) = (0.0000, 0.0001, 0.0025)
+2026-06-24 05:34:38,091 - Reshaping measurements to shape = [256, 128, 128]
+2026-06-24 05:34:38,091 - Flipping measurements with [flipud, fliplr, transpose] = [0, 0, 0]
+2026-06-24 05:34:38,095 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:34:38,095 - Normalizing measurements with mode = 'max_at_one' and value = 'None'
+2026-06-24 05:34:38,095 - Normalizing by max of the 2D mean pattern intensity: 0.0017028856
+2026-06-24 05:34:38,096 - meausrements shape / dtype = (256, 128, 128), dtype = float32
+2026-06-24 05:34:38,097 - meausrements int. statistics (min, mean, max) = (0.0000, 0.0356, 1.4864)
+2026-06-24 05:34:38,098 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:34:38,099 - Pattern total int. statistics (min, mean, max)        = (578.5764, 582.7809, 588.2803), with min/max = 98.4%
+2026-06-24 05:34:38,100 - Global meausrements int. statistics (min, mean, max)  = (0.0000, 0.0356, 1.4864)
+2026-06-24 05:34:38,100 - measurements                              (N, Ky, Kx) = float32, (256, 128, 128)
+2026-06-24 05:34:38,100 -  
+2026-06-24 05:34:38,100 - ### Setting up calibration ###
+2026-06-24 05:34:38,100 - meas_calibration mode = 'kMax', value = 2.5
+2026-06-24 05:34:38,101 - Using loaded raw averaged measurement (before crop/pad/resample) to fit RBF as a part of the meas calibration
+2026-06-24 05:34:38,101 - Radius of fitted bright field disk (RBF) = 13.04 px with meas_Npix = 128
+2026-06-24 05:34:38,101 - Suggested probe_mask_k radius (RBF*2/Npix) > 0.2037
+2026-06-24 05:34:38,101 - Fitting raw averaged measurement with center, radius, and Gaussian blur std as a sanity check
+2026-06-24 05:34:38,101 - Note that the fitted Gaussian blur std (detector blur) would be affected by overlapping Bragg disks
+2026-06-24 05:34:38,329 - Initial guess: center=(63.61, 63.63), radius=13.04, Gaussian blur std=0.50
+2026-06-24 05:34:38,341 - Final fit: center=(63.61, 63.63), radius=13.04, Gaussian blur std=0.66
+2026-06-24 05:34:38,341 - Using init_params, the inferred RBF (conv_angle / 1e3 * Npix * dx / wavelength) = 13.06 px with Npix = 128
+2026-06-24 05:34:38,341 - dx (real space pixel size of probe and object) set to 0.2000 Ang with Npix = 128
+2026-06-24 05:34:38,341 -  
+2026-06-24 05:34:38,341 - ### Setting init_variables dict ###
+2026-06-24 05:34:38,341 - Derived values given input init_params:
+2026-06-24 05:34:38,341 -   kv          = 200.0 kV
+2026-06-24 05:34:38,341 -   wavelength  = 0.0251 Ang
+2026-06-24 05:34:38,341 -   conv_angle  = 12.791192054748535 mrad
+2026-06-24 05:34:38,341 -   Npix        = 128 px
+2026-06-24 05:34:38,341 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:34:38,341 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:34:38,341 -   da          = 0.9797 mrad
+2026-06-24 05:34:38,341 -   angleMax    = 62.6984 mrad
+2026-06-24 05:34:38,341 -   RBF         = 13.0567 px (Inferred from the given calibration, NOT necessarily from the loaded measurement data)
+2026-06-24 05:34:38,341 -   n_alpha     = 4.9017 (# conv_angle)
+2026-06-24 05:34:38,341 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:34:38,342 -   Rayleigh-limited resolution  = 1.1960 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:34:38,342 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:34:38,342 -  
+2026-06-24 05:34:38,342 - ### Initializing probe ###
+2026-06-24 05:34:38,342 - Loading probe from source = 'simu'
+2026-06-24 05:34:38,342 - Using experimental parameters specified by 'init_params' for initial probe simulation.
+2026-06-24 05:34:38,342 - Start simulating STEM probe
+2026-06-24 05:34:38,342 -   kv          = 200.0 kV
+2026-06-24 05:34:38,342 -   wavelength  = 0.0251 Ang
+2026-06-24 05:34:38,342 -   conv_angle  = 12.791192054748535 mrad
+2026-06-24 05:34:38,342 -   Npix        = 128 px
+2026-06-24 05:34:38,342 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:34:38,342 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:34:38,342 -   alpha_max   = 62.6984 mrad
+2026-06-24 05:34:38,342 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:34:38,342 -   Rayleigh-limited resolution  = 1.1960 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:34:38,342 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:34:38,342 - Krivanek Haider     Magnitude         Angle (°)  Description
+2026-06-24 05:34:38,342 - -------------------------------------------------------------------------------------
+2026-06-24 05:34:38,342 - C1,0     C1         -4.8240           -          Defocus (C10 = -df)
+2026-06-24 05:34:38,342 - C1,2     A1         0.3270            59.70      2-fold astigmatism
+2026-06-24 05:34:38,342 - C2,1     3*B2       134.8210          127.14     Axial coma
+2026-06-24 05:34:38,342 - C2,3     A2         251.0010          21.11      3-fold astigmatism
+2026-06-24 05:34:38,342 - C3,0     C3         2393991.0000      -          Spherical aberration
+2026-06-24 05:34:38,342 - C3,2     4*S3       -5988.5080        313.77     Axial star aberration
+2026-06-24 05:34:38,343 - C3,4     A3         -4476.0310        67.07      4-fold astigmatism
+2026-06-24 05:34:38,343 - C4,1     4*B4       -304813.3120      327.11     Axial coma(4th)
+2026-06-24 05:34:38,343 - C4,3     4*D4       364127.7500       77.76      3-lobe aberration
+2026-06-24 05:34:38,343 - C4,5     A4         -97685.3590       184.90     5-fold astigmatism
+2026-06-24 05:34:38,343 - C5,0     C5         -410699616.0000   -          Spherical aberration (5th)
+2026-06-24 05:34:38,343 - C5,2     6*S5       -18312922.0000    64.88      Axial star aberration(5th)
+2026-06-24 05:34:38,343 - C5,4     6*R5       -17729770.0000    177.94     4-lobe aberration
+2026-06-24 05:34:38,343 - C5,6     A5         -5185558.0000     49.28      6-fold astigmatism
+2026-06-24 05:34:38,346 - Loaded probe shape = (1, 128, 128), dtype = complex128
+2026-06-24 05:34:38,346 - pmode_now: 1 and pmode_max: 6, padding the pmode.
+2026-06-24 05:34:38,346 - Creating 5 new probe modes from the major mode
+2026-06-24 05:34:38,346 - Start making mixed-state STEM probe with 6 incoherent probe modes
+2026-06-24 05:34:38,450 - Relative power of probe modes = [0.9  0.02 0.02 0.02 0.02 0.02]
+2026-06-24 05:34:38,450 - Orthogonalizing 6 pmodes
+2026-06-24 05:34:38,452 - Sorting 6 pmodes by their intensities
+2026-06-24 05:34:38,453 - Normalizing probe intensity with mode = 'mean_total_ints' and value = 'None'
+2026-06-24 05:34:38,454 - sum(|probe_data|**2) = 582.78, while meas_total_ints (min, mean, max) = (578.5764, 582.7809, 588.2803)
+2026-06-24 05:34:38,454 - probe                         (pmode, Ny, Nx) = complex64, (6, 128, 128)
+2026-06-24 05:34:38,454 -  
+2026-06-24 05:34:38,454 - ### Initializing probe positions ###
+2026-06-24 05:34:38,454 - Loading probe positions from source = 'simu'
+2026-06-24 05:34:38,454 - Using experimental parameters specified by 'init_params' (dx, scan_step size, N_scan_slow, N_scan_fast) for initial position simulation.
+2026-06-24 05:34:38,454 - Simulating probe positions with dx = 0.2000, scan_step_size = 0.4756, N_scan_fast = 16, N_scan_slow = 16
+2026-06-24 05:34:38,454 - Applying Gaussian distributed random displacement with std = 0.15 px to scan positions
+2026-06-24 05:34:38,456 - crop_pos                                (N,2) = int16, (256, 2)
+2026-06-24 05:34:38,456 - crop_pos 1st and last px coords (y,x)         = ([16, 16], [52, 52])
+2026-06-24 05:34:38,456 - crop_pos extent (Ang)                         = [7.2 7.2]
+2026-06-24 05:34:38,456 - probe_pos_shifts                        (N,2) = float32, (256, 2)
+2026-06-24 05:34:38,457 -  
+2026-06-24 05:34:38,457 - ### Initializing object ###
+2026-06-24 05:34:38,457 - Loading object from source = 'simu'
+2026-06-24 05:34:38,457 - Using experimental parameters specified by 'init_params' for initial object simulation.
+2026-06-24 05:34:38,458 - omode_now: 1 and omode_max: 1, leaving the omode unchanged.
+2026-06-24 05:34:38,458 - object                    (omode, Nz, Ny, Nx) = complex64, (1, 1, 198, 198)
+2026-06-24 05:34:38,458 - object extent                 (Z, Y, X) (Ang) = [20.  39.6 39.6]
+2026-06-24 05:34:38,458 -  
+2026-06-24 05:34:38,458 - ### Initializing omode_occu from 'uniform' ###
+2026-06-24 05:34:38,458 - omode_occu                            (omode) = float32, (1,)
+2026-06-24 05:34:38,458 -  
+2026-06-24 05:34:38,458 - ### Initializing H (Fresnel propagator) ###
+2026-06-24 05:34:38,459 - Calculating H with probe_shape = (128, 128) px, dx = 0.2000 Ang, slice_thickness = 20.0000 Ang, lambd = 0.0251 Ang
+2026-06-24 05:34:38,459 - H                                    (Ky, Kx) = complex64, (128, 128)
+2026-06-24 05:34:38,459 -  
+2026-06-24 05:34:38,459 - ### Initializing obj tilts from = 'simu' ###
+2026-06-24 05:34:38,459 - Initialized obj_tilts with init_tilts = [[0, 0]] (theta_y, theta_x) mrad
+2026-06-24 05:34:38,459 - obj_tilts                              (N, 2) = float32, (1, 2)
+2026-06-24 05:34:38,459 -  
+2026-06-24 05:34:38,459 - ### Checking consistency between input params with the initialized variables ###
+2026-06-24 05:34:38,459 - meas_Npix, simu_Npix, DP measurements, probe, and H shapes are consistent as '128'
+2026-06-24 05:34:38,459 - N_scans, len(meas), N_scan_slow*N_scan_fast, len(crop_pos), and len(probe_pos_shifts) are consistent as '256'
+2026-06-24 05:34:38,459 - obj.shape[0] is consistent with len(omode_occu) as '1'
+2026-06-24 05:34:38,460 - obj.shape[1] is consistent with Nlayer as '1'
+2026-06-24 05:34:38,460 - crop positions (yx_min=[16 16], yx_max=[180 180]) are well contained inside object canvas (Ny,Nx) = (198, 198).
+2026-06-24 05:34:38,460 - obj_tilts is consistent with either 1 or N_scans
+2026-06-24 05:34:38,460 - Pass the consistency check of initialized variables, initialization is done!
+2026-06-24 05:34:38,460 -  
+2026-06-24 05:34:38,460 - ### Collecting reconstruction provenance ###
+2026-06-24 05:34:38,460 - Reconstruction provenance is collected and initialized.
+2026-06-24 05:34:38,460 -  
+2026-06-24 05:34:38,460 - ### Initializing loss function ###
+2026-06-24 05:34:38,460 - Active loss types:
+2026-06-24 05:34:38,460 -   loss_single : {'state': True, 'weight': 1.0, 'dp_pow': 0.5}
+2026-06-24 05:34:38,460 -  
+2026-06-24 05:34:38,460 - ### Initializing constraint function ###
+2026-06-24 05:34:38,460 - Active constraint types:
+2026-06-24 05:34:38,460 -   ortho_pmode   : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:34:38,460 -   fix_probe_int : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:34:38,461 -   obj_zblur     : {'start_iter': 1, 'step': 1, 'end_iter': None, 'obj_type': 'both', 'kernel_size': 5, 'std': 1.0}
+2026-06-24 05:34:38,461 -   obja_thresh   : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0, 'thresh': [0.96, 1.04]}
+2026-06-24 05:34:38,461 -   pos_recenter  : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0}
+2026-06-24 05:34:38,461 -  
+2026-06-24 05:34:38,461 - ### Done initializing PtyRADSolver ###
+2026-06-24 05:34:38,461 -  
+2026-06-24 05:34:38,586 - ### Starting the PtyRADSolver in reconstruct mode ###
+2026-06-24 05:34:38,586 -  
+2026-06-24 05:34:38,587 - ### Initializing PtychoModel model ###
+2026-06-24 05:34:38,659 - ### PtychoModel optimizable variables ###
+2026-06-24 05:34:38,659 - obja            : torch.Size([1, 1, 198, 198])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:34:38,659 - objp            : torch.Size([1, 1, 198, 198])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:34:38,659 - obj_tilts       : torch.Size([1, 2])              , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:34:38,659 - slice_thickness : torch.Size([])                  , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:34:38,659 - probe           : torch.Size([6, 128, 128, 2])    , torch.float32   , device:cuda:0, grad:True , lr:1e-04
+2026-06-24 05:34:38,659 - probe_pos_shifts: torch.Size([256, 2])            , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:34:38,659 -  
+2026-06-24 05:34:38,659 - ### Optimizable variables statitsics ###
+2026-06-24 05:34:38,659 - Total measurement values  : 4,194,304
+2026-06-24 05:34:38,659 - Total optimizing variables: 275,528
+2026-06-24 05:34:38,659 - Overdetermined ratio      : 15.22
+2026-06-24 05:34:38,659 -  
+2026-06-24 05:34:38,659 - ### Model behavior ###
+2026-06-24 05:34:38,659 - Tilt propagator           : False
+2026-06-24 05:34:38,660 - Change slice thickness    : False
+2026-06-24 05:34:38,660 - Detector blur             : False
+2026-06-24 05:34:38,660 - Preload data              : True
+2026-06-24 05:34:38,660 - On-the-fly meas padding   : False
+2026-06-24 05:34:38,660 - On-the-fly meas resample  : False
+2026-06-24 05:34:38,660 - On-the-fly simu match mode: None
+2026-06-24 05:34:38,660 -  
+2026-06-24 05:34:38,701 - ### Done initializing PtychoModel model ###
+2026-06-24 05:34:38,702 -  
+2026-06-24 05:34:38,702 - ### Creating PyTorch 'Adam' optimizer with configs = {} ###
+2026-06-24 05:34:38,702 -  
+2026-06-24 05:34:38,702 - ### Generating indices, batches, and output_path ###
+2026-06-24 05:34:38,703 - d90 = 45.000 px or 9.000 Ang
+2026-06-24 05:34:38,703 - Selecting indices with the 'full' mode 
+2026-06-24 05:34:39,161 - Generated 8 'random' groups of ~32 scan positions in 0.000 sec
+2026-06-24 05:34:39,236 - The effective batch size (i.e., how many probe positions are simultaneously used for 1 update of ptychographic parameters) is batch_size * grad_accumulation = 32 * 1 = 32
+2026-06-24 05:34:39,236 - Original recon_dir_affixes = ['default']
+2026-06-24 05:34:39,237 - Expanded recon_dir_affixes = ['indices', 'meas', 'batch', 'pmode', 'omode', 'nlayer', 'lr', 'model', 'constraint', 'loss', 'affine', 'tilt', 'aberrations']
+2026-06-24 05:34:39,243 - Path corrected for compatibility:
+2026-06-24 05:34:39,243 -   Original: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000002/c30_small__simulation_data1__sample_000002_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-4.82_C12_0.327_phi12_59.7_C21_135_phi21_127_C23_251_phi23_21.1_C30_2.39e+06_C32_-5.99e+03_phi32_314_C34_-4.48e+03_phi34_67.1_C41_-3.05e+05_phi41_327_C43_3.64e+05_phi43_77.8_C45_-9.77e+04_phi45_185_C50_-4.11e+08_C52_-1.83e+07_phi52_64.9_C54_-1.77e+07_phi54_178_C56_-5.19e+06_phi56_49.3
+2026-06-24 05:34:39,243 -   Corrected: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000002/c30_small__simulation_data1__sample_000002_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-4.82_C12_0.327_phi12_59.7_C21_135_phi21_127_C23_251_phi23_21.1_C30_2.39e+06_C32_-5.99e+03_phi32.3
+2026-06-24 05:34:39,244 - output_path = '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000002/c30_small__simulation_data1__sample_000002_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-4.82_C12_0.327_phi12_59.7_C21_135_phi21_127_C23_251_phi23_21.1_C30_2.39e+06_C32_-5.99e+03_phi32.3' is generated!
+2026-06-24 05:34:39,397 -  
+2026-06-24 05:34:39,398 - ### Log file is flushed (created) as /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000002/c30_small__simulation_data1__sample_000002_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-4.82_C12_0.327_phi12_59.7_C21_135_phi21_127_C23_251_phi23_21.1_C30_2.39e+06_C32_-5.99e+03_phi32.3/20260624_053439_ptyrad_log.txt ###
+2026-06-24 05:34:39,398 -  
+2026-06-24 05:34:39,400 - ### Creating ConvergenceMonitor with {'tensors': ['obja', 'objp', 'probe', 'probe_pos_shifts'], 'every_n_iters': None, 'percentile_range': [15.0, 85.0]} ### 
+2026-06-24 05:34:39,400 - ### Start the PtyRAD iterative ptycho reconstruction ###
+2026-06-24 05:34:39,400 - Setting up PyTorch compiler with {'fullgraph': False, 'dynamic': None, 'backend': 'inductor', 'mode': 'default', 'options': None, 'disable': True}
+2026-06-24 05:34:40,573 - Iter: 1, Total Loss: 0.8783, loss_single: 0.8783, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.433 sec
+2026-06-24 05:34:40,611 - Iter: 2, Total Loss: 0.8536, loss_single: 0.8536, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:34:40,647 - Iter: 3, Total Loss: 0.8352, loss_single: 0.8352, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:40,700 - Iter: 4, Total Loss: 0.8159, loss_single: 0.8159, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:40,735 - Iter: 5, Total Loss: 0.7939, loss_single: 0.7939, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:40,770 - Iter: 6, Total Loss: 0.7720, loss_single: 0.7720, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:40,806 - Iter: 7, Total Loss: 0.7499, loss_single: 0.7499, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:40,842 - Iter: 8, Total Loss: 0.7290, loss_single: 0.7290, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:40,877 - Iter: 9, Total Loss: 0.7123, loss_single: 0.7123, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:40,912 - Iter: 10, Total Loss: 0.6963, loss_single: 0.6963, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:40,947 - Iter: 11, Total Loss: 0.6788, loss_single: 0.6788, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:40,983 - Iter: 12, Total Loss: 0.6609, loss_single: 0.6609, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,018 - Iter: 13, Total Loss: 0.6436, loss_single: 0.6436, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,053 - Iter: 14, Total Loss: 0.6272, loss_single: 0.6272, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,088 - Iter: 15, Total Loss: 0.6106, loss_single: 0.6106, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,124 - Iter: 16, Total Loss: 0.5934, loss_single: 0.5934, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,159 - Iter: 17, Total Loss: 0.5780, loss_single: 0.5780, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,194 - Iter: 18, Total Loss: 0.5628, loss_single: 0.5628, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,229 - Iter: 19, Total Loss: 0.5466, loss_single: 0.5466, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,265 - Iter: 20, Total Loss: 0.5320, loss_single: 0.5320, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,300 - Iter: 21, Total Loss: 0.5164, loss_single: 0.5164, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,335 - Iter: 22, Total Loss: 0.5010, loss_single: 0.5010, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,371 - Iter: 23, Total Loss: 0.4863, loss_single: 0.4863, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,405 - Iter: 24, Total Loss: 0.4707, loss_single: 0.4707, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,440 - Iter: 25, Total Loss: 0.4543, loss_single: 0.4543, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,475 - Iter: 26, Total Loss: 0.4394, loss_single: 0.4394, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,511 - Iter: 27, Total Loss: 0.4282, loss_single: 0.4282, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,546 - Iter: 28, Total Loss: 0.4177, loss_single: 0.4177, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,581 - Iter: 29, Total Loss: 0.4071, loss_single: 0.4071, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,617 - Iter: 30, Total Loss: 0.3976, loss_single: 0.3976, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,652 - Iter: 31, Total Loss: 0.3865, loss_single: 0.3865, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,687 - Iter: 32, Total Loss: 0.3758, loss_single: 0.3758, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,722 - Iter: 33, Total Loss: 0.3677, loss_single: 0.3677, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,757 - Iter: 34, Total Loss: 0.3609, loss_single: 0.3609, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,793 - Iter: 35, Total Loss: 0.3544, loss_single: 0.3544, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,828 - Iter: 36, Total Loss: 0.3485, loss_single: 0.3485, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,863 - Iter: 37, Total Loss: 0.3427, loss_single: 0.3427, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,899 - Iter: 38, Total Loss: 0.3365, loss_single: 0.3365, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,934 - Iter: 39, Total Loss: 0.3316, loss_single: 0.3316, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:41,970 - Iter: 40, Total Loss: 0.3263, loss_single: 0.3263, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,008 - Iter: 41, Total Loss: 0.3197, loss_single: 0.3197, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,043 - Iter: 42, Total Loss: 0.3145, loss_single: 0.3145, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,079 - Iter: 43, Total Loss: 0.3112, loss_single: 0.3112, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,114 - Iter: 44, Total Loss: 0.3084, loss_single: 0.3084, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,149 - Iter: 45, Total Loss: 0.3050, loss_single: 0.3050, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,185 - Iter: 46, Total Loss: 0.3018, loss_single: 0.3018, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,220 - Iter: 47, Total Loss: 0.2993, loss_single: 0.2993, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,256 - Iter: 48, Total Loss: 0.2964, loss_single: 0.2964, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,291 - Iter: 49, Total Loss: 0.2937, loss_single: 0.2937, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,326 - Iter: 50, Total Loss: 0.2912, loss_single: 0.2912, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,362 - Iter: 51, Total Loss: 0.2889, loss_single: 0.2889, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,397 - Iter: 52, Total Loss: 0.2868, loss_single: 0.2868, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,433 - Iter: 53, Total Loss: 0.2846, loss_single: 0.2846, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,469 - Iter: 54, Total Loss: 0.2828, loss_single: 0.2828, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,504 - Iter: 55, Total Loss: 0.2814, loss_single: 0.2814, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,540 - Iter: 56, Total Loss: 0.2801, loss_single: 0.2801, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,575 - Iter: 57, Total Loss: 0.2787, loss_single: 0.2787, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,610 - Iter: 58, Total Loss: 0.2776, loss_single: 0.2776, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,646 - Iter: 59, Total Loss: 0.2766, loss_single: 0.2766, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,681 - Iter: 60, Total Loss: 0.2757, loss_single: 0.2757, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,717 - Iter: 61, Total Loss: 0.2750, loss_single: 0.2750, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,753 - Iter: 62, Total Loss: 0.2742, loss_single: 0.2742, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,789 - Iter: 63, Total Loss: 0.2734, loss_single: 0.2734, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:34:42,824 - Iter: 64, Total Loss: 0.2726, loss_single: 0.2726, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,859 - Iter: 65, Total Loss: 0.2718, loss_single: 0.2718, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,895 - Iter: 66, Total Loss: 0.2711, loss_single: 0.2711, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,930 - Iter: 67, Total Loss: 0.2704, loss_single: 0.2704, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:42,966 - Iter: 68, Total Loss: 0.2700, loss_single: 0.2700, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,002 - Iter: 69, Total Loss: 0.2696, loss_single: 0.2696, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:34:43,037 - Iter: 70, Total Loss: 0.2691, loss_single: 0.2691, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,073 - Iter: 71, Total Loss: 0.2688, loss_single: 0.2688, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:34:43,108 - Iter: 72, Total Loss: 0.2684, loss_single: 0.2684, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,143 - Iter: 73, Total Loss: 0.2681, loss_single: 0.2681, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,178 - Iter: 74, Total Loss: 0.2677, loss_single: 0.2677, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,214 - Iter: 75, Total Loss: 0.2674, loss_single: 0.2674, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:34:43,250 - Iter: 76, Total Loss: 0.2672, loss_single: 0.2672, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,285 - Iter: 77, Total Loss: 0.2669, loss_single: 0.2669, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,320 - Iter: 78, Total Loss: 0.2666, loss_single: 0.2666, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,356 - Iter: 79, Total Loss: 0.2664, loss_single: 0.2664, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,391 - Iter: 80, Total Loss: 0.2661, loss_single: 0.2661, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,427 - Iter: 81, Total Loss: 0.2659, loss_single: 0.2659, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,462 - Iter: 82, Total Loss: 0.2657, loss_single: 0.2657, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,497 - Iter: 83, Total Loss: 0.2655, loss_single: 0.2655, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,532 - Iter: 84, Total Loss: 0.2653, loss_single: 0.2653, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,568 - Iter: 85, Total Loss: 0.2652, loss_single: 0.2652, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,603 - Iter: 86, Total Loss: 0.2650, loss_single: 0.2650, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,638 - Iter: 87, Total Loss: 0.2648, loss_single: 0.2648, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,674 - Iter: 88, Total Loss: 0.2647, loss_single: 0.2647, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,709 - Iter: 89, Total Loss: 0.2646, loss_single: 0.2646, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,745 - Iter: 90, Total Loss: 0.2644, loss_single: 0.2644, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,780 - Iter: 91, Total Loss: 0.2643, loss_single: 0.2643, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,816 - Iter: 92, Total Loss: 0.2642, loss_single: 0.2642, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,851 - Iter: 93, Total Loss: 0.2641, loss_single: 0.2641, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,887 - Iter: 94, Total Loss: 0.2640, loss_single: 0.2640, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,923 - Iter: 95, Total Loss: 0.2639, loss_single: 0.2639, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,958 - Iter: 96, Total Loss: 0.2637, loss_single: 0.2637, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:43,994 - Iter: 97, Total Loss: 0.2636, loss_single: 0.2636, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:44,030 - Iter: 98, Total Loss: 0.2635, loss_single: 0.2635, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:44,065 - Iter: 99, Total Loss: 0.2634, loss_single: 0.2634, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:44,101 - Iter: 100, Total Loss: 0.2634, loss_single: 0.2634, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:34:44,211 - Saving summary figures for iter 100
+2026-06-24 05:34:47,856 - ### Finished 100 iterations, averaged iter_t = 0.038983 with std = 0.040 ###
+2026-06-24 05:34:47,856 -  
+2026-06-24 05:34:47,856 - ### The PtyRADSolver is finished in 9.270 sec ###
+2026-06-24 05:34:47,856 -  

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000002/c30_small__simulation_data1__sample_000002_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-4.82_C12_0.327_phi12_59.7_C21_135_phi21_127_C23_251_phi23_21.1_C30_2.39e+06_C32_-5.99e+03_phi32.3/model_iter0100.hdf5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce4124c9e1e9838a85b78cc4666f852952fc2444cef05839ce8b5acd3921d895
+size 1631656

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000002/c30_small__simulation_data1__sample_000002_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-4.82_C12_0.327_phi12_59.7_C21_135_phi21_127_C23_251_phi23_21.1_C30_2.39e+06_C32_-5.99e+03_phi32.3/simulation_data1__sample_000002.yaml

@@ -0,0 +1,140 @@
+init_params:
+  probe_kv: 200.0
+  probe_conv_angle: 12.791192054748535
+  probe_aberrations:
+    C10: -4.823992729187012
+    C12: 0.3268119692802429
+    C21: 134.8209686279297
+    C23: 251.00131225585938
+    C30: 2393991.0
+    C32: -5988.50830078125
+    C34: -4476.03076171875
+    C41: -304813.3125
+    C43: 364127.75
+    C45: -97685.359375
+    C50: -410699616.0
+    C52: -18312922.0
+    C54: -17729770.0
+    C56: -5185558.0
+    phi12: 59.69768383660901
+    phi21: 127.13888973904487
+    phi23: 21.108124186486833
+    phi32: 313.7700112244648
+    phi34: 67.07319530995558
+    phi41: 327.1123486359778
+    phi43: 77.75825445506432
+    phi45: 184.9026601536781
+    phi52: 64.88032892597714
+    phi54: 177.94459618202438
+    phi56: 49.277547238837556
+  meas_Npix: 128
+  pos_N_scan_slow: 16
+  pos_N_scan_fast: 16
+  pos_scan_step_size: 0.4755619168281555
+  probe_pmode_max: 6
+  obj_Nlayer: 1
+  obj_slice_thickness: 20.0
+  meas_permute: null
+  meas_reshape:
+  - 256
+  - 128
+  - 128
+  meas_flipT:
+  - 0
+  - 0
+  - 0
+  meas_crop: null
+  meas_pad: null
+  meas_resample: null
+  pos_scan_affine: null
+  meas_params:
+    path: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000002_measurement.h5
+    key: measurement
+  random_seed: 20261846
+  obj_omode_max: 1
+  meas_calibration:
+    mode: kMax
+    value: 2.5
+model_params:
+  detector_blur_std: null
+  update_params:
+    obja:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    objp:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    probe:
+      start_iter: 1
+      lr: 0.0001
+      end_iter: null
+    probe_pos_shifts:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+loss_params:
+  loss_single:
+    state: true
+    weight: 1.0
+    dp_pow: 0.5
+  loss_poissn:
+    state: false
+    weight: 1.0
+    dp_pow: 1.0
+    eps: 1.0e-06
+  loss_sparse:
+    state: false
+    weight: 0.1
+    ln_order: 1
+constraint_params:
+  ortho_pmode:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  fix_probe_int:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  obj_rblur:
+    start_iter: null
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 0.4
+  obj_zblur:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 1
+  obja_thresh:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    relax: 0
+    thresh:
+    - 0.96
+    - 1.04
+  objp_postiv:
+    start_iter: null
+    step: 1
+    end_iter: null
+    relax: 0
+recon_params:
+  NITER: 100
+  BATCH_SIZE:
+    size: 32
+    grad_accumulation: 1
+  SAVE_ITERS: 100
+  output_dir: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000002
+  recon_dir_affixes:
+  - default
+  prefix: c30_small__simulation_data1__sample_000002
+  postfix: ''
+  compiler_configs:
+    enable: false
+  prefix_time: false

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000004/c30_small__simulation_data1__sample_000004_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-309_C12_0.25_phi12_218_C21_883_phi21_66.5_C23_293_phi23_121_C30_3.42e+06_C32_775_phi32_235_C34_.6/20260624_053508_ptyrad_log.txt

@@ -0,0 +1,417 @@
+2026-06-24 05:35:07,173 - ### PtyRAD LoggingManager configuration ###
+2026-06-24 05:35:07,173 - log_file       = 'ptyrad_log.txt'. If log_file = None, no log file will be created.
+2026-06-24 05:35:07,173 - log_dir        = 'auto'. If log_dir = 'auto', then log will be saved to `output_path` or 'logs/'.
+2026-06-24 05:35:07,173 - flush_file     = True. Automatically set to True if `log_file is not None`
+2026-06-24 05:35:07,173 - prefix_time    = datetime. If true, preset strings ('date', 'time', 'datetime'), or a string of time format, a datetime str is prefixed to the `log_file`.
+2026-06-24 05:35:07,173 - prefix_jobid   = '0'. If not 0, it'll be prefixed to the log file. This is used for hypertune mode with multiple GPUs.
+2026-06-24 05:35:07,173 - append_to_file = True. If true, logs will be appended to the existing file. If false, the log file will be overwritten.
+2026-06-24 05:35:07,173 - show_timestamp = True. If true, the printed information will contain a timestamp.
+2026-06-24 05:35:07,173 -  
+2026-06-24 05:35:07,283 - ### Initializing HuggingFace accelerator ###
+2026-06-24 05:35:07,283 - Accelerator.distributed_type = DistributedType.NO
+2026-06-24 05:35:07,283 - Accelerator.num_process      = 1
+2026-06-24 05:35:07,283 - Accelerator.mixed_precision  = no
+2026-06-24 05:35:07,307 - 'accelerate' is available but NOT using distributed mode or mixed precision
+2026-06-24 05:35:07,307 - If you want to utilize 'accelerate' for multiGPU or mixed precision, 
+2026-06-24 05:35:07,307 - Run `accelerate launch --multi_gpu --num_processes=2 --mixed_precision='no' -m ptyrad run <PTYRAD_ARGUMENTS> --gpuid 'acc'` in your terminal
+2026-06-24 05:35:07,307 -  
+2026-06-24 05:35:07,307 - ### System information ###
+2026-06-24 05:35:07,307 - Platform: Linux-4.18.0-553.69.1.el8_10.x86_64-x86_64-with-glibc2.28
+2026-06-24 05:35:07,307 - Operating System: Linux 4.18.0-553.69.1.el8_10.x86_64
+2026-06-24 05:35:07,307 - OS Version: #1 SMP Thu Aug 7 18:10:00 EDT 2025
+2026-06-24 05:35:07,307 - Machine: x86_64
+2026-06-24 05:35:07,307 - Processor: x86_64
+2026-06-24 05:35:07,307 - Available CPU cores: 8
+2026-06-24 05:35:07,307 - SLURM-Allocated Total Memory: 78.12 GB
+2026-06-24 05:35:07,307 -  
+2026-06-24 05:35:07,307 - ### GPU information ###
+2026-06-24 05:35:07,307 - CUDA Available: True
+2026-06-24 05:35:07,307 - CUDA Version: 13.0
+2026-06-24 05:35:07,308 - Available CUDA GPUs: ['NVIDIA A100 80GB PCIe']
+2026-06-24 05:35:07,308 - CUDA Compute Capability: ['8.0']
+2026-06-24 05:35:07,308 -   INFO: For torch.compile with Triton, you'll need CUDA GPU with Compute Capability >= 7.0.
+2026-06-24 05:35:07,308 -         In addition, Triton does not directly support Windows.
+2026-06-24 05:35:07,308 -         For Windows users, please follow the instruction and download `triton-windows` from https://github.com/woct0rdho/triton-windows.
+2026-06-24 05:35:07,337 - MIG (Multi-Instance GPU) mode = False
+2026-06-24 05:35:07,337 -   INFO: MIG splits a physical GPU into multiple GPU slices, but multiGPU does not support these MIG slices.
+2026-06-24 05:35:07,337 -         In addition, multiGPU is currently only available on Linux due to the limited NCCL support.
+2026-06-24 05:35:07,337 -       -> If you're doing normal reconstruction/hypertune, you can safely ignore this.
+2026-06-24 05:35:07,337 -       -> If you want to do multiGPU, you must provide multiple 'full' GPUs that are not in MIG mode.
+2026-06-24 05:35:07,337 -  
+2026-06-24 05:35:07,337 - ### Python information ###
+2026-06-24 05:35:07,338 - Python Executable: /home/tnguye11/anaconda3/envs/ptyrad/bin/python3.12
+2026-06-24 05:35:07,338 - Python Version: 3.12.13 | packaged by Anaconda, Inc. | (main, Mar 19 2026, 20:20:58) [GCC 14.3.0]
+2026-06-24 05:35:07,338 -  
+2026-06-24 05:35:07,338 - ### Packages information ###
+2026-06-24 05:35:07,338 - Numpy Version (metadata): 2.4.6
+2026-06-24 05:35:07,339 - PyTorch Version (metadata): 2.12.0
+2026-06-24 05:35:07,340 - Optuna Version (metadata): 4.9.0
+2026-06-24 05:35:07,340 - Accelerate Version (metadata): 1.13.0
+2026-06-24 05:35:07,341 - PtyRAD Version (ptyrad/__init__.py): 1.0.0
+2026-06-24 05:35:07,341 - PtyRAD is located at: /home/tnguye11/anaconda3/envs/ptyrad/lib/python3.12/site-packages/ptyrad/__init__.py
+2026-06-24 05:35:07,341 -  
+2026-06-24 05:35:07,341 - ### Loading params file ###
+2026-06-24 05:35:07,341 - params_path = /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/params/c30_small/simulation_data1__sample_000004.yaml
+2026-06-24 05:35:07,348 - validate = True: Filling defaults and validating the params file...
+2026-06-24 05:35:07,349 - Success! Params file validated and defaults applied.
+2026-06-24 05:35:07,349 -  
+2026-06-24 05:35:07,349 - ### Setting GPU Device ###
+2026-06-24 05:35:07,349 - Selected GPU device: cuda:0 (NVIDIA A100 80GB PCIe)
+2026-06-24 05:35:07,349 -  
+2026-06-24 05:35:07,349 - Random seed: 20261848 provided by params file
+2026-06-24 05:35:07,349 - ### Initializing Initializer ###
+2026-06-24 05:35:07,349 - init_params are displayed below:
+2026-06-24 05:35:07,350 -   random_seed: 20261848
+2026-06-24 05:35:07,350 -   probe_illum_type: electron
+2026-06-24 05:35:07,350 -   probe_kv: 120.0
+2026-06-24 05:35:07,350 -   probe_conv_angle: 15.071081161499023
+2026-06-24 05:35:07,350 -   probe_aberrations: {'C10': -308.916, 'C12': 0.25, 'phi12': 217.782, 'C21': 883.007, 'phi21': 66.513, 'C23': 292.989, 'phi23': 120.944, 'C30': 3418372.25, 'C32': 775.45, 'phi32': 235.401, 'C34': 11457.985, 'phi34': 78.242, 'C41': 57754.867, 'phi41': 123.47, 'C43': -1213.19, 'phi43': 218.041, 'C45': -325368.656, 'phi45': 168.413, 'C50': 491943200.0, 'C52': 12746455.0, 'phi52': 352.811, 'C54': -13085720.0, 'phi54': 109.581, 'C56': 19757192.0, 'phi56': 45.596}
+2026-06-24 05:35:07,350 -   beam_kev: None
+2026-06-24 05:35:07,350 -   probe_dRn: None
+2026-06-24 05:35:07,350 -   probe_Rn: None
+2026-06-24 05:35:07,350 -   probe_D_H: None
+2026-06-24 05:35:07,350 -   probe_D_FZP: None
+2026-06-24 05:35:07,350 -   probe_Ls: None
+2026-06-24 05:35:07,350 -   meas_Npix: 128
+2026-06-24 05:35:07,350 -   pos_N_scans: 256
+2026-06-24 05:35:07,350 -   pos_N_scan_slow: 16
+2026-06-24 05:35:07,350 -   pos_N_scan_fast: 16
+2026-06-24 05:35:07,350 -   pos_scan_step_size: 0.3673764169216156
+2026-06-24 05:35:07,350 -   meas_calibration: {'mode': 'kMax', 'value': 2.5}
+2026-06-24 05:35:07,350 -   probe_pmode_max: 6
+2026-06-24 05:35:07,350 -   probe_pmode_init_pows: [0.02]
+2026-06-24 05:35:07,350 -   obj_omode_max: 1
+2026-06-24 05:35:07,350 -   obj_omode_init_occu: {'occu_type': 'uniform', 'init_occu': None}
+2026-06-24 05:35:07,350 -   obj_Nlayer: 1
+2026-06-24 05:35:07,350 -   obj_slice_thickness: 20.0
+2026-06-24 05:35:07,350 -   simu_Npix: None
+2026-06-24 05:35:07,350 -   simu_match_mode: None
+2026-06-24 05:35:07,350 -   meas_permute: None
+2026-06-24 05:35:07,350 -   meas_reshape: [256, 128, 128]
+2026-06-24 05:35:07,350 -   meas_flipT: [0, 0, 0]
+2026-06-24 05:35:07,350 -   meas_crop: None
+2026-06-24 05:35:07,350 -   meas_pad: None
+2026-06-24 05:35:07,350 -   meas_resample: None
+2026-06-24 05:35:07,350 -   meas_add_source_size: None
+2026-06-24 05:35:07,350 -   meas_add_detector_blur: None
+2026-06-24 05:35:07,350 -   meas_remove_neg_values: {'mode': 'clip_neg', 'value': None, 'force': False}
+2026-06-24 05:35:07,350 -   meas_normalization: {'mode': 'max_at_one', 'value': None}
+2026-06-24 05:35:07,350 -   meas_add_poisson_noise: None
+2026-06-24 05:35:07,350 -   meas_export: None
+2026-06-24 05:35:07,350 -   probe_permute: None
+2026-06-24 05:35:07,350 -   probe_z_shift: None
+2026-06-24 05:35:07,350 -   probe_normalization: {'mode': 'mean_total_ints', 'value': None}
+2026-06-24 05:35:07,351 -   pos_scan_flipT: None
+2026-06-24 05:35:07,351 -   pos_scan_affine: None
+2026-06-24 05:35:07,351 -   pos_scan_rand_std: 0.15
+2026-06-24 05:35:07,351 -   obj_z_crop: None
+2026-06-24 05:35:07,351 -   obj_z_pad: None
+2026-06-24 05:35:07,351 -   obj_z_resample: None
+2026-06-24 05:35:07,351 -   meas_source: file
+2026-06-24 05:35:07,351 -   meas_params: {'path': '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000004_measurement.h5', 'key': 'measurement', 'shape': None, 'offset': None, 'gap': None, 'selection': None, 'zarr_kwargs': None}
+2026-06-24 05:35:07,351 -   probe_source: simu
+2026-06-24 05:35:07,351 -   probe_params: None
+2026-06-24 05:35:07,351 -   pos_source: simu
+2026-06-24 05:35:07,351 -   pos_params: None
+2026-06-24 05:35:07,351 -   obj_source: simu
+2026-06-24 05:35:07,351 -   obj_params: None
+2026-06-24 05:35:07,351 -   tilt_source: simu
+2026-06-24 05:35:07,351 -   tilt_params: {'tilt_type': 'all', 'init_tilts': [[0, 0]]}
+2026-06-24 05:35:07,351 -  
+2026-06-24 05:35:07,351 - ### Initializing cache ###
+2026-06-24 05:35:07,351 - use_cached_obj   = False
+2026-06-24 05:35:07,351 - use_cached_probe = False
+2026-06-24 05:35:07,351 - use_cached_pos   = False
+2026-06-24 05:35:07,351 -  
+2026-06-24 05:35:07,351 - ### Initializing measurements ###
+2026-06-24 05:35:07,351 - Loading measurements from source = 'file'
+2026-06-24 05:35:07,351 - Detected measurement file type = '.h5'
+2026-06-24 05:35:07,430 - Original measurements dtype is float32, casting to float32 (single precision) for computational efficiency.
+2026-06-24 05:35:07,430 - Imported meausrements shape / dtype = (16, 16, 128, 128), dtype = float32
+2026-06-24 05:35:07,432 - Imported meausrements int. statistics (min, mean, max) = (0.0000, 0.0001, 0.0036)
+2026-06-24 05:35:07,432 - Reshaping measurements to shape = [256, 128, 128]
+2026-06-24 05:35:07,432 - Flipping measurements with [flipud, fliplr, transpose] = [0, 0, 0]
+2026-06-24 05:35:07,434 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:35:07,434 - Normalizing measurements with mode = 'max_at_one' and value = 'None'
+2026-06-24 05:35:07,435 - Normalizing by max of the 2D mean pattern intensity: 0.0021835053
+2026-06-24 05:35:07,435 - meausrements shape / dtype = (256, 128, 128), dtype = float32
+2026-06-24 05:35:07,436 - meausrements int. statistics (min, mean, max) = (0.0000, 0.0277, 1.6535)
+2026-06-24 05:35:07,437 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:35:07,438 - Pattern total int. statistics (min, mean, max)        = (447.2535, 453.0605, 457.8853), with min/max = 97.7%
+2026-06-24 05:35:07,440 - Global meausrements int. statistics (min, mean, max)  = (0.0000, 0.0277, 1.6535)
+2026-06-24 05:35:07,440 - measurements                              (N, Ky, Kx) = float32, (256, 128, 128)
+2026-06-24 05:35:07,440 -  
+2026-06-24 05:35:07,440 - ### Setting up calibration ###
+2026-06-24 05:35:07,440 - meas_calibration mode = 'kMax', value = 2.5
+2026-06-24 05:35:07,440 - Using loaded raw averaged measurement (before crop/pad/resample) to fit RBF as a part of the meas calibration
+2026-06-24 05:35:07,440 - Radius of fitted bright field disk (RBF) = 11.37 px with meas_Npix = 128
+2026-06-24 05:35:07,440 - Suggested probe_mask_k radius (RBF*2/Npix) > 0.1776
+2026-06-24 05:35:07,440 - Fitting raw averaged measurement with center, radius, and Gaussian blur std as a sanity check
+2026-06-24 05:35:07,440 - Note that the fitted Gaussian blur std (detector blur) would be affected by overlapping Bragg disks
+2026-06-24 05:35:07,663 - Initial guess: center=(63.77, 63.61), radius=11.37, Gaussian blur std=0.50
+2026-06-24 05:35:07,677 - Final fit: center=(63.77, 63.61), radius=11.37, Gaussian blur std=0.68
+2026-06-24 05:35:07,677 - Using init_params, the inferred RBF (conv_angle / 1e3 * Npix * dx / wavelength) = 11.52 px with Npix = 128
+2026-06-24 05:35:07,677 - dx (real space pixel size of probe and object) set to 0.2000 Ang with Npix = 128
+2026-06-24 05:35:07,678 -  
+2026-06-24 05:35:07,678 - ### Setting init_variables dict ###
+2026-06-24 05:35:07,678 - Derived values given input init_params:
+2026-06-24 05:35:07,678 -   kv          = 120.0 kV
+2026-06-24 05:35:07,678 -   wavelength  = 0.0335 Ang
+2026-06-24 05:35:07,678 -   conv_angle  = 15.071081161499023 mrad
+2026-06-24 05:35:07,678 -   Npix        = 128 px
+2026-06-24 05:35:07,678 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:35:07,678 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:35:07,678 -   da          = 1.3083 mrad
+2026-06-24 05:35:07,678 -   angleMax    = 83.7304 mrad
+2026-06-24 05:35:07,678 -   RBF         = 11.5197 px (Inferred from the given calibration, NOT necessarily from the loaded measurement data)
+2026-06-24 05:35:07,678 -   n_alpha     = 5.5557 (# conv_angle)
+2026-06-24 05:35:07,678 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:35:07,678 -   Rayleigh-limited resolution  = 1.3556 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:35:07,678 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:35:07,678 -  
+2026-06-24 05:35:07,678 - ### Initializing probe ###
+2026-06-24 05:35:07,678 - Loading probe from source = 'simu'
+2026-06-24 05:35:07,678 - Using experimental parameters specified by 'init_params' for initial probe simulation.
+2026-06-24 05:35:07,678 - Start simulating STEM probe
+2026-06-24 05:35:07,678 -   kv          = 120.0 kV
+2026-06-24 05:35:07,678 -   wavelength  = 0.0335 Ang
+2026-06-24 05:35:07,678 -   conv_angle  = 15.071081161499023 mrad
+2026-06-24 05:35:07,678 -   Npix        = 128 px
+2026-06-24 05:35:07,679 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:35:07,679 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:35:07,679 -   alpha_max   = 83.7304 mrad
+2026-06-24 05:35:07,679 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:35:07,679 -   Rayleigh-limited resolution  = 1.3556 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:35:07,679 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:35:07,679 - Krivanek Haider     Magnitude        Angle (°)  Description
+2026-06-24 05:35:07,679 - ------------------------------------------------------------------------------------
+2026-06-24 05:35:07,679 - C1,0     C1         -308.9160        -          Defocus (C10 = -df)
+2026-06-24 05:35:07,679 - C1,2     A1         0.2500           217.78     2-fold astigmatism
+2026-06-24 05:35:07,679 - C2,1     3*B2       883.0070         66.51      Axial coma
+2026-06-24 05:35:07,679 - C2,3     A2         292.9890         120.94     3-fold astigmatism
+2026-06-24 05:35:07,679 - C3,0     C3         3418372.2500     -          Spherical aberration
+2026-06-24 05:35:07,679 - C3,2     4*S3       775.4500         235.40     Axial star aberration
+2026-06-24 05:35:07,679 - C3,4     A3         11457.9850       78.24      4-fold astigmatism
+2026-06-24 05:35:07,679 - C4,1     4*B4       57754.8670       123.47     Axial coma(4th)
+2026-06-24 05:35:07,679 - C4,3     4*D4       -1213.1900       218.04     3-lobe aberration
+2026-06-24 05:35:07,679 - C4,5     A4         -325368.6560     168.41     5-fold astigmatism
+2026-06-24 05:35:07,679 - C5,0     C5         491943200.0000   -          Spherical aberration (5th)
+2026-06-24 05:35:07,679 - C5,2     6*S5       12746455.0000    352.81     Axial star aberration(5th)
+2026-06-24 05:35:07,679 - C5,4     6*R5       -13085720.0000   109.58     4-lobe aberration
+2026-06-24 05:35:07,679 - C5,6     A5         19757192.0000    45.60      6-fold astigmatism
+2026-06-24 05:35:07,682 - Loaded probe shape = (1, 128, 128), dtype = complex128
+2026-06-24 05:35:07,682 - pmode_now: 1 and pmode_max: 6, padding the pmode.
+2026-06-24 05:35:07,682 - Creating 5 new probe modes from the major mode
+2026-06-24 05:35:07,682 - Start making mixed-state STEM probe with 6 incoherent probe modes
+2026-06-24 05:35:07,852 - Relative power of probe modes = [0.9  0.02 0.02 0.02 0.02 0.02]
+2026-06-24 05:35:07,853 - Orthogonalizing 6 pmodes
+2026-06-24 05:35:07,855 - Sorting 6 pmodes by their intensities
+2026-06-24 05:35:07,855 - Normalizing probe intensity with mode = 'mean_total_ints' and value = 'None'
+2026-06-24 05:35:07,856 - sum(|probe_data|**2) = 453.06, while meas_total_ints (min, mean, max) = (447.2535, 453.0605, 457.8853)
+2026-06-24 05:35:07,856 - probe                         (pmode, Ny, Nx) = complex64, (6, 128, 128)
+2026-06-24 05:35:07,856 -  
+2026-06-24 05:35:07,856 - ### Initializing probe positions ###
+2026-06-24 05:35:07,856 - Loading probe positions from source = 'simu'
+2026-06-24 05:35:07,856 - Using experimental parameters specified by 'init_params' (dx, scan_step size, N_scan_slow, N_scan_fast) for initial position simulation.
+2026-06-24 05:35:07,856 - Simulating probe positions with dx = 0.2000, scan_step_size = 0.3674, N_scan_fast = 16, N_scan_slow = 16
+2026-06-24 05:35:07,856 - Applying Gaussian distributed random displacement with std = 0.15 px to scan positions
+2026-06-24 05:35:07,859 - crop_pos                                (N,2) = int16, (256, 2)
+2026-06-24 05:35:07,859 - crop_pos 1st and last px coords (y,x)         = ([15, 15], [43, 43])
+2026-06-24 05:35:07,859 - crop_pos extent (Ang)                         = [5.6 5.6]
+2026-06-24 05:35:07,859 - probe_pos_shifts                        (N,2) = float32, (256, 2)
+2026-06-24 05:35:07,859 -  
+2026-06-24 05:35:07,859 - ### Initializing object ###
+2026-06-24 05:35:07,859 - Loading object from source = 'simu'
+2026-06-24 05:35:07,859 - Using experimental parameters specified by 'init_params' for initial object simulation.
+2026-06-24 05:35:07,860 - omode_now: 1 and omode_max: 1, leaving the omode unchanged.
+2026-06-24 05:35:07,861 - object                    (omode, Nz, Ny, Nx) = complex64, (1, 1, 188, 188)
+2026-06-24 05:35:07,861 - object extent                 (Z, Y, X) (Ang) = [20.  37.6 37.6]
+2026-06-24 05:35:07,861 -  
+2026-06-24 05:35:07,861 - ### Initializing omode_occu from 'uniform' ###
+2026-06-24 05:35:07,861 - omode_occu                            (omode) = float32, (1,)
+2026-06-24 05:35:07,861 -  
+2026-06-24 05:35:07,861 - ### Initializing H (Fresnel propagator) ###
+2026-06-24 05:35:07,861 - Calculating H with probe_shape = (128, 128) px, dx = 0.2000 Ang, slice_thickness = 20.0000 Ang, lambd = 0.0335 Ang
+2026-06-24 05:35:07,862 - H                                    (Ky, Kx) = complex64, (128, 128)
+2026-06-24 05:35:07,862 -  
+2026-06-24 05:35:07,862 - ### Initializing obj tilts from = 'simu' ###
+2026-06-24 05:35:07,862 - Initialized obj_tilts with init_tilts = [[0, 0]] (theta_y, theta_x) mrad
+2026-06-24 05:35:07,862 - obj_tilts                              (N, 2) = float32, (1, 2)
+2026-06-24 05:35:07,862 -  
+2026-06-24 05:35:07,862 - ### Checking consistency between input params with the initialized variables ###
+2026-06-24 05:35:07,862 - meas_Npix, simu_Npix, DP measurements, probe, and H shapes are consistent as '128'
+2026-06-24 05:35:07,862 - N_scans, len(meas), N_scan_slow*N_scan_fast, len(crop_pos), and len(probe_pos_shifts) are consistent as '256'
+2026-06-24 05:35:07,862 - obj.shape[0] is consistent with len(omode_occu) as '1'
+2026-06-24 05:35:07,862 - obj.shape[1] is consistent with Nlayer as '1'
+2026-06-24 05:35:07,862 - crop positions (yx_min=[15 15], yx_max=[171 171]) are well contained inside object canvas (Ny,Nx) = (188, 188).
+2026-06-24 05:35:07,862 - obj_tilts is consistent with either 1 or N_scans
+2026-06-24 05:35:07,862 - Pass the consistency check of initialized variables, initialization is done!
+2026-06-24 05:35:07,862 -  
+2026-06-24 05:35:07,862 - ### Collecting reconstruction provenance ###
+2026-06-24 05:35:07,862 - Reconstruction provenance is collected and initialized.
+2026-06-24 05:35:07,862 -  
+2026-06-24 05:35:07,862 - ### Initializing loss function ###
+2026-06-24 05:35:07,862 - Active loss types:
+2026-06-24 05:35:07,863 -   loss_single : {'state': True, 'weight': 1.0, 'dp_pow': 0.5}
+2026-06-24 05:35:07,863 -  
+2026-06-24 05:35:07,863 - ### Initializing constraint function ###
+2026-06-24 05:35:07,863 - Active constraint types:
+2026-06-24 05:35:07,863 -   ortho_pmode   : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:35:07,863 -   fix_probe_int : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:35:07,863 -   obj_zblur     : {'start_iter': 1, 'step': 1, 'end_iter': None, 'obj_type': 'both', 'kernel_size': 5, 'std': 1.0}
+2026-06-24 05:35:07,863 -   obja_thresh   : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0, 'thresh': [0.96, 1.04]}
+2026-06-24 05:35:07,863 -   pos_recenter  : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0}
+2026-06-24 05:35:07,863 -  
+2026-06-24 05:35:07,863 - ### Done initializing PtyRADSolver ###
+2026-06-24 05:35:07,863 -  
+2026-06-24 05:35:07,987 - ### Starting the PtyRADSolver in reconstruct mode ###
+2026-06-24 05:35:07,987 -  
+2026-06-24 05:35:07,987 - ### Initializing PtychoModel model ###
+2026-06-24 05:35:08,059 - ### PtychoModel optimizable variables ###
+2026-06-24 05:35:08,059 - obja            : torch.Size([1, 1, 188, 188])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:35:08,059 - objp            : torch.Size([1, 1, 188, 188])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:35:08,059 - obj_tilts       : torch.Size([1, 2])              , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:35:08,059 - slice_thickness : torch.Size([])                  , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:35:08,059 - probe           : torch.Size([6, 128, 128, 2])    , torch.float32   , device:cuda:0, grad:True , lr:1e-04
+2026-06-24 05:35:08,059 - probe_pos_shifts: torch.Size([256, 2])            , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:35:08,060 -  
+2026-06-24 05:35:08,060 - ### Optimizable variables statitsics ###
+2026-06-24 05:35:08,060 - Total measurement values  : 4,194,304
+2026-06-24 05:35:08,060 - Total optimizing variables: 267,808
+2026-06-24 05:35:08,060 - Overdetermined ratio      : 15.66
+2026-06-24 05:35:08,060 -  
+2026-06-24 05:35:08,060 - ### Model behavior ###
+2026-06-24 05:35:08,060 - Tilt propagator           : False
+2026-06-24 05:35:08,060 - Change slice thickness    : False
+2026-06-24 05:35:08,060 - Detector blur             : False
+2026-06-24 05:35:08,060 - Preload data              : True
+2026-06-24 05:35:08,060 - On-the-fly meas padding   : False
+2026-06-24 05:35:08,060 - On-the-fly meas resample  : False
+2026-06-24 05:35:08,060 - On-the-fly simu match mode: None
+2026-06-24 05:35:08,060 -  
+2026-06-24 05:35:08,101 - ### Done initializing PtychoModel model ###
+2026-06-24 05:35:08,102 -  
+2026-06-24 05:35:08,102 - ### Creating PyTorch 'Adam' optimizer with configs = {} ###
+2026-06-24 05:35:08,102 -  
+2026-06-24 05:35:08,102 - ### Generating indices, batches, and output_path ###
+2026-06-24 05:35:08,103 - d90 = 57.000 px or 11.400 Ang
+2026-06-24 05:35:08,103 - Selecting indices with the 'full' mode 
+2026-06-24 05:35:08,558 - Generated 8 'random' groups of ~32 scan positions in 0.000 sec
+2026-06-24 05:35:08,630 - The effective batch size (i.e., how many probe positions are simultaneously used for 1 update of ptychographic parameters) is batch_size * grad_accumulation = 32 * 1 = 32
+2026-06-24 05:35:08,630 - Original recon_dir_affixes = ['default']
+2026-06-24 05:35:08,630 - Expanded recon_dir_affixes = ['indices', 'meas', 'batch', 'pmode', 'omode', 'nlayer', 'lr', 'model', 'constraint', 'loss', 'affine', 'tilt', 'aberrations']
+2026-06-24 05:35:08,636 - Path corrected for compatibility:
+2026-06-24 05:35:08,636 -   Original: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000004/c30_small__simulation_data1__sample_000004_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-309_C12_0.25_phi12_218_C21_883_phi21_66.5_C23_293_phi23_121_C30_3.42e+06_C32_775_phi32_235_C34_1.15e+04_phi34_78.2_C41_5.78e+04_phi41_123_C43_-1.21e+03_phi43_218_C45_-3.25e+05_phi45_168_C50_4.92e+08_C52_1.27e+07_phi52_353_C54_-1.31e+07_phi54_110_C56_1.98e+07_phi56_45.6
+2026-06-24 05:35:08,636 -   Corrected: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000004/c30_small__simulation_data1__sample_000004_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-309_C12_0.25_phi12_218_C21_883_phi21_66.5_C23_293_phi23_121_C30_3.42e+06_C32_775_phi32_235_C34_.6
+2026-06-24 05:35:08,637 - output_path = '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000004/c30_small__simulation_data1__sample_000004_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-309_C12_0.25_phi12_218_C21_883_phi21_66.5_C23_293_phi23_121_C30_3.42e+06_C32_775_phi32_235_C34_.6' is generated!
+2026-06-24 05:35:08,789 -  
+2026-06-24 05:35:08,790 - ### Log file is flushed (created) as /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000004/c30_small__simulation_data1__sample_000004_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-309_C12_0.25_phi12_218_C21_883_phi21_66.5_C23_293_phi23_121_C30_3.42e+06_C32_775_phi32_235_C34_.6/20260624_053508_ptyrad_log.txt ###
+2026-06-24 05:35:08,790 -  
+2026-06-24 05:35:08,792 - ### Creating ConvergenceMonitor with {'tensors': ['obja', 'objp', 'probe', 'probe_pos_shifts'], 'every_n_iters': None, 'percentile_range': [15.0, 85.0]} ### 
+2026-06-24 05:35:08,792 - ### Start the PtyRAD iterative ptycho reconstruction ###
+2026-06-24 05:35:08,792 - Setting up PyTorch compiler with {'fullgraph': False, 'dynamic': None, 'backend': 'inductor', 'mode': 'default', 'options': None, 'disable': True}
+2026-06-24 05:35:09,986 - Iter: 1, Total Loss: 1.0463, loss_single: 1.0463, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.441 sec
+2026-06-24 05:35:10,023 - Iter: 2, Total Loss: 1.0182, loss_single: 1.0182, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:35:10,059 - Iter: 3, Total Loss: 0.9947, loss_single: 0.9947, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,095 - Iter: 4, Total Loss: 0.9718, loss_single: 0.9718, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,130 - Iter: 5, Total Loss: 0.9497, loss_single: 0.9497, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,165 - Iter: 6, Total Loss: 0.9293, loss_single: 0.9293, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,201 - Iter: 7, Total Loss: 0.9071, loss_single: 0.9071, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:35:10,237 - Iter: 8, Total Loss: 0.8827, loss_single: 0.8827, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,272 - Iter: 9, Total Loss: 0.8579, loss_single: 0.8579, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,308 - Iter: 10, Total Loss: 0.8315, loss_single: 0.8315, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,343 - Iter: 11, Total Loss: 0.8068, loss_single: 0.8068, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,379 - Iter: 12, Total Loss: 0.7830, loss_single: 0.7830, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,414 - Iter: 13, Total Loss: 0.7590, loss_single: 0.7590, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,450 - Iter: 14, Total Loss: 0.7358, loss_single: 0.7358, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,485 - Iter: 15, Total Loss: 0.7123, loss_single: 0.7123, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,521 - Iter: 16, Total Loss: 0.6890, loss_single: 0.6890, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,556 - Iter: 17, Total Loss: 0.6661, loss_single: 0.6661, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,592 - Iter: 18, Total Loss: 0.6433, loss_single: 0.6433, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,627 - Iter: 19, Total Loss: 0.6224, loss_single: 0.6224, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,662 - Iter: 20, Total Loss: 0.6038, loss_single: 0.6038, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,715 - Iter: 21, Total Loss: 0.5867, loss_single: 0.5867, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,750 - Iter: 22, Total Loss: 0.5696, loss_single: 0.5696, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,786 - Iter: 23, Total Loss: 0.5533, loss_single: 0.5533, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,821 - Iter: 24, Total Loss: 0.5378, loss_single: 0.5378, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,857 - Iter: 25, Total Loss: 0.5234, loss_single: 0.5234, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,892 - Iter: 26, Total Loss: 0.5097, loss_single: 0.5097, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,928 - Iter: 27, Total Loss: 0.4958, loss_single: 0.4958, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,963 - Iter: 28, Total Loss: 0.4832, loss_single: 0.4832, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:10,999 - Iter: 29, Total Loss: 0.4717, loss_single: 0.4717, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,034 - Iter: 30, Total Loss: 0.4618, loss_single: 0.4618, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,069 - Iter: 31, Total Loss: 0.4532, loss_single: 0.4532, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,105 - Iter: 32, Total Loss: 0.4449, loss_single: 0.4449, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,140 - Iter: 33, Total Loss: 0.4375, loss_single: 0.4375, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,176 - Iter: 34, Total Loss: 0.4309, loss_single: 0.4309, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,212 - Iter: 35, Total Loss: 0.4248, loss_single: 0.4248, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,247 - Iter: 36, Total Loss: 0.4195, loss_single: 0.4195, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,283 - Iter: 37, Total Loss: 0.4151, loss_single: 0.4151, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,318 - Iter: 38, Total Loss: 0.4113, loss_single: 0.4113, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,353 - Iter: 39, Total Loss: 0.4080, loss_single: 0.4080, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:35:11,388 - Iter: 40, Total Loss: 0.4052, loss_single: 0.4052, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,426 - Iter: 41, Total Loss: 0.4027, loss_single: 0.4027, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:35:11,461 - Iter: 42, Total Loss: 0.4004, loss_single: 0.4004, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,497 - Iter: 43, Total Loss: 0.3984, loss_single: 0.3984, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,532 - Iter: 44, Total Loss: 0.3968, loss_single: 0.3968, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,568 - Iter: 45, Total Loss: 0.3954, loss_single: 0.3954, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,603 - Iter: 46, Total Loss: 0.3942, loss_single: 0.3942, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,638 - Iter: 47, Total Loss: 0.3932, loss_single: 0.3932, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,673 - Iter: 48, Total Loss: 0.3924, loss_single: 0.3924, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,708 - Iter: 49, Total Loss: 0.3919, loss_single: 0.3919, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,744 - Iter: 50, Total Loss: 0.3912, loss_single: 0.3912, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,779 - Iter: 51, Total Loss: 0.3906, loss_single: 0.3906, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,814 - Iter: 52, Total Loss: 0.3900, loss_single: 0.3900, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,849 - Iter: 53, Total Loss: 0.3894, loss_single: 0.3894, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,884 - Iter: 54, Total Loss: 0.3889, loss_single: 0.3889, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,920 - Iter: 55, Total Loss: 0.3884, loss_single: 0.3884, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,955 - Iter: 56, Total Loss: 0.3880, loss_single: 0.3880, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:11,990 - Iter: 57, Total Loss: 0.3875, loss_single: 0.3875, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,025 - Iter: 58, Total Loss: 0.3871, loss_single: 0.3871, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,060 - Iter: 59, Total Loss: 0.3867, loss_single: 0.3867, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,096 - Iter: 60, Total Loss: 0.3863, loss_single: 0.3863, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,132 - Iter: 61, Total Loss: 0.3859, loss_single: 0.3859, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,167 - Iter: 62, Total Loss: 0.3855, loss_single: 0.3855, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,202 - Iter: 63, Total Loss: 0.3852, loss_single: 0.3852, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,238 - Iter: 64, Total Loss: 0.3848, loss_single: 0.3848, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,273 - Iter: 65, Total Loss: 0.3845, loss_single: 0.3845, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,308 - Iter: 66, Total Loss: 0.3842, loss_single: 0.3842, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,344 - Iter: 67, Total Loss: 0.3838, loss_single: 0.3838, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,379 - Iter: 68, Total Loss: 0.3835, loss_single: 0.3835, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,415 - Iter: 69, Total Loss: 0.3832, loss_single: 0.3832, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,450 - Iter: 70, Total Loss: 0.3829, loss_single: 0.3829, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,485 - Iter: 71, Total Loss: 0.3825, loss_single: 0.3825, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,521 - Iter: 72, Total Loss: 0.3822, loss_single: 0.3822, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,556 - Iter: 73, Total Loss: 0.3819, loss_single: 0.3819, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,591 - Iter: 74, Total Loss: 0.3816, loss_single: 0.3816, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,627 - Iter: 75, Total Loss: 0.3813, loss_single: 0.3813, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,662 - Iter: 76, Total Loss: 0.3810, loss_single: 0.3810, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,697 - Iter: 77, Total Loss: 0.3807, loss_single: 0.3807, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,732 - Iter: 78, Total Loss: 0.3804, loss_single: 0.3804, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,767 - Iter: 79, Total Loss: 0.3801, loss_single: 0.3801, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,802 - Iter: 80, Total Loss: 0.3797, loss_single: 0.3797, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,838 - Iter: 81, Total Loss: 0.3794, loss_single: 0.3794, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,873 - Iter: 82, Total Loss: 0.3791, loss_single: 0.3791, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,908 - Iter: 83, Total Loss: 0.3788, loss_single: 0.3788, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,943 - Iter: 84, Total Loss: 0.3785, loss_single: 0.3785, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:12,979 - Iter: 85, Total Loss: 0.3782, loss_single: 0.3782, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,014 - Iter: 86, Total Loss: 0.3779, loss_single: 0.3779, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,049 - Iter: 87, Total Loss: 0.3776, loss_single: 0.3776, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,084 - Iter: 88, Total Loss: 0.3773, loss_single: 0.3773, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,120 - Iter: 89, Total Loss: 0.3770, loss_single: 0.3770, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,155 - Iter: 90, Total Loss: 0.3767, loss_single: 0.3767, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,193 - Iter: 91, Total Loss: 0.3765, loss_single: 0.3765, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:35:13,228 - Iter: 92, Total Loss: 0.3762, loss_single: 0.3762, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,264 - Iter: 93, Total Loss: 0.3759, loss_single: 0.3759, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,299 - Iter: 94, Total Loss: 0.3756, loss_single: 0.3756, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,335 - Iter: 95, Total Loss: 0.3753, loss_single: 0.3753, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,370 - Iter: 96, Total Loss: 0.3751, loss_single: 0.3751, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,406 - Iter: 97, Total Loss: 0.3748, loss_single: 0.3748, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,441 - Iter: 98, Total Loss: 0.3745, loss_single: 0.3745, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,476 - Iter: 99, Total Loss: 0.3742, loss_single: 0.3742, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,512 - Iter: 100, Total Loss: 0.3739, loss_single: 0.3739, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:13,623 - Saving summary figures for iter 100
+2026-06-24 05:35:17,253 - ### Finished 100 iterations, averaged iter_t = 0.039048 with std = 0.040 ###
+2026-06-24 05:35:17,254 -  
+2026-06-24 05:35:17,254 - ### The PtyRADSolver is finished in 9.267 sec ###
+2026-06-24 05:35:17,254 -  

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000004/c30_small__simulation_data1__sample_000004_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-309_C12_0.25_phi12_218_C21_883_phi21_66.5_C23_293_phi23_121_C30_3.42e+06_C32_775_phi32_235_C34_.6/model_iter0100.hdf5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e34f1d684b627a9a2be2aeee45ff8eb9b2c4707bebeec9af3543baf26b55456b
+size 1600768

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000004/c30_small__simulation_data1__sample_000004_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-309_C12_0.25_phi12_218_C21_883_phi21_66.5_C23_293_phi23_121_C30_3.42e+06_C32_775_phi32_235_C34_.6/simulation_data1__sample_000004.yaml

@@ -0,0 +1,140 @@
+init_params:
+  probe_kv: 120.0
+  probe_conv_angle: 15.071081161499023
+  probe_aberrations:
+    C10: -308.9161682128906
+    C12: 0.24956321716308594
+    C21: 883.0072021484375
+    C23: 292.9891357421875
+    C30: 3418372.25
+    C32: 775.450439453125
+    C34: 11457.9853515625
+    C41: 57754.8671875
+    C43: -1213.1898193359375
+    C45: -325368.65625
+    C50: 491943200.0
+    C52: 12746455.0
+    C54: -13085720.0
+    C56: 19757192.0
+    phi12: 217.78222093125737
+    phi21: 66.5133520244463
+    phi23: 120.94399012411334
+    phi32: 235.40126763119554
+    phi34: 78.24228264078644
+    phi41: 123.46997166410247
+    phi43: 218.0407982328557
+    phi45: 168.41271202543476
+    phi52: 352.81123591647423
+    phi54: 109.58061806234147
+    phi56: 45.59583621957973
+  meas_Npix: 128
+  pos_N_scan_slow: 16
+  pos_N_scan_fast: 16
+  pos_scan_step_size: 0.3673764169216156
+  probe_pmode_max: 6
+  obj_Nlayer: 1
+  obj_slice_thickness: 20.0
+  meas_permute: null
+  meas_reshape:
+  - 256
+  - 128
+  - 128
+  meas_flipT:
+  - 0
+  - 0
+  - 0
+  meas_crop: null
+  meas_pad: null
+  meas_resample: null
+  pos_scan_affine: null
+  meas_params:
+    path: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000004_measurement.h5
+    key: measurement
+  random_seed: 20261848
+  obj_omode_max: 1
+  meas_calibration:
+    mode: kMax
+    value: 2.5
+model_params:
+  detector_blur_std: null
+  update_params:
+    obja:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    objp:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    probe:
+      start_iter: 1
+      lr: 0.0001
+      end_iter: null
+    probe_pos_shifts:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+loss_params:
+  loss_single:
+    state: true
+    weight: 1.0
+    dp_pow: 0.5
+  loss_poissn:
+    state: false
+    weight: 1.0
+    dp_pow: 1.0
+    eps: 1.0e-06
+  loss_sparse:
+    state: false
+    weight: 0.1
+    ln_order: 1
+constraint_params:
+  ortho_pmode:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  fix_probe_int:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  obj_rblur:
+    start_iter: null
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 0.4
+  obj_zblur:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 1
+  obja_thresh:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    relax: 0
+    thresh:
+    - 0.96
+    - 1.04
+  objp_postiv:
+    start_iter: null
+    step: 1
+    end_iter: null
+    relax: 0
+recon_params:
+  NITER: 100
+  BATCH_SIZE:
+    size: 32
+    grad_accumulation: 1
+  SAVE_ITERS: 100
+  output_dir: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000004
+  recon_dir_affixes:
+  - default
+  prefix: c30_small__simulation_data1__sample_000004
+  postfix: ''
+  compiler_configs:
+    enable: false
+  prefix_time: false

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000006/c30_small__simulation_data1__sample_000006_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_13.5_C12_-30.3_phi12_57.8_C21_907_phi21_174_C23_-613_phi23_113_C30_2.81e+06_C32_-2.62e+03_phi32.06/20260624_053538_ptyrad_log.txt

@@ -0,0 +1,417 @@
+2026-06-24 05:35:36,649 - ### PtyRAD LoggingManager configuration ###
+2026-06-24 05:35:36,649 - log_file       = 'ptyrad_log.txt'. If log_file = None, no log file will be created.
+2026-06-24 05:35:36,649 - log_dir        = 'auto'. If log_dir = 'auto', then log will be saved to `output_path` or 'logs/'.
+2026-06-24 05:35:36,649 - flush_file     = True. Automatically set to True if `log_file is not None`
+2026-06-24 05:35:36,649 - prefix_time    = datetime. If true, preset strings ('date', 'time', 'datetime'), or a string of time format, a datetime str is prefixed to the `log_file`.
+2026-06-24 05:35:36,649 - prefix_jobid   = '0'. If not 0, it'll be prefixed to the log file. This is used for hypertune mode with multiple GPUs.
+2026-06-24 05:35:36,649 - append_to_file = True. If true, logs will be appended to the existing file. If false, the log file will be overwritten.
+2026-06-24 05:35:36,649 - show_timestamp = True. If true, the printed information will contain a timestamp.
+2026-06-24 05:35:36,649 -  
+2026-06-24 05:35:36,754 - ### Initializing HuggingFace accelerator ###
+2026-06-24 05:35:36,754 - Accelerator.distributed_type = DistributedType.NO
+2026-06-24 05:35:36,754 - Accelerator.num_process      = 1
+2026-06-24 05:35:36,754 - Accelerator.mixed_precision  = no
+2026-06-24 05:35:36,765 - 'accelerate' is available but NOT using distributed mode or mixed precision
+2026-06-24 05:35:36,765 - If you want to utilize 'accelerate' for multiGPU or mixed precision, 
+2026-06-24 05:35:36,765 - Run `accelerate launch --multi_gpu --num_processes=2 --mixed_precision='no' -m ptyrad run <PTYRAD_ARGUMENTS> --gpuid 'acc'` in your terminal
+2026-06-24 05:35:36,765 -  
+2026-06-24 05:35:36,765 - ### System information ###
+2026-06-24 05:35:36,765 - Platform: Linux-4.18.0-553.69.1.el8_10.x86_64-x86_64-with-glibc2.28
+2026-06-24 05:35:36,765 - Operating System: Linux 4.18.0-553.69.1.el8_10.x86_64
+2026-06-24 05:35:36,765 - OS Version: #1 SMP Thu Aug 7 18:10:00 EDT 2025
+2026-06-24 05:35:36,765 - Machine: x86_64
+2026-06-24 05:35:36,766 - Processor: x86_64
+2026-06-24 05:35:36,766 - Available CPU cores: 8
+2026-06-24 05:35:36,766 - SLURM-Allocated Total Memory: 78.12 GB
+2026-06-24 05:35:36,766 -  
+2026-06-24 05:35:36,766 - ### GPU information ###
+2026-06-24 05:35:36,766 - CUDA Available: True
+2026-06-24 05:35:36,766 - CUDA Version: 13.0
+2026-06-24 05:35:36,766 - Available CUDA GPUs: ['NVIDIA A100 80GB PCIe']
+2026-06-24 05:35:36,766 - CUDA Compute Capability: ['8.0']
+2026-06-24 05:35:36,766 -   INFO: For torch.compile with Triton, you'll need CUDA GPU with Compute Capability >= 7.0.
+2026-06-24 05:35:36,766 -         In addition, Triton does not directly support Windows.
+2026-06-24 05:35:36,766 -         For Windows users, please follow the instruction and download `triton-windows` from https://github.com/woct0rdho/triton-windows.
+2026-06-24 05:35:36,794 - MIG (Multi-Instance GPU) mode = False
+2026-06-24 05:35:36,794 -   INFO: MIG splits a physical GPU into multiple GPU slices, but multiGPU does not support these MIG slices.
+2026-06-24 05:35:36,794 -         In addition, multiGPU is currently only available on Linux due to the limited NCCL support.
+2026-06-24 05:35:36,794 -       -> If you're doing normal reconstruction/hypertune, you can safely ignore this.
+2026-06-24 05:35:36,794 -       -> If you want to do multiGPU, you must provide multiple 'full' GPUs that are not in MIG mode.
+2026-06-24 05:35:36,794 -  
+2026-06-24 05:35:36,794 - ### Python information ###
+2026-06-24 05:35:36,796 - Python Executable: /home/tnguye11/anaconda3/envs/ptyrad/bin/python3.12
+2026-06-24 05:35:36,796 - Python Version: 3.12.13 | packaged by Anaconda, Inc. | (main, Mar 19 2026, 20:20:58) [GCC 14.3.0]
+2026-06-24 05:35:36,796 -  
+2026-06-24 05:35:36,796 - ### Packages information ###
+2026-06-24 05:35:36,797 - Numpy Version (metadata): 2.4.6
+2026-06-24 05:35:36,797 - PyTorch Version (metadata): 2.12.0
+2026-06-24 05:35:36,798 - Optuna Version (metadata): 4.9.0
+2026-06-24 05:35:36,799 - Accelerate Version (metadata): 1.13.0
+2026-06-24 05:35:36,799 - PtyRAD Version (ptyrad/__init__.py): 1.0.0
+2026-06-24 05:35:36,799 - PtyRAD is located at: /home/tnguye11/anaconda3/envs/ptyrad/lib/python3.12/site-packages/ptyrad/__init__.py
+2026-06-24 05:35:36,799 -  
+2026-06-24 05:35:36,799 - ### Loading params file ###
+2026-06-24 05:35:36,799 - params_path = /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/params/c30_small/simulation_data1__sample_000006.yaml
+2026-06-24 05:35:36,806 - validate = True: Filling defaults and validating the params file...
+2026-06-24 05:35:36,807 - Success! Params file validated and defaults applied.
+2026-06-24 05:35:36,807 -  
+2026-06-24 05:35:36,807 - ### Setting GPU Device ###
+2026-06-24 05:35:36,807 - Selected GPU device: cuda:0 (NVIDIA A100 80GB PCIe)
+2026-06-24 05:35:36,807 -  
+2026-06-24 05:35:36,807 - Random seed: 20261850 provided by params file
+2026-06-24 05:35:36,808 - ### Initializing Initializer ###
+2026-06-24 05:35:36,808 - init_params are displayed below:
+2026-06-24 05:35:36,808 -   random_seed: 20261850
+2026-06-24 05:35:36,808 -   probe_illum_type: electron
+2026-06-24 05:35:36,808 -   probe_kv: 100.0
+2026-06-24 05:35:36,808 -   probe_conv_angle: 12.753067016601562
+2026-06-24 05:35:36,808 -   probe_aberrations: {'C10': 13.468, 'C12': -30.276, 'phi12': 57.833, 'C21': 906.579, 'phi21': 174.005, 'C23': -612.995, 'phi23': 112.89, 'C30': 2811631.5, 'C32': -2623.853, 'phi32': 300.321, 'C34': 676.396, 'phi34': 209.604, 'C41': -310396.312, 'phi41': 107.555, 'C43': -289293.562, 'phi43': 60.982, 'C45': -353779.719, 'phi45': 42.212, 'C50': -19880760.0, 'C52': 17803310.0, 'phi52': 185.556, 'C54': 4128458.75, 'phi54': 17.824, 'C56': -9667259.0, 'phi56': 2.063}
+2026-06-24 05:35:36,808 -   beam_kev: None
+2026-06-24 05:35:36,808 -   probe_dRn: None
+2026-06-24 05:35:36,808 -   probe_Rn: None
+2026-06-24 05:35:36,808 -   probe_D_H: None
+2026-06-24 05:35:36,808 -   probe_D_FZP: None
+2026-06-24 05:35:36,808 -   probe_Ls: None
+2026-06-24 05:35:36,808 -   meas_Npix: 128
+2026-06-24 05:35:36,808 -   pos_N_scans: 256
+2026-06-24 05:35:36,808 -   pos_N_scan_slow: 16
+2026-06-24 05:35:36,808 -   pos_N_scan_fast: 16
+2026-06-24 05:35:36,808 -   pos_scan_step_size: 0.37385937571525574
+2026-06-24 05:35:36,808 -   meas_calibration: {'mode': 'kMax', 'value': 2.5}
+2026-06-24 05:35:36,808 -   probe_pmode_max: 6
+2026-06-24 05:35:36,808 -   probe_pmode_init_pows: [0.02]
+2026-06-24 05:35:36,808 -   obj_omode_max: 1
+2026-06-24 05:35:36,809 -   obj_omode_init_occu: {'occu_type': 'uniform', 'init_occu': None}
+2026-06-24 05:35:36,809 -   obj_Nlayer: 1
+2026-06-24 05:35:36,809 -   obj_slice_thickness: 20.0
+2026-06-24 05:35:36,809 -   simu_Npix: None
+2026-06-24 05:35:36,809 -   simu_match_mode: None
+2026-06-24 05:35:36,809 -   meas_permute: None
+2026-06-24 05:35:36,809 -   meas_reshape: [256, 128, 128]
+2026-06-24 05:35:36,809 -   meas_flipT: [0, 0, 0]
+2026-06-24 05:35:36,809 -   meas_crop: None
+2026-06-24 05:35:36,809 -   meas_pad: None
+2026-06-24 05:35:36,809 -   meas_resample: None
+2026-06-24 05:35:36,809 -   meas_add_source_size: None
+2026-06-24 05:35:36,809 -   meas_add_detector_blur: None
+2026-06-24 05:35:36,809 -   meas_remove_neg_values: {'mode': 'clip_neg', 'value': None, 'force': False}
+2026-06-24 05:35:36,809 -   meas_normalization: {'mode': 'max_at_one', 'value': None}
+2026-06-24 05:35:36,809 -   meas_add_poisson_noise: None
+2026-06-24 05:35:36,809 -   meas_export: None
+2026-06-24 05:35:36,809 -   probe_permute: None
+2026-06-24 05:35:36,809 -   probe_z_shift: None
+2026-06-24 05:35:36,809 -   probe_normalization: {'mode': 'mean_total_ints', 'value': None}
+2026-06-24 05:35:36,809 -   pos_scan_flipT: None
+2026-06-24 05:35:36,809 -   pos_scan_affine: None
+2026-06-24 05:35:36,809 -   pos_scan_rand_std: 0.15
+2026-06-24 05:35:36,809 -   obj_z_crop: None
+2026-06-24 05:35:36,809 -   obj_z_pad: None
+2026-06-24 05:35:36,809 -   obj_z_resample: None
+2026-06-24 05:35:36,809 -   meas_source: file
+2026-06-24 05:35:36,809 -   meas_params: {'path': '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000006_measurement.h5', 'key': 'measurement', 'shape': None, 'offset': None, 'gap': None, 'selection': None, 'zarr_kwargs': None}
+2026-06-24 05:35:36,809 -   probe_source: simu
+2026-06-24 05:35:36,809 -   probe_params: None
+2026-06-24 05:35:36,809 -   pos_source: simu
+2026-06-24 05:35:36,809 -   pos_params: None
+2026-06-24 05:35:36,809 -   obj_source: simu
+2026-06-24 05:35:36,809 -   obj_params: None
+2026-06-24 05:35:36,809 -   tilt_source: simu
+2026-06-24 05:35:36,809 -   tilt_params: {'tilt_type': 'all', 'init_tilts': [[0, 0]]}
+2026-06-24 05:35:36,809 -  
+2026-06-24 05:35:36,809 - ### Initializing cache ###
+2026-06-24 05:35:36,809 - use_cached_obj   = False
+2026-06-24 05:35:36,810 - use_cached_probe = False
+2026-06-24 05:35:36,810 - use_cached_pos   = False
+2026-06-24 05:35:36,810 -  
+2026-06-24 05:35:36,810 - ### Initializing measurements ###
+2026-06-24 05:35:36,810 - Loading measurements from source = 'file'
+2026-06-24 05:35:36,810 - Detected measurement file type = '.h5'
+2026-06-24 05:35:36,889 - Original measurements dtype is float32, casting to float32 (single precision) for computational efficiency.
+2026-06-24 05:35:36,889 - Imported meausrements shape / dtype = (16, 16, 128, 128), dtype = float32
+2026-06-24 05:35:36,891 - Imported meausrements int. statistics (min, mean, max) = (0.0000, 0.0001, 0.0055)
+2026-06-24 05:35:36,891 - Reshaping measurements to shape = [256, 128, 128]
+2026-06-24 05:35:36,891 - Flipping measurements with [flipud, fliplr, transpose] = [0, 0, 0]
+2026-06-24 05:35:36,892 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:35:36,892 - Normalizing measurements with mode = 'max_at_one' and value = 'None'
+2026-06-24 05:35:36,893 - Normalizing by max of the 2D mean pattern intensity: 0.0034246303
+2026-06-24 05:35:36,894 - meausrements shape / dtype = (256, 128, 128), dtype = float32
+2026-06-24 05:35:36,895 - meausrements int. statistics (min, mean, max) = (0.0000, 0.0176, 1.6097)
+2026-06-24 05:35:36,895 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:35:36,897 - Pattern total int. statistics (min, mean, max)        = (284.3222, 287.9081, 290.7194), with min/max = 97.8%
+2026-06-24 05:35:36,898 - Global meausrements int. statistics (min, mean, max)  = (0.0000, 0.0176, 1.6097)
+2026-06-24 05:35:36,898 - measurements                              (N, Ky, Kx) = float32, (256, 128, 128)
+2026-06-24 05:35:36,898 -  
+2026-06-24 05:35:36,898 - ### Setting up calibration ###
+2026-06-24 05:35:36,898 - meas_calibration mode = 'kMax', value = 2.5
+2026-06-24 05:35:36,898 - Using loaded raw averaged measurement (before crop/pad/resample) to fit RBF as a part of the meas calibration
+2026-06-24 05:35:36,898 - Radius of fitted bright field disk (RBF) = 8.72 px with meas_Npix = 128
+2026-06-24 05:35:36,898 - Suggested probe_mask_k radius (RBF*2/Npix) > 0.1363
+2026-06-24 05:35:36,898 - Fitting raw averaged measurement with center, radius, and Gaussian blur std as a sanity check
+2026-06-24 05:35:36,898 - Note that the fitted Gaussian blur std (detector blur) would be affected by overlapping Bragg disks
+2026-06-24 05:35:37,120 - Initial guess: center=(63.64, 63.62), radius=8.72, Gaussian blur std=0.50
+2026-06-24 05:35:37,134 - Final fit: center=(63.64, 63.62), radius=8.72, Gaussian blur std=0.78
+2026-06-24 05:35:37,134 - Using init_params, the inferred RBF (conv_angle / 1e3 * Npix * dx / wavelength) = 8.82 px with Npix = 128
+2026-06-24 05:35:37,134 - dx (real space pixel size of probe and object) set to 0.2000 Ang with Npix = 128
+2026-06-24 05:35:37,134 -  
+2026-06-24 05:35:37,134 - ### Setting init_variables dict ###
+2026-06-24 05:35:37,134 - Derived values given input init_params:
+2026-06-24 05:35:37,134 -   kv          = 100.0 kV
+2026-06-24 05:35:37,134 -   wavelength  = 0.0370 Ang
+2026-06-24 05:35:37,134 -   conv_angle  = 12.753067016601562 mrad
+2026-06-24 05:35:37,134 -   Npix        = 128 px
+2026-06-24 05:35:37,134 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:35:37,135 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:35:37,135 -   da          = 1.4459 mrad
+2026-06-24 05:35:37,135 -   angleMax    = 92.5359 mrad
+2026-06-24 05:35:37,135 -   RBF         = 8.8203 px (Inferred from the given calibration, NOT necessarily from the loaded measurement data)
+2026-06-24 05:35:37,135 -   n_alpha     = 7.2560 (# conv_angle)
+2026-06-24 05:35:37,135 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:35:37,135 -   Rayleigh-limited resolution  = 1.7705 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:35:37,135 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:35:37,135 -  
+2026-06-24 05:35:37,135 - ### Initializing probe ###
+2026-06-24 05:35:37,135 - Loading probe from source = 'simu'
+2026-06-24 05:35:37,135 - Using experimental parameters specified by 'init_params' for initial probe simulation.
+2026-06-24 05:35:37,135 - Start simulating STEM probe
+2026-06-24 05:35:37,135 -   kv          = 100.0 kV
+2026-06-24 05:35:37,135 -   wavelength  = 0.0370 Ang
+2026-06-24 05:35:37,135 -   conv_angle  = 12.753067016601562 mrad
+2026-06-24 05:35:37,135 -   Npix        = 128 px
+2026-06-24 05:35:37,135 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:35:37,135 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:35:37,135 -   alpha_max   = 92.5359 mrad
+2026-06-24 05:35:37,135 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:35:37,135 -   Rayleigh-limited resolution  = 1.7705 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:35:37,135 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:35:37,135 - Krivanek Haider     Magnitude        Angle (°)  Description
+2026-06-24 05:35:37,136 - ------------------------------------------------------------------------------------
+2026-06-24 05:35:37,136 - C1,0     C1         13.4680          -          Defocus (C10 = -df)
+2026-06-24 05:35:37,136 - C1,2     A1         -30.2760         57.83      2-fold astigmatism
+2026-06-24 05:35:37,136 - C2,1     3*B2       906.5790         174.00     Axial coma
+2026-06-24 05:35:37,136 - C2,3     A2         -612.9950        112.89     3-fold astigmatism
+2026-06-24 05:35:37,136 - C3,0     C3         2811631.5000     -          Spherical aberration
+2026-06-24 05:35:37,136 - C3,2     4*S3       -2623.8530       300.32     Axial star aberration
+2026-06-24 05:35:37,136 - C3,4     A3         676.3960         209.60     4-fold astigmatism
+2026-06-24 05:35:37,136 - C4,1     4*B4       -310396.3120     107.56     Axial coma(4th)
+2026-06-24 05:35:37,136 - C4,3     4*D4       -289293.5620     60.98      3-lobe aberration
+2026-06-24 05:35:37,136 - C4,5     A4         -353779.7190     42.21      5-fold astigmatism
+2026-06-24 05:35:37,136 - C5,0     C5         -19880760.0000   -          Spherical aberration (5th)
+2026-06-24 05:35:37,136 - C5,2     6*S5       17803310.0000    185.56     Axial star aberration(5th)
+2026-06-24 05:35:37,136 - C5,4     6*R5       4128458.7500     17.82      4-lobe aberration
+2026-06-24 05:35:37,136 - C5,6     A5         -9667259.0000    2.06       6-fold astigmatism
+2026-06-24 05:35:37,139 - Loaded probe shape = (1, 128, 128), dtype = complex128
+2026-06-24 05:35:37,139 - pmode_now: 1 and pmode_max: 6, padding the pmode.
+2026-06-24 05:35:37,139 - Creating 5 new probe modes from the major mode
+2026-06-24 05:35:37,139 - Start making mixed-state STEM probe with 6 incoherent probe modes
+2026-06-24 05:35:37,332 - Relative power of probe modes = [0.9  0.02 0.02 0.02 0.02 0.02]
+2026-06-24 05:35:37,333 - Orthogonalizing 6 pmodes
+2026-06-24 05:35:37,335 - Sorting 6 pmodes by their intensities
+2026-06-24 05:35:37,336 - Normalizing probe intensity with mode = 'mean_total_ints' and value = 'None'
+2026-06-24 05:35:37,337 - sum(|probe_data|**2) = 287.91, while meas_total_ints (min, mean, max) = (284.3222, 287.9081, 290.7194)
+2026-06-24 05:35:37,337 - probe                         (pmode, Ny, Nx) = complex64, (6, 128, 128)
+2026-06-24 05:35:37,337 -  
+2026-06-24 05:35:37,337 - ### Initializing probe positions ###
+2026-06-24 05:35:37,337 - Loading probe positions from source = 'simu'
+2026-06-24 05:35:37,337 - Using experimental parameters specified by 'init_params' (dx, scan_step size, N_scan_slow, N_scan_fast) for initial position simulation.
+2026-06-24 05:35:37,337 - Simulating probe positions with dx = 0.2000, scan_step_size = 0.3739, N_scan_fast = 16, N_scan_slow = 16
+2026-06-24 05:35:37,337 - Applying Gaussian distributed random displacement with std = 0.15 px to scan positions
+2026-06-24 05:35:37,339 - crop_pos                                (N,2) = int16, (256, 2)
+2026-06-24 05:35:37,339 - crop_pos 1st and last px coords (y,x)         = ([16, 16], [44, 44])
+2026-06-24 05:35:37,339 - crop_pos extent (Ang)                         = [5.6 5.6]
+2026-06-24 05:35:37,339 - probe_pos_shifts                        (N,2) = float32, (256, 2)
+2026-06-24 05:35:37,339 -  
+2026-06-24 05:35:37,339 - ### Initializing object ###
+2026-06-24 05:35:37,339 - Loading object from source = 'simu'
+2026-06-24 05:35:37,339 - Using experimental parameters specified by 'init_params' for initial object simulation.
+2026-06-24 05:35:37,341 - omode_now: 1 and omode_max: 1, leaving the omode unchanged.
+2026-06-24 05:35:37,341 - object                    (omode, Nz, Ny, Nx) = complex64, (1, 1, 188, 188)
+2026-06-24 05:35:37,341 - object extent                 (Z, Y, X) (Ang) = [20.  37.6 37.6]
+2026-06-24 05:35:37,341 -  
+2026-06-24 05:35:37,341 - ### Initializing omode_occu from 'uniform' ###
+2026-06-24 05:35:37,341 - omode_occu                            (omode) = float32, (1,)
+2026-06-24 05:35:37,341 -  
+2026-06-24 05:35:37,341 - ### Initializing H (Fresnel propagator) ###
+2026-06-24 05:35:37,341 - Calculating H with probe_shape = (128, 128) px, dx = 0.2000 Ang, slice_thickness = 20.0000 Ang, lambd = 0.0370 Ang
+2026-06-24 05:35:37,342 - H                                    (Ky, Kx) = complex64, (128, 128)
+2026-06-24 05:35:37,342 -  
+2026-06-24 05:35:37,342 - ### Initializing obj tilts from = 'simu' ###
+2026-06-24 05:35:37,342 - Initialized obj_tilts with init_tilts = [[0, 0]] (theta_y, theta_x) mrad
+2026-06-24 05:35:37,342 - obj_tilts                              (N, 2) = float32, (1, 2)
+2026-06-24 05:35:37,342 -  
+2026-06-24 05:35:37,342 - ### Checking consistency between input params with the initialized variables ###
+2026-06-24 05:35:37,342 - meas_Npix, simu_Npix, DP measurements, probe, and H shapes are consistent as '128'
+2026-06-24 05:35:37,342 - N_scans, len(meas), N_scan_slow*N_scan_fast, len(crop_pos), and len(probe_pos_shifts) are consistent as '256'
+2026-06-24 05:35:37,342 - obj.shape[0] is consistent with len(omode_occu) as '1'
+2026-06-24 05:35:37,342 - obj.shape[1] is consistent with Nlayer as '1'
+2026-06-24 05:35:37,342 - crop positions (yx_min=[16 16], yx_max=[172 172]) are well contained inside object canvas (Ny,Nx) = (188, 188).
+2026-06-24 05:35:37,342 - obj_tilts is consistent with either 1 or N_scans
+2026-06-24 05:35:37,343 - Pass the consistency check of initialized variables, initialization is done!
+2026-06-24 05:35:37,343 -  
+2026-06-24 05:35:37,343 - ### Collecting reconstruction provenance ###
+2026-06-24 05:35:37,343 - Reconstruction provenance is collected and initialized.
+2026-06-24 05:35:37,343 -  
+2026-06-24 05:35:37,343 - ### Initializing loss function ###
+2026-06-24 05:35:37,343 - Active loss types:
+2026-06-24 05:35:37,343 -   loss_single : {'state': True, 'weight': 1.0, 'dp_pow': 0.5}
+2026-06-24 05:35:37,343 -  
+2026-06-24 05:35:37,343 - ### Initializing constraint function ###
+2026-06-24 05:35:37,343 - Active constraint types:
+2026-06-24 05:35:37,343 -   ortho_pmode   : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:35:37,343 -   fix_probe_int : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:35:37,343 -   obj_zblur     : {'start_iter': 1, 'step': 1, 'end_iter': None, 'obj_type': 'both', 'kernel_size': 5, 'std': 1.0}
+2026-06-24 05:35:37,343 -   obja_thresh   : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0, 'thresh': [0.96, 1.04]}
+2026-06-24 05:35:37,343 -   pos_recenter  : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0}
+2026-06-24 05:35:37,343 -  
+2026-06-24 05:35:37,343 - ### Done initializing PtyRADSolver ###
+2026-06-24 05:35:37,343 -  
+2026-06-24 05:35:37,469 - ### Starting the PtyRADSolver in reconstruct mode ###
+2026-06-24 05:35:37,469 -  
+2026-06-24 05:35:37,470 - ### Initializing PtychoModel model ###
+2026-06-24 05:35:37,540 - ### PtychoModel optimizable variables ###
+2026-06-24 05:35:37,541 - obja            : torch.Size([1, 1, 188, 188])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:35:37,541 - objp            : torch.Size([1, 1, 188, 188])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:35:37,541 - obj_tilts       : torch.Size([1, 2])              , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:35:37,541 - slice_thickness : torch.Size([])                  , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:35:37,541 - probe           : torch.Size([6, 128, 128, 2])    , torch.float32   , device:cuda:0, grad:True , lr:1e-04
+2026-06-24 05:35:37,541 - probe_pos_shifts: torch.Size([256, 2])            , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:35:37,541 -  
+2026-06-24 05:35:37,541 - ### Optimizable variables statitsics ###
+2026-06-24 05:35:37,541 - Total measurement values  : 4,194,304
+2026-06-24 05:35:37,541 - Total optimizing variables: 267,808
+2026-06-24 05:35:37,541 - Overdetermined ratio      : 15.66
+2026-06-24 05:35:37,541 -  
+2026-06-24 05:35:37,541 - ### Model behavior ###
+2026-06-24 05:35:37,541 - Tilt propagator           : False
+2026-06-24 05:35:37,541 - Change slice thickness    : False
+2026-06-24 05:35:37,541 - Detector blur             : False
+2026-06-24 05:35:37,541 - Preload data              : True
+2026-06-24 05:35:37,541 - On-the-fly meas padding   : False
+2026-06-24 05:35:37,541 - On-the-fly meas resample  : False
+2026-06-24 05:35:37,541 - On-the-fly simu match mode: None
+2026-06-24 05:35:37,541 -  
+2026-06-24 05:35:37,582 - ### Done initializing PtychoModel model ###
+2026-06-24 05:35:37,582 -  
+2026-06-24 05:35:37,583 - ### Creating PyTorch 'Adam' optimizer with configs = {} ###
+2026-06-24 05:35:37,583 -  
+2026-06-24 05:35:37,583 - ### Generating indices, batches, and output_path ###
+2026-06-24 05:35:37,584 - d90 = 57.000 px or 11.400 Ang
+2026-06-24 05:35:37,584 - Selecting indices with the 'full' mode 
+2026-06-24 05:35:38,051 - Generated 8 'random' groups of ~32 scan positions in 0.000 sec
+2026-06-24 05:35:38,124 - The effective batch size (i.e., how many probe positions are simultaneously used for 1 update of ptychographic parameters) is batch_size * grad_accumulation = 32 * 1 = 32
+2026-06-24 05:35:38,124 - Original recon_dir_affixes = ['default']
+2026-06-24 05:35:38,124 - Expanded recon_dir_affixes = ['indices', 'meas', 'batch', 'pmode', 'omode', 'nlayer', 'lr', 'model', 'constraint', 'loss', 'affine', 'tilt', 'aberrations']
+2026-06-24 05:35:38,130 - Path corrected for compatibility:
+2026-06-24 05:35:38,130 -   Original: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000006/c30_small__simulation_data1__sample_000006_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_13.5_C12_-30.3_phi12_57.8_C21_907_phi21_174_C23_-613_phi23_113_C30_2.81e+06_C32_-2.62e+03_phi32_300_C34_676_phi34_210_C41_-3.1e+05_phi41_108_C43_-2.89e+05_phi43_61_C45_-3.54e+05_phi45_42.2_C50_-1.99e+07_C52_1.78e+07_phi52_186_C54_4.13e+06_phi54_17.8_C56_-9.67e+06_phi56_2.06
+2026-06-24 05:35:38,130 -   Corrected: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000006/c30_small__simulation_data1__sample_000006_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_13.5_C12_-30.3_phi12_57.8_C21_907_phi21_174_C23_-613_phi23_113_C30_2.81e+06_C32_-2.62e+03_phi32.06
+2026-06-24 05:35:38,132 - output_path = '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000006/c30_small__simulation_data1__sample_000006_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_13.5_C12_-30.3_phi12_57.8_C21_907_phi21_174_C23_-613_phi23_113_C30_2.81e+06_C32_-2.62e+03_phi32.06' is generated!
+2026-06-24 05:35:38,278 -  
+2026-06-24 05:35:38,279 - ### Log file is flushed (created) as /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000006/c30_small__simulation_data1__sample_000006_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_13.5_C12_-30.3_phi12_57.8_C21_907_phi21_174_C23_-613_phi23_113_C30_2.81e+06_C32_-2.62e+03_phi32.06/20260624_053538_ptyrad_log.txt ###
+2026-06-24 05:35:38,279 -  
+2026-06-24 05:35:38,281 - ### Creating ConvergenceMonitor with {'tensors': ['obja', 'objp', 'probe', 'probe_pos_shifts'], 'every_n_iters': None, 'percentile_range': [15.0, 85.0]} ### 
+2026-06-24 05:35:38,281 - ### Start the PtyRAD iterative ptycho reconstruction ###
+2026-06-24 05:35:38,281 - Setting up PyTorch compiler with {'fullgraph': False, 'dynamic': None, 'backend': 'inductor', 'mode': 'default', 'options': None, 'disable': True}
+2026-06-24 05:35:39,475 - Iter: 1, Total Loss: 1.1613, loss_single: 1.1613, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.436 sec
+2026-06-24 05:35:39,512 - Iter: 2, Total Loss: 1.1273, loss_single: 1.1273, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:35:39,548 - Iter: 3, Total Loss: 1.0952, loss_single: 1.0952, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,583 - Iter: 4, Total Loss: 1.0668, loss_single: 1.0668, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,618 - Iter: 5, Total Loss: 1.0430, loss_single: 1.0430, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,653 - Iter: 6, Total Loss: 1.0186, loss_single: 1.0186, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,689 - Iter: 7, Total Loss: 0.9935, loss_single: 0.9935, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,724 - Iter: 8, Total Loss: 0.9708, loss_single: 0.9708, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,759 - Iter: 9, Total Loss: 0.9497, loss_single: 0.9497, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,794 - Iter: 10, Total Loss: 0.9265, loss_single: 0.9265, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,829 - Iter: 11, Total Loss: 0.9039, loss_single: 0.9039, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,864 - Iter: 12, Total Loss: 0.8820, loss_single: 0.8820, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,900 - Iter: 13, Total Loss: 0.8592, loss_single: 0.8592, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:35:39,935 - Iter: 14, Total Loss: 0.8341, loss_single: 0.8341, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:39,971 - Iter: 15, Total Loss: 0.8085, loss_single: 0.8085, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,006 - Iter: 16, Total Loss: 0.7852, loss_single: 0.7852, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,041 - Iter: 17, Total Loss: 0.7626, loss_single: 0.7626, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,077 - Iter: 18, Total Loss: 0.7376, loss_single: 0.7376, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,112 - Iter: 19, Total Loss: 0.7110, loss_single: 0.7110, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,148 - Iter: 20, Total Loss: 0.6893, loss_single: 0.6893, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,183 - Iter: 21, Total Loss: 0.6689, loss_single: 0.6689, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,218 - Iter: 22, Total Loss: 0.6505, loss_single: 0.6505, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,254 - Iter: 23, Total Loss: 0.6355, loss_single: 0.6355, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,289 - Iter: 24, Total Loss: 0.6166, loss_single: 0.6166, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,324 - Iter: 25, Total Loss: 0.5999, loss_single: 0.5999, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,359 - Iter: 26, Total Loss: 0.5858, loss_single: 0.5858, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:35:40,397 - Iter: 27, Total Loss: 0.5696, loss_single: 0.5696, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.038 sec
+2026-06-24 05:35:40,433 - Iter: 28, Total Loss: 0.5563, loss_single: 0.5563, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:35:40,468 - Iter: 29, Total Loss: 0.5453, loss_single: 0.5453, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,503 - Iter: 30, Total Loss: 0.5340, loss_single: 0.5340, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:35:40,539 - Iter: 31, Total Loss: 0.5225, loss_single: 0.5225, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,574 - Iter: 32, Total Loss: 0.5108, loss_single: 0.5108, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,609 - Iter: 33, Total Loss: 0.4996, loss_single: 0.4996, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,644 - Iter: 34, Total Loss: 0.4878, loss_single: 0.4878, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,679 - Iter: 35, Total Loss: 0.4775, loss_single: 0.4775, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,735 - Iter: 36, Total Loss: 0.4706, loss_single: 0.4706, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,770 - Iter: 37, Total Loss: 0.4642, loss_single: 0.4642, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,806 - Iter: 38, Total Loss: 0.4580, loss_single: 0.4580, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,841 - Iter: 39, Total Loss: 0.4521, loss_single: 0.4521, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,877 - Iter: 40, Total Loss: 0.4466, loss_single: 0.4466, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,913 - Iter: 41, Total Loss: 0.4425, loss_single: 0.4425, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,948 - Iter: 42, Total Loss: 0.4394, loss_single: 0.4394, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:40,984 - Iter: 43, Total Loss: 0.4364, loss_single: 0.4364, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,020 - Iter: 44, Total Loss: 0.4336, loss_single: 0.4336, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:35:41,055 - Iter: 45, Total Loss: 0.4313, loss_single: 0.4313, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,090 - Iter: 46, Total Loss: 0.4288, loss_single: 0.4288, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,126 - Iter: 47, Total Loss: 0.4259, loss_single: 0.4259, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,161 - Iter: 48, Total Loss: 0.4230, loss_single: 0.4230, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,197 - Iter: 49, Total Loss: 0.4211, loss_single: 0.4211, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,232 - Iter: 50, Total Loss: 0.4196, loss_single: 0.4196, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,268 - Iter: 51, Total Loss: 0.4178, loss_single: 0.4178, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,303 - Iter: 52, Total Loss: 0.4160, loss_single: 0.4160, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,338 - Iter: 53, Total Loss: 0.4145, loss_single: 0.4145, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,373 - Iter: 54, Total Loss: 0.4133, loss_single: 0.4133, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,409 - Iter: 55, Total Loss: 0.4126, loss_single: 0.4126, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,444 - Iter: 56, Total Loss: 0.4119, loss_single: 0.4119, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,479 - Iter: 57, Total Loss: 0.4113, loss_single: 0.4113, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,515 - Iter: 58, Total Loss: 0.4107, loss_single: 0.4107, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,550 - Iter: 59, Total Loss: 0.4102, loss_single: 0.4102, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,586 - Iter: 60, Total Loss: 0.4096, loss_single: 0.4096, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,621 - Iter: 61, Total Loss: 0.4089, loss_single: 0.4089, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,656 - Iter: 62, Total Loss: 0.4083, loss_single: 0.4083, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,692 - Iter: 63, Total Loss: 0.4078, loss_single: 0.4078, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,728 - Iter: 64, Total Loss: 0.4073, loss_single: 0.4073, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,763 - Iter: 65, Total Loss: 0.4069, loss_single: 0.4069, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,799 - Iter: 66, Total Loss: 0.4064, loss_single: 0.4064, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,834 - Iter: 67, Total Loss: 0.4061, loss_single: 0.4061, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,869 - Iter: 68, Total Loss: 0.4057, loss_single: 0.4057, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,904 - Iter: 69, Total Loss: 0.4053, loss_single: 0.4053, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:41,940 - Iter: 70, Total Loss: 0.4050, loss_single: 0.4050, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:35:41,976 - Iter: 71, Total Loss: 0.4047, loss_single: 0.4047, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,011 - Iter: 72, Total Loss: 0.4043, loss_single: 0.4043, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,046 - Iter: 73, Total Loss: 0.4040, loss_single: 0.4040, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,081 - Iter: 74, Total Loss: 0.4037, loss_single: 0.4037, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,116 - Iter: 75, Total Loss: 0.4034, loss_single: 0.4034, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,152 - Iter: 76, Total Loss: 0.4031, loss_single: 0.4031, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,188 - Iter: 77, Total Loss: 0.4028, loss_single: 0.4028, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,223 - Iter: 78, Total Loss: 0.4025, loss_single: 0.4025, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,258 - Iter: 79, Total Loss: 0.4022, loss_single: 0.4022, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:35:42,293 - Iter: 80, Total Loss: 0.4019, loss_single: 0.4019, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,329 - Iter: 81, Total Loss: 0.4016, loss_single: 0.4016, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,364 - Iter: 82, Total Loss: 0.4013, loss_single: 0.4013, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,399 - Iter: 83, Total Loss: 0.4010, loss_single: 0.4010, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,435 - Iter: 84, Total Loss: 0.4007, loss_single: 0.4007, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,470 - Iter: 85, Total Loss: 0.4004, loss_single: 0.4004, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,505 - Iter: 86, Total Loss: 0.4001, loss_single: 0.4001, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,540 - Iter: 87, Total Loss: 0.3999, loss_single: 0.3999, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,576 - Iter: 88, Total Loss: 0.3996, loss_single: 0.3996, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,610 - Iter: 89, Total Loss: 0.3993, loss_single: 0.3993, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,645 - Iter: 90, Total Loss: 0.3990, loss_single: 0.3990, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:35:42,681 - Iter: 91, Total Loss: 0.3987, loss_single: 0.3987, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,718 - Iter: 92, Total Loss: 0.3984, loss_single: 0.3984, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:35:42,754 - Iter: 93, Total Loss: 0.3981, loss_single: 0.3981, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,789 - Iter: 94, Total Loss: 0.3978, loss_single: 0.3978, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,824 - Iter: 95, Total Loss: 0.3975, loss_single: 0.3975, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,860 - Iter: 96, Total Loss: 0.3972, loss_single: 0.3972, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,895 - Iter: 97, Total Loss: 0.3969, loss_single: 0.3969, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,931 - Iter: 98, Total Loss: 0.3966, loss_single: 0.3966, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:42,966 - Iter: 99, Total Loss: 0.3963, loss_single: 0.3963, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:43,001 - Iter: 100, Total Loss: 0.3959, loss_single: 0.3959, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:35:43,115 - Saving summary figures for iter 100
+2026-06-24 05:35:46,825 - ### Finished 100 iterations, averaged iter_t = 0.038995 with std = 0.040 ###
+2026-06-24 05:35:46,825 -  
+2026-06-24 05:35:46,825 - ### The PtyRADSolver is finished in 9.356 sec ###
+2026-06-24 05:35:46,825 -  

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000006/c30_small__simulation_data1__sample_000006_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_13.5_C12_-30.3_phi12_57.8_C21_907_phi21_174_C23_-613_phi23_113_C30_2.81e+06_C32_-2.62e+03_phi32.06/simulation_data1__sample_000006.yaml

@@ -0,0 +1,140 @@
+init_params:
+  probe_kv: 100.0
+  probe_conv_angle: 12.753067016601562
+  probe_aberrations:
+    C10: 13.468389511108398
+    C12: -30.27562141418457
+    C21: 906.5794067382812
+    C23: -612.9951171875
+    C30: 2811631.5
+    C32: -2623.8525390625
+    C34: 676.3961791992188
+    C41: -310396.3125
+    C43: -289293.5625
+    C45: -353779.71875
+    C50: -19880760.0
+    C52: 17803310.0
+    C54: 4128458.75
+    C56: -9667259.0
+    phi12: 57.8331173252764
+    phi21: 174.0047518234643
+    phi23: 112.89017256429281
+    phi32: 300.3208769750894
+    phi34: 209.6041349096776
+    phi41: 107.55490689792074
+    phi43: 60.9820121175707
+    phi45: 42.21222197645795
+    phi52: 185.55604971005988
+    phi54: 17.823547686330233
+    phi56: 2.0629896965804164
+  meas_Npix: 128
+  pos_N_scan_slow: 16
+  pos_N_scan_fast: 16
+  pos_scan_step_size: 0.37385937571525574
+  probe_pmode_max: 6
+  obj_Nlayer: 1
+  obj_slice_thickness: 20.0
+  meas_permute: null
+  meas_reshape:
+  - 256
+  - 128
+  - 128
+  meas_flipT:
+  - 0
+  - 0
+  - 0
+  meas_crop: null
+  meas_pad: null
+  meas_resample: null
+  pos_scan_affine: null
+  meas_params:
+    path: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000006_measurement.h5
+    key: measurement
+  random_seed: 20261850
+  obj_omode_max: 1
+  meas_calibration:
+    mode: kMax
+    value: 2.5
+model_params:
+  detector_blur_std: null
+  update_params:
+    obja:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    objp:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    probe:
+      start_iter: 1
+      lr: 0.0001
+      end_iter: null
+    probe_pos_shifts:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+loss_params:
+  loss_single:
+    state: true
+    weight: 1.0
+    dp_pow: 0.5
+  loss_poissn:
+    state: false
+    weight: 1.0
+    dp_pow: 1.0
+    eps: 1.0e-06
+  loss_sparse:
+    state: false
+    weight: 0.1
+    ln_order: 1
+constraint_params:
+  ortho_pmode:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  fix_probe_int:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  obj_rblur:
+    start_iter: null
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 0.4
+  obj_zblur:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 1
+  obja_thresh:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    relax: 0
+    thresh:
+    - 0.96
+    - 1.04
+  objp_postiv:
+    start_iter: null
+    step: 1
+    end_iter: null
+    relax: 0
+recon_params:
+  NITER: 100
+  BATCH_SIZE:
+    size: 32
+    grad_accumulation: 1
+  SAVE_ITERS: 100
+  output_dir: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000006
+  recon_dir_affixes:
+  - default
+  prefix: c30_small__simulation_data1__sample_000006
+  postfix: ''
+  compiler_configs:
+    enable: false
+  prefix_time: false

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000008/c30_small__simulation_data1__sample_000008_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-308_C12_-23.6_phi12_165_C21_868_phi21_90.9_C23_-872_phi23_356_C30_-8.97e+05_C32_.57e+07_phi56_344/20260624_053608_ptyrad_log.txt

@@ -0,0 +1,417 @@
+2026-06-24 05:36:06,341 - ### PtyRAD LoggingManager configuration ###
+2026-06-24 05:36:06,342 - log_file       = 'ptyrad_log.txt'. If log_file = None, no log file will be created.
+2026-06-24 05:36:06,342 - log_dir        = 'auto'. If log_dir = 'auto', then log will be saved to `output_path` or 'logs/'.
+2026-06-24 05:36:06,342 - flush_file     = True. Automatically set to True if `log_file is not None`
+2026-06-24 05:36:06,342 - prefix_time    = datetime. If true, preset strings ('date', 'time', 'datetime'), or a string of time format, a datetime str is prefixed to the `log_file`.
+2026-06-24 05:36:06,342 - prefix_jobid   = '0'. If not 0, it'll be prefixed to the log file. This is used for hypertune mode with multiple GPUs.
+2026-06-24 05:36:06,342 - append_to_file = True. If true, logs will be appended to the existing file. If false, the log file will be overwritten.
+2026-06-24 05:36:06,342 - show_timestamp = True. If true, the printed information will contain a timestamp.
+2026-06-24 05:36:06,342 -  
+2026-06-24 05:36:06,449 - ### Initializing HuggingFace accelerator ###
+2026-06-24 05:36:06,449 - Accelerator.distributed_type = DistributedType.NO
+2026-06-24 05:36:06,449 - Accelerator.num_process      = 1
+2026-06-24 05:36:06,449 - Accelerator.mixed_precision  = no
+2026-06-24 05:36:06,471 - 'accelerate' is available but NOT using distributed mode or mixed precision
+2026-06-24 05:36:06,471 - If you want to utilize 'accelerate' for multiGPU or mixed precision, 
+2026-06-24 05:36:06,471 - Run `accelerate launch --multi_gpu --num_processes=2 --mixed_precision='no' -m ptyrad run <PTYRAD_ARGUMENTS> --gpuid 'acc'` in your terminal
+2026-06-24 05:36:06,471 -  
+2026-06-24 05:36:06,471 - ### System information ###
+2026-06-24 05:36:06,471 - Platform: Linux-4.18.0-553.69.1.el8_10.x86_64-x86_64-with-glibc2.28
+2026-06-24 05:36:06,471 - Operating System: Linux 4.18.0-553.69.1.el8_10.x86_64
+2026-06-24 05:36:06,471 - OS Version: #1 SMP Thu Aug 7 18:10:00 EDT 2025
+2026-06-24 05:36:06,471 - Machine: x86_64
+2026-06-24 05:36:06,471 - Processor: x86_64
+2026-06-24 05:36:06,471 - Available CPU cores: 8
+2026-06-24 05:36:06,471 - SLURM-Allocated Total Memory: 78.12 GB
+2026-06-24 05:36:06,471 -  
+2026-06-24 05:36:06,471 - ### GPU information ###
+2026-06-24 05:36:06,472 - CUDA Available: True
+2026-06-24 05:36:06,472 - CUDA Version: 13.0
+2026-06-24 05:36:06,472 - Available CUDA GPUs: ['NVIDIA A100 80GB PCIe']
+2026-06-24 05:36:06,472 - CUDA Compute Capability: ['8.0']
+2026-06-24 05:36:06,472 -   INFO: For torch.compile with Triton, you'll need CUDA GPU with Compute Capability >= 7.0.
+2026-06-24 05:36:06,472 -         In addition, Triton does not directly support Windows.
+2026-06-24 05:36:06,472 -         For Windows users, please follow the instruction and download `triton-windows` from https://github.com/woct0rdho/triton-windows.
+2026-06-24 05:36:06,487 - MIG (Multi-Instance GPU) mode = False
+2026-06-24 05:36:06,487 -   INFO: MIG splits a physical GPU into multiple GPU slices, but multiGPU does not support these MIG slices.
+2026-06-24 05:36:06,487 -         In addition, multiGPU is currently only available on Linux due to the limited NCCL support.
+2026-06-24 05:36:06,487 -       -> If you're doing normal reconstruction/hypertune, you can safely ignore this.
+2026-06-24 05:36:06,487 -       -> If you want to do multiGPU, you must provide multiple 'full' GPUs that are not in MIG mode.
+2026-06-24 05:36:06,487 -  
+2026-06-24 05:36:06,487 - ### Python information ###
+2026-06-24 05:36:06,489 - Python Executable: /home/tnguye11/anaconda3/envs/ptyrad/bin/python3.12
+2026-06-24 05:36:06,489 - Python Version: 3.12.13 | packaged by Anaconda, Inc. | (main, Mar 19 2026, 20:20:58) [GCC 14.3.0]
+2026-06-24 05:36:06,489 -  
+2026-06-24 05:36:06,489 - ### Packages information ###
+2026-06-24 05:36:06,490 - Numpy Version (metadata): 2.4.6
+2026-06-24 05:36:06,491 - PyTorch Version (metadata): 2.12.0
+2026-06-24 05:36:06,491 - Optuna Version (metadata): 4.9.0
+2026-06-24 05:36:06,492 - Accelerate Version (metadata): 1.13.0
+2026-06-24 05:36:06,492 - PtyRAD Version (ptyrad/__init__.py): 1.0.0
+2026-06-24 05:36:06,492 - PtyRAD is located at: /home/tnguye11/anaconda3/envs/ptyrad/lib/python3.12/site-packages/ptyrad/__init__.py
+2026-06-24 05:36:06,492 -  
+2026-06-24 05:36:06,492 - ### Loading params file ###
+2026-06-24 05:36:06,492 - params_path = /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/params/c30_small/simulation_data1__sample_000008.yaml
+2026-06-24 05:36:06,499 - validate = True: Filling defaults and validating the params file...
+2026-06-24 05:36:06,500 - Success! Params file validated and defaults applied.
+2026-06-24 05:36:06,500 -  
+2026-06-24 05:36:06,500 - ### Setting GPU Device ###
+2026-06-24 05:36:06,500 - Selected GPU device: cuda:0 (NVIDIA A100 80GB PCIe)
+2026-06-24 05:36:06,500 -  
+2026-06-24 05:36:06,500 - Random seed: 20261852 provided by params file
+2026-06-24 05:36:06,500 - ### Initializing Initializer ###
+2026-06-24 05:36:06,501 - init_params are displayed below:
+2026-06-24 05:36:06,501 -   random_seed: 20261852
+2026-06-24 05:36:06,501 -   probe_illum_type: electron
+2026-06-24 05:36:06,501 -   probe_kv: 300.0
+2026-06-24 05:36:06,501 -   probe_conv_angle: 12.933050155639648
+2026-06-24 05:36:06,501 -   probe_aberrations: {'C10': -307.544, 'C12': -23.624, 'phi12': 165.476, 'C21': 867.661, 'phi21': 90.944, 'C23': -871.995, 'phi23': 355.572, 'C30': -897010.875, 'C32': 5510.824, 'phi32': 174.58, 'C34': -10106.436, 'phi34': 300.845, 'C41': 422961.812, 'phi41': 233.117, 'C43': -401591.344, 'phi43': 264.827, 'C45': -308371.031, 'phi45': 279.55, 'C50': -448757952.0, 'C52': -18148220.0, 'phi52': 207.609, 'C54': -12083272.0, 'phi54': 49.521, 'C56': 15654723.0, 'phi56': 343.861}
+2026-06-24 05:36:06,501 -   beam_kev: None
+2026-06-24 05:36:06,501 -   probe_dRn: None
+2026-06-24 05:36:06,501 -   probe_Rn: None
+2026-06-24 05:36:06,501 -   probe_D_H: None
+2026-06-24 05:36:06,501 -   probe_D_FZP: None
+2026-06-24 05:36:06,501 -   probe_Ls: None
+2026-06-24 05:36:06,501 -   meas_Npix: 128
+2026-06-24 05:36:06,501 -   pos_N_scans: 256
+2026-06-24 05:36:06,501 -   pos_N_scan_slow: 16
+2026-06-24 05:36:06,501 -   pos_N_scan_fast: 16
+2026-06-24 05:36:06,501 -   pos_scan_step_size: 0.4713726341724396
+2026-06-24 05:36:06,501 -   meas_calibration: {'mode': 'kMax', 'value': 2.5}
+2026-06-24 05:36:06,501 -   probe_pmode_max: 6
+2026-06-24 05:36:06,501 -   probe_pmode_init_pows: [0.02]
+2026-06-24 05:36:06,501 -   obj_omode_max: 1
+2026-06-24 05:36:06,501 -   obj_omode_init_occu: {'occu_type': 'uniform', 'init_occu': None}
+2026-06-24 05:36:06,501 -   obj_Nlayer: 1
+2026-06-24 05:36:06,501 -   obj_slice_thickness: 20.0
+2026-06-24 05:36:06,501 -   simu_Npix: None
+2026-06-24 05:36:06,501 -   simu_match_mode: None
+2026-06-24 05:36:06,501 -   meas_permute: None
+2026-06-24 05:36:06,501 -   meas_reshape: [256, 128, 128]
+2026-06-24 05:36:06,501 -   meas_flipT: [0, 0, 0]
+2026-06-24 05:36:06,501 -   meas_crop: None
+2026-06-24 05:36:06,501 -   meas_pad: None
+2026-06-24 05:36:06,501 -   meas_resample: None
+2026-06-24 05:36:06,501 -   meas_add_source_size: None
+2026-06-24 05:36:06,501 -   meas_add_detector_blur: None
+2026-06-24 05:36:06,501 -   meas_remove_neg_values: {'mode': 'clip_neg', 'value': None, 'force': False}
+2026-06-24 05:36:06,502 -   meas_normalization: {'mode': 'max_at_one', 'value': None}
+2026-06-24 05:36:06,502 -   meas_add_poisson_noise: None
+2026-06-24 05:36:06,502 -   meas_export: None
+2026-06-24 05:36:06,502 -   probe_permute: None
+2026-06-24 05:36:06,502 -   probe_z_shift: None
+2026-06-24 05:36:06,502 -   probe_normalization: {'mode': 'mean_total_ints', 'value': None}
+2026-06-24 05:36:06,502 -   pos_scan_flipT: None
+2026-06-24 05:36:06,502 -   pos_scan_affine: None
+2026-06-24 05:36:06,502 -   pos_scan_rand_std: 0.15
+2026-06-24 05:36:06,502 -   obj_z_crop: None
+2026-06-24 05:36:06,502 -   obj_z_pad: None
+2026-06-24 05:36:06,502 -   obj_z_resample: None
+2026-06-24 05:36:06,502 -   meas_source: file
+2026-06-24 05:36:06,502 -   meas_params: {'path': '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000008_measurement.h5', 'key': 'measurement', 'shape': None, 'offset': None, 'gap': None, 'selection': None, 'zarr_kwargs': None}
+2026-06-24 05:36:06,502 -   probe_source: simu
+2026-06-24 05:36:06,502 -   probe_params: None
+2026-06-24 05:36:06,502 -   pos_source: simu
+2026-06-24 05:36:06,502 -   pos_params: None
+2026-06-24 05:36:06,502 -   obj_source: simu
+2026-06-24 05:36:06,502 -   obj_params: None
+2026-06-24 05:36:06,502 -   tilt_source: simu
+2026-06-24 05:36:06,502 -   tilt_params: {'tilt_type': 'all', 'init_tilts': [[0, 0]]}
+2026-06-24 05:36:06,502 -  
+2026-06-24 05:36:06,502 - ### Initializing cache ###
+2026-06-24 05:36:06,502 - use_cached_obj   = False
+2026-06-24 05:36:06,502 - use_cached_probe = False
+2026-06-24 05:36:06,502 - use_cached_pos   = False
+2026-06-24 05:36:06,502 -  
+2026-06-24 05:36:06,502 - ### Initializing measurements ###
+2026-06-24 05:36:06,502 - Loading measurements from source = 'file'
+2026-06-24 05:36:06,502 - Detected measurement file type = '.h5'
+2026-06-24 05:36:06,581 - Original measurements dtype is float32, casting to float32 (single precision) for computational efficiency.
+2026-06-24 05:36:06,581 - Imported meausrements shape / dtype = (16, 16, 128, 128), dtype = float32
+2026-06-24 05:36:06,583 - Imported meausrements int. statistics (min, mean, max) = (0.0000, 0.0001, 0.0017)
+2026-06-24 05:36:06,583 - Reshaping measurements to shape = [256, 128, 128]
+2026-06-24 05:36:06,583 - Flipping measurements with [flipud, fliplr, transpose] = [0, 0, 0]
+2026-06-24 05:36:06,584 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:36:06,584 - Normalizing measurements with mode = 'max_at_one' and value = 'None'
+2026-06-24 05:36:06,585 - Normalizing by max of the 2D mean pattern intensity: 0.0010798314
+2026-06-24 05:36:06,585 - meausrements shape / dtype = (256, 128, 128), dtype = float32
+2026-06-24 05:36:06,587 - meausrements int. statistics (min, mean, max) = (0.0000, 0.0563, 1.5565)
+2026-06-24 05:36:06,587 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:36:06,589 - Pattern total int. statistics (min, mean, max)        = (912.6027, 921.7542, 928.0841), with min/max = 98.3%
+2026-06-24 05:36:06,590 - Global meausrements int. statistics (min, mean, max)  = (0.0000, 0.0563, 1.5565)
+2026-06-24 05:36:06,590 - measurements                              (N, Ky, Kx) = float32, (256, 128, 128)
+2026-06-24 05:36:06,590 -  
+2026-06-24 05:36:06,590 - ### Setting up calibration ###
+2026-06-24 05:36:06,590 - meas_calibration mode = 'kMax', value = 2.5
+2026-06-24 05:36:06,590 - Using loaded raw averaged measurement (before crop/pad/resample) to fit RBF as a part of the meas calibration
+2026-06-24 05:36:06,590 - Radius of fitted bright field disk (RBF) = 16.69 px with meas_Npix = 128
+2026-06-24 05:36:06,590 - Suggested probe_mask_k radius (RBF*2/Npix) > 0.2608
+2026-06-24 05:36:06,590 - Fitting raw averaged measurement with center, radius, and Gaussian blur std as a sanity check
+2026-06-24 05:36:06,590 - Note that the fitted Gaussian blur std (detector blur) would be affected by overlapping Bragg disks
+2026-06-24 05:36:06,811 - Initial guess: center=(63.61, 63.62), radius=16.69, Gaussian blur std=0.50
+2026-06-24 05:36:06,826 - Final fit: center=(63.61, 63.62), radius=16.69, Gaussian blur std=0.76
+2026-06-24 05:36:06,826 - Using init_params, the inferred RBF (conv_angle / 1e3 * Npix * dx / wavelength) = 16.82 px with Npix = 128
+2026-06-24 05:36:06,826 - dx (real space pixel size of probe and object) set to 0.2000 Ang with Npix = 128
+2026-06-24 05:36:06,826 -  
+2026-06-24 05:36:06,826 - ### Setting init_variables dict ###
+2026-06-24 05:36:06,826 - Derived values given input init_params:
+2026-06-24 05:36:06,826 -   kv          = 300.0 kV
+2026-06-24 05:36:06,826 -   wavelength  = 0.0197 Ang
+2026-06-24 05:36:06,826 -   conv_angle  = 12.933050155639648 mrad
+2026-06-24 05:36:06,826 -   Npix        = 128 px
+2026-06-24 05:36:06,826 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:36:06,826 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:36:06,826 -   da          = 0.7690 mrad
+2026-06-24 05:36:06,826 -   angleMax    = 49.2187 mrad
+2026-06-24 05:36:06,826 -   RBF         = 16.8171 px (Inferred from the given calibration, NOT necessarily from the loaded measurement data)
+2026-06-24 05:36:06,826 -   n_alpha     = 3.8057 (# conv_angle)
+2026-06-24 05:36:06,826 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:36:06,826 -   Rayleigh-limited resolution  = 0.9286 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:36:06,826 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:36:06,826 -  
+2026-06-24 05:36:06,826 - ### Initializing probe ###
+2026-06-24 05:36:06,827 - Loading probe from source = 'simu'
+2026-06-24 05:36:06,827 - Using experimental parameters specified by 'init_params' for initial probe simulation.
+2026-06-24 05:36:06,827 - Start simulating STEM probe
+2026-06-24 05:36:06,827 -   kv          = 300.0 kV
+2026-06-24 05:36:06,827 -   wavelength  = 0.0197 Ang
+2026-06-24 05:36:06,827 -   conv_angle  = 12.933050155639648 mrad
+2026-06-24 05:36:06,827 -   Npix        = 128 px
+2026-06-24 05:36:06,827 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:36:06,827 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:36:06,827 -   alpha_max   = 49.2187 mrad
+2026-06-24 05:36:06,827 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:36:06,827 -   Rayleigh-limited resolution  = 0.9286 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:36:06,827 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:36:06,827 - Krivanek Haider     Magnitude         Angle (°)  Description
+2026-06-24 05:36:06,827 - -------------------------------------------------------------------------------------
+2026-06-24 05:36:06,827 - C1,0     C1         -307.5440         -          Defocus (C10 = -df)
+2026-06-24 05:36:06,827 - C1,2     A1         -23.6240          165.48     2-fold astigmatism
+2026-06-24 05:36:06,827 - C2,1     3*B2       867.6610          90.94      Axial coma
+2026-06-24 05:36:06,827 - C2,3     A2         -871.9950         355.57     3-fold astigmatism
+2026-06-24 05:36:06,827 - C3,0     C3         -897010.8750      -          Spherical aberration
+2026-06-24 05:36:06,827 - C3,2     4*S3       5510.8240         174.58     Axial star aberration
+2026-06-24 05:36:06,827 - C3,4     A3         -10106.4360       300.85     4-fold astigmatism
+2026-06-24 05:36:06,827 - C4,1     4*B4       422961.8120       233.12     Axial coma(4th)
+2026-06-24 05:36:06,827 - C4,3     4*D4       -401591.3440      264.83     3-lobe aberration
+2026-06-24 05:36:06,827 - C4,5     A4         -308371.0310      279.55     5-fold astigmatism
+2026-06-24 05:36:06,828 - C5,0     C5         -448757952.0000   -          Spherical aberration (5th)
+2026-06-24 05:36:06,828 - C5,2     6*S5       -18148220.0000    207.61     Axial star aberration(5th)
+2026-06-24 05:36:06,828 - C5,4     6*R5       -12083272.0000    49.52      4-lobe aberration
+2026-06-24 05:36:06,828 - C5,6     A5         15654723.0000     343.86     6-fold astigmatism
+2026-06-24 05:36:06,830 - Loaded probe shape = (1, 128, 128), dtype = complex128
+2026-06-24 05:36:06,830 - pmode_now: 1 and pmode_max: 6, padding the pmode.
+2026-06-24 05:36:06,831 - Creating 5 new probe modes from the major mode
+2026-06-24 05:36:06,831 - Start making mixed-state STEM probe with 6 incoherent probe modes
+2026-06-24 05:36:07,066 - Relative power of probe modes = [0.9  0.02 0.02 0.02 0.02 0.02]
+2026-06-24 05:36:07,067 - Orthogonalizing 6 pmodes
+2026-06-24 05:36:07,080 - Sorting 6 pmodes by their intensities
+2026-06-24 05:36:07,080 - Normalizing probe intensity with mode = 'mean_total_ints' and value = 'None'
+2026-06-24 05:36:07,081 - sum(|probe_data|**2) = 921.75, while meas_total_ints (min, mean, max) = (912.6027, 921.7542, 928.0841)
+2026-06-24 05:36:07,081 - probe                         (pmode, Ny, Nx) = complex64, (6, 128, 128)
+2026-06-24 05:36:07,081 -  
+2026-06-24 05:36:07,081 - ### Initializing probe positions ###
+2026-06-24 05:36:07,081 - Loading probe positions from source = 'simu'
+2026-06-24 05:36:07,081 - Using experimental parameters specified by 'init_params' (dx, scan_step size, N_scan_slow, N_scan_fast) for initial position simulation.
+2026-06-24 05:36:07,081 - Simulating probe positions with dx = 0.2000, scan_step_size = 0.4714, N_scan_fast = 16, N_scan_slow = 16
+2026-06-24 05:36:07,081 - Applying Gaussian distributed random displacement with std = 0.15 px to scan positions
+2026-06-24 05:36:07,083 - crop_pos                                (N,2) = int16, (256, 2)
+2026-06-24 05:36:07,083 - crop_pos 1st and last px coords (y,x)         = ([17, 16], [51, 52])
+2026-06-24 05:36:07,084 - crop_pos extent (Ang)                         = [7.2 7.2]
+2026-06-24 05:36:07,084 - probe_pos_shifts                        (N,2) = float32, (256, 2)
+2026-06-24 05:36:07,084 -  
+2026-06-24 05:36:07,084 - ### Initializing object ###
+2026-06-24 05:36:07,084 - Loading object from source = 'simu'
+2026-06-24 05:36:07,084 - Using experimental parameters specified by 'init_params' for initial object simulation.
+2026-06-24 05:36:07,085 - omode_now: 1 and omode_max: 1, leaving the omode unchanged.
+2026-06-24 05:36:07,085 - object                    (omode, Nz, Ny, Nx) = complex64, (1, 1, 196, 196)
+2026-06-24 05:36:07,085 - object extent                 (Z, Y, X) (Ang) = [20.  39.2 39.2]
+2026-06-24 05:36:07,086 -  
+2026-06-24 05:36:07,086 - ### Initializing omode_occu from 'uniform' ###
+2026-06-24 05:36:07,086 - omode_occu                            (omode) = float32, (1,)
+2026-06-24 05:36:07,086 -  
+2026-06-24 05:36:07,086 - ### Initializing H (Fresnel propagator) ###
+2026-06-24 05:36:07,086 - Calculating H with probe_shape = (128, 128) px, dx = 0.2000 Ang, slice_thickness = 20.0000 Ang, lambd = 0.0197 Ang
+2026-06-24 05:36:07,086 - H                                    (Ky, Kx) = complex64, (128, 128)
+2026-06-24 05:36:07,086 -  
+2026-06-24 05:36:07,086 - ### Initializing obj tilts from = 'simu' ###
+2026-06-24 05:36:07,086 - Initialized obj_tilts with init_tilts = [[0, 0]] (theta_y, theta_x) mrad
+2026-06-24 05:36:07,087 - obj_tilts                              (N, 2) = float32, (1, 2)
+2026-06-24 05:36:07,087 -  
+2026-06-24 05:36:07,087 - ### Checking consistency between input params with the initialized variables ###
+2026-06-24 05:36:07,087 - meas_Npix, simu_Npix, DP measurements, probe, and H shapes are consistent as '128'
+2026-06-24 05:36:07,087 - N_scans, len(meas), N_scan_slow*N_scan_fast, len(crop_pos), and len(probe_pos_shifts) are consistent as '256'
+2026-06-24 05:36:07,087 - obj.shape[0] is consistent with len(omode_occu) as '1'
+2026-06-24 05:36:07,087 - obj.shape[1] is consistent with Nlayer as '1'
+2026-06-24 05:36:07,087 - crop positions (yx_min=[16 16], yx_max=[180 180]) are well contained inside object canvas (Ny,Nx) = (196, 196).
+2026-06-24 05:36:07,087 - obj_tilts is consistent with either 1 or N_scans
+2026-06-24 05:36:07,087 - Pass the consistency check of initialized variables, initialization is done!
+2026-06-24 05:36:07,087 -  
+2026-06-24 05:36:07,087 - ### Collecting reconstruction provenance ###
+2026-06-24 05:36:07,087 - Reconstruction provenance is collected and initialized.
+2026-06-24 05:36:07,087 -  
+2026-06-24 05:36:07,087 - ### Initializing loss function ###
+2026-06-24 05:36:07,087 - Active loss types:
+2026-06-24 05:36:07,087 -   loss_single : {'state': True, 'weight': 1.0, 'dp_pow': 0.5}
+2026-06-24 05:36:07,087 -  
+2026-06-24 05:36:07,087 - ### Initializing constraint function ###
+2026-06-24 05:36:07,087 - Active constraint types:
+2026-06-24 05:36:07,087 -   ortho_pmode   : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:36:07,087 -   fix_probe_int : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:36:07,088 -   obj_zblur     : {'start_iter': 1, 'step': 1, 'end_iter': None, 'obj_type': 'both', 'kernel_size': 5, 'std': 1.0}
+2026-06-24 05:36:07,088 -   obja_thresh   : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0, 'thresh': [0.96, 1.04]}
+2026-06-24 05:36:07,088 -   pos_recenter  : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0}
+2026-06-24 05:36:07,088 -  
+2026-06-24 05:36:07,088 - ### Done initializing PtyRADSolver ###
+2026-06-24 05:36:07,088 -  
+2026-06-24 05:36:07,212 - ### Starting the PtyRADSolver in reconstruct mode ###
+2026-06-24 05:36:07,212 -  
+2026-06-24 05:36:07,212 - ### Initializing PtychoModel model ###
+2026-06-24 05:36:07,284 - ### PtychoModel optimizable variables ###
+2026-06-24 05:36:07,284 - obja            : torch.Size([1, 1, 196, 196])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:36:07,284 - objp            : torch.Size([1, 1, 196, 196])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:36:07,284 - obj_tilts       : torch.Size([1, 2])              , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:36:07,285 - slice_thickness : torch.Size([])                  , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:36:07,285 - probe           : torch.Size([6, 128, 128, 2])    , torch.float32   , device:cuda:0, grad:True , lr:1e-04
+2026-06-24 05:36:07,285 - probe_pos_shifts: torch.Size([256, 2])            , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:36:07,285 -  
+2026-06-24 05:36:07,285 - ### Optimizable variables statitsics ###
+2026-06-24 05:36:07,285 - Total measurement values  : 4,194,304
+2026-06-24 05:36:07,285 - Total optimizing variables: 273,952
+2026-06-24 05:36:07,285 - Overdetermined ratio      : 15.31
+2026-06-24 05:36:07,285 -  
+2026-06-24 05:36:07,285 - ### Model behavior ###
+2026-06-24 05:36:07,285 - Tilt propagator           : False
+2026-06-24 05:36:07,285 - Change slice thickness    : False
+2026-06-24 05:36:07,285 - Detector blur             : False
+2026-06-24 05:36:07,285 - Preload data              : True
+2026-06-24 05:36:07,285 - On-the-fly meas padding   : False
+2026-06-24 05:36:07,285 - On-the-fly meas resample  : False
+2026-06-24 05:36:07,285 - On-the-fly simu match mode: None
+2026-06-24 05:36:07,285 -  
+2026-06-24 05:36:07,326 - ### Done initializing PtychoModel model ###
+2026-06-24 05:36:07,327 -  
+2026-06-24 05:36:07,327 - ### Creating PyTorch 'Adam' optimizer with configs = {} ###
+2026-06-24 05:36:07,327 -  
+2026-06-24 05:36:07,327 - ### Generating indices, batches, and output_path ###
+2026-06-24 05:36:07,328 - d90 = 57.000 px or 11.400 Ang
+2026-06-24 05:36:07,328 - Selecting indices with the 'full' mode 
+2026-06-24 05:36:07,793 - Generated 8 'random' groups of ~32 scan positions in 0.000 sec
+2026-06-24 05:36:07,868 - The effective batch size (i.e., how many probe positions are simultaneously used for 1 update of ptychographic parameters) is batch_size * grad_accumulation = 32 * 1 = 32
+2026-06-24 05:36:07,869 - Original recon_dir_affixes = ['default']
+2026-06-24 05:36:07,869 - Expanded recon_dir_affixes = ['indices', 'meas', 'batch', 'pmode', 'omode', 'nlayer', 'lr', 'model', 'constraint', 'loss', 'affine', 'tilt', 'aberrations']
+2026-06-24 05:36:07,874 - Path corrected for compatibility:
+2026-06-24 05:36:07,874 -   Original: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000008/c30_small__simulation_data1__sample_000008_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-308_C12_-23.6_phi12_165_C21_868_phi21_90.9_C23_-872_phi23_356_C30_-8.97e+05_C32_5.51e+03_phi32_175_C34_-1.01e+04_phi34_301_C41_4.23e+05_phi41_233_C43_-4.02e+05_phi43_265_C45_-3.08e+05_phi45_280_C50_-4.49e+08_C52_-1.81e+07_phi52_208_C54_-1.21e+07_phi54_49.5_C56_1.57e+07_phi56_344
+2026-06-24 05:36:07,874 -   Corrected: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000008/c30_small__simulation_data1__sample_000008_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-308_C12_-23.6_phi12_165_C21_868_phi21_90.9_C23_-872_phi23_356_C30_-8.97e+05_C32_.57e+07_phi56_344
+2026-06-24 05:36:07,875 - output_path = '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000008/c30_small__simulation_data1__sample_000008_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-308_C12_-23.6_phi12_165_C21_868_phi21_90.9_C23_-872_phi23_356_C30_-8.97e+05_C32_.57e+07_phi56_344' is generated!
+2026-06-24 05:36:08,031 -  
+2026-06-24 05:36:08,032 - ### Log file is flushed (created) as /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000008/c30_small__simulation_data1__sample_000008_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-308_C12_-23.6_phi12_165_C21_868_phi21_90.9_C23_-872_phi23_356_C30_-8.97e+05_C32_.57e+07_phi56_344/20260624_053608_ptyrad_log.txt ###
+2026-06-24 05:36:08,032 -  
+2026-06-24 05:36:08,033 - ### Creating ConvergenceMonitor with {'tensors': ['obja', 'objp', 'probe', 'probe_pos_shifts'], 'every_n_iters': None, 'percentile_range': [15.0, 85.0]} ### 
+2026-06-24 05:36:08,034 - ### Start the PtyRAD iterative ptycho reconstruction ###
+2026-06-24 05:36:08,034 - Setting up PyTorch compiler with {'fullgraph': False, 'dynamic': None, 'backend': 'inductor', 'mode': 'default', 'options': None, 'disable': True}
+2026-06-24 05:36:09,227 - Iter: 1, Total Loss: 0.7269, loss_single: 0.7269, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.434 sec
+2026-06-24 05:36:09,264 - Iter: 2, Total Loss: 0.7053, loss_single: 0.7053, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:36:09,300 - Iter: 3, Total Loss: 0.6886, loss_single: 0.6886, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,335 - Iter: 4, Total Loss: 0.6742, loss_single: 0.6742, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,370 - Iter: 5, Total Loss: 0.6580, loss_single: 0.6580, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,406 - Iter: 6, Total Loss: 0.6436, loss_single: 0.6436, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,441 - Iter: 7, Total Loss: 0.6307, loss_single: 0.6307, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,476 - Iter: 8, Total Loss: 0.6167, loss_single: 0.6167, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,512 - Iter: 9, Total Loss: 0.6013, loss_single: 0.6013, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,547 - Iter: 10, Total Loss: 0.5865, loss_single: 0.5865, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,582 - Iter: 11, Total Loss: 0.5721, loss_single: 0.5721, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:09,617 - Iter: 12, Total Loss: 0.5578, loss_single: 0.5578, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,653 - Iter: 13, Total Loss: 0.5437, loss_single: 0.5437, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,688 - Iter: 14, Total Loss: 0.5305, loss_single: 0.5305, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,723 - Iter: 15, Total Loss: 0.5167, loss_single: 0.5167, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:09,758 - Iter: 16, Total Loss: 0.5012, loss_single: 0.5012, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,793 - Iter: 17, Total Loss: 0.4881, loss_single: 0.4881, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,829 - Iter: 18, Total Loss: 0.4770, loss_single: 0.4770, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,864 - Iter: 19, Total Loss: 0.4659, loss_single: 0.4659, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,899 - Iter: 20, Total Loss: 0.4557, loss_single: 0.4557, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:09,934 - Iter: 21, Total Loss: 0.4458, loss_single: 0.4458, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:09,970 - Iter: 22, Total Loss: 0.4368, loss_single: 0.4368, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,009 - Iter: 23, Total Loss: 0.4289, loss_single: 0.4289, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.039 sec
+2026-06-24 05:36:10,044 - Iter: 24, Total Loss: 0.4185, loss_single: 0.4185, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,079 - Iter: 25, Total Loss: 0.4059, loss_single: 0.4059, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,115 - Iter: 26, Total Loss: 0.3938, loss_single: 0.3938, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,150 - Iter: 27, Total Loss: 0.3828, loss_single: 0.3828, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,186 - Iter: 28, Total Loss: 0.3735, loss_single: 0.3735, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,221 - Iter: 29, Total Loss: 0.3649, loss_single: 0.3649, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,256 - Iter: 30, Total Loss: 0.3562, loss_single: 0.3562, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,291 - Iter: 31, Total Loss: 0.3485, loss_single: 0.3485, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:10,326 - Iter: 32, Total Loss: 0.3415, loss_single: 0.3415, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,361 - Iter: 33, Total Loss: 0.3345, loss_single: 0.3345, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,397 - Iter: 34, Total Loss: 0.3284, loss_single: 0.3284, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,432 - Iter: 35, Total Loss: 0.3226, loss_single: 0.3226, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,467 - Iter: 36, Total Loss: 0.3146, loss_single: 0.3146, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,502 - Iter: 37, Total Loss: 0.3078, loss_single: 0.3078, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,537 - Iter: 38, Total Loss: 0.3022, loss_single: 0.3022, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,572 - Iter: 39, Total Loss: 0.2955, loss_single: 0.2955, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,608 - Iter: 40, Total Loss: 0.2891, loss_single: 0.2891, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,646 - Iter: 41, Total Loss: 0.2851, loss_single: 0.2851, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,682 - Iter: 42, Total Loss: 0.2826, loss_single: 0.2826, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,718 - Iter: 43, Total Loss: 0.2785, loss_single: 0.2785, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:36:10,761 - Iter: 44, Total Loss: 0.2741, loss_single: 0.2741, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,796 - Iter: 45, Total Loss: 0.2703, loss_single: 0.2703, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,832 - Iter: 46, Total Loss: 0.2667, loss_single: 0.2667, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:36:10,867 - Iter: 47, Total Loss: 0.2644, loss_single: 0.2644, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,902 - Iter: 48, Total Loss: 0.2622, loss_single: 0.2622, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,938 - Iter: 49, Total Loss: 0.2601, loss_single: 0.2601, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:10,973 - Iter: 50, Total Loss: 0.2583, loss_single: 0.2583, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,008 - Iter: 51, Total Loss: 0.2559, loss_single: 0.2559, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,044 - Iter: 52, Total Loss: 0.2542, loss_single: 0.2542, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,079 - Iter: 53, Total Loss: 0.2520, loss_single: 0.2520, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,115 - Iter: 54, Total Loss: 0.2497, loss_single: 0.2497, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,150 - Iter: 55, Total Loss: 0.2475, loss_single: 0.2475, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,185 - Iter: 56, Total Loss: 0.2457, loss_single: 0.2457, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,221 - Iter: 57, Total Loss: 0.2442, loss_single: 0.2442, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,256 - Iter: 58, Total Loss: 0.2424, loss_single: 0.2424, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,292 - Iter: 59, Total Loss: 0.2403, loss_single: 0.2403, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,327 - Iter: 60, Total Loss: 0.2389, loss_single: 0.2389, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,362 - Iter: 61, Total Loss: 0.2378, loss_single: 0.2378, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:11,397 - Iter: 62, Total Loss: 0.2365, loss_single: 0.2365, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,432 - Iter: 63, Total Loss: 0.2353, loss_single: 0.2353, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,467 - Iter: 64, Total Loss: 0.2343, loss_single: 0.2343, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,503 - Iter: 65, Total Loss: 0.2332, loss_single: 0.2332, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,538 - Iter: 66, Total Loss: 0.2319, loss_single: 0.2319, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,573 - Iter: 67, Total Loss: 0.2307, loss_single: 0.2307, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,609 - Iter: 68, Total Loss: 0.2299, loss_single: 0.2299, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,644 - Iter: 69, Total Loss: 0.2290, loss_single: 0.2290, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,679 - Iter: 70, Total Loss: 0.2282, loss_single: 0.2282, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,715 - Iter: 71, Total Loss: 0.2274, loss_single: 0.2274, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,750 - Iter: 72, Total Loss: 0.2268, loss_single: 0.2268, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,785 - Iter: 73, Total Loss: 0.2262, loss_single: 0.2262, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,821 - Iter: 74, Total Loss: 0.2257, loss_single: 0.2257, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,856 - Iter: 75, Total Loss: 0.2251, loss_single: 0.2251, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,891 - Iter: 76, Total Loss: 0.2247, loss_single: 0.2247, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,926 - Iter: 77, Total Loss: 0.2242, loss_single: 0.2242, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,961 - Iter: 78, Total Loss: 0.2238, loss_single: 0.2238, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:11,996 - Iter: 79, Total Loss: 0.2235, loss_single: 0.2235, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,031 - Iter: 80, Total Loss: 0.2231, loss_single: 0.2231, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,066 - Iter: 81, Total Loss: 0.2228, loss_single: 0.2228, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:12,101 - Iter: 82, Total Loss: 0.2225, loss_single: 0.2225, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,136 - Iter: 83, Total Loss: 0.2222, loss_single: 0.2222, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,172 - Iter: 84, Total Loss: 0.2219, loss_single: 0.2219, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,207 - Iter: 85, Total Loss: 0.2217, loss_single: 0.2217, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,243 - Iter: 86, Total Loss: 0.2215, loss_single: 0.2215, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,277 - Iter: 87, Total Loss: 0.2213, loss_single: 0.2213, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:12,313 - Iter: 88, Total Loss: 0.2211, loss_single: 0.2211, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,348 - Iter: 89, Total Loss: 0.2209, loss_single: 0.2209, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,383 - Iter: 90, Total Loss: 0.2207, loss_single: 0.2207, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,420 - Iter: 91, Total Loss: 0.2206, loss_single: 0.2206, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:36:12,456 - Iter: 92, Total Loss: 0.2204, loss_single: 0.2204, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,491 - Iter: 93, Total Loss: 0.2203, loss_single: 0.2203, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,526 - Iter: 94, Total Loss: 0.2202, loss_single: 0.2202, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,561 - Iter: 95, Total Loss: 0.2200, loss_single: 0.2200, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,596 - Iter: 96, Total Loss: 0.2199, loss_single: 0.2199, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,631 - Iter: 97, Total Loss: 0.2198, loss_single: 0.2198, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,667 - Iter: 98, Total Loss: 0.2197, loss_single: 0.2197, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,702 - Iter: 99, Total Loss: 0.2196, loss_single: 0.2196, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,737 - Iter: 100, Total Loss: 0.2196, loss_single: 0.2196, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:12,850 - Saving summary figures for iter 100
+2026-06-24 05:36:16,637 - ### Finished 100 iterations, averaged iter_t = 0.038919 with std = 0.040 ###
+2026-06-24 05:36:16,638 -  
+2026-06-24 05:36:16,638 - ### The PtyRADSolver is finished in 9.426 sec ###
+2026-06-24 05:36:16,638 -  

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000008/c30_small__simulation_data1__sample_000008_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-308_C12_-23.6_phi12_165_C21_868_phi21_90.9_C23_-872_phi23_356_C30_-8.97e+05_C32_.57e+07_phi56_344/simulation_data1__sample_000008.yaml

@@ -0,0 +1,140 @@
+init_params:
+  probe_kv: 300.0
+  probe_conv_angle: 12.933050155639648
+  probe_aberrations:
+    C10: -307.54443359375
+    C12: -23.624135971069336
+    C21: 867.66064453125
+    C23: -871.9953002929688
+    C30: -897010.875
+    C32: 5510.82421875
+    C34: -10106.435546875
+    C41: 422961.8125
+    C43: -401591.34375
+    C45: -308371.03125
+    C50: -448757952.0
+    C52: -18148220.0
+    C54: -12083272.0
+    C56: 15654723.0
+    phi12: 165.47571702195094
+    phi21: 90.94372974209882
+    phi23: 355.5720803412634
+    phi32: 174.58019526103786
+    phi34: 300.8448617674561
+    phi41: 233.1174982595534
+    phi43: 264.8271159342012
+    phi45: 279.5500531430272
+    phi52: 207.60881808706347
+    phi54: 49.52084540726258
+    phi56: 343.8609833488462
+  meas_Npix: 128
+  pos_N_scan_slow: 16
+  pos_N_scan_fast: 16
+  pos_scan_step_size: 0.4713726341724396
+  probe_pmode_max: 6
+  obj_Nlayer: 1
+  obj_slice_thickness: 20.0
+  meas_permute: null
+  meas_reshape:
+  - 256
+  - 128
+  - 128
+  meas_flipT:
+  - 0
+  - 0
+  - 0
+  meas_crop: null
+  meas_pad: null
+  meas_resample: null
+  pos_scan_affine: null
+  meas_params:
+    path: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000008_measurement.h5
+    key: measurement
+  random_seed: 20261852
+  obj_omode_max: 1
+  meas_calibration:
+    mode: kMax
+    value: 2.5
+model_params:
+  detector_blur_std: null
+  update_params:
+    obja:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    objp:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    probe:
+      start_iter: 1
+      lr: 0.0001
+      end_iter: null
+    probe_pos_shifts:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+loss_params:
+  loss_single:
+    state: true
+    weight: 1.0
+    dp_pow: 0.5
+  loss_poissn:
+    state: false
+    weight: 1.0
+    dp_pow: 1.0
+    eps: 1.0e-06
+  loss_sparse:
+    state: false
+    weight: 0.1
+    ln_order: 1
+constraint_params:
+  ortho_pmode:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  fix_probe_int:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  obj_rblur:
+    start_iter: null
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 0.4
+  obj_zblur:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 1
+  obja_thresh:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    relax: 0
+    thresh:
+    - 0.96
+    - 1.04
+  objp_postiv:
+    start_iter: null
+    step: 1
+    end_iter: null
+    relax: 0
+recon_params:
+  NITER: 100
+  BATCH_SIZE:
+    size: 32
+    grad_accumulation: 1
+  SAVE_ITERS: 100
+  output_dir: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000008
+  recon_dir_affixes:
+  - default
+  prefix: c30_small__simulation_data1__sample_000008
+  postfix: ''
+  compiler_configs:
+    enable: false
+  prefix_time: false

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000009/c30_small__simulation_data1__sample_000009_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-267_C12_18.2_phi12_230_C21_758_phi21_299_C23_882_phi23_3.72_C30_1.33e+05_C32_9.9.88e+05_phi56_333/20260624_053622_ptyrad_log.txt

@@ -0,0 +1,417 @@
+2026-06-24 05:36:21,408 - ### PtyRAD LoggingManager configuration ###
+2026-06-24 05:36:21,408 - log_file       = 'ptyrad_log.txt'. If log_file = None, no log file will be created.
+2026-06-24 05:36:21,408 - log_dir        = 'auto'. If log_dir = 'auto', then log will be saved to `output_path` or 'logs/'.
+2026-06-24 05:36:21,408 - flush_file     = True. Automatically set to True if `log_file is not None`
+2026-06-24 05:36:21,408 - prefix_time    = datetime. If true, preset strings ('date', 'time', 'datetime'), or a string of time format, a datetime str is prefixed to the `log_file`.
+2026-06-24 05:36:21,408 - prefix_jobid   = '0'. If not 0, it'll be prefixed to the log file. This is used for hypertune mode with multiple GPUs.
+2026-06-24 05:36:21,408 - append_to_file = True. If true, logs will be appended to the existing file. If false, the log file will be overwritten.
+2026-06-24 05:36:21,408 - show_timestamp = True. If true, the printed information will contain a timestamp.
+2026-06-24 05:36:21,408 -  
+2026-06-24 05:36:21,519 - ### Initializing HuggingFace accelerator ###
+2026-06-24 05:36:21,519 - Accelerator.distributed_type = DistributedType.NO
+2026-06-24 05:36:21,519 - Accelerator.num_process      = 1
+2026-06-24 05:36:21,519 - Accelerator.mixed_precision  = no
+2026-06-24 05:36:21,530 - 'accelerate' is available but NOT using distributed mode or mixed precision
+2026-06-24 05:36:21,530 - If you want to utilize 'accelerate' for multiGPU or mixed precision, 
+2026-06-24 05:36:21,530 - Run `accelerate launch --multi_gpu --num_processes=2 --mixed_precision='no' -m ptyrad run <PTYRAD_ARGUMENTS> --gpuid 'acc'` in your terminal
+2026-06-24 05:36:21,530 -  
+2026-06-24 05:36:21,530 - ### System information ###
+2026-06-24 05:36:21,530 - Platform: Linux-4.18.0-553.69.1.el8_10.x86_64-x86_64-with-glibc2.28
+2026-06-24 05:36:21,530 - Operating System: Linux 4.18.0-553.69.1.el8_10.x86_64
+2026-06-24 05:36:21,530 - OS Version: #1 SMP Thu Aug 7 18:10:00 EDT 2025
+2026-06-24 05:36:21,530 - Machine: x86_64
+2026-06-24 05:36:21,530 - Processor: x86_64
+2026-06-24 05:36:21,530 - Available CPU cores: 8
+2026-06-24 05:36:21,530 - SLURM-Allocated Total Memory: 78.12 GB
+2026-06-24 05:36:21,530 -  
+2026-06-24 05:36:21,530 - ### GPU information ###
+2026-06-24 05:36:21,530 - CUDA Available: True
+2026-06-24 05:36:21,530 - CUDA Version: 13.0
+2026-06-24 05:36:21,531 - Available CUDA GPUs: ['NVIDIA A100 80GB PCIe']
+2026-06-24 05:36:21,531 - CUDA Compute Capability: ['8.0']
+2026-06-24 05:36:21,531 -   INFO: For torch.compile with Triton, you'll need CUDA GPU with Compute Capability >= 7.0.
+2026-06-24 05:36:21,531 -         In addition, Triton does not directly support Windows.
+2026-06-24 05:36:21,531 -         For Windows users, please follow the instruction and download `triton-windows` from https://github.com/woct0rdho/triton-windows.
+2026-06-24 05:36:21,547 - MIG (Multi-Instance GPU) mode = False
+2026-06-24 05:36:21,547 -   INFO: MIG splits a physical GPU into multiple GPU slices, but multiGPU does not support these MIG slices.
+2026-06-24 05:36:21,547 -         In addition, multiGPU is currently only available on Linux due to the limited NCCL support.
+2026-06-24 05:36:21,547 -       -> If you're doing normal reconstruction/hypertune, you can safely ignore this.
+2026-06-24 05:36:21,547 -       -> If you want to do multiGPU, you must provide multiple 'full' GPUs that are not in MIG mode.
+2026-06-24 05:36:21,547 -  
+2026-06-24 05:36:21,547 - ### Python information ###
+2026-06-24 05:36:21,549 - Python Executable: /home/tnguye11/anaconda3/envs/ptyrad/bin/python3.12
+2026-06-24 05:36:21,549 - Python Version: 3.12.13 | packaged by Anaconda, Inc. | (main, Mar 19 2026, 20:20:58) [GCC 14.3.0]
+2026-06-24 05:36:21,549 -  
+2026-06-24 05:36:21,549 - ### Packages information ###
+2026-06-24 05:36:21,550 - Numpy Version (metadata): 2.4.6
+2026-06-24 05:36:21,551 - PyTorch Version (metadata): 2.12.0
+2026-06-24 05:36:21,551 - Optuna Version (metadata): 4.9.0
+2026-06-24 05:36:21,552 - Accelerate Version (metadata): 1.13.0
+2026-06-24 05:36:21,552 - PtyRAD Version (ptyrad/__init__.py): 1.0.0
+2026-06-24 05:36:21,552 - PtyRAD is located at: /home/tnguye11/anaconda3/envs/ptyrad/lib/python3.12/site-packages/ptyrad/__init__.py
+2026-06-24 05:36:21,552 -  
+2026-06-24 05:36:21,552 - ### Loading params file ###
+2026-06-24 05:36:21,553 - params_path = /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/params/c30_small/simulation_data1__sample_000009.yaml
+2026-06-24 05:36:21,559 - validate = True: Filling defaults and validating the params file...
+2026-06-24 05:36:21,560 - Success! Params file validated and defaults applied.
+2026-06-24 05:36:21,560 -  
+2026-06-24 05:36:21,560 - ### Setting GPU Device ###
+2026-06-24 05:36:21,560 - Selected GPU device: cuda:0 (NVIDIA A100 80GB PCIe)
+2026-06-24 05:36:21,560 -  
+2026-06-24 05:36:21,560 - Random seed: 20261853 provided by params file
+2026-06-24 05:36:21,561 - ### Initializing Initializer ###
+2026-06-24 05:36:21,561 - init_params are displayed below:
+2026-06-24 05:36:21,561 -   random_seed: 20261853
+2026-06-24 05:36:21,561 -   probe_illum_type: electron
+2026-06-24 05:36:21,561 -   probe_kv: 300.0
+2026-06-24 05:36:21,561 -   probe_conv_angle: 22.653152465820312
+2026-06-24 05:36:21,561 -   probe_aberrations: {'C10': -266.603, 'C12': 18.172, 'phi12': 229.657, 'C21': 757.701, 'phi21': 298.59, 'C23': 881.624, 'phi23': 3.72, 'C30': 133419.656, 'C32': 9907.309, 'phi32': 59.121, 'C34': -1121.24, 'phi34': 318.911, 'C41': -363432.125, 'phi41': 242.013, 'C43': -177851.125, 'phi43': 339.07, 'C45': 41571.309, 'phi45': 55.737, 'C50': 269112576.0, 'C52': 18410188.0, 'phi52': 311.7, 'C54': 17605530.0, 'phi54': 199.502, 'C56': -788044.688, 'phi56': 333.474}
+2026-06-24 05:36:21,561 -   beam_kev: None
+2026-06-24 05:36:21,561 -   probe_dRn: None
+2026-06-24 05:36:21,561 -   probe_Rn: None
+2026-06-24 05:36:21,561 -   probe_D_H: None
+2026-06-24 05:36:21,561 -   probe_D_FZP: None
+2026-06-24 05:36:21,561 -   probe_Ls: None
+2026-06-24 05:36:21,561 -   meas_Npix: 128
+2026-06-24 05:36:21,561 -   pos_N_scans: 256
+2026-06-24 05:36:21,561 -   pos_N_scan_slow: 16
+2026-06-24 05:36:21,561 -   pos_N_scan_fast: 16
+2026-06-24 05:36:21,561 -   pos_scan_step_size: 0.43370673060417175
+2026-06-24 05:36:21,561 -   meas_calibration: {'mode': 'kMax', 'value': 2.5}
+2026-06-24 05:36:21,561 -   probe_pmode_max: 6
+2026-06-24 05:36:21,561 -   probe_pmode_init_pows: [0.02]
+2026-06-24 05:36:21,561 -   obj_omode_max: 1
+2026-06-24 05:36:21,561 -   obj_omode_init_occu: {'occu_type': 'uniform', 'init_occu': None}
+2026-06-24 05:36:21,561 -   obj_Nlayer: 1
+2026-06-24 05:36:21,561 -   obj_slice_thickness: 20.0
+2026-06-24 05:36:21,561 -   simu_Npix: None
+2026-06-24 05:36:21,561 -   simu_match_mode: None
+2026-06-24 05:36:21,561 -   meas_permute: None
+2026-06-24 05:36:21,561 -   meas_reshape: [256, 128, 128]
+2026-06-24 05:36:21,561 -   meas_flipT: [0, 0, 0]
+2026-06-24 05:36:21,561 -   meas_crop: None
+2026-06-24 05:36:21,561 -   meas_pad: None
+2026-06-24 05:36:21,561 -   meas_resample: None
+2026-06-24 05:36:21,562 -   meas_add_source_size: None
+2026-06-24 05:36:21,562 -   meas_add_detector_blur: None
+2026-06-24 05:36:21,562 -   meas_remove_neg_values: {'mode': 'clip_neg', 'value': None, 'force': False}
+2026-06-24 05:36:21,562 -   meas_normalization: {'mode': 'max_at_one', 'value': None}
+2026-06-24 05:36:21,562 -   meas_add_poisson_noise: None
+2026-06-24 05:36:21,562 -   meas_export: None
+2026-06-24 05:36:21,562 -   probe_permute: None
+2026-06-24 05:36:21,562 -   probe_z_shift: None
+2026-06-24 05:36:21,562 -   probe_normalization: {'mode': 'mean_total_ints', 'value': None}
+2026-06-24 05:36:21,562 -   pos_scan_flipT: None
+2026-06-24 05:36:21,562 -   pos_scan_affine: None
+2026-06-24 05:36:21,562 -   pos_scan_rand_std: 0.15
+2026-06-24 05:36:21,562 -   obj_z_crop: None
+2026-06-24 05:36:21,562 -   obj_z_pad: None
+2026-06-24 05:36:21,562 -   obj_z_resample: None
+2026-06-24 05:36:21,562 -   meas_source: file
+2026-06-24 05:36:21,562 -   meas_params: {'path': '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000009_measurement.h5', 'key': 'measurement', 'shape': None, 'offset': None, 'gap': None, 'selection': None, 'zarr_kwargs': None}
+2026-06-24 05:36:21,562 -   probe_source: simu
+2026-06-24 05:36:21,562 -   probe_params: None
+2026-06-24 05:36:21,562 -   pos_source: simu
+2026-06-24 05:36:21,562 -   pos_params: None
+2026-06-24 05:36:21,562 -   obj_source: simu
+2026-06-24 05:36:21,562 -   obj_params: None
+2026-06-24 05:36:21,562 -   tilt_source: simu
+2026-06-24 05:36:21,562 -   tilt_params: {'tilt_type': 'all', 'init_tilts': [[0, 0]]}
+2026-06-24 05:36:21,562 -  
+2026-06-24 05:36:21,562 - ### Initializing cache ###
+2026-06-24 05:36:21,562 - use_cached_obj   = False
+2026-06-24 05:36:21,562 - use_cached_probe = False
+2026-06-24 05:36:21,562 - use_cached_pos   = False
+2026-06-24 05:36:21,562 -  
+2026-06-24 05:36:21,562 - ### Initializing measurements ###
+2026-06-24 05:36:21,562 - Loading measurements from source = 'file'
+2026-06-24 05:36:21,562 - Detected measurement file type = '.h5'
+2026-06-24 05:36:21,644 - Original measurements dtype is float32, casting to float32 (single precision) for computational efficiency.
+2026-06-24 05:36:21,645 - Imported meausrements shape / dtype = (16, 16, 128, 128), dtype = float32
+2026-06-24 05:36:21,646 - Imported meausrements int. statistics (min, mean, max) = (0.0000, 0.0001, 0.0005)
+2026-06-24 05:36:21,646 - Reshaping measurements to shape = [256, 128, 128]
+2026-06-24 05:36:21,646 - Flipping measurements with [flipud, fliplr, transpose] = [0, 0, 0]
+2026-06-24 05:36:21,648 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:36:21,648 - Normalizing measurements with mode = 'max_at_one' and value = 'None'
+2026-06-24 05:36:21,649 - Normalizing by max of the 2D mean pattern intensity: 0.00036771825
+2026-06-24 05:36:21,650 - meausrements shape / dtype = (256, 128, 128), dtype = float32
+2026-06-24 05:36:21,651 - meausrements int. statistics (min, mean, max) = (0.0000, 0.1651, 1.4306)
+2026-06-24 05:36:21,651 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:36:21,653 - Pattern total int. statistics (min, mean, max)        = (2691.6978, 2705.3447, 2721.6626), with min/max = 98.9%
+2026-06-24 05:36:21,654 - Global meausrements int. statistics (min, mean, max)  = (0.0000, 0.1651, 1.4306)
+2026-06-24 05:36:21,654 - measurements                              (N, Ky, Kx) = float32, (256, 128, 128)
+2026-06-24 05:36:21,654 -  
+2026-06-24 05:36:21,654 - ### Setting up calibration ###
+2026-06-24 05:36:21,654 - meas_calibration mode = 'kMax', value = 2.5
+2026-06-24 05:36:21,654 - Using loaded raw averaged measurement (before crop/pad/resample) to fit RBF as a part of the meas calibration
+2026-06-24 05:36:21,654 - Radius of fitted bright field disk (RBF) = 29.22 px with meas_Npix = 128
+2026-06-24 05:36:21,654 - Suggested probe_mask_k radius (RBF*2/Npix) > 0.4565
+2026-06-24 05:36:21,654 - Fitting raw averaged measurement with center, radius, and Gaussian blur std as a sanity check
+2026-06-24 05:36:21,654 - Note that the fitted Gaussian blur std (detector blur) would be affected by overlapping Bragg disks
+2026-06-24 05:36:21,883 - Initial guess: center=(63.62, 63.63), radius=29.22, Gaussian blur std=0.50
+2026-06-24 05:36:21,899 - Final fit: center=(63.62, 63.63), radius=29.22, Gaussian blur std=0.80
+2026-06-24 05:36:21,899 - Using init_params, the inferred RBF (conv_angle / 1e3 * Npix * dx / wavelength) = 29.46 px with Npix = 128
+2026-06-24 05:36:21,899 - dx (real space pixel size of probe and object) set to 0.2000 Ang with Npix = 128
+2026-06-24 05:36:21,899 -  
+2026-06-24 05:36:21,899 - ### Setting init_variables dict ###
+2026-06-24 05:36:21,900 - Derived values given input init_params:
+2026-06-24 05:36:21,900 -   kv          = 300.0 kV
+2026-06-24 05:36:21,900 -   wavelength  = 0.0197 Ang
+2026-06-24 05:36:21,900 -   conv_angle  = 22.653152465820312 mrad
+2026-06-24 05:36:21,900 -   Npix        = 128 px
+2026-06-24 05:36:21,900 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:36:21,900 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:36:21,900 -   da          = 0.7690 mrad
+2026-06-24 05:36:21,900 -   angleMax    = 49.2187 mrad
+2026-06-24 05:36:21,900 -   RBF         = 29.4563 px (Inferred from the given calibration, NOT necessarily from the loaded measurement data)
+2026-06-24 05:36:21,900 -   n_alpha     = 2.1727 (# conv_angle)
+2026-06-24 05:36:21,900 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:36:21,900 -   Rayleigh-limited resolution  = 0.5301 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:36:21,900 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:36:21,900 -  
+2026-06-24 05:36:21,900 - ### Initializing probe ###
+2026-06-24 05:36:21,900 - Loading probe from source = 'simu'
+2026-06-24 05:36:21,900 - Using experimental parameters specified by 'init_params' for initial probe simulation.
+2026-06-24 05:36:21,900 - Start simulating STEM probe
+2026-06-24 05:36:21,900 -   kv          = 300.0 kV
+2026-06-24 05:36:21,900 -   wavelength  = 0.0197 Ang
+2026-06-24 05:36:21,900 -   conv_angle  = 22.653152465820312 mrad
+2026-06-24 05:36:21,900 -   Npix        = 128 px
+2026-06-24 05:36:21,900 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:36:21,901 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:36:21,901 -   alpha_max   = 49.2187 mrad
+2026-06-24 05:36:21,901 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:36:21,901 -   Rayleigh-limited resolution  = 0.5301 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:36:21,901 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:36:21,901 - Krivanek Haider     Magnitude        Angle (°)  Description
+2026-06-24 05:36:21,901 - ------------------------------------------------------------------------------------
+2026-06-24 05:36:21,901 - C1,0     C1         -266.6030        -          Defocus (C10 = -df)
+2026-06-24 05:36:21,901 - C1,2     A1         18.1720          229.66     2-fold astigmatism
+2026-06-24 05:36:21,901 - C2,1     3*B2       757.7010         298.59     Axial coma
+2026-06-24 05:36:21,901 - C2,3     A2         881.6240         3.72       3-fold astigmatism
+2026-06-24 05:36:21,901 - C3,0     C3         133419.6560      -          Spherical aberration
+2026-06-24 05:36:21,901 - C3,2     4*S3       9907.3090        59.12      Axial star aberration
+2026-06-24 05:36:21,901 - C3,4     A3         -1121.2400       318.91     4-fold astigmatism
+2026-06-24 05:36:21,901 - C4,1     4*B4       -363432.1250     242.01     Axial coma(4th)
+2026-06-24 05:36:21,901 - C4,3     4*D4       -177851.1250     339.07     3-lobe aberration
+2026-06-24 05:36:21,901 - C4,5     A4         41571.3090       55.74      5-fold astigmatism
+2026-06-24 05:36:21,901 - C5,0     C5         269112576.0000   -          Spherical aberration (5th)
+2026-06-24 05:36:21,901 - C5,2     6*S5       18410188.0000    311.70     Axial star aberration(5th)
+2026-06-24 05:36:21,901 - C5,4     6*R5       17605530.0000    199.50     4-lobe aberration
+2026-06-24 05:36:21,901 - C5,6     A5         -788044.6880     333.47     6-fold astigmatism
+2026-06-24 05:36:21,904 - Loaded probe shape = (1, 128, 128), dtype = complex128
+2026-06-24 05:36:21,904 - pmode_now: 1 and pmode_max: 6, padding the pmode.
+2026-06-24 05:36:21,904 - Creating 5 new probe modes from the major mode
+2026-06-24 05:36:21,904 - Start making mixed-state STEM probe with 6 incoherent probe modes
+2026-06-24 05:36:22,050 - Relative power of probe modes = [0.9  0.02 0.02 0.02 0.02 0.02]
+2026-06-24 05:36:22,051 - Orthogonalizing 6 pmodes
+2026-06-24 05:36:22,054 - Sorting 6 pmodes by their intensities
+2026-06-24 05:36:22,054 - Normalizing probe intensity with mode = 'mean_total_ints' and value = 'None'
+2026-06-24 05:36:22,055 - sum(|probe_data|**2) = 2705.34, while meas_total_ints (min, mean, max) = (2691.6978, 2705.3447, 2721.6626)
+2026-06-24 05:36:22,055 - probe                         (pmode, Ny, Nx) = complex64, (6, 128, 128)
+2026-06-24 05:36:22,055 -  
+2026-06-24 05:36:22,055 - ### Initializing probe positions ###
+2026-06-24 05:36:22,055 - Loading probe positions from source = 'simu'
+2026-06-24 05:36:22,055 - Using experimental parameters specified by 'init_params' (dx, scan_step size, N_scan_slow, N_scan_fast) for initial position simulation.
+2026-06-24 05:36:22,055 - Simulating probe positions with dx = 0.2000, scan_step_size = 0.4337, N_scan_fast = 16, N_scan_slow = 16
+2026-06-24 05:36:22,055 - Applying Gaussian distributed random displacement with std = 0.15 px to scan positions
+2026-06-24 05:36:22,057 - crop_pos                                (N,2) = int16, (256, 2)
+2026-06-24 05:36:22,057 - crop_pos 1st and last px coords (y,x)         = ([16, 16], [48, 48])
+2026-06-24 05:36:22,057 - crop_pos extent (Ang)                         = [6.8 6.8]
+2026-06-24 05:36:22,057 - probe_pos_shifts                        (N,2) = float32, (256, 2)
+2026-06-24 05:36:22,057 -  
+2026-06-24 05:36:22,057 - ### Initializing object ###
+2026-06-24 05:36:22,057 - Loading object from source = 'simu'
+2026-06-24 05:36:22,058 - Using experimental parameters specified by 'init_params' for initial object simulation.
+2026-06-24 05:36:22,059 - omode_now: 1 and omode_max: 1, leaving the omode unchanged.
+2026-06-24 05:36:22,059 - object                    (omode, Nz, Ny, Nx) = complex64, (1, 1, 194, 194)
+2026-06-24 05:36:22,059 - object extent                 (Z, Y, X) (Ang) = [20.  38.8 38.8]
+2026-06-24 05:36:22,059 -  
+2026-06-24 05:36:22,059 - ### Initializing omode_occu from 'uniform' ###
+2026-06-24 05:36:22,059 - omode_occu                            (omode) = float32, (1,)
+2026-06-24 05:36:22,059 -  
+2026-06-24 05:36:22,059 - ### Initializing H (Fresnel propagator) ###
+2026-06-24 05:36:22,059 - Calculating H with probe_shape = (128, 128) px, dx = 0.2000 Ang, slice_thickness = 20.0000 Ang, lambd = 0.0197 Ang
+2026-06-24 05:36:22,060 - H                                    (Ky, Kx) = complex64, (128, 128)
+2026-06-24 05:36:22,060 -  
+2026-06-24 05:36:22,060 - ### Initializing obj tilts from = 'simu' ###
+2026-06-24 05:36:22,060 - Initialized obj_tilts with init_tilts = [[0, 0]] (theta_y, theta_x) mrad
+2026-06-24 05:36:22,060 - obj_tilts                              (N, 2) = float32, (1, 2)
+2026-06-24 05:36:22,060 -  
+2026-06-24 05:36:22,060 - ### Checking consistency between input params with the initialized variables ###
+2026-06-24 05:36:22,060 - meas_Npix, simu_Npix, DP measurements, probe, and H shapes are consistent as '128'
+2026-06-24 05:36:22,060 - N_scans, len(meas), N_scan_slow*N_scan_fast, len(crop_pos), and len(probe_pos_shifts) are consistent as '256'
+2026-06-24 05:36:22,060 - obj.shape[0] is consistent with len(omode_occu) as '1'
+2026-06-24 05:36:22,060 - obj.shape[1] is consistent with Nlayer as '1'
+2026-06-24 05:36:22,061 - crop positions (yx_min=[15 15], yx_max=[177 177]) are well contained inside object canvas (Ny,Nx) = (194, 194).
+2026-06-24 05:36:22,061 - obj_tilts is consistent with either 1 or N_scans
+2026-06-24 05:36:22,061 - Pass the consistency check of initialized variables, initialization is done!
+2026-06-24 05:36:22,061 -  
+2026-06-24 05:36:22,061 - ### Collecting reconstruction provenance ###
+2026-06-24 05:36:22,061 - Reconstruction provenance is collected and initialized.
+2026-06-24 05:36:22,061 -  
+2026-06-24 05:36:22,061 - ### Initializing loss function ###
+2026-06-24 05:36:22,061 - Active loss types:
+2026-06-24 05:36:22,061 -   loss_single : {'state': True, 'weight': 1.0, 'dp_pow': 0.5}
+2026-06-24 05:36:22,061 -  
+2026-06-24 05:36:22,061 - ### Initializing constraint function ###
+2026-06-24 05:36:22,061 - Active constraint types:
+2026-06-24 05:36:22,061 -   ortho_pmode   : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:36:22,061 -   fix_probe_int : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:36:22,061 -   obj_zblur     : {'start_iter': 1, 'step': 1, 'end_iter': None, 'obj_type': 'both', 'kernel_size': 5, 'std': 1.0}
+2026-06-24 05:36:22,061 -   obja_thresh   : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0, 'thresh': [0.96, 1.04]}
+2026-06-24 05:36:22,061 -   pos_recenter  : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0}
+2026-06-24 05:36:22,061 -  
+2026-06-24 05:36:22,061 - ### Done initializing PtyRADSolver ###
+2026-06-24 05:36:22,061 -  
+2026-06-24 05:36:22,184 - ### Starting the PtyRADSolver in reconstruct mode ###
+2026-06-24 05:36:22,184 -  
+2026-06-24 05:36:22,184 - ### Initializing PtychoModel model ###
+2026-06-24 05:36:22,257 - ### PtychoModel optimizable variables ###
+2026-06-24 05:36:22,257 - obja            : torch.Size([1, 1, 194, 194])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:36:22,257 - objp            : torch.Size([1, 1, 194, 194])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:36:22,257 - obj_tilts       : torch.Size([1, 2])              , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:36:22,257 - slice_thickness : torch.Size([])                  , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:36:22,257 - probe           : torch.Size([6, 128, 128, 2])    , torch.float32   , device:cuda:0, grad:True , lr:1e-04
+2026-06-24 05:36:22,257 - probe_pos_shifts: torch.Size([256, 2])            , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:36:22,257 -  
+2026-06-24 05:36:22,257 - ### Optimizable variables statitsics ###
+2026-06-24 05:36:22,257 - Total measurement values  : 4,194,304
+2026-06-24 05:36:22,257 - Total optimizing variables: 272,392
+2026-06-24 05:36:22,257 - Overdetermined ratio      : 15.40
+2026-06-24 05:36:22,257 -  
+2026-06-24 05:36:22,257 - ### Model behavior ###
+2026-06-24 05:36:22,257 - Tilt propagator           : False
+2026-06-24 05:36:22,257 - Change slice thickness    : False
+2026-06-24 05:36:22,257 - Detector blur             : False
+2026-06-24 05:36:22,257 - Preload data              : True
+2026-06-24 05:36:22,257 - On-the-fly meas padding   : False
+2026-06-24 05:36:22,257 - On-the-fly meas resample  : False
+2026-06-24 05:36:22,257 - On-the-fly simu match mode: None
+2026-06-24 05:36:22,258 -  
+2026-06-24 05:36:22,299 - ### Done initializing PtychoModel model ###
+2026-06-24 05:36:22,299 -  
+2026-06-24 05:36:22,299 - ### Creating PyTorch 'Adam' optimizer with configs = {} ###
+2026-06-24 05:36:22,299 -  
+2026-06-24 05:36:22,299 - ### Generating indices, batches, and output_path ###
+2026-06-24 05:36:22,300 - d90 = 41.000 px or 8.200 Ang
+2026-06-24 05:36:22,300 - Selecting indices with the 'full' mode 
+2026-06-24 05:36:22,761 - Generated 8 'random' groups of ~32 scan positions in 0.000 sec
+2026-06-24 05:36:22,834 - The effective batch size (i.e., how many probe positions are simultaneously used for 1 update of ptychographic parameters) is batch_size * grad_accumulation = 32 * 1 = 32
+2026-06-24 05:36:22,834 - Original recon_dir_affixes = ['default']
+2026-06-24 05:36:22,834 - Expanded recon_dir_affixes = ['indices', 'meas', 'batch', 'pmode', 'omode', 'nlayer', 'lr', 'model', 'constraint', 'loss', 'affine', 'tilt', 'aberrations']
+2026-06-24 05:36:22,841 - Path corrected for compatibility:
+2026-06-24 05:36:22,841 -   Original: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000009/c30_small__simulation_data1__sample_000009_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-267_C12_18.2_phi12_230_C21_758_phi21_299_C23_882_phi23_3.72_C30_1.33e+05_C32_9.91e+03_phi32_59.1_C34_-1.12e+03_phi34_319_C41_-3.63e+05_phi41_242_C43_-1.78e+05_phi43_339_C45_4.16e+04_phi45_55.7_C50_2.69e+08_C52_1.84e+07_phi52_312_C54_1.76e+07_phi54_200_C56_-7.88e+05_phi56_333
+2026-06-24 05:36:22,841 -   Corrected: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000009/c30_small__simulation_data1__sample_000009_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-267_C12_18.2_phi12_230_C21_758_phi21_299_C23_882_phi23_3.72_C30_1.33e+05_C32_9.9.88e+05_phi56_333
+2026-06-24 05:36:22,842 - output_path = '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000009/c30_small__simulation_data1__sample_000009_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-267_C12_18.2_phi12_230_C21_758_phi21_299_C23_882_phi23_3.72_C30_1.33e+05_C32_9.9.88e+05_phi56_333' is generated!
+2026-06-24 05:36:22,991 -  
+2026-06-24 05:36:22,992 - ### Log file is flushed (created) as /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000009/c30_small__simulation_data1__sample_000009_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-267_C12_18.2_phi12_230_C21_758_phi21_299_C23_882_phi23_3.72_C30_1.33e+05_C32_9.9.88e+05_phi56_333/20260624_053622_ptyrad_log.txt ###
+2026-06-24 05:36:22,992 -  
+2026-06-24 05:36:22,993 - ### Creating ConvergenceMonitor with {'tensors': ['obja', 'objp', 'probe', 'probe_pos_shifts'], 'every_n_iters': None, 'percentile_range': [15.0, 85.0]} ### 
+2026-06-24 05:36:22,993 - ### Start the PtyRAD iterative ptycho reconstruction ###
+2026-06-24 05:36:22,993 - Setting up PyTorch compiler with {'fullgraph': False, 'dynamic': None, 'backend': 'inductor', 'mode': 'default', 'options': None, 'disable': True}
+2026-06-24 05:36:24,185 - Iter: 1, Total Loss: 0.3887, loss_single: 0.3887, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.437 sec
+2026-06-24 05:36:24,222 - Iter: 2, Total Loss: 0.3786, loss_single: 0.3786, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:36:24,257 - Iter: 3, Total Loss: 0.3696, loss_single: 0.3696, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,293 - Iter: 4, Total Loss: 0.3604, loss_single: 0.3604, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,328 - Iter: 5, Total Loss: 0.3520, loss_single: 0.3520, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,363 - Iter: 6, Total Loss: 0.3446, loss_single: 0.3446, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,399 - Iter: 7, Total Loss: 0.3370, loss_single: 0.3370, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,434 - Iter: 8, Total Loss: 0.3301, loss_single: 0.3301, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,469 - Iter: 9, Total Loss: 0.3232, loss_single: 0.3232, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,506 - Iter: 10, Total Loss: 0.3161, loss_single: 0.3161, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:36:24,541 - Iter: 11, Total Loss: 0.3095, loss_single: 0.3095, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,577 - Iter: 12, Total Loss: 0.3038, loss_single: 0.3038, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,612 - Iter: 13, Total Loss: 0.2985, loss_single: 0.2985, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,647 - Iter: 14, Total Loss: 0.2931, loss_single: 0.2931, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,682 - Iter: 15, Total Loss: 0.2875, loss_single: 0.2875, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,718 - Iter: 16, Total Loss: 0.2823, loss_single: 0.2823, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:36:24,753 - Iter: 17, Total Loss: 0.2770, loss_single: 0.2770, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,789 - Iter: 18, Total Loss: 0.2723, loss_single: 0.2723, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,824 - Iter: 19, Total Loss: 0.2683, loss_single: 0.2683, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,859 - Iter: 20, Total Loss: 0.2635, loss_single: 0.2635, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,895 - Iter: 21, Total Loss: 0.2585, loss_single: 0.2585, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,930 - Iter: 22, Total Loss: 0.2539, loss_single: 0.2539, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:24,966 - Iter: 23, Total Loss: 0.2492, loss_single: 0.2492, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,001 - Iter: 24, Total Loss: 0.2438, loss_single: 0.2438, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,036 - Iter: 25, Total Loss: 0.2390, loss_single: 0.2390, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,071 - Iter: 26, Total Loss: 0.2349, loss_single: 0.2349, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,107 - Iter: 27, Total Loss: 0.2305, loss_single: 0.2305, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,142 - Iter: 28, Total Loss: 0.2262, loss_single: 0.2262, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,177 - Iter: 29, Total Loss: 0.2220, loss_single: 0.2220, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,212 - Iter: 30, Total Loss: 0.2181, loss_single: 0.2181, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,247 - Iter: 31, Total Loss: 0.2140, loss_single: 0.2140, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,283 - Iter: 32, Total Loss: 0.2102, loss_single: 0.2102, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,318 - Iter: 33, Total Loss: 0.2069, loss_single: 0.2069, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,353 - Iter: 34, Total Loss: 0.2035, loss_single: 0.2035, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,389 - Iter: 35, Total Loss: 0.1999, loss_single: 0.1999, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,424 - Iter: 36, Total Loss: 0.1966, loss_single: 0.1966, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,460 - Iter: 37, Total Loss: 0.1933, loss_single: 0.1933, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,496 - Iter: 38, Total Loss: 0.1899, loss_single: 0.1899, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:36:25,531 - Iter: 39, Total Loss: 0.1868, loss_single: 0.1868, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,566 - Iter: 40, Total Loss: 0.1840, loss_single: 0.1840, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,603 - Iter: 41, Total Loss: 0.1812, loss_single: 0.1812, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,638 - Iter: 42, Total Loss: 0.1787, loss_single: 0.1787, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,674 - Iter: 43, Total Loss: 0.1765, loss_single: 0.1765, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,709 - Iter: 44, Total Loss: 0.1743, loss_single: 0.1743, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,786 - Iter: 45, Total Loss: 0.1722, loss_single: 0.1722, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,822 - Iter: 46, Total Loss: 0.1698, loss_single: 0.1698, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,857 - Iter: 47, Total Loss: 0.1675, loss_single: 0.1675, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,892 - Iter: 48, Total Loss: 0.1657, loss_single: 0.1657, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,928 - Iter: 49, Total Loss: 0.1643, loss_single: 0.1643, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,964 - Iter: 50, Total Loss: 0.1624, loss_single: 0.1624, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:25,998 - Iter: 51, Total Loss: 0.1604, loss_single: 0.1604, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:26,034 - Iter: 52, Total Loss: 0.1587, loss_single: 0.1587, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,070 - Iter: 53, Total Loss: 0.1571, loss_single: 0.1571, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,105 - Iter: 54, Total Loss: 0.1557, loss_single: 0.1557, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,141 - Iter: 55, Total Loss: 0.1545, loss_single: 0.1545, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,176 - Iter: 56, Total Loss: 0.1533, loss_single: 0.1533, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,212 - Iter: 57, Total Loss: 0.1518, loss_single: 0.1518, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,247 - Iter: 58, Total Loss: 0.1503, loss_single: 0.1503, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,282 - Iter: 59, Total Loss: 0.1489, loss_single: 0.1489, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,318 - Iter: 60, Total Loss: 0.1475, loss_single: 0.1475, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,353 - Iter: 61, Total Loss: 0.1463, loss_single: 0.1463, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,388 - Iter: 62, Total Loss: 0.1453, loss_single: 0.1453, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,424 - Iter: 63, Total Loss: 0.1444, loss_single: 0.1444, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,459 - Iter: 64, Total Loss: 0.1435, loss_single: 0.1435, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,494 - Iter: 65, Total Loss: 0.1426, loss_single: 0.1426, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,530 - Iter: 66, Total Loss: 0.1417, loss_single: 0.1417, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,565 - Iter: 67, Total Loss: 0.1409, loss_single: 0.1409, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,600 - Iter: 68, Total Loss: 0.1402, loss_single: 0.1402, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,635 - Iter: 69, Total Loss: 0.1394, loss_single: 0.1394, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,671 - Iter: 70, Total Loss: 0.1386, loss_single: 0.1386, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,706 - Iter: 71, Total Loss: 0.1380, loss_single: 0.1380, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,741 - Iter: 72, Total Loss: 0.1376, loss_single: 0.1376, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,776 - Iter: 73, Total Loss: 0.1370, loss_single: 0.1370, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,811 - Iter: 74, Total Loss: 0.1365, loss_single: 0.1365, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,846 - Iter: 75, Total Loss: 0.1360, loss_single: 0.1360, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,882 - Iter: 76, Total Loss: 0.1355, loss_single: 0.1355, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,917 - Iter: 77, Total Loss: 0.1351, loss_single: 0.1351, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,953 - Iter: 78, Total Loss: 0.1346, loss_single: 0.1346, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:26,988 - Iter: 79, Total Loss: 0.1342, loss_single: 0.1342, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,023 - Iter: 80, Total Loss: 0.1337, loss_single: 0.1337, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,059 - Iter: 81, Total Loss: 0.1335, loss_single: 0.1335, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,094 - Iter: 82, Total Loss: 0.1332, loss_single: 0.1332, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,129 - Iter: 83, Total Loss: 0.1330, loss_single: 0.1330, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,164 - Iter: 84, Total Loss: 0.1328, loss_single: 0.1328, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,199 - Iter: 85, Total Loss: 0.1325, loss_single: 0.1325, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,234 - Iter: 86, Total Loss: 0.1324, loss_single: 0.1324, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,269 - Iter: 87, Total Loss: 0.1322, loss_single: 0.1322, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:27,304 - Iter: 88, Total Loss: 0.1320, loss_single: 0.1320, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:27,339 - Iter: 89, Total Loss: 0.1318, loss_single: 0.1318, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,374 - Iter: 90, Total Loss: 0.1317, loss_single: 0.1317, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,411 - Iter: 91, Total Loss: 0.1315, loss_single: 0.1315, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:36:27,446 - Iter: 92, Total Loss: 0.1313, loss_single: 0.1313, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,481 - Iter: 93, Total Loss: 0.1312, loss_single: 0.1312, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:27,516 - Iter: 94, Total Loss: 0.1310, loss_single: 0.1310, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,551 - Iter: 95, Total Loss: 0.1309, loss_single: 0.1309, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,586 - Iter: 96, Total Loss: 0.1308, loss_single: 0.1308, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,622 - Iter: 97, Total Loss: 0.1307, loss_single: 0.1307, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,657 - Iter: 98, Total Loss: 0.1306, loss_single: 0.1306, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,692 - Iter: 99, Total Loss: 0.1305, loss_single: 0.1305, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,727 - Iter: 100, Total Loss: 0.1304, loss_single: 0.1304, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:27,838 - Saving summary figures for iter 100
+2026-06-24 05:36:31,613 - ### Finished 100 iterations, averaged iter_t = 0.03893 with std = 0.040 ###
+2026-06-24 05:36:31,613 -  
+2026-06-24 05:36:31,613 - ### The PtyRADSolver is finished in 9.429 sec ###
+2026-06-24 05:36:31,613 -  

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000009/c30_small__simulation_data1__sample_000009_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-267_C12_18.2_phi12_230_C21_758_phi21_299_C23_882_phi23_3.72_C30_1.33e+05_C32_9.9.88e+05_phi56_333/model_iter0100.hdf5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4939f980582ee608b14e7875dd0cc9e97b9ca99d229ceac8187c7ed7ead55ecb
+size 1619112

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000009/c30_small__simulation_data1__sample_000009_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-267_C12_18.2_phi12_230_C21_758_phi21_299_C23_882_phi23_3.72_C30_1.33e+05_C32_9.9.88e+05_phi56_333/simulation_data1__sample_000009.yaml

@@ -0,0 +1,140 @@
+init_params:
+  probe_kv: 300.0
+  probe_conv_angle: 22.653152465820312
+  probe_aberrations:
+    C10: -266.6025390625
+    C12: 18.172391891479492
+    C21: 757.7008056640625
+    C23: 881.6241455078125
+    C30: 133419.65625
+    C32: 9907.30859375
+    C34: -1121.2396240234375
+    C41: -363432.125
+    C43: -177851.125
+    C45: 41571.30859375
+    C50: 269112576.0
+    C52: 18410188.0
+    C54: 17605530.0
+    C56: -788044.6875
+    phi12: 229.65743370905165
+    phi21: 298.59016168092154
+    phi23: 3.719704373402842
+    phi32: 59.121017795174
+    phi34: 318.9114497825702
+    phi41: 242.0133825213881
+    phi43: 339.069714929109
+    phi45: 55.73730451430925
+    phi52: 311.7001633776274
+    phi54: 199.50210754231034
+    phi56: 333.4743419987645
+  meas_Npix: 128
+  pos_N_scan_slow: 16
+  pos_N_scan_fast: 16
+  pos_scan_step_size: 0.43370673060417175
+  probe_pmode_max: 6
+  obj_Nlayer: 1
+  obj_slice_thickness: 20.0
+  meas_permute: null
+  meas_reshape:
+  - 256
+  - 128
+  - 128
+  meas_flipT:
+  - 0
+  - 0
+  - 0
+  meas_crop: null
+  meas_pad: null
+  meas_resample: null
+  pos_scan_affine: null
+  meas_params:
+    path: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000009_measurement.h5
+    key: measurement
+  random_seed: 20261853
+  obj_omode_max: 1
+  meas_calibration:
+    mode: kMax
+    value: 2.5
+model_params:
+  detector_blur_std: null
+  update_params:
+    obja:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    objp:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    probe:
+      start_iter: 1
+      lr: 0.0001
+      end_iter: null
+    probe_pos_shifts:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+loss_params:
+  loss_single:
+    state: true
+    weight: 1.0
+    dp_pow: 0.5
+  loss_poissn:
+    state: false
+    weight: 1.0
+    dp_pow: 1.0
+    eps: 1.0e-06
+  loss_sparse:
+    state: false
+    weight: 0.1
+    ln_order: 1
+constraint_params:
+  ortho_pmode:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  fix_probe_int:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  obj_rblur:
+    start_iter: null
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 0.4
+  obj_zblur:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 1
+  obja_thresh:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    relax: 0
+    thresh:
+    - 0.96
+    - 1.04
+  objp_postiv:
+    start_iter: null
+    step: 1
+    end_iter: null
+    relax: 0
+recon_params:
+  NITER: 100
+  BATCH_SIZE:
+    size: 32
+    grad_accumulation: 1
+  SAVE_ITERS: 100
+  output_dir: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000009
+  recon_dir_affixes:
+  - default
+  prefix: c30_small__simulation_data1__sample_000009
+  postfix: ''
+  compiler_configs:
+    enable: false
+  prefix_time: false

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000010/c30_small__simulation_data1__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_6.82_C12_-46.2_phi12_189_C21_698_phi21_28_C23_-991_phi23_186_C30_2.87e+06_C32_-7..72e+07_phi56_117/20260624_053637_ptyrad_log.txt

@@ -0,0 +1,417 @@
+2026-06-24 05:36:36,278 - ### PtyRAD LoggingManager configuration ###
+2026-06-24 05:36:36,278 - log_file       = 'ptyrad_log.txt'. If log_file = None, no log file will be created.
+2026-06-24 05:36:36,278 - log_dir        = 'auto'. If log_dir = 'auto', then log will be saved to `output_path` or 'logs/'.
+2026-06-24 05:36:36,278 - flush_file     = True. Automatically set to True if `log_file is not None`
+2026-06-24 05:36:36,278 - prefix_time    = datetime. If true, preset strings ('date', 'time', 'datetime'), or a string of time format, a datetime str is prefixed to the `log_file`.
+2026-06-24 05:36:36,278 - prefix_jobid   = '0'. If not 0, it'll be prefixed to the log file. This is used for hypertune mode with multiple GPUs.
+2026-06-24 05:36:36,278 - append_to_file = True. If true, logs will be appended to the existing file. If false, the log file will be overwritten.
+2026-06-24 05:36:36,278 - show_timestamp = True. If true, the printed information will contain a timestamp.
+2026-06-24 05:36:36,278 -  
+2026-06-24 05:36:36,385 - ### Initializing HuggingFace accelerator ###
+2026-06-24 05:36:36,385 - Accelerator.distributed_type = DistributedType.NO
+2026-06-24 05:36:36,385 - Accelerator.num_process      = 1
+2026-06-24 05:36:36,385 - Accelerator.mixed_precision  = no
+2026-06-24 05:36:36,409 - 'accelerate' is available but NOT using distributed mode or mixed precision
+2026-06-24 05:36:36,409 - If you want to utilize 'accelerate' for multiGPU or mixed precision, 
+2026-06-24 05:36:36,409 - Run `accelerate launch --multi_gpu --num_processes=2 --mixed_precision='no' -m ptyrad run <PTYRAD_ARGUMENTS> --gpuid 'acc'` in your terminal
+2026-06-24 05:36:36,409 -  
+2026-06-24 05:36:36,409 - ### System information ###
+2026-06-24 05:36:36,409 - Platform: Linux-4.18.0-553.69.1.el8_10.x86_64-x86_64-with-glibc2.28
+2026-06-24 05:36:36,409 - Operating System: Linux 4.18.0-553.69.1.el8_10.x86_64
+2026-06-24 05:36:36,409 - OS Version: #1 SMP Thu Aug 7 18:10:00 EDT 2025
+2026-06-24 05:36:36,409 - Machine: x86_64
+2026-06-24 05:36:36,409 - Processor: x86_64
+2026-06-24 05:36:36,409 - Available CPU cores: 8
+2026-06-24 05:36:36,410 - SLURM-Allocated Total Memory: 78.12 GB
+2026-06-24 05:36:36,410 -  
+2026-06-24 05:36:36,410 - ### GPU information ###
+2026-06-24 05:36:36,410 - CUDA Available: True
+2026-06-24 05:36:36,410 - CUDA Version: 13.0
+2026-06-24 05:36:36,410 - Available CUDA GPUs: ['NVIDIA A100 80GB PCIe']
+2026-06-24 05:36:36,410 - CUDA Compute Capability: ['8.0']
+2026-06-24 05:36:36,410 -   INFO: For torch.compile with Triton, you'll need CUDA GPU with Compute Capability >= 7.0.
+2026-06-24 05:36:36,410 -         In addition, Triton does not directly support Windows.
+2026-06-24 05:36:36,410 -         For Windows users, please follow the instruction and download `triton-windows` from https://github.com/woct0rdho/triton-windows.
+2026-06-24 05:36:36,426 - MIG (Multi-Instance GPU) mode = False
+2026-06-24 05:36:36,426 -   INFO: MIG splits a physical GPU into multiple GPU slices, but multiGPU does not support these MIG slices.
+2026-06-24 05:36:36,426 -         In addition, multiGPU is currently only available on Linux due to the limited NCCL support.
+2026-06-24 05:36:36,426 -       -> If you're doing normal reconstruction/hypertune, you can safely ignore this.
+2026-06-24 05:36:36,426 -       -> If you want to do multiGPU, you must provide multiple 'full' GPUs that are not in MIG mode.
+2026-06-24 05:36:36,426 -  
+2026-06-24 05:36:36,426 - ### Python information ###
+2026-06-24 05:36:36,428 - Python Executable: /home/tnguye11/anaconda3/envs/ptyrad/bin/python3.12
+2026-06-24 05:36:36,428 - Python Version: 3.12.13 | packaged by Anaconda, Inc. | (main, Mar 19 2026, 20:20:58) [GCC 14.3.0]
+2026-06-24 05:36:36,428 -  
+2026-06-24 05:36:36,428 - ### Packages information ###
+2026-06-24 05:36:36,428 - Numpy Version (metadata): 2.4.6
+2026-06-24 05:36:36,429 - PyTorch Version (metadata): 2.12.0
+2026-06-24 05:36:36,430 - Optuna Version (metadata): 4.9.0
+2026-06-24 05:36:36,431 - Accelerate Version (metadata): 1.13.0
+2026-06-24 05:36:36,431 - PtyRAD Version (ptyrad/__init__.py): 1.0.0
+2026-06-24 05:36:36,431 - PtyRAD is located at: /home/tnguye11/anaconda3/envs/ptyrad/lib/python3.12/site-packages/ptyrad/__init__.py
+2026-06-24 05:36:36,431 -  
+2026-06-24 05:36:36,431 - ### Loading params file ###
+2026-06-24 05:36:36,431 - params_path = /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/params/c30_small/simulation_data1__sample_000010.yaml
+2026-06-24 05:36:36,438 - validate = True: Filling defaults and validating the params file...
+2026-06-24 05:36:36,439 - Success! Params file validated and defaults applied.
+2026-06-24 05:36:36,439 -  
+2026-06-24 05:36:36,439 - ### Setting GPU Device ###
+2026-06-24 05:36:36,439 - Selected GPU device: cuda:0 (NVIDIA A100 80GB PCIe)
+2026-06-24 05:36:36,439 -  
+2026-06-24 05:36:36,439 - Random seed: 20261854 provided by params file
+2026-06-24 05:36:36,439 - ### Initializing Initializer ###
+2026-06-24 05:36:36,440 - init_params are displayed below:
+2026-06-24 05:36:36,440 -   random_seed: 20261854
+2026-06-24 05:36:36,440 -   probe_illum_type: electron
+2026-06-24 05:36:36,440 -   probe_kv: 60.0
+2026-06-24 05:36:36,440 -   probe_conv_angle: 12.819171905517578
+2026-06-24 05:36:36,440 -   probe_aberrations: {'C10': 6.817, 'C12': -46.184, 'phi12': 189.019, 'C21': 698.398, 'phi21': 27.966, 'C23': -991.115, 'phi23': 185.988, 'C30': 2867064.25, 'C32': -7273.422, 'phi32': 270.523, 'C34': 4690.002, 'phi34': 102.292, 'C41': -229145.469, 'phi41': 278.888, 'C43': 67091.086, 'phi43': 347.245, 'C45': 209215.375, 'phi45': 265.808, 'C50': 364242912.0, 'C52': 1600665.625, 'phi52': 83.851, 'C54': 10428370.0, 'phi54': 185.356, 'C56': 17220394.0, 'phi56': 117.476}
+2026-06-24 05:36:36,440 -   beam_kev: None
+2026-06-24 05:36:36,440 -   probe_dRn: None
+2026-06-24 05:36:36,440 -   probe_Rn: None
+2026-06-24 05:36:36,440 -   probe_D_H: None
+2026-06-24 05:36:36,440 -   probe_D_FZP: None
+2026-06-24 05:36:36,440 -   probe_Ls: None
+2026-06-24 05:36:36,440 -   meas_Npix: 128
+2026-06-24 05:36:36,440 -   pos_N_scans: 256
+2026-06-24 05:36:36,440 -   pos_N_scan_slow: 16
+2026-06-24 05:36:36,440 -   pos_N_scan_fast: 16
+2026-06-24 05:36:36,440 -   pos_scan_step_size: 0.36360010504722595
+2026-06-24 05:36:36,440 -   meas_calibration: {'mode': 'kMax', 'value': 2.5}
+2026-06-24 05:36:36,440 -   probe_pmode_max: 6
+2026-06-24 05:36:36,440 -   probe_pmode_init_pows: [0.02]
+2026-06-24 05:36:36,440 -   obj_omode_max: 1
+2026-06-24 05:36:36,440 -   obj_omode_init_occu: {'occu_type': 'uniform', 'init_occu': None}
+2026-06-24 05:36:36,440 -   obj_Nlayer: 1
+2026-06-24 05:36:36,440 -   obj_slice_thickness: 20.0
+2026-06-24 05:36:36,440 -   simu_Npix: None
+2026-06-24 05:36:36,440 -   simu_match_mode: None
+2026-06-24 05:36:36,440 -   meas_permute: None
+2026-06-24 05:36:36,440 -   meas_reshape: [256, 128, 128]
+2026-06-24 05:36:36,440 -   meas_flipT: [0, 0, 0]
+2026-06-24 05:36:36,440 -   meas_crop: None
+2026-06-24 05:36:36,440 -   meas_pad: None
+2026-06-24 05:36:36,440 -   meas_resample: None
+2026-06-24 05:36:36,440 -   meas_add_source_size: None
+2026-06-24 05:36:36,441 -   meas_add_detector_blur: None
+2026-06-24 05:36:36,441 -   meas_remove_neg_values: {'mode': 'clip_neg', 'value': None, 'force': False}
+2026-06-24 05:36:36,441 -   meas_normalization: {'mode': 'max_at_one', 'value': None}
+2026-06-24 05:36:36,441 -   meas_add_poisson_noise: None
+2026-06-24 05:36:36,441 -   meas_export: None
+2026-06-24 05:36:36,441 -   probe_permute: None
+2026-06-24 05:36:36,441 -   probe_z_shift: None
+2026-06-24 05:36:36,441 -   probe_normalization: {'mode': 'mean_total_ints', 'value': None}
+2026-06-24 05:36:36,441 -   pos_scan_flipT: None
+2026-06-24 05:36:36,441 -   pos_scan_affine: None
+2026-06-24 05:36:36,441 -   pos_scan_rand_std: 0.15
+2026-06-24 05:36:36,441 -   obj_z_crop: None
+2026-06-24 05:36:36,441 -   obj_z_pad: None
+2026-06-24 05:36:36,441 -   obj_z_resample: None
+2026-06-24 05:36:36,441 -   meas_source: file
+2026-06-24 05:36:36,441 -   meas_params: {'path': '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000010_measurement.h5', 'key': 'measurement', 'shape': None, 'offset': None, 'gap': None, 'selection': None, 'zarr_kwargs': None}
+2026-06-24 05:36:36,441 -   probe_source: simu
+2026-06-24 05:36:36,441 -   probe_params: None
+2026-06-24 05:36:36,441 -   pos_source: simu
+2026-06-24 05:36:36,441 -   pos_params: None
+2026-06-24 05:36:36,441 -   obj_source: simu
+2026-06-24 05:36:36,441 -   obj_params: None
+2026-06-24 05:36:36,441 -   tilt_source: simu
+2026-06-24 05:36:36,441 -   tilt_params: {'tilt_type': 'all', 'init_tilts': [[0, 0]]}
+2026-06-24 05:36:36,441 -  
+2026-06-24 05:36:36,441 - ### Initializing cache ###
+2026-06-24 05:36:36,441 - use_cached_obj   = False
+2026-06-24 05:36:36,441 - use_cached_probe = False
+2026-06-24 05:36:36,441 - use_cached_pos   = False
+2026-06-24 05:36:36,441 -  
+2026-06-24 05:36:36,441 - ### Initializing measurements ###
+2026-06-24 05:36:36,441 - Loading measurements from source = 'file'
+2026-06-24 05:36:36,441 - Detected measurement file type = '.h5'
+2026-06-24 05:36:36,520 - Original measurements dtype is float32, casting to float32 (single precision) for computational efficiency.
+2026-06-24 05:36:36,520 - Imported meausrements shape / dtype = (16, 16, 128, 128), dtype = float32
+2026-06-24 05:36:36,522 - Imported meausrements int. statistics (min, mean, max) = (0.0000, 0.0001, 0.0085)
+2026-06-24 05:36:36,522 - Reshaping measurements to shape = [256, 128, 128]
+2026-06-24 05:36:36,522 - Flipping measurements with [flipud, fliplr, transpose] = [0, 0, 0]
+2026-06-24 05:36:36,523 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:36:36,523 - Normalizing measurements with mode = 'max_at_one' and value = 'None'
+2026-06-24 05:36:36,523 - Normalizing by max of the 2D mean pattern intensity: 0.0066038887
+2026-06-24 05:36:36,524 - meausrements shape / dtype = (256, 128, 128), dtype = float32
+2026-06-24 05:36:36,525 - meausrements int. statistics (min, mean, max) = (0.0000, 0.0092, 1.2810)
+2026-06-24 05:36:36,526 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:36:36,527 - Pattern total int. statistics (min, mean, max)        = (150.1019, 151.0493, 152.2493), with min/max = 98.6%
+2026-06-24 05:36:36,528 - Global meausrements int. statistics (min, mean, max)  = (0.0000, 0.0092, 1.2810)
+2026-06-24 05:36:36,528 - measurements                              (N, Ky, Kx) = float32, (256, 128, 128)
+2026-06-24 05:36:36,528 -  
+2026-06-24 05:36:36,529 - ### Setting up calibration ###
+2026-06-24 05:36:36,529 - meas_calibration mode = 'kMax', value = 2.5
+2026-06-24 05:36:36,529 - Using loaded raw averaged measurement (before crop/pad/resample) to fit RBF as a part of the meas calibration
+2026-06-24 05:36:36,529 - Radius of fitted bright field disk (RBF) = 6.68 px with meas_Npix = 128
+2026-06-24 05:36:36,529 - Suggested probe_mask_k radius (RBF*2/Npix) > 0.1043
+2026-06-24 05:36:36,529 - Fitting raw averaged measurement with center, radius, and Gaussian blur std as a sanity check
+2026-06-24 05:36:36,529 - Note that the fitted Gaussian blur std (detector blur) would be affected by overlapping Bragg disks
+2026-06-24 05:36:36,752 - Initial guess: center=(63.63, 63.64), radius=6.68, Gaussian blur std=0.50
+2026-06-24 05:36:36,764 - Final fit: center=(63.63, 63.64), radius=6.68, Gaussian blur std=0.66
+2026-06-24 05:36:36,764 - Using init_params, the inferred RBF (conv_angle / 1e3 * Npix * dx / wavelength) = 6.74 px with Npix = 128
+2026-06-24 05:36:36,764 - dx (real space pixel size of probe and object) set to 0.2000 Ang with Npix = 128
+2026-06-24 05:36:36,764 -  
+2026-06-24 05:36:36,764 - ### Setting init_variables dict ###
+2026-06-24 05:36:36,764 - Derived values given input init_params:
+2026-06-24 05:36:36,764 -   kv          = 60.0 kV
+2026-06-24 05:36:36,764 -   wavelength  = 0.0487 Ang
+2026-06-24 05:36:36,764 -   conv_angle  = 12.819171905517578 mrad
+2026-06-24 05:36:36,764 -   Npix        = 128 px
+2026-06-24 05:36:36,764 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:36:36,764 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:36:36,764 -   da          = 1.9008 mrad
+2026-06-24 05:36:36,764 -   angleMax    = 121.6515 mrad
+2026-06-24 05:36:36,764 -   RBF         = 6.7441 px (Inferred from the given calibration, NOT necessarily from the loaded measurement data)
+2026-06-24 05:36:36,764 -   n_alpha     = 9.4898 (# conv_angle)
+2026-06-24 05:36:36,764 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:36:36,764 -   Rayleigh-limited resolution  = 2.3155 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:36:36,764 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:36:36,764 -  
+2026-06-24 05:36:36,764 - ### Initializing probe ###
+2026-06-24 05:36:36,764 - Loading probe from source = 'simu'
+2026-06-24 05:36:36,764 - Using experimental parameters specified by 'init_params' for initial probe simulation.
+2026-06-24 05:36:36,765 - Start simulating STEM probe
+2026-06-24 05:36:36,765 -   kv          = 60.0 kV
+2026-06-24 05:36:36,765 -   wavelength  = 0.0487 Ang
+2026-06-24 05:36:36,765 -   conv_angle  = 12.819171905517578 mrad
+2026-06-24 05:36:36,765 -   Npix        = 128 px
+2026-06-24 05:36:36,765 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:36:36,765 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:36:36,765 -   alpha_max   = 121.6515 mrad
+2026-06-24 05:36:36,765 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:36:36,765 -   Rayleigh-limited resolution  = 2.3155 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:36:36,765 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:36:36,765 - Krivanek Haider     Magnitude        Angle (°)  Description
+2026-06-24 05:36:36,765 - ------------------------------------------------------------------------------------
+2026-06-24 05:36:36,765 - C1,0     C1         6.8170           -          Defocus (C10 = -df)
+2026-06-24 05:36:36,765 - C1,2     A1         -46.1840         189.02     2-fold astigmatism
+2026-06-24 05:36:36,765 - C2,1     3*B2       698.3980         27.97      Axial coma
+2026-06-24 05:36:36,765 - C2,3     A2         -991.1150        185.99     3-fold astigmatism
+2026-06-24 05:36:36,765 - C3,0     C3         2867064.2500     -          Spherical aberration
+2026-06-24 05:36:36,765 - C3,2     4*S3       -7273.4220       270.52     Axial star aberration
+2026-06-24 05:36:36,765 - C3,4     A3         4690.0020        102.29     4-fold astigmatism
+2026-06-24 05:36:36,765 - C4,1     4*B4       -229145.4690     278.89     Axial coma(4th)
+2026-06-24 05:36:36,765 - C4,3     4*D4       67091.0860       347.25     3-lobe aberration
+2026-06-24 05:36:36,765 - C4,5     A4         209215.3750      265.81     5-fold astigmatism
+2026-06-24 05:36:36,765 - C5,0     C5         364242912.0000   -          Spherical aberration (5th)
+2026-06-24 05:36:36,765 - C5,2     6*S5       1600665.6250     83.85      Axial star aberration(5th)
+2026-06-24 05:36:36,765 - C5,4     6*R5       10428370.0000    185.36     4-lobe aberration
+2026-06-24 05:36:36,765 - C5,6     A5         17220394.0000    117.48     6-fold astigmatism
+2026-06-24 05:36:36,768 - Loaded probe shape = (1, 128, 128), dtype = complex128
+2026-06-24 05:36:36,768 - pmode_now: 1 and pmode_max: 6, padding the pmode.
+2026-06-24 05:36:36,768 - Creating 5 new probe modes from the major mode
+2026-06-24 05:36:36,768 - Start making mixed-state STEM probe with 6 incoherent probe modes
+2026-06-24 05:36:36,960 - Relative power of probe modes = [0.9  0.02 0.02 0.02 0.02 0.02]
+2026-06-24 05:36:36,961 - Orthogonalizing 6 pmodes
+2026-06-24 05:36:36,964 - Sorting 6 pmodes by their intensities
+2026-06-24 05:36:36,964 - Normalizing probe intensity with mode = 'mean_total_ints' and value = 'None'
+2026-06-24 05:36:36,965 - sum(|probe_data|**2) = 151.05, while meas_total_ints (min, mean, max) = (150.1019, 151.0493, 152.2493)
+2026-06-24 05:36:36,965 - probe                         (pmode, Ny, Nx) = complex64, (6, 128, 128)
+2026-06-24 05:36:36,965 -  
+2026-06-24 05:36:36,965 - ### Initializing probe positions ###
+2026-06-24 05:36:36,965 - Loading probe positions from source = 'simu'
+2026-06-24 05:36:36,965 - Using experimental parameters specified by 'init_params' (dx, scan_step size, N_scan_slow, N_scan_fast) for initial position simulation.
+2026-06-24 05:36:36,965 - Simulating probe positions with dx = 0.2000, scan_step_size = 0.3636, N_scan_fast = 16, N_scan_slow = 16
+2026-06-24 05:36:36,965 - Applying Gaussian distributed random displacement with std = 0.15 px to scan positions
+2026-06-24 05:36:36,967 - crop_pos                                (N,2) = int16, (256, 2)
+2026-06-24 05:36:36,967 - crop_pos 1st and last px coords (y,x)         = ([16, 16], [42, 43])
+2026-06-24 05:36:36,968 - crop_pos extent (Ang)                         = [5.6 5.6]
+2026-06-24 05:36:36,968 - probe_pos_shifts                        (N,2) = float32, (256, 2)
+2026-06-24 05:36:36,968 -  
+2026-06-24 05:36:36,968 - ### Initializing object ###
+2026-06-24 05:36:36,968 - Loading object from source = 'simu'
+2026-06-24 05:36:36,968 - Using experimental parameters specified by 'init_params' for initial object simulation.
+2026-06-24 05:36:36,969 - omode_now: 1 and omode_max: 1, leaving the omode unchanged.
+2026-06-24 05:36:36,969 - object                    (omode, Nz, Ny, Nx) = complex64, (1, 1, 187, 187)
+2026-06-24 05:36:36,969 - object extent                 (Z, Y, X) (Ang) = [20.  37.4 37.4]
+2026-06-24 05:36:36,969 -  
+2026-06-24 05:36:36,969 - ### Initializing omode_occu from 'uniform' ###
+2026-06-24 05:36:36,969 - omode_occu                            (omode) = float32, (1,)
+2026-06-24 05:36:36,970 -  
+2026-06-24 05:36:36,970 - ### Initializing H (Fresnel propagator) ###
+2026-06-24 05:36:36,970 - Calculating H with probe_shape = (128, 128) px, dx = 0.2000 Ang, slice_thickness = 20.0000 Ang, lambd = 0.0487 Ang
+2026-06-24 05:36:36,970 - H                                    (Ky, Kx) = complex64, (128, 128)
+2026-06-24 05:36:36,970 -  
+2026-06-24 05:36:36,970 - ### Initializing obj tilts from = 'simu' ###
+2026-06-24 05:36:36,970 - Initialized obj_tilts with init_tilts = [[0, 0]] (theta_y, theta_x) mrad
+2026-06-24 05:36:36,970 - obj_tilts                              (N, 2) = float32, (1, 2)
+2026-06-24 05:36:36,970 -  
+2026-06-24 05:36:36,970 - ### Checking consistency between input params with the initialized variables ###
+2026-06-24 05:36:36,970 - meas_Npix, simu_Npix, DP measurements, probe, and H shapes are consistent as '128'
+2026-06-24 05:36:36,971 - N_scans, len(meas), N_scan_slow*N_scan_fast, len(crop_pos), and len(probe_pos_shifts) are consistent as '256'
+2026-06-24 05:36:36,971 - obj.shape[0] is consistent with len(omode_occu) as '1'
+2026-06-24 05:36:36,971 - obj.shape[1] is consistent with Nlayer as '1'
+2026-06-24 05:36:36,971 - crop positions (yx_min=[15 15], yx_max=[171 171]) are well contained inside object canvas (Ny,Nx) = (187, 187).
+2026-06-24 05:36:36,971 - obj_tilts is consistent with either 1 or N_scans
+2026-06-24 05:36:36,971 - Pass the consistency check of initialized variables, initialization is done!
+2026-06-24 05:36:36,971 -  
+2026-06-24 05:36:36,971 - ### Collecting reconstruction provenance ###
+2026-06-24 05:36:36,971 - Reconstruction provenance is collected and initialized.
+2026-06-24 05:36:36,971 -  
+2026-06-24 05:36:36,971 - ### Initializing loss function ###
+2026-06-24 05:36:36,971 - Active loss types:
+2026-06-24 05:36:36,971 -   loss_single : {'state': True, 'weight': 1.0, 'dp_pow': 0.5}
+2026-06-24 05:36:36,971 -  
+2026-06-24 05:36:36,971 - ### Initializing constraint function ###
+2026-06-24 05:36:36,971 - Active constraint types:
+2026-06-24 05:36:36,971 -   ortho_pmode   : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:36:36,971 -   fix_probe_int : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:36:36,971 -   obj_zblur     : {'start_iter': 1, 'step': 1, 'end_iter': None, 'obj_type': 'both', 'kernel_size': 5, 'std': 1.0}
+2026-06-24 05:36:36,971 -   obja_thresh   : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0, 'thresh': [0.96, 1.04]}
+2026-06-24 05:36:36,971 -   pos_recenter  : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0}
+2026-06-24 05:36:36,972 -  
+2026-06-24 05:36:36,972 - ### Done initializing PtyRADSolver ###
+2026-06-24 05:36:36,972 -  
+2026-06-24 05:36:37,095 - ### Starting the PtyRADSolver in reconstruct mode ###
+2026-06-24 05:36:37,096 -  
+2026-06-24 05:36:37,096 - ### Initializing PtychoModel model ###
+2026-06-24 05:36:37,167 - ### PtychoModel optimizable variables ###
+2026-06-24 05:36:37,167 - obja            : torch.Size([1, 1, 187, 187])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:36:37,167 - objp            : torch.Size([1, 1, 187, 187])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:36:37,167 - obj_tilts       : torch.Size([1, 2])              , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:36:37,167 - slice_thickness : torch.Size([])                  , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:36:37,167 - probe           : torch.Size([6, 128, 128, 2])    , torch.float32   , device:cuda:0, grad:True , lr:1e-04
+2026-06-24 05:36:37,167 - probe_pos_shifts: torch.Size([256, 2])            , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:36:37,167 -  
+2026-06-24 05:36:37,167 - ### Optimizable variables statitsics ###
+2026-06-24 05:36:37,167 - Total measurement values  : 4,194,304
+2026-06-24 05:36:37,167 - Total optimizing variables: 267,058
+2026-06-24 05:36:37,167 - Overdetermined ratio      : 15.71
+2026-06-24 05:36:37,167 -  
+2026-06-24 05:36:37,167 - ### Model behavior ###
+2026-06-24 05:36:37,167 - Tilt propagator           : False
+2026-06-24 05:36:37,168 - Change slice thickness    : False
+2026-06-24 05:36:37,168 - Detector blur             : False
+2026-06-24 05:36:37,168 - Preload data              : True
+2026-06-24 05:36:37,168 - On-the-fly meas padding   : False
+2026-06-24 05:36:37,168 - On-the-fly meas resample  : False
+2026-06-24 05:36:37,168 - On-the-fly simu match mode: None
+2026-06-24 05:36:37,168 -  
+2026-06-24 05:36:37,208 - ### Done initializing PtychoModel model ###
+2026-06-24 05:36:37,209 -  
+2026-06-24 05:36:37,209 - ### Creating PyTorch 'Adam' optimizer with configs = {} ###
+2026-06-24 05:36:37,209 -  
+2026-06-24 05:36:37,209 - ### Generating indices, batches, and output_path ###
+2026-06-24 05:36:37,210 - d90 = 65.000 px or 13.000 Ang
+2026-06-24 05:36:37,210 - Selecting indices with the 'full' mode 
+2026-06-24 05:36:37,678 - Generated 8 'random' groups of ~32 scan positions in 0.000 sec
+2026-06-24 05:36:37,751 - The effective batch size (i.e., how many probe positions are simultaneously used for 1 update of ptychographic parameters) is batch_size * grad_accumulation = 32 * 1 = 32
+2026-06-24 05:36:37,752 - Original recon_dir_affixes = ['default']
+2026-06-24 05:36:37,752 - Expanded recon_dir_affixes = ['indices', 'meas', 'batch', 'pmode', 'omode', 'nlayer', 'lr', 'model', 'constraint', 'loss', 'affine', 'tilt', 'aberrations']
+2026-06-24 05:36:37,758 - Path corrected for compatibility:
+2026-06-24 05:36:37,758 -   Original: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000010/c30_small__simulation_data1__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_6.82_C12_-46.2_phi12_189_C21_698_phi21_28_C23_-991_phi23_186_C30_2.87e+06_C32_-7.27e+03_phi32_271_C34_4.69e+03_phi34_102_C41_-2.29e+05_phi41_279_C43_6.71e+04_phi43_347_C45_2.09e+05_phi45_266_C50_3.64e+08_C52_1.6e+06_phi52_83.9_C54_1.04e+07_phi54_185_C56_1.72e+07_phi56_117
+2026-06-24 05:36:37,758 -   Corrected: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000010/c30_small__simulation_data1__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_6.82_C12_-46.2_phi12_189_C21_698_phi21_28_C23_-991_phi23_186_C30_2.87e+06_C32_-7..72e+07_phi56_117
+2026-06-24 05:36:37,759 - output_path = '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000010/c30_small__simulation_data1__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_6.82_C12_-46.2_phi12_189_C21_698_phi21_28_C23_-991_phi23_186_C30_2.87e+06_C32_-7..72e+07_phi56_117' is generated!
+2026-06-24 05:36:37,903 -  
+2026-06-24 05:36:37,904 - ### Log file is flushed (created) as /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000010/c30_small__simulation_data1__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_6.82_C12_-46.2_phi12_189_C21_698_phi21_28_C23_-991_phi23_186_C30_2.87e+06_C32_-7..72e+07_phi56_117/20260624_053637_ptyrad_log.txt ###
+2026-06-24 05:36:37,904 -  
+2026-06-24 05:36:37,905 - ### Creating ConvergenceMonitor with {'tensors': ['obja', 'objp', 'probe', 'probe_pos_shifts'], 'every_n_iters': None, 'percentile_range': [15.0, 85.0]} ### 
+2026-06-24 05:36:37,905 - ### Start the PtyRAD iterative ptycho reconstruction ###
+2026-06-24 05:36:37,905 - Setting up PyTorch compiler with {'fullgraph': False, 'dynamic': None, 'backend': 'inductor', 'mode': 'default', 'options': None, 'disable': True}
+2026-06-24 05:36:39,099 - Iter: 1, Total Loss: 1.2446, loss_single: 1.2446, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.436 sec
+2026-06-24 05:36:39,136 - Iter: 2, Total Loss: 1.1707, loss_single: 1.1707, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:36:39,172 - Iter: 3, Total Loss: 1.1140, loss_single: 1.1140, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,208 - Iter: 4, Total Loss: 1.0532, loss_single: 1.0532, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,243 - Iter: 5, Total Loss: 0.9869, loss_single: 0.9869, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,279 - Iter: 6, Total Loss: 0.9253, loss_single: 0.9253, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,313 - Iter: 7, Total Loss: 0.8804, loss_single: 0.8804, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:39,349 - Iter: 8, Total Loss: 0.8428, loss_single: 0.8428, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,385 - Iter: 9, Total Loss: 0.7982, loss_single: 0.7982, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,420 - Iter: 10, Total Loss: 0.7609, loss_single: 0.7609, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,455 - Iter: 11, Total Loss: 0.7257, loss_single: 0.7257, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,490 - Iter: 12, Total Loss: 0.6827, loss_single: 0.6827, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,525 - Iter: 13, Total Loss: 0.6416, loss_single: 0.6416, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,560 - Iter: 14, Total Loss: 0.6026, loss_single: 0.6026, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,596 - Iter: 15, Total Loss: 0.5681, loss_single: 0.5681, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,632 - Iter: 16, Total Loss: 0.5397, loss_single: 0.5397, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,667 - Iter: 17, Total Loss: 0.5132, loss_single: 0.5132, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,702 - Iter: 18, Total Loss: 0.4897, loss_single: 0.4897, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:39,738 - Iter: 19, Total Loss: 0.4690, loss_single: 0.4690, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,773 - Iter: 20, Total Loss: 0.4497, loss_single: 0.4497, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,809 - Iter: 21, Total Loss: 0.4322, loss_single: 0.4322, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,844 - Iter: 22, Total Loss: 0.4191, loss_single: 0.4191, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,880 - Iter: 23, Total Loss: 0.4101, loss_single: 0.4101, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,915 - Iter: 24, Total Loss: 0.4040, loss_single: 0.4040, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,950 - Iter: 25, Total Loss: 0.4004, loss_single: 0.4004, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:39,989 - Iter: 26, Total Loss: 0.3984, loss_single: 0.3984, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.038 sec
+2026-06-24 05:36:40,024 - Iter: 27, Total Loss: 0.3972, loss_single: 0.3972, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,060 - Iter: 28, Total Loss: 0.3963, loss_single: 0.3963, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,095 - Iter: 29, Total Loss: 0.3956, loss_single: 0.3956, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,131 - Iter: 30, Total Loss: 0.3950, loss_single: 0.3950, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,166 - Iter: 31, Total Loss: 0.3945, loss_single: 0.3945, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,201 - Iter: 32, Total Loss: 0.3940, loss_single: 0.3940, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:40,236 - Iter: 33, Total Loss: 0.3936, loss_single: 0.3936, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,272 - Iter: 34, Total Loss: 0.3931, loss_single: 0.3931, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,307 - Iter: 35, Total Loss: 0.3927, loss_single: 0.3927, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,342 - Iter: 36, Total Loss: 0.3923, loss_single: 0.3923, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,378 - Iter: 37, Total Loss: 0.3920, loss_single: 0.3920, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,413 - Iter: 38, Total Loss: 0.3917, loss_single: 0.3917, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:36:40,449 - Iter: 39, Total Loss: 0.3914, loss_single: 0.3914, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,484 - Iter: 40, Total Loss: 0.3911, loss_single: 0.3911, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,522 - Iter: 41, Total Loss: 0.3908, loss_single: 0.3908, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,558 - Iter: 42, Total Loss: 0.3905, loss_single: 0.3905, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,593 - Iter: 43, Total Loss: 0.3902, loss_single: 0.3902, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,628 - Iter: 44, Total Loss: 0.3900, loss_single: 0.3900, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,663 - Iter: 45, Total Loss: 0.3897, loss_single: 0.3897, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,698 - Iter: 46, Total Loss: 0.3894, loss_single: 0.3894, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,733 - Iter: 47, Total Loss: 0.3892, loss_single: 0.3892, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,769 - Iter: 48, Total Loss: 0.3890, loss_single: 0.3890, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,804 - Iter: 49, Total Loss: 0.3887, loss_single: 0.3887, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,840 - Iter: 50, Total Loss: 0.3885, loss_single: 0.3885, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,876 - Iter: 51, Total Loss: 0.3882, loss_single: 0.3882, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,911 - Iter: 52, Total Loss: 0.3880, loss_single: 0.3880, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,947 - Iter: 53, Total Loss: 0.3877, loss_single: 0.3877, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:40,982 - Iter: 54, Total Loss: 0.3875, loss_single: 0.3875, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,017 - Iter: 55, Total Loss: 0.3873, loss_single: 0.3873, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,053 - Iter: 56, Total Loss: 0.3870, loss_single: 0.3870, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,088 - Iter: 57, Total Loss: 0.3868, loss_single: 0.3868, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,123 - Iter: 58, Total Loss: 0.3865, loss_single: 0.3865, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,158 - Iter: 59, Total Loss: 0.3863, loss_single: 0.3863, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:41,193 - Iter: 60, Total Loss: 0.3860, loss_single: 0.3860, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,228 - Iter: 61, Total Loss: 0.3858, loss_single: 0.3858, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,263 - Iter: 62, Total Loss: 0.3855, loss_single: 0.3855, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:41,298 - Iter: 63, Total Loss: 0.3853, loss_single: 0.3853, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,333 - Iter: 64, Total Loss: 0.3850, loss_single: 0.3850, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,369 - Iter: 65, Total Loss: 0.3848, loss_single: 0.3848, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,404 - Iter: 66, Total Loss: 0.3845, loss_single: 0.3845, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,439 - Iter: 67, Total Loss: 0.3843, loss_single: 0.3843, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,474 - Iter: 68, Total Loss: 0.3840, loss_single: 0.3840, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,509 - Iter: 69, Total Loss: 0.3838, loss_single: 0.3838, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,545 - Iter: 70, Total Loss: 0.3835, loss_single: 0.3835, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,580 - Iter: 71, Total Loss: 0.3832, loss_single: 0.3832, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,615 - Iter: 72, Total Loss: 0.3830, loss_single: 0.3830, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,650 - Iter: 73, Total Loss: 0.3827, loss_single: 0.3827, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,685 - Iter: 74, Total Loss: 0.3825, loss_single: 0.3825, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:41,720 - Iter: 75, Total Loss: 0.3822, loss_single: 0.3822, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,755 - Iter: 76, Total Loss: 0.3819, loss_single: 0.3819, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,790 - Iter: 77, Total Loss: 0.3817, loss_single: 0.3817, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,825 - Iter: 78, Total Loss: 0.3814, loss_single: 0.3814, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,861 - Iter: 79, Total Loss: 0.3811, loss_single: 0.3811, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,896 - Iter: 80, Total Loss: 0.3809, loss_single: 0.3809, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,932 - Iter: 81, Total Loss: 0.3806, loss_single: 0.3806, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:41,967 - Iter: 82, Total Loss: 0.3803, loss_single: 0.3803, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,002 - Iter: 83, Total Loss: 0.3801, loss_single: 0.3801, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,037 - Iter: 84, Total Loss: 0.3798, loss_single: 0.3798, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,072 - Iter: 85, Total Loss: 0.3795, loss_single: 0.3795, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,107 - Iter: 86, Total Loss: 0.3793, loss_single: 0.3793, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,142 - Iter: 87, Total Loss: 0.3790, loss_single: 0.3790, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:42,177 - Iter: 88, Total Loss: 0.3787, loss_single: 0.3787, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,212 - Iter: 89, Total Loss: 0.3784, loss_single: 0.3784, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,247 - Iter: 90, Total Loss: 0.3782, loss_single: 0.3782, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:42,283 - Iter: 91, Total Loss: 0.3779, loss_single: 0.3779, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,320 - Iter: 92, Total Loss: 0.3776, loss_single: 0.3776, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:36:42,355 - Iter: 93, Total Loss: 0.3774, loss_single: 0.3774, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,390 - Iter: 94, Total Loss: 0.3771, loss_single: 0.3771, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,426 - Iter: 95, Total Loss: 0.3768, loss_single: 0.3768, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,460 - Iter: 96, Total Loss: 0.3766, loss_single: 0.3766, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:36:42,495 - Iter: 97, Total Loss: 0.3763, loss_single: 0.3763, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,531 - Iter: 98, Total Loss: 0.3760, loss_single: 0.3760, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,566 - Iter: 99, Total Loss: 0.3757, loss_single: 0.3757, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,601 - Iter: 100, Total Loss: 0.3755, loss_single: 0.3755, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:36:42,711 - Saving summary figures for iter 100
+2026-06-24 05:36:46,280 - ### Finished 100 iterations, averaged iter_t = 0.038926 with std = 0.040 ###
+2026-06-24 05:36:46,280 -  
+2026-06-24 05:36:46,280 - ### The PtyRADSolver is finished in 9.184 sec ###
+2026-06-24 05:36:46,280 -  

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000010/c30_small__simulation_data1__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_6.82_C12_-46.2_phi12_189_C21_698_phi21_28_C23_-991_phi23_186_C30_2.87e+06_C32_-7..72e+07_phi56_117/model_iter0100.hdf5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f41e26475e6b64da92435d65acbdc319220186c7dc110f94b41ea35fcf64be2b
+size 1597800

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000010/c30_small__simulation_data1__sample_000010_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_6.82_C12_-46.2_phi12_189_C21_698_phi21_28_C23_-991_phi23_186_C30_2.87e+06_C32_-7..72e+07_phi56_117/simulation_data1__sample_000010.yaml

@@ -0,0 +1,140 @@
+init_params:
+  probe_kv: 60.0
+  probe_conv_angle: 12.819171905517578
+  probe_aberrations:
+    C10: 6.817439556121826
+    C12: -46.1841926574707
+    C21: 698.3984985351562
+    C23: -991.1152954101562
+    C30: 2867064.25
+    C32: -7273.42236328125
+    C34: 4690.001953125
+    C41: -229145.46875
+    C43: 67091.0859375
+    C45: 209215.375
+    C50: 364242912.0
+    C52: 1600665.625
+    C54: 10428370.0
+    C56: 17220394.0
+    phi12: 189.01854580790615
+    phi21: 27.966190773596992
+    phi23: 185.98835970338666
+    phi32: 270.5234663812351
+    phi34: 102.29181584050822
+    phi41: 278.8881258501829
+    phi43: 347.2451235165341
+    phi45: 265.80825894809516
+    phi52: 83.85078319471569
+    phi54: 185.35621203532367
+    phi56: 117.47605719568774
+  meas_Npix: 128
+  pos_N_scan_slow: 16
+  pos_N_scan_fast: 16
+  pos_scan_step_size: 0.36360010504722595
+  probe_pmode_max: 6
+  obj_Nlayer: 1
+  obj_slice_thickness: 20.0
+  meas_permute: null
+  meas_reshape:
+  - 256
+  - 128
+  - 128
+  meas_flipT:
+  - 0
+  - 0
+  - 0
+  meas_crop: null
+  meas_pad: null
+  meas_resample: null
+  pos_scan_affine: null
+  meas_params:
+    path: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000010_measurement.h5
+    key: measurement
+  random_seed: 20261854
+  obj_omode_max: 1
+  meas_calibration:
+    mode: kMax
+    value: 2.5
+model_params:
+  detector_blur_std: null
+  update_params:
+    obja:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    objp:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    probe:
+      start_iter: 1
+      lr: 0.0001
+      end_iter: null
+    probe_pos_shifts:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+loss_params:
+  loss_single:
+    state: true
+    weight: 1.0
+    dp_pow: 0.5
+  loss_poissn:
+    state: false
+    weight: 1.0
+    dp_pow: 1.0
+    eps: 1.0e-06
+  loss_sparse:
+    state: false
+    weight: 0.1
+    ln_order: 1
+constraint_params:
+  ortho_pmode:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  fix_probe_int:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  obj_rblur:
+    start_iter: null
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 0.4
+  obj_zblur:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 1
+  obja_thresh:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    relax: 0
+    thresh:
+    - 0.96
+    - 1.04
+  objp_postiv:
+    start_iter: null
+    step: 1
+    end_iter: null
+    relax: 0
+recon_params:
+  NITER: 100
+  BATCH_SIZE:
+    size: 32
+    grad_accumulation: 1
+  SAVE_ITERS: 100
+  output_dir: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000010
+  recon_dir_affixes:
+  - default
+  prefix: c30_small__simulation_data1__sample_000010
+  postfix: ''
+  compiler_configs:
+    enable: false
+  prefix_time: false

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000013/c30_small__simulation_data1__sample_000013_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-560_C12_35.3_phi12_331_C21_810_phi21_131_C23_-495_phi23_41.6_C30_3.96e+06_C32_-7.06e+03_phi32_2.6/20260624_053722_ptyrad_log.txt

@@ -0,0 +1,417 @@
+2026-06-24 05:37:20,480 - ### PtyRAD LoggingManager configuration ###
+2026-06-24 05:37:20,480 - log_file       = 'ptyrad_log.txt'. If log_file = None, no log file will be created.
+2026-06-24 05:37:20,480 - log_dir        = 'auto'. If log_dir = 'auto', then log will be saved to `output_path` or 'logs/'.
+2026-06-24 05:37:20,480 - flush_file     = True. Automatically set to True if `log_file is not None`
+2026-06-24 05:37:20,480 - prefix_time    = datetime. If true, preset strings ('date', 'time', 'datetime'), or a string of time format, a datetime str is prefixed to the `log_file`.
+2026-06-24 05:37:20,480 - prefix_jobid   = '0'. If not 0, it'll be prefixed to the log file. This is used for hypertune mode with multiple GPUs.
+2026-06-24 05:37:20,480 - append_to_file = True. If true, logs will be appended to the existing file. If false, the log file will be overwritten.
+2026-06-24 05:37:20,480 - show_timestamp = True. If true, the printed information will contain a timestamp.
+2026-06-24 05:37:20,480 -  
+2026-06-24 05:37:20,587 - ### Initializing HuggingFace accelerator ###
+2026-06-24 05:37:20,587 - Accelerator.distributed_type = DistributedType.NO
+2026-06-24 05:37:20,587 - Accelerator.num_process      = 1
+2026-06-24 05:37:20,587 - Accelerator.mixed_precision  = no
+2026-06-24 05:37:20,598 - 'accelerate' is available but NOT using distributed mode or mixed precision
+2026-06-24 05:37:20,598 - If you want to utilize 'accelerate' for multiGPU or mixed precision, 
+2026-06-24 05:37:20,598 - Run `accelerate launch --multi_gpu --num_processes=2 --mixed_precision='no' -m ptyrad run <PTYRAD_ARGUMENTS> --gpuid 'acc'` in your terminal
+2026-06-24 05:37:20,598 -  
+2026-06-24 05:37:20,598 - ### System information ###
+2026-06-24 05:37:20,598 - Platform: Linux-4.18.0-553.69.1.el8_10.x86_64-x86_64-with-glibc2.28
+2026-06-24 05:37:20,598 - Operating System: Linux 4.18.0-553.69.1.el8_10.x86_64
+2026-06-24 05:37:20,598 - OS Version: #1 SMP Thu Aug 7 18:10:00 EDT 2025
+2026-06-24 05:37:20,598 - Machine: x86_64
+2026-06-24 05:37:20,599 - Processor: x86_64
+2026-06-24 05:37:20,599 - Available CPU cores: 8
+2026-06-24 05:37:20,599 - SLURM-Allocated Total Memory: 78.12 GB
+2026-06-24 05:37:20,599 -  
+2026-06-24 05:37:20,599 - ### GPU information ###
+2026-06-24 05:37:20,599 - CUDA Available: True
+2026-06-24 05:37:20,599 - CUDA Version: 13.0
+2026-06-24 05:37:20,599 - Available CUDA GPUs: ['NVIDIA A100 80GB PCIe']
+2026-06-24 05:37:20,599 - CUDA Compute Capability: ['8.0']
+2026-06-24 05:37:20,599 -   INFO: For torch.compile with Triton, you'll need CUDA GPU with Compute Capability >= 7.0.
+2026-06-24 05:37:20,599 -         In addition, Triton does not directly support Windows.
+2026-06-24 05:37:20,599 -         For Windows users, please follow the instruction and download `triton-windows` from https://github.com/woct0rdho/triton-windows.
+2026-06-24 05:37:20,615 - MIG (Multi-Instance GPU) mode = False
+2026-06-24 05:37:20,615 -   INFO: MIG splits a physical GPU into multiple GPU slices, but multiGPU does not support these MIG slices.
+2026-06-24 05:37:20,615 -         In addition, multiGPU is currently only available on Linux due to the limited NCCL support.
+2026-06-24 05:37:20,615 -       -> If you're doing normal reconstruction/hypertune, you can safely ignore this.
+2026-06-24 05:37:20,615 -       -> If you want to do multiGPU, you must provide multiple 'full' GPUs that are not in MIG mode.
+2026-06-24 05:37:20,615 -  
+2026-06-24 05:37:20,615 - ### Python information ###
+2026-06-24 05:37:20,617 - Python Executable: /home/tnguye11/anaconda3/envs/ptyrad/bin/python3.12
+2026-06-24 05:37:20,617 - Python Version: 3.12.13 | packaged by Anaconda, Inc. | (main, Mar 19 2026, 20:20:58) [GCC 14.3.0]
+2026-06-24 05:37:20,617 -  
+2026-06-24 05:37:20,617 - ### Packages information ###
+2026-06-24 05:37:20,618 - Numpy Version (metadata): 2.4.6
+2026-06-24 05:37:20,618 - PyTorch Version (metadata): 2.12.0
+2026-06-24 05:37:20,619 - Optuna Version (metadata): 4.9.0
+2026-06-24 05:37:20,620 - Accelerate Version (metadata): 1.13.0
+2026-06-24 05:37:20,620 - PtyRAD Version (ptyrad/__init__.py): 1.0.0
+2026-06-24 05:37:20,620 - PtyRAD is located at: /home/tnguye11/anaconda3/envs/ptyrad/lib/python3.12/site-packages/ptyrad/__init__.py
+2026-06-24 05:37:20,620 -  
+2026-06-24 05:37:20,620 - ### Loading params file ###
+2026-06-24 05:37:20,620 - params_path = /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/params/c30_small/simulation_data1__sample_000013.yaml
+2026-06-24 05:37:20,627 - validate = True: Filling defaults and validating the params file...
+2026-06-24 05:37:20,628 - Success! Params file validated and defaults applied.
+2026-06-24 05:37:20,628 -  
+2026-06-24 05:37:20,628 - ### Setting GPU Device ###
+2026-06-24 05:37:20,628 - Selected GPU device: cuda:0 (NVIDIA A100 80GB PCIe)
+2026-06-24 05:37:20,628 -  
+2026-06-24 05:37:20,628 - Random seed: 20261857 provided by params file
+2026-06-24 05:37:20,628 - ### Initializing Initializer ###
+2026-06-24 05:37:20,629 - init_params are displayed below:
+2026-06-24 05:37:20,629 -   random_seed: 20261857
+2026-06-24 05:37:20,629 -   probe_illum_type: electron
+2026-06-24 05:37:20,629 -   probe_kv: 80.0
+2026-06-24 05:37:20,629 -   probe_conv_angle: 16.1203670501709
+2026-06-24 05:37:20,629 -   probe_aberrations: {'C10': -560.077, 'C12': 35.326, 'phi12': 330.623, 'C21': 809.986, 'phi21': 130.577, 'C23': -494.813, 'phi23': 41.647, 'C30': 3962327.75, 'C32': -7056.434, 'phi32': 211.292, 'C34': -5681.441, 'phi34': 145.407, 'C41': -223541.406, 'phi41': 123.519, 'C43': 129071.148, 'phi43': 245.656, 'C45': 124501.633, 'phi45': 192.589, 'C50': -174077264.0, 'C52': -7003224.5, 'phi52': 181.199, 'C54': 9173544.0, 'phi54': 199.834, 'C56': 11069104.0, 'phi56': 12.622}
+2026-06-24 05:37:20,629 -   beam_kev: None
+2026-06-24 05:37:20,629 -   probe_dRn: None
+2026-06-24 05:37:20,629 -   probe_Rn: None
+2026-06-24 05:37:20,629 -   probe_D_H: None
+2026-06-24 05:37:20,629 -   probe_D_FZP: None
+2026-06-24 05:37:20,629 -   probe_Ls: None
+2026-06-24 05:37:20,629 -   meas_Npix: 128
+2026-06-24 05:37:20,629 -   pos_N_scans: 256
+2026-06-24 05:37:20,629 -   pos_N_scan_slow: 16
+2026-06-24 05:37:20,629 -   pos_N_scan_fast: 16
+2026-06-24 05:37:20,629 -   pos_scan_step_size: 0.361175537109375
+2026-06-24 05:37:20,629 -   meas_calibration: {'mode': 'kMax', 'value': 2.5}
+2026-06-24 05:37:20,629 -   probe_pmode_max: 6
+2026-06-24 05:37:20,629 -   probe_pmode_init_pows: [0.02]
+2026-06-24 05:37:20,629 -   obj_omode_max: 1
+2026-06-24 05:37:20,629 -   obj_omode_init_occu: {'occu_type': 'uniform', 'init_occu': None}
+2026-06-24 05:37:20,629 -   obj_Nlayer: 1
+2026-06-24 05:37:20,629 -   obj_slice_thickness: 20.0
+2026-06-24 05:37:20,629 -   simu_Npix: None
+2026-06-24 05:37:20,629 -   simu_match_mode: None
+2026-06-24 05:37:20,629 -   meas_permute: None
+2026-06-24 05:37:20,629 -   meas_reshape: [256, 128, 128]
+2026-06-24 05:37:20,629 -   meas_flipT: [0, 0, 0]
+2026-06-24 05:37:20,629 -   meas_crop: None
+2026-06-24 05:37:20,629 -   meas_pad: None
+2026-06-24 05:37:20,629 -   meas_resample: None
+2026-06-24 05:37:20,629 -   meas_add_source_size: None
+2026-06-24 05:37:20,629 -   meas_add_detector_blur: None
+2026-06-24 05:37:20,629 -   meas_remove_neg_values: {'mode': 'clip_neg', 'value': None, 'force': False}
+2026-06-24 05:37:20,630 -   meas_normalization: {'mode': 'max_at_one', 'value': None}
+2026-06-24 05:37:20,630 -   meas_add_poisson_noise: None
+2026-06-24 05:37:20,630 -   meas_export: None
+2026-06-24 05:37:20,630 -   probe_permute: None
+2026-06-24 05:37:20,630 -   probe_z_shift: None
+2026-06-24 05:37:20,630 -   probe_normalization: {'mode': 'mean_total_ints', 'value': None}
+2026-06-24 05:37:20,630 -   pos_scan_flipT: None
+2026-06-24 05:37:20,630 -   pos_scan_affine: None
+2026-06-24 05:37:20,630 -   pos_scan_rand_std: 0.15
+2026-06-24 05:37:20,630 -   obj_z_crop: None
+2026-06-24 05:37:20,630 -   obj_z_pad: None
+2026-06-24 05:37:20,630 -   obj_z_resample: None
+2026-06-24 05:37:20,630 -   meas_source: file
+2026-06-24 05:37:20,630 -   meas_params: {'path': '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000013_measurement.h5', 'key': 'measurement', 'shape': None, 'offset': None, 'gap': None, 'selection': None, 'zarr_kwargs': None}
+2026-06-24 05:37:20,630 -   probe_source: simu
+2026-06-24 05:37:20,630 -   probe_params: None
+2026-06-24 05:37:20,630 -   pos_source: simu
+2026-06-24 05:37:20,630 -   pos_params: None
+2026-06-24 05:37:20,630 -   obj_source: simu
+2026-06-24 05:37:20,630 -   obj_params: None
+2026-06-24 05:37:20,630 -   tilt_source: simu
+2026-06-24 05:37:20,630 -   tilt_params: {'tilt_type': 'all', 'init_tilts': [[0, 0]]}
+2026-06-24 05:37:20,630 -  
+2026-06-24 05:37:20,630 - ### Initializing cache ###
+2026-06-24 05:37:20,630 - use_cached_obj   = False
+2026-06-24 05:37:20,630 - use_cached_probe = False
+2026-06-24 05:37:20,630 - use_cached_pos   = False
+2026-06-24 05:37:20,630 -  
+2026-06-24 05:37:20,630 - ### Initializing measurements ###
+2026-06-24 05:37:20,630 - Loading measurements from source = 'file'
+2026-06-24 05:37:20,630 - Detected measurement file type = '.h5'
+2026-06-24 05:37:20,706 - Original measurements dtype is float32, casting to float32 (single precision) for computational efficiency.
+2026-06-24 05:37:20,706 - Imported meausrements shape / dtype = (16, 16, 128, 128), dtype = float32
+2026-06-24 05:37:20,708 - Imported meausrements int. statistics (min, mean, max) = (0.0000, 0.0001, 0.0047)
+2026-06-24 05:37:20,708 - Reshaping measurements to shape = [256, 128, 128]
+2026-06-24 05:37:20,708 - Flipping measurements with [flipud, fliplr, transpose] = [0, 0, 0]
+2026-06-24 05:37:20,709 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:37:20,709 - Normalizing measurements with mode = 'max_at_one' and value = 'None'
+2026-06-24 05:37:20,710 - Normalizing by max of the 2D mean pattern intensity: 0.0029769363
+2026-06-24 05:37:20,710 - meausrements shape / dtype = (256, 128, 128), dtype = float32
+2026-06-24 05:37:20,712 - meausrements int. statistics (min, mean, max) = (0.0000, 0.0204, 1.5802)
+2026-06-24 05:37:20,712 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:37:20,713 - Pattern total int. statistics (min, mean, max)        = (327.9684, 334.6317, 339.9153), with min/max = 96.5%
+2026-06-24 05:37:20,715 - Global meausrements int. statistics (min, mean, max)  = (0.0000, 0.0204, 1.5802)
+2026-06-24 05:37:20,715 - measurements                              (N, Ky, Kx) = float32, (256, 128, 128)
+2026-06-24 05:37:20,715 -  
+2026-06-24 05:37:20,715 - ### Setting up calibration ###
+2026-06-24 05:37:20,715 - meas_calibration mode = 'kMax', value = 2.5
+2026-06-24 05:37:20,715 - Using loaded raw averaged measurement (before crop/pad/resample) to fit RBF as a part of the meas calibration
+2026-06-24 05:37:20,715 - Radius of fitted bright field disk (RBF) = 9.84 px with meas_Npix = 128
+2026-06-24 05:37:20,715 - Suggested probe_mask_k radius (RBF*2/Npix) > 0.1537
+2026-06-24 05:37:20,715 - Fitting raw averaged measurement with center, radius, and Gaussian blur std as a sanity check
+2026-06-24 05:37:20,715 - Note that the fitted Gaussian blur std (detector blur) would be affected by overlapping Bragg disks
+2026-06-24 05:37:20,934 - Initial guess: center=(63.61, 63.64), radius=9.84, Gaussian blur std=0.50
+2026-06-24 05:37:20,949 - Final fit: center=(63.61, 63.64), radius=9.84, Gaussian blur std=0.70
+2026-06-24 05:37:20,949 - Using init_params, the inferred RBF (conv_angle / 1e3 * Npix * dx / wavelength) = 9.88 px with Npix = 128
+2026-06-24 05:37:20,949 - dx (real space pixel size of probe and object) set to 0.2000 Ang with Npix = 128
+2026-06-24 05:37:20,949 -  
+2026-06-24 05:37:20,949 - ### Setting init_variables dict ###
+2026-06-24 05:37:20,949 - Derived values given input init_params:
+2026-06-24 05:37:20,949 -   kv          = 80.0 kV
+2026-06-24 05:37:20,949 -   wavelength  = 0.0418 Ang
+2026-06-24 05:37:20,949 -   conv_angle  = 16.1203670501709 mrad
+2026-06-24 05:37:20,949 -   Npix        = 128 px
+2026-06-24 05:37:20,949 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:37:20,949 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:37:20,949 -   da          = 1.6311 mrad
+2026-06-24 05:37:20,950 -   angleMax    = 104.3929 mrad
+2026-06-24 05:37:20,950 -   RBF         = 9.8829 px (Inferred from the given calibration, NOT necessarily from the loaded measurement data)
+2026-06-24 05:37:20,950 -   n_alpha     = 6.4758 (# conv_angle)
+2026-06-24 05:37:20,950 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:37:20,950 -   Rayleigh-limited resolution  = 1.5801 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:37:20,950 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:37:20,950 -  
+2026-06-24 05:37:20,950 - ### Initializing probe ###
+2026-06-24 05:37:20,950 - Loading probe from source = 'simu'
+2026-06-24 05:37:20,950 - Using experimental parameters specified by 'init_params' for initial probe simulation.
+2026-06-24 05:37:20,950 - Start simulating STEM probe
+2026-06-24 05:37:20,950 -   kv          = 80.0 kV
+2026-06-24 05:37:20,950 -   wavelength  = 0.0418 Ang
+2026-06-24 05:37:20,950 -   conv_angle  = 16.1203670501709 mrad
+2026-06-24 05:37:20,950 -   Npix        = 128 px
+2026-06-24 05:37:20,950 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:37:20,950 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:37:20,950 -   alpha_max   = 104.3929 mrad
+2026-06-24 05:37:20,950 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:37:20,950 -   Rayleigh-limited resolution  = 1.5801 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:37:20,950 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:37:20,950 - Krivanek Haider     Magnitude         Angle (°)  Description
+2026-06-24 05:37:20,950 - -------------------------------------------------------------------------------------
+2026-06-24 05:37:20,950 - C1,0     C1         -560.0770         -          Defocus (C10 = -df)
+2026-06-24 05:37:20,951 - C1,2     A1         35.3260           330.62     2-fold astigmatism
+2026-06-24 05:37:20,951 - C2,1     3*B2       809.9860          130.58     Axial coma
+2026-06-24 05:37:20,951 - C2,3     A2         -494.8130         41.65      3-fold astigmatism
+2026-06-24 05:37:20,951 - C3,0     C3         3962327.7500      -          Spherical aberration
+2026-06-24 05:37:20,951 - C3,2     4*S3       -7056.4340        211.29     Axial star aberration
+2026-06-24 05:37:20,951 - C3,4     A3         -5681.4410        145.41     4-fold astigmatism
+2026-06-24 05:37:20,951 - C4,1     4*B4       -223541.4060      123.52     Axial coma(4th)
+2026-06-24 05:37:20,951 - C4,3     4*D4       129071.1480       245.66     3-lobe aberration
+2026-06-24 05:37:20,951 - C4,5     A4         124501.6330       192.59     5-fold astigmatism
+2026-06-24 05:37:20,951 - C5,0     C5         -174077264.0000   -          Spherical aberration (5th)
+2026-06-24 05:37:20,951 - C5,2     6*S5       -7003224.5000     181.20     Axial star aberration(5th)
+2026-06-24 05:37:20,951 - C5,4     6*R5       9173544.0000      199.83     4-lobe aberration
+2026-06-24 05:37:20,951 - C5,6     A5         11069104.0000     12.62      6-fold astigmatism
+2026-06-24 05:37:20,954 - Loaded probe shape = (1, 128, 128), dtype = complex128
+2026-06-24 05:37:20,954 - pmode_now: 1 and pmode_max: 6, padding the pmode.
+2026-06-24 05:37:20,954 - Creating 5 new probe modes from the major mode
+2026-06-24 05:37:20,954 - Start making mixed-state STEM probe with 6 incoherent probe modes
+2026-06-24 05:37:21,050 - Relative power of probe modes = [0.9  0.02 0.02 0.02 0.02 0.02]
+2026-06-24 05:37:21,051 - Orthogonalizing 6 pmodes
+2026-06-24 05:37:21,053 - Sorting 6 pmodes by their intensities
+2026-06-24 05:37:21,054 - Normalizing probe intensity with mode = 'mean_total_ints' and value = 'None'
+2026-06-24 05:37:21,054 - sum(|probe_data|**2) = 334.63, while meas_total_ints (min, mean, max) = (327.9684, 334.6317, 339.9153)
+2026-06-24 05:37:21,055 - probe                         (pmode, Ny, Nx) = complex64, (6, 128, 128)
+2026-06-24 05:37:21,055 -  
+2026-06-24 05:37:21,055 - ### Initializing probe positions ###
+2026-06-24 05:37:21,055 - Loading probe positions from source = 'simu'
+2026-06-24 05:37:21,055 - Using experimental parameters specified by 'init_params' (dx, scan_step size, N_scan_slow, N_scan_fast) for initial position simulation.
+2026-06-24 05:37:21,055 - Simulating probe positions with dx = 0.2000, scan_step_size = 0.3612, N_scan_fast = 16, N_scan_slow = 16
+2026-06-24 05:37:21,055 - Applying Gaussian distributed random displacement with std = 0.15 px to scan positions
+2026-06-24 05:37:21,057 - crop_pos                                (N,2) = int16, (256, 2)
+2026-06-24 05:37:21,057 - crop_pos 1st and last px coords (y,x)         = ([15, 15], [43, 43])
+2026-06-24 05:37:21,057 - crop_pos extent (Ang)                         = [5.6 5.6]
+2026-06-24 05:37:21,057 - probe_pos_shifts                        (N,2) = float32, (256, 2)
+2026-06-24 05:37:21,057 -  
+2026-06-24 05:37:21,057 - ### Initializing object ###
+2026-06-24 05:37:21,057 - Loading object from source = 'simu'
+2026-06-24 05:37:21,057 - Using experimental parameters specified by 'init_params' for initial object simulation.
+2026-06-24 05:37:21,059 - omode_now: 1 and omode_max: 1, leaving the omode unchanged.
+2026-06-24 05:37:21,059 - object                    (omode, Nz, Ny, Nx) = complex64, (1, 1, 187, 187)
+2026-06-24 05:37:21,059 - object extent                 (Z, Y, X) (Ang) = [20.  37.4 37.4]
+2026-06-24 05:37:21,059 -  
+2026-06-24 05:37:21,059 - ### Initializing omode_occu from 'uniform' ###
+2026-06-24 05:37:21,059 - omode_occu                            (omode) = float32, (1,)
+2026-06-24 05:37:21,059 -  
+2026-06-24 05:37:21,059 - ### Initializing H (Fresnel propagator) ###
+2026-06-24 05:37:21,059 - Calculating H with probe_shape = (128, 128) px, dx = 0.2000 Ang, slice_thickness = 20.0000 Ang, lambd = 0.0418 Ang
+2026-06-24 05:37:21,060 - H                                    (Ky, Kx) = complex64, (128, 128)
+2026-06-24 05:37:21,060 -  
+2026-06-24 05:37:21,060 - ### Initializing obj tilts from = 'simu' ###
+2026-06-24 05:37:21,060 - Initialized obj_tilts with init_tilts = [[0, 0]] (theta_y, theta_x) mrad
+2026-06-24 05:37:21,060 - obj_tilts                              (N, 2) = float32, (1, 2)
+2026-06-24 05:37:21,060 -  
+2026-06-24 05:37:21,060 - ### Checking consistency between input params with the initialized variables ###
+2026-06-24 05:37:21,060 - meas_Npix, simu_Npix, DP measurements, probe, and H shapes are consistent as '128'
+2026-06-24 05:37:21,060 - N_scans, len(meas), N_scan_slow*N_scan_fast, len(crop_pos), and len(probe_pos_shifts) are consistent as '256'
+2026-06-24 05:37:21,060 - obj.shape[0] is consistent with len(omode_occu) as '1'
+2026-06-24 05:37:21,060 - obj.shape[1] is consistent with Nlayer as '1'
+2026-06-24 05:37:21,060 - crop positions (yx_min=[15 15], yx_max=[171 171]) are well contained inside object canvas (Ny,Nx) = (187, 187).
+2026-06-24 05:37:21,060 - obj_tilts is consistent with either 1 or N_scans
+2026-06-24 05:37:21,060 - Pass the consistency check of initialized variables, initialization is done!
+2026-06-24 05:37:21,060 -  
+2026-06-24 05:37:21,061 - ### Collecting reconstruction provenance ###
+2026-06-24 05:37:21,061 - Reconstruction provenance is collected and initialized.
+2026-06-24 05:37:21,061 -  
+2026-06-24 05:37:21,061 - ### Initializing loss function ###
+2026-06-24 05:37:21,061 - Active loss types:
+2026-06-24 05:37:21,061 -   loss_single : {'state': True, 'weight': 1.0, 'dp_pow': 0.5}
+2026-06-24 05:37:21,061 -  
+2026-06-24 05:37:21,061 - ### Initializing constraint function ###
+2026-06-24 05:37:21,061 - Active constraint types:
+2026-06-24 05:37:21,061 -   ortho_pmode   : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:37:21,061 -   fix_probe_int : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:37:21,061 -   obj_zblur     : {'start_iter': 1, 'step': 1, 'end_iter': None, 'obj_type': 'both', 'kernel_size': 5, 'std': 1.0}
+2026-06-24 05:37:21,061 -   obja_thresh   : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0, 'thresh': [0.96, 1.04]}
+2026-06-24 05:37:21,061 -   pos_recenter  : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0}
+2026-06-24 05:37:21,061 -  
+2026-06-24 05:37:21,061 - ### Done initializing PtyRADSolver ###
+2026-06-24 05:37:21,061 -  
+2026-06-24 05:37:21,185 - ### Starting the PtyRADSolver in reconstruct mode ###
+2026-06-24 05:37:21,185 -  
+2026-06-24 05:37:21,185 - ### Initializing PtychoModel model ###
+2026-06-24 05:37:21,268 - ### PtychoModel optimizable variables ###
+2026-06-24 05:37:21,268 - obja            : torch.Size([1, 1, 187, 187])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:37:21,268 - objp            : torch.Size([1, 1, 187, 187])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:37:21,268 - obj_tilts       : torch.Size([1, 2])              , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:37:21,268 - slice_thickness : torch.Size([])                  , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:37:21,268 - probe           : torch.Size([6, 128, 128, 2])    , torch.float32   , device:cuda:0, grad:True , lr:1e-04
+2026-06-24 05:37:21,268 - probe_pos_shifts: torch.Size([256, 2])            , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:37:21,268 -  
+2026-06-24 05:37:21,268 - ### Optimizable variables statitsics ###
+2026-06-24 05:37:21,268 - Total measurement values  : 4,194,304
+2026-06-24 05:37:21,268 - Total optimizing variables: 267,058
+2026-06-24 05:37:21,268 - Overdetermined ratio      : 15.71
+2026-06-24 05:37:21,268 -  
+2026-06-24 05:37:21,268 - ### Model behavior ###
+2026-06-24 05:37:21,268 - Tilt propagator           : False
+2026-06-24 05:37:21,268 - Change slice thickness    : False
+2026-06-24 05:37:21,268 - Detector blur             : False
+2026-06-24 05:37:21,268 - Preload data              : True
+2026-06-24 05:37:21,268 - On-the-fly meas padding   : False
+2026-06-24 05:37:21,268 - On-the-fly meas resample  : False
+2026-06-24 05:37:21,269 - On-the-fly simu match mode: None
+2026-06-24 05:37:21,269 -  
+2026-06-24 05:37:21,310 - ### Done initializing PtychoModel model ###
+2026-06-24 05:37:21,310 -  
+2026-06-24 05:37:21,310 - ### Creating PyTorch 'Adam' optimizer with configs = {} ###
+2026-06-24 05:37:21,310 -  
+2026-06-24 05:37:21,310 - ### Generating indices, batches, and output_path ###
+2026-06-24 05:37:21,312 - d90 = 49.000 px or 9.800 Ang
+2026-06-24 05:37:21,312 - Selecting indices with the 'full' mode 
+2026-06-24 05:37:21,771 - Generated 8 'random' groups of ~32 scan positions in 0.000 sec
+2026-06-24 05:37:21,846 - The effective batch size (i.e., how many probe positions are simultaneously used for 1 update of ptychographic parameters) is batch_size * grad_accumulation = 32 * 1 = 32
+2026-06-24 05:37:21,847 - Original recon_dir_affixes = ['default']
+2026-06-24 05:37:21,847 - Expanded recon_dir_affixes = ['indices', 'meas', 'batch', 'pmode', 'omode', 'nlayer', 'lr', 'model', 'constraint', 'loss', 'affine', 'tilt', 'aberrations']
+2026-06-24 05:37:21,853 - Path corrected for compatibility:
+2026-06-24 05:37:21,853 -   Original: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000013/c30_small__simulation_data1__sample_000013_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-560_C12_35.3_phi12_331_C21_810_phi21_131_C23_-495_phi23_41.6_C30_3.96e+06_C32_-7.06e+03_phi32_211_C34_-5.68e+03_phi34_145_C41_-2.24e+05_phi41_124_C43_1.29e+05_phi43_246_C45_1.25e+05_phi45_193_C50_-1.74e+08_C52_-7e+06_phi52_181_C54_9.17e+06_phi54_200_C56_1.11e+07_phi56_12.6
+2026-06-24 05:37:21,853 -   Corrected: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000013/c30_small__simulation_data1__sample_000013_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-560_C12_35.3_phi12_331_C21_810_phi21_131_C23_-495_phi23_41.6_C30_3.96e+06_C32_-7.06e+03_phi32_2.6
+2026-06-24 05:37:21,854 - output_path = '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000013/c30_small__simulation_data1__sample_000013_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-560_C12_35.3_phi12_331_C21_810_phi21_131_C23_-495_phi23_41.6_C30_3.96e+06_C32_-7.06e+03_phi32_2.6' is generated!
+2026-06-24 05:37:22,001 -  
+2026-06-24 05:37:22,002 - ### Log file is flushed (created) as /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000013/c30_small__simulation_data1__sample_000013_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-560_C12_35.3_phi12_331_C21_810_phi21_131_C23_-495_phi23_41.6_C30_3.96e+06_C32_-7.06e+03_phi32_2.6/20260624_053722_ptyrad_log.txt ###
+2026-06-24 05:37:22,002 -  
+2026-06-24 05:37:22,003 - ### Creating ConvergenceMonitor with {'tensors': ['obja', 'objp', 'probe', 'probe_pos_shifts'], 'every_n_iters': None, 'percentile_range': [15.0, 85.0]} ### 
+2026-06-24 05:37:22,003 - ### Start the PtyRAD iterative ptycho reconstruction ###
+2026-06-24 05:37:22,003 - Setting up PyTorch compiler with {'fullgraph': False, 'dynamic': None, 'backend': 'inductor', 'mode': 'default', 'options': None, 'disable': True}
+2026-06-24 05:37:23,194 - Iter: 1, Total Loss: 1.1104, loss_single: 1.1104, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.435 sec
+2026-06-24 05:37:23,231 - Iter: 2, Total Loss: 1.0764, loss_single: 1.0764, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:37:23,267 - Iter: 3, Total Loss: 1.0504, loss_single: 1.0504, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,302 - Iter: 4, Total Loss: 1.0240, loss_single: 1.0240, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,337 - Iter: 5, Total Loss: 0.9998, loss_single: 0.9998, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,372 - Iter: 6, Total Loss: 0.9790, loss_single: 0.9790, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,408 - Iter: 7, Total Loss: 0.9556, loss_single: 0.9556, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,444 - Iter: 8, Total Loss: 0.9310, loss_single: 0.9310, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,479 - Iter: 9, Total Loss: 0.9070, loss_single: 0.9070, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,515 - Iter: 10, Total Loss: 0.8827, loss_single: 0.8827, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,550 - Iter: 11, Total Loss: 0.8608, loss_single: 0.8608, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,585 - Iter: 12, Total Loss: 0.8432, loss_single: 0.8432, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,620 - Iter: 13, Total Loss: 0.8253, loss_single: 0.8253, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,655 - Iter: 14, Total Loss: 0.8073, loss_single: 0.8073, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,691 - Iter: 15, Total Loss: 0.7904, loss_single: 0.7904, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,727 - Iter: 16, Total Loss: 0.7702, loss_single: 0.7702, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,762 - Iter: 17, Total Loss: 0.7521, loss_single: 0.7521, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,798 - Iter: 18, Total Loss: 0.7381, loss_single: 0.7381, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,833 - Iter: 19, Total Loss: 0.7238, loss_single: 0.7238, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,869 - Iter: 20, Total Loss: 0.7099, loss_single: 0.7099, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,904 - Iter: 21, Total Loss: 0.6969, loss_single: 0.6969, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,939 - Iter: 22, Total Loss: 0.6858, loss_single: 0.6858, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:23,975 - Iter: 23, Total Loss: 0.6741, loss_single: 0.6741, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:37:24,010 - Iter: 24, Total Loss: 0.6619, loss_single: 0.6619, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:37:24,045 - Iter: 25, Total Loss: 0.6503, loss_single: 0.6503, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,081 - Iter: 26, Total Loss: 0.6399, loss_single: 0.6399, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,116 - Iter: 27, Total Loss: 0.6320, loss_single: 0.6320, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.034 sec
+2026-06-24 05:37:24,151 - Iter: 28, Total Loss: 0.6250, loss_single: 0.6250, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,187 - Iter: 29, Total Loss: 0.6170, loss_single: 0.6170, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,222 - Iter: 30, Total Loss: 0.6096, loss_single: 0.6096, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,257 - Iter: 31, Total Loss: 0.6029, loss_single: 0.6029, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,296 - Iter: 32, Total Loss: 0.5957, loss_single: 0.5957, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.038 sec
+2026-06-24 05:37:24,332 - Iter: 33, Total Loss: 0.5903, loss_single: 0.5903, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,367 - Iter: 34, Total Loss: 0.5861, loss_single: 0.5861, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,402 - Iter: 35, Total Loss: 0.5811, loss_single: 0.5811, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,437 - Iter: 36, Total Loss: 0.5764, loss_single: 0.5764, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,473 - Iter: 37, Total Loss: 0.5724, loss_single: 0.5724, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,509 - Iter: 38, Total Loss: 0.5688, loss_single: 0.5688, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:37:24,544 - Iter: 39, Total Loss: 0.5653, loss_single: 0.5653, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,579 - Iter: 40, Total Loss: 0.5618, loss_single: 0.5618, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,615 - Iter: 41, Total Loss: 0.5585, loss_single: 0.5585, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,651 - Iter: 42, Total Loss: 0.5560, loss_single: 0.5560, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,686 - Iter: 43, Total Loss: 0.5537, loss_single: 0.5537, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,721 - Iter: 44, Total Loss: 0.5512, loss_single: 0.5512, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,756 - Iter: 45, Total Loss: 0.5491, loss_single: 0.5491, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,792 - Iter: 46, Total Loss: 0.5474, loss_single: 0.5474, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,827 - Iter: 47, Total Loss: 0.5459, loss_single: 0.5459, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,862 - Iter: 48, Total Loss: 0.5444, loss_single: 0.5444, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,898 - Iter: 49, Total Loss: 0.5433, loss_single: 0.5433, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,933 - Iter: 50, Total Loss: 0.5426, loss_single: 0.5426, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:24,969 - Iter: 51, Total Loss: 0.5418, loss_single: 0.5418, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,004 - Iter: 52, Total Loss: 0.5410, loss_single: 0.5410, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,039 - Iter: 53, Total Loss: 0.5403, loss_single: 0.5403, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,075 - Iter: 54, Total Loss: 0.5397, loss_single: 0.5397, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,111 - Iter: 55, Total Loss: 0.5392, loss_single: 0.5392, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,146 - Iter: 56, Total Loss: 0.5387, loss_single: 0.5387, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,181 - Iter: 57, Total Loss: 0.5383, loss_single: 0.5383, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,216 - Iter: 58, Total Loss: 0.5379, loss_single: 0.5379, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,252 - Iter: 59, Total Loss: 0.5376, loss_single: 0.5376, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,287 - Iter: 60, Total Loss: 0.5373, loss_single: 0.5373, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,323 - Iter: 61, Total Loss: 0.5370, loss_single: 0.5370, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,358 - Iter: 62, Total Loss: 0.5367, loss_single: 0.5367, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,394 - Iter: 63, Total Loss: 0.5364, loss_single: 0.5364, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,429 - Iter: 64, Total Loss: 0.5361, loss_single: 0.5361, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,464 - Iter: 65, Total Loss: 0.5359, loss_single: 0.5359, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,499 - Iter: 66, Total Loss: 0.5357, loss_single: 0.5357, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,535 - Iter: 67, Total Loss: 0.5355, loss_single: 0.5355, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:37:25,571 - Iter: 68, Total Loss: 0.5353, loss_single: 0.5353, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,606 - Iter: 69, Total Loss: 0.5351, loss_single: 0.5351, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,641 - Iter: 70, Total Loss: 0.5349, loss_single: 0.5349, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,677 - Iter: 71, Total Loss: 0.5347, loss_single: 0.5347, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,712 - Iter: 72, Total Loss: 0.5346, loss_single: 0.5346, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,747 - Iter: 73, Total Loss: 0.5344, loss_single: 0.5344, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,783 - Iter: 74, Total Loss: 0.5342, loss_single: 0.5342, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,818 - Iter: 75, Total Loss: 0.5341, loss_single: 0.5341, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,853 - Iter: 76, Total Loss: 0.5339, loss_single: 0.5339, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,888 - Iter: 77, Total Loss: 0.5338, loss_single: 0.5338, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,924 - Iter: 78, Total Loss: 0.5337, loss_single: 0.5337, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,959 - Iter: 79, Total Loss: 0.5336, loss_single: 0.5336, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:25,995 - Iter: 80, Total Loss: 0.5334, loss_single: 0.5334, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,030 - Iter: 81, Total Loss: 0.5333, loss_single: 0.5333, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,065 - Iter: 82, Total Loss: 0.5332, loss_single: 0.5332, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,101 - Iter: 83, Total Loss: 0.5330, loss_single: 0.5330, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,136 - Iter: 84, Total Loss: 0.5329, loss_single: 0.5329, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,172 - Iter: 85, Total Loss: 0.5328, loss_single: 0.5328, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,208 - Iter: 86, Total Loss: 0.5327, loss_single: 0.5327, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:37:26,243 - Iter: 87, Total Loss: 0.5325, loss_single: 0.5325, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,278 - Iter: 88, Total Loss: 0.5324, loss_single: 0.5324, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,314 - Iter: 89, Total Loss: 0.5323, loss_single: 0.5323, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,349 - Iter: 90, Total Loss: 0.5322, loss_single: 0.5322, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,386 - Iter: 91, Total Loss: 0.5320, loss_single: 0.5320, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:37:26,422 - Iter: 92, Total Loss: 0.5319, loss_single: 0.5319, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,457 - Iter: 93, Total Loss: 0.5318, loss_single: 0.5318, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,492 - Iter: 94, Total Loss: 0.5317, loss_single: 0.5317, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,527 - Iter: 95, Total Loss: 0.5316, loss_single: 0.5316, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,563 - Iter: 96, Total Loss: 0.5314, loss_single: 0.5314, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,598 - Iter: 97, Total Loss: 0.5313, loss_single: 0.5313, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,634 - Iter: 98, Total Loss: 0.5312, loss_single: 0.5312, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,669 - Iter: 99, Total Loss: 0.5310, loss_single: 0.5310, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,704 - Iter: 100, Total Loss: 0.5309, loss_single: 0.5309, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:37:26,814 - Saving summary figures for iter 100
+2026-06-24 05:37:30,442 - ### Finished 100 iterations, averaged iter_t = 0.039031 with std = 0.040 ###
+2026-06-24 05:37:30,443 -  
+2026-06-24 05:37:30,443 - ### The PtyRADSolver is finished in 9.258 sec ###
+2026-06-24 05:37:30,443 -  

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000013/c30_small__simulation_data1__sample_000013_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-560_C12_35.3_phi12_331_C21_810_phi21_131_C23_-495_phi23_41.6_C30_3.96e+06_C32_-7.06e+03_phi32_2.6/simulation_data1__sample_000013.yaml

@@ -0,0 +1,140 @@
+init_params:
+  probe_kv: 80.0
+  probe_conv_angle: 16.1203670501709
+  probe_aberrations:
+    C10: -560.0769653320312
+    C12: 35.326045989990234
+    C21: 809.986083984375
+    C23: -494.8126220703125
+    C30: 3962327.75
+    C32: -7056.43408203125
+    C34: -5681.44140625
+    C41: -223541.40625
+    C43: 129071.1484375
+    C45: 124501.6328125
+    C50: -174077264.0
+    C52: -7003224.5
+    C54: 9173544.0
+    C56: 11069104.0
+    phi12: 330.6234483599579
+    phi21: 130.5769337596625
+    phi23: 41.64716725661196
+    phi32: 211.29175171018272
+    phi34: 145.4065858440808
+    phi41: 123.51869823364135
+    phi43: 245.65630489634944
+    phi45: 192.58945066825885
+    phi52: 181.19882615169863
+    phi54: 199.83397277370292
+    phi56: 12.622103355045292
+  meas_Npix: 128
+  pos_N_scan_slow: 16
+  pos_N_scan_fast: 16
+  pos_scan_step_size: 0.361175537109375
+  probe_pmode_max: 6
+  obj_Nlayer: 1
+  obj_slice_thickness: 20.0
+  meas_permute: null
+  meas_reshape:
+  - 256
+  - 128
+  - 128
+  meas_flipT:
+  - 0
+  - 0
+  - 0
+  meas_crop: null
+  meas_pad: null
+  meas_resample: null
+  pos_scan_affine: null
+  meas_params:
+    path: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000013_measurement.h5
+    key: measurement
+  random_seed: 20261857
+  obj_omode_max: 1
+  meas_calibration:
+    mode: kMax
+    value: 2.5
+model_params:
+  detector_blur_std: null
+  update_params:
+    obja:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    objp:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    probe:
+      start_iter: 1
+      lr: 0.0001
+      end_iter: null
+    probe_pos_shifts:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+loss_params:
+  loss_single:
+    state: true
+    weight: 1.0
+    dp_pow: 0.5
+  loss_poissn:
+    state: false
+    weight: 1.0
+    dp_pow: 1.0
+    eps: 1.0e-06
+  loss_sparse:
+    state: false
+    weight: 0.1
+    ln_order: 1
+constraint_params:
+  ortho_pmode:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  fix_probe_int:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  obj_rblur:
+    start_iter: null
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 0.4
+  obj_zblur:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 1
+  obja_thresh:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    relax: 0
+    thresh:
+    - 0.96
+    - 1.04
+  objp_postiv:
+    start_iter: null
+    step: 1
+    end_iter: null
+    relax: 0
+recon_params:
+  NITER: 100
+  BATCH_SIZE:
+    size: 32
+    grad_accumulation: 1
+  SAVE_ITERS: 100
+  output_dir: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000013
+  recon_dir_affixes:
+  - default
+  prefix: c30_small__simulation_data1__sample_000013
+  postfix: ''
+  compiler_configs:
+    enable: false
+  prefix_time: false

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000016/c30_small__simulation_data1__sample_000016_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-644_C12_-38.8_phi12_109_C21_-35.9_phi21_346_C23_643_phi23_271_C30_1.46e+06_C32_4.66e+07_phi56_303/20260624_053806_ptyrad_log.txt

@@ -0,0 +1,417 @@
+2026-06-24 05:38:04,925 - ### PtyRAD LoggingManager configuration ###
+2026-06-24 05:38:04,925 - log_file       = 'ptyrad_log.txt'. If log_file = None, no log file will be created.
+2026-06-24 05:38:04,925 - log_dir        = 'auto'. If log_dir = 'auto', then log will be saved to `output_path` or 'logs/'.
+2026-06-24 05:38:04,925 - flush_file     = True. Automatically set to True if `log_file is not None`
+2026-06-24 05:38:04,925 - prefix_time    = datetime. If true, preset strings ('date', 'time', 'datetime'), or a string of time format, a datetime str is prefixed to the `log_file`.
+2026-06-24 05:38:04,925 - prefix_jobid   = '0'. If not 0, it'll be prefixed to the log file. This is used for hypertune mode with multiple GPUs.
+2026-06-24 05:38:04,925 - append_to_file = True. If true, logs will be appended to the existing file. If false, the log file will be overwritten.
+2026-06-24 05:38:04,925 - show_timestamp = True. If true, the printed information will contain a timestamp.
+2026-06-24 05:38:04,925 -  
+2026-06-24 05:38:05,031 - ### Initializing HuggingFace accelerator ###
+2026-06-24 05:38:05,031 - Accelerator.distributed_type = DistributedType.NO
+2026-06-24 05:38:05,031 - Accelerator.num_process      = 1
+2026-06-24 05:38:05,031 - Accelerator.mixed_precision  = no
+2026-06-24 05:38:05,054 - 'accelerate' is available but NOT using distributed mode or mixed precision
+2026-06-24 05:38:05,054 - If you want to utilize 'accelerate' for multiGPU or mixed precision, 
+2026-06-24 05:38:05,054 - Run `accelerate launch --multi_gpu --num_processes=2 --mixed_precision='no' -m ptyrad run <PTYRAD_ARGUMENTS> --gpuid 'acc'` in your terminal
+2026-06-24 05:38:05,054 -  
+2026-06-24 05:38:05,054 - ### System information ###
+2026-06-24 05:38:05,054 - Platform: Linux-4.18.0-553.69.1.el8_10.x86_64-x86_64-with-glibc2.28
+2026-06-24 05:38:05,054 - Operating System: Linux 4.18.0-553.69.1.el8_10.x86_64
+2026-06-24 05:38:05,054 - OS Version: #1 SMP Thu Aug 7 18:10:00 EDT 2025
+2026-06-24 05:38:05,054 - Machine: x86_64
+2026-06-24 05:38:05,054 - Processor: x86_64
+2026-06-24 05:38:05,054 - Available CPU cores: 8
+2026-06-24 05:38:05,054 - SLURM-Allocated Total Memory: 78.12 GB
+2026-06-24 05:38:05,054 -  
+2026-06-24 05:38:05,054 - ### GPU information ###
+2026-06-24 05:38:05,054 - CUDA Available: True
+2026-06-24 05:38:05,054 - CUDA Version: 13.0
+2026-06-24 05:38:05,055 - Available CUDA GPUs: ['NVIDIA A100 80GB PCIe']
+2026-06-24 05:38:05,055 - CUDA Compute Capability: ['8.0']
+2026-06-24 05:38:05,055 -   INFO: For torch.compile with Triton, you'll need CUDA GPU with Compute Capability >= 7.0.
+2026-06-24 05:38:05,055 -         In addition, Triton does not directly support Windows.
+2026-06-24 05:38:05,055 -         For Windows users, please follow the instruction and download `triton-windows` from https://github.com/woct0rdho/triton-windows.
+2026-06-24 05:38:05,070 - MIG (Multi-Instance GPU) mode = False
+2026-06-24 05:38:05,070 -   INFO: MIG splits a physical GPU into multiple GPU slices, but multiGPU does not support these MIG slices.
+2026-06-24 05:38:05,070 -         In addition, multiGPU is currently only available on Linux due to the limited NCCL support.
+2026-06-24 05:38:05,070 -       -> If you're doing normal reconstruction/hypertune, you can safely ignore this.
+2026-06-24 05:38:05,070 -       -> If you want to do multiGPU, you must provide multiple 'full' GPUs that are not in MIG mode.
+2026-06-24 05:38:05,070 -  
+2026-06-24 05:38:05,070 - ### Python information ###
+2026-06-24 05:38:05,071 - Python Executable: /home/tnguye11/anaconda3/envs/ptyrad/bin/python3.12
+2026-06-24 05:38:05,071 - Python Version: 3.12.13 | packaged by Anaconda, Inc. | (main, Mar 19 2026, 20:20:58) [GCC 14.3.0]
+2026-06-24 05:38:05,071 -  
+2026-06-24 05:38:05,071 - ### Packages information ###
+2026-06-24 05:38:05,072 - Numpy Version (metadata): 2.4.6
+2026-06-24 05:38:05,073 - PyTorch Version (metadata): 2.12.0
+2026-06-24 05:38:05,073 - Optuna Version (metadata): 4.9.0
+2026-06-24 05:38:05,074 - Accelerate Version (metadata): 1.13.0
+2026-06-24 05:38:05,074 - PtyRAD Version (ptyrad/__init__.py): 1.0.0
+2026-06-24 05:38:05,074 - PtyRAD is located at: /home/tnguye11/anaconda3/envs/ptyrad/lib/python3.12/site-packages/ptyrad/__init__.py
+2026-06-24 05:38:05,074 -  
+2026-06-24 05:38:05,074 - ### Loading params file ###
+2026-06-24 05:38:05,075 - params_path = /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/params/c30_small/simulation_data1__sample_000016.yaml
+2026-06-24 05:38:05,081 - validate = True: Filling defaults and validating the params file...
+2026-06-24 05:38:05,082 - Success! Params file validated and defaults applied.
+2026-06-24 05:38:05,082 -  
+2026-06-24 05:38:05,082 - ### Setting GPU Device ###
+2026-06-24 05:38:05,082 - Selected GPU device: cuda:0 (NVIDIA A100 80GB PCIe)
+2026-06-24 05:38:05,082 -  
+2026-06-24 05:38:05,082 - Random seed: 20261860 provided by params file
+2026-06-24 05:38:05,082 - ### Initializing Initializer ###
+2026-06-24 05:38:05,082 - init_params are displayed below:
+2026-06-24 05:38:05,083 -   random_seed: 20261860
+2026-06-24 05:38:05,083 -   probe_illum_type: electron
+2026-06-24 05:38:05,083 -   probe_kv: 300.0
+2026-06-24 05:38:05,083 -   probe_conv_angle: 10.52404499053955
+2026-06-24 05:38:05,083 -   probe_aberrations: {'C10': -644.5, 'C12': -38.844, 'phi12': 109.382, 'C21': -35.942, 'phi21': 346.359, 'C23': 643.489, 'phi23': 270.924, 'C30': 1462981.375, 'C32': 4877.956, 'phi32': 147.219, 'C34': 393.031, 'phi34': 258.074, 'C41': -405787.281, 'phi41': 352.498, 'C43': -96929.461, 'phi43': 47.073, 'C45': 479755.188, 'phi45': 124.592, 'C50': 332991232.0, 'C52': -666431.125, 'phi52': 256.261, 'C54': 6416499.5, 'phi54': 101.013, 'C56': 16632105.0, 'phi56': 302.969}
+2026-06-24 05:38:05,083 -   beam_kev: None
+2026-06-24 05:38:05,083 -   probe_dRn: None
+2026-06-24 05:38:05,083 -   probe_Rn: None
+2026-06-24 05:38:05,083 -   probe_D_H: None
+2026-06-24 05:38:05,083 -   probe_D_FZP: None
+2026-06-24 05:38:05,083 -   probe_Ls: None
+2026-06-24 05:38:05,083 -   meas_Npix: 128
+2026-06-24 05:38:05,083 -   pos_N_scans: 256
+2026-06-24 05:38:05,083 -   pos_N_scan_slow: 16
+2026-06-24 05:38:05,083 -   pos_N_scan_fast: 16
+2026-06-24 05:38:05,083 -   pos_scan_step_size: 0.41889888048171997
+2026-06-24 05:38:05,083 -   meas_calibration: {'mode': 'kMax', 'value': 2.5}
+2026-06-24 05:38:05,083 -   probe_pmode_max: 6
+2026-06-24 05:38:05,083 -   probe_pmode_init_pows: [0.02]
+2026-06-24 05:38:05,083 -   obj_omode_max: 1
+2026-06-24 05:38:05,083 -   obj_omode_init_occu: {'occu_type': 'uniform', 'init_occu': None}
+2026-06-24 05:38:05,083 -   obj_Nlayer: 1
+2026-06-24 05:38:05,083 -   obj_slice_thickness: 20.0
+2026-06-24 05:38:05,083 -   simu_Npix: None
+2026-06-24 05:38:05,083 -   simu_match_mode: None
+2026-06-24 05:38:05,083 -   meas_permute: None
+2026-06-24 05:38:05,083 -   meas_reshape: [256, 128, 128]
+2026-06-24 05:38:05,083 -   meas_flipT: [0, 0, 0]
+2026-06-24 05:38:05,083 -   meas_crop: None
+2026-06-24 05:38:05,083 -   meas_pad: None
+2026-06-24 05:38:05,083 -   meas_resample: None
+2026-06-24 05:38:05,083 -   meas_add_source_size: None
+2026-06-24 05:38:05,083 -   meas_add_detector_blur: None
+2026-06-24 05:38:05,083 -   meas_remove_neg_values: {'mode': 'clip_neg', 'value': None, 'force': False}
+2026-06-24 05:38:05,083 -   meas_normalization: {'mode': 'max_at_one', 'value': None}
+2026-06-24 05:38:05,083 -   meas_add_poisson_noise: None
+2026-06-24 05:38:05,083 -   meas_export: None
+2026-06-24 05:38:05,083 -   probe_permute: None
+2026-06-24 05:38:05,083 -   probe_z_shift: None
+2026-06-24 05:38:05,084 -   probe_normalization: {'mode': 'mean_total_ints', 'value': None}
+2026-06-24 05:38:05,084 -   pos_scan_flipT: None
+2026-06-24 05:38:05,084 -   pos_scan_affine: None
+2026-06-24 05:38:05,084 -   pos_scan_rand_std: 0.15
+2026-06-24 05:38:05,084 -   obj_z_crop: None
+2026-06-24 05:38:05,084 -   obj_z_pad: None
+2026-06-24 05:38:05,084 -   obj_z_resample: None
+2026-06-24 05:38:05,084 -   meas_source: file
+2026-06-24 05:38:05,084 -   meas_params: {'path': '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000016_measurement.h5', 'key': 'measurement', 'shape': None, 'offset': None, 'gap': None, 'selection': None, 'zarr_kwargs': None}
+2026-06-24 05:38:05,084 -   probe_source: simu
+2026-06-24 05:38:05,084 -   probe_params: None
+2026-06-24 05:38:05,084 -   pos_source: simu
+2026-06-24 05:38:05,084 -   pos_params: None
+2026-06-24 05:38:05,084 -   obj_source: simu
+2026-06-24 05:38:05,084 -   obj_params: None
+2026-06-24 05:38:05,084 -   tilt_source: simu
+2026-06-24 05:38:05,084 -   tilt_params: {'tilt_type': 'all', 'init_tilts': [[0, 0]]}
+2026-06-24 05:38:05,084 -  
+2026-06-24 05:38:05,084 - ### Initializing cache ###
+2026-06-24 05:38:05,084 - use_cached_obj   = False
+2026-06-24 05:38:05,084 - use_cached_probe = False
+2026-06-24 05:38:05,084 - use_cached_pos   = False
+2026-06-24 05:38:05,084 -  
+2026-06-24 05:38:05,084 - ### Initializing measurements ###
+2026-06-24 05:38:05,084 - Loading measurements from source = 'file'
+2026-06-24 05:38:05,084 - Detected measurement file type = '.h5'
+2026-06-24 05:38:05,160 - Original measurements dtype is float32, casting to float32 (single precision) for computational efficiency.
+2026-06-24 05:38:05,160 - Imported meausrements shape / dtype = (16, 16, 128, 128), dtype = float32
+2026-06-24 05:38:05,162 - Imported meausrements int. statistics (min, mean, max) = (0.0000, 0.0001, 0.0024)
+2026-06-24 05:38:05,162 - Reshaping measurements to shape = [256, 128, 128]
+2026-06-24 05:38:05,162 - Flipping measurements with [flipud, fliplr, transpose] = [0, 0, 0]
+2026-06-24 05:38:05,164 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:38:05,164 - Normalizing measurements with mode = 'max_at_one' and value = 'None'
+2026-06-24 05:38:05,164 - Normalizing by max of the 2D mean pattern intensity: 0.0017218026
+2026-06-24 05:38:05,165 - meausrements shape / dtype = (256, 128, 128), dtype = float32
+2026-06-24 05:38:05,166 - meausrements int. statistics (min, mean, max) = (0.0000, 0.0354, 1.4042)
+2026-06-24 05:38:05,167 - No negative values found in measurements. Skipping non-neg correction.
+2026-06-24 05:38:05,168 - Pattern total int. statistics (min, mean, max)        = (575.9019, 580.2035, 584.2603), with min/max = 98.6%
+2026-06-24 05:38:05,169 - Global meausrements int. statistics (min, mean, max)  = (0.0000, 0.0354, 1.4042)
+2026-06-24 05:38:05,169 - measurements                              (N, Ky, Kx) = float32, (256, 128, 128)
+2026-06-24 05:38:05,169 -  
+2026-06-24 05:38:05,169 - ### Setting up calibration ###
+2026-06-24 05:38:05,169 - meas_calibration mode = 'kMax', value = 2.5
+2026-06-24 05:38:05,169 - Using loaded raw averaged measurement (before crop/pad/resample) to fit RBF as a part of the meas calibration
+2026-06-24 05:38:05,169 - Radius of fitted bright field disk (RBF) = 13.54 px with meas_Npix = 128
+2026-06-24 05:38:05,169 - Suggested probe_mask_k radius (RBF*2/Npix) > 0.2116
+2026-06-24 05:38:05,169 - Fitting raw averaged measurement with center, radius, and Gaussian blur std as a sanity check
+2026-06-24 05:38:05,169 - Note that the fitted Gaussian blur std (detector blur) would be affected by overlapping Bragg disks
+2026-06-24 05:38:05,389 - Initial guess: center=(63.62, 63.63), radius=13.54, Gaussian blur std=0.50
+2026-06-24 05:38:05,399 - Final fit: center=(63.62, 63.63), radius=13.54, Gaussian blur std=0.57
+2026-06-24 05:38:05,399 - Using init_params, the inferred RBF (conv_angle / 1e3 * Npix * dx / wavelength) = 13.68 px with Npix = 128
+2026-06-24 05:38:05,399 - dx (real space pixel size of probe and object) set to 0.2000 Ang with Npix = 128
+2026-06-24 05:38:05,399 -  
+2026-06-24 05:38:05,399 - ### Setting init_variables dict ###
+2026-06-24 05:38:05,399 - Derived values given input init_params:
+2026-06-24 05:38:05,399 -   kv          = 300.0 kV
+2026-06-24 05:38:05,399 -   wavelength  = 0.0197 Ang
+2026-06-24 05:38:05,399 -   conv_angle  = 10.52404499053955 mrad
+2026-06-24 05:38:05,399 -   Npix        = 128 px
+2026-06-24 05:38:05,399 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:38:05,399 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:38:05,400 -   da          = 0.7690 mrad
+2026-06-24 05:38:05,400 -   angleMax    = 49.2187 mrad
+2026-06-24 05:38:05,400 -   RBF         = 13.6846 px (Inferred from the given calibration, NOT necessarily from the loaded measurement data)
+2026-06-24 05:38:05,400 -   n_alpha     = 4.6768 (# conv_angle)
+2026-06-24 05:38:05,400 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:38:05,400 -   Rayleigh-limited resolution  = 1.1411 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:38:05,400 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:38:05,400 -  
+2026-06-24 05:38:05,400 - ### Initializing probe ###
+2026-06-24 05:38:05,400 - Loading probe from source = 'simu'
+2026-06-24 05:38:05,400 - Using experimental parameters specified by 'init_params' for initial probe simulation.
+2026-06-24 05:38:05,400 - Start simulating STEM probe
+2026-06-24 05:38:05,400 -   kv          = 300.0 kV
+2026-06-24 05:38:05,400 -   wavelength  = 0.0197 Ang
+2026-06-24 05:38:05,400 -   conv_angle  = 10.52404499053955 mrad
+2026-06-24 05:38:05,400 -   Npix        = 128 px
+2026-06-24 05:38:05,400 -   dk          = 0.0391 Ang^-1
+2026-06-24 05:38:05,400 -   kMax        = 2.5000 Ang^-1
+2026-06-24 05:38:05,400 -   alpha_max   = 49.2187 mrad
+2026-06-24 05:38:05,400 -   dx          = 0.2000 Ang, Nyquist-limited dmin = 2*dx = 0.4000 Ang
+2026-06-24 05:38:05,400 -   Rayleigh-limited resolution  = 1.1411 Ang (0.61*lambda/alpha for focused probe )
+2026-06-24 05:38:05,400 -   Real space probe extent = 25.6000 Ang
+2026-06-24 05:38:05,400 - Krivanek Haider     Magnitude        Angle (°)  Description
+2026-06-24 05:38:05,400 - ------------------------------------------------------------------------------------
+2026-06-24 05:38:05,401 - C1,0     C1         -644.5000        -          Defocus (C10 = -df)
+2026-06-24 05:38:05,401 - C1,2     A1         -38.8440         109.38     2-fold astigmatism
+2026-06-24 05:38:05,401 - C2,1     3*B2       -35.9420         346.36     Axial coma
+2026-06-24 05:38:05,401 - C2,3     A2         643.4890         270.92     3-fold astigmatism
+2026-06-24 05:38:05,401 - C3,0     C3         1462981.3750     -          Spherical aberration
+2026-06-24 05:38:05,401 - C3,2     4*S3       4877.9560        147.22     Axial star aberration
+2026-06-24 05:38:05,401 - C3,4     A3         393.0310         258.07     4-fold astigmatism
+2026-06-24 05:38:05,401 - C4,1     4*B4       -405787.2810     352.50     Axial coma(4th)
+2026-06-24 05:38:05,401 - C4,3     4*D4       -96929.4610      47.07      3-lobe aberration
+2026-06-24 05:38:05,401 - C4,5     A4         479755.1880      124.59     5-fold astigmatism
+2026-06-24 05:38:05,401 - C5,0     C5         332991232.0000   -          Spherical aberration (5th)
+2026-06-24 05:38:05,401 - C5,2     6*S5       -666431.1250     256.26     Axial star aberration(5th)
+2026-06-24 05:38:05,401 - C5,4     6*R5       6416499.5000     101.01     4-lobe aberration
+2026-06-24 05:38:05,401 - C5,6     A5         16632105.0000    302.97     6-fold astigmatism
+2026-06-24 05:38:05,404 - Loaded probe shape = (1, 128, 128), dtype = complex128
+2026-06-24 05:38:05,404 - pmode_now: 1 and pmode_max: 6, padding the pmode.
+2026-06-24 05:38:05,404 - Creating 5 new probe modes from the major mode
+2026-06-24 05:38:05,404 - Start making mixed-state STEM probe with 6 incoherent probe modes
+2026-06-24 05:38:05,508 - Relative power of probe modes = [0.9  0.02 0.02 0.02 0.02 0.02]
+2026-06-24 05:38:05,508 - Orthogonalizing 6 pmodes
+2026-06-24 05:38:05,511 - Sorting 6 pmodes by their intensities
+2026-06-24 05:38:05,511 - Normalizing probe intensity with mode = 'mean_total_ints' and value = 'None'
+2026-06-24 05:38:05,512 - sum(|probe_data|**2) = 580.20, while meas_total_ints (min, mean, max) = (575.9019, 580.2035, 584.2603)
+2026-06-24 05:38:05,512 - probe                         (pmode, Ny, Nx) = complex64, (6, 128, 128)
+2026-06-24 05:38:05,512 -  
+2026-06-24 05:38:05,512 - ### Initializing probe positions ###
+2026-06-24 05:38:05,512 - Loading probe positions from source = 'simu'
+2026-06-24 05:38:05,512 - Using experimental parameters specified by 'init_params' (dx, scan_step size, N_scan_slow, N_scan_fast) for initial position simulation.
+2026-06-24 05:38:05,512 - Simulating probe positions with dx = 0.2000, scan_step_size = 0.4189, N_scan_fast = 16, N_scan_slow = 16
+2026-06-24 05:38:05,512 - Applying Gaussian distributed random displacement with std = 0.15 px to scan positions
+2026-06-24 05:38:05,514 - crop_pos                                (N,2) = int16, (256, 2)
+2026-06-24 05:38:05,514 - crop_pos 1st and last px coords (y,x)         = ([16, 16], [48, 48])
+2026-06-24 05:38:05,515 - crop_pos extent (Ang)                         = [6.4 6.4]
+2026-06-24 05:38:05,515 - probe_pos_shifts                        (N,2) = float32, (256, 2)
+2026-06-24 05:38:05,515 -  
+2026-06-24 05:38:05,515 - ### Initializing object ###
+2026-06-24 05:38:05,515 - Loading object from source = 'simu'
+2026-06-24 05:38:05,515 - Using experimental parameters specified by 'init_params' for initial object simulation.
+2026-06-24 05:38:05,516 - omode_now: 1 and omode_max: 1, leaving the omode unchanged.
+2026-06-24 05:38:05,516 - object                    (omode, Nz, Ny, Nx) = complex64, (1, 1, 193, 193)
+2026-06-24 05:38:05,516 - object extent                 (Z, Y, X) (Ang) = [20.  38.6 38.6]
+2026-06-24 05:38:05,517 -  
+2026-06-24 05:38:05,517 - ### Initializing omode_occu from 'uniform' ###
+2026-06-24 05:38:05,517 - omode_occu                            (omode) = float32, (1,)
+2026-06-24 05:38:05,517 -  
+2026-06-24 05:38:05,517 - ### Initializing H (Fresnel propagator) ###
+2026-06-24 05:38:05,517 - Calculating H with probe_shape = (128, 128) px, dx = 0.2000 Ang, slice_thickness = 20.0000 Ang, lambd = 0.0197 Ang
+2026-06-24 05:38:05,517 - H                                    (Ky, Kx) = complex64, (128, 128)
+2026-06-24 05:38:05,517 -  
+2026-06-24 05:38:05,517 - ### Initializing obj tilts from = 'simu' ###
+2026-06-24 05:38:05,517 - Initialized obj_tilts with init_tilts = [[0, 0]] (theta_y, theta_x) mrad
+2026-06-24 05:38:05,517 - obj_tilts                              (N, 2) = float32, (1, 2)
+2026-06-24 05:38:05,518 -  
+2026-06-24 05:38:05,518 - ### Checking consistency between input params with the initialized variables ###
+2026-06-24 05:38:05,518 - meas_Npix, simu_Npix, DP measurements, probe, and H shapes are consistent as '128'
+2026-06-24 05:38:05,518 - N_scans, len(meas), N_scan_slow*N_scan_fast, len(crop_pos), and len(probe_pos_shifts) are consistent as '256'
+2026-06-24 05:38:05,518 - obj.shape[0] is consistent with len(omode_occu) as '1'
+2026-06-24 05:38:05,518 - obj.shape[1] is consistent with Nlayer as '1'
+2026-06-24 05:38:05,518 - crop positions (yx_min=[16 16], yx_max=[176 176]) are well contained inside object canvas (Ny,Nx) = (193, 193).
+2026-06-24 05:38:05,518 - obj_tilts is consistent with either 1 or N_scans
+2026-06-24 05:38:05,518 - Pass the consistency check of initialized variables, initialization is done!
+2026-06-24 05:38:05,518 -  
+2026-06-24 05:38:05,518 - ### Collecting reconstruction provenance ###
+2026-06-24 05:38:05,518 - Reconstruction provenance is collected and initialized.
+2026-06-24 05:38:05,518 -  
+2026-06-24 05:38:05,518 - ### Initializing loss function ###
+2026-06-24 05:38:05,518 - Active loss types:
+2026-06-24 05:38:05,518 -   loss_single : {'state': True, 'weight': 1.0, 'dp_pow': 0.5}
+2026-06-24 05:38:05,518 -  
+2026-06-24 05:38:05,518 - ### Initializing constraint function ###
+2026-06-24 05:38:05,518 - Active constraint types:
+2026-06-24 05:38:05,518 -   ortho_pmode   : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:38:05,518 -   fix_probe_int : {'start_iter': 1, 'step': 1, 'end_iter': None}
+2026-06-24 05:38:05,519 -   obj_zblur     : {'start_iter': 1, 'step': 1, 'end_iter': None, 'obj_type': 'both', 'kernel_size': 5, 'std': 1.0}
+2026-06-24 05:38:05,519 -   obja_thresh   : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0, 'thresh': [0.96, 1.04]}
+2026-06-24 05:38:05,519 -   pos_recenter  : {'start_iter': 1, 'step': 1, 'end_iter': None, 'relax': 0.0}
+2026-06-24 05:38:05,519 -  
+2026-06-24 05:38:05,519 - ### Done initializing PtyRADSolver ###
+2026-06-24 05:38:05,519 -  
+2026-06-24 05:38:05,646 - ### Starting the PtyRADSolver in reconstruct mode ###
+2026-06-24 05:38:05,646 -  
+2026-06-24 05:38:05,646 - ### Initializing PtychoModel model ###
+2026-06-24 05:38:05,717 - ### PtychoModel optimizable variables ###
+2026-06-24 05:38:05,717 - obja            : torch.Size([1, 1, 193, 193])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:38:05,717 - objp            : torch.Size([1, 1, 193, 193])    , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:38:05,718 - obj_tilts       : torch.Size([1, 2])              , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:38:05,718 - slice_thickness : torch.Size([])                  , torch.float32   , device:cuda:0, grad:False, lr:0e+00
+2026-06-24 05:38:05,718 - probe           : torch.Size([6, 128, 128, 2])    , torch.float32   , device:cuda:0, grad:True , lr:1e-04
+2026-06-24 05:38:05,718 - probe_pos_shifts: torch.Size([256, 2])            , torch.float32   , device:cuda:0, grad:True , lr:5e-04
+2026-06-24 05:38:05,718 -  
+2026-06-24 05:38:05,718 - ### Optimizable variables statitsics ###
+2026-06-24 05:38:05,718 - Total measurement values  : 4,194,304
+2026-06-24 05:38:05,718 - Total optimizing variables: 271,618
+2026-06-24 05:38:05,718 - Overdetermined ratio      : 15.44
+2026-06-24 05:38:05,718 -  
+2026-06-24 05:38:05,718 - ### Model behavior ###
+2026-06-24 05:38:05,718 - Tilt propagator           : False
+2026-06-24 05:38:05,718 - Change slice thickness    : False
+2026-06-24 05:38:05,718 - Detector blur             : False
+2026-06-24 05:38:05,718 - Preload data              : True
+2026-06-24 05:38:05,718 - On-the-fly meas padding   : False
+2026-06-24 05:38:05,718 - On-the-fly meas resample  : False
+2026-06-24 05:38:05,718 - On-the-fly simu match mode: None
+2026-06-24 05:38:05,718 -  
+2026-06-24 05:38:05,760 - ### Done initializing PtychoModel model ###
+2026-06-24 05:38:05,760 -  
+2026-06-24 05:38:05,761 - ### Creating PyTorch 'Adam' optimizer with configs = {} ###
+2026-06-24 05:38:05,761 -  
+2026-06-24 05:38:05,761 - ### Generating indices, batches, and output_path ###
+2026-06-24 05:38:05,762 - d90 = 59.000 px or 11.800 Ang
+2026-06-24 05:38:05,762 - Selecting indices with the 'full' mode 
+2026-06-24 05:38:06,220 - Generated 8 'random' groups of ~32 scan positions in 0.000 sec
+2026-06-24 05:38:06,296 - The effective batch size (i.e., how many probe positions are simultaneously used for 1 update of ptychographic parameters) is batch_size * grad_accumulation = 32 * 1 = 32
+2026-06-24 05:38:06,297 - Original recon_dir_affixes = ['default']
+2026-06-24 05:38:06,297 - Expanded recon_dir_affixes = ['indices', 'meas', 'batch', 'pmode', 'omode', 'nlayer', 'lr', 'model', 'constraint', 'loss', 'affine', 'tilt', 'aberrations']
+2026-06-24 05:38:06,303 - Path corrected for compatibility:
+2026-06-24 05:38:06,303 -   Original: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000016/c30_small__simulation_data1__sample_000016_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-644_C12_-38.8_phi12_109_C21_-35.9_phi21_346_C23_643_phi23_271_C30_1.46e+06_C32_4.88e+03_phi32_147_C34_393_phi34_258_C41_-4.06e+05_phi41_352_C43_-9.69e+04_phi43_47.1_C45_4.8e+05_phi45_125_C50_3.33e+08_C52_-6.66e+05_phi52_256_C54_6.42e+06_phi54_101_C56_1.66e+07_phi56_303
+2026-06-24 05:38:06,303 -   Corrected: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000016/c30_small__simulation_data1__sample_000016_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-644_C12_-38.8_phi12_109_C21_-35.9_phi21_346_C23_643_phi23_271_C30_1.46e+06_C32_4.66e+07_phi56_303
+2026-06-24 05:38:06,304 - output_path = '/gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000016/c30_small__simulation_data1__sample_000016_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-644_C12_-38.8_phi12_109_C21_-35.9_phi21_346_C23_643_phi23_271_C30_1.46e+06_C32_4.66e+07_phi56_303' is generated!
+2026-06-24 05:38:06,455 -  
+2026-06-24 05:38:06,456 - ### Log file is flushed (created) as /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000016/c30_small__simulation_data1__sample_000016_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-644_C12_-38.8_phi12_109_C21_-35.9_phi21_346_C23_643_phi23_271_C30_1.46e+06_C32_4.66e+07_phi56_303/20260624_053806_ptyrad_log.txt ###
+2026-06-24 05:38:06,456 -  
+2026-06-24 05:38:06,457 - ### Creating ConvergenceMonitor with {'tensors': ['obja', 'objp', 'probe', 'probe_pos_shifts'], 'every_n_iters': None, 'percentile_range': [15.0, 85.0]} ### 
+2026-06-24 05:38:06,457 - ### Start the PtyRAD iterative ptycho reconstruction ###
+2026-06-24 05:38:06,457 - Setting up PyTorch compiler with {'fullgraph': False, 'dynamic': None, 'backend': 'inductor', 'mode': 'default', 'options': None, 'disable': True}
+2026-06-24 05:38:07,655 - Iter: 1, Total Loss: 0.7657, loss_single: 0.7657, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.436 sec
+2026-06-24 05:38:07,693 - Iter: 2, Total Loss: 0.7309, loss_single: 0.7309, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:07,728 - Iter: 3, Total Loss: 0.7068, loss_single: 0.7068, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:07,764 - Iter: 4, Total Loss: 0.6859, loss_single: 0.6859, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:07,799 - Iter: 5, Total Loss: 0.6643, loss_single: 0.6643, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:07,835 - Iter: 6, Total Loss: 0.6432, loss_single: 0.6432, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:07,870 - Iter: 7, Total Loss: 0.6194, loss_single: 0.6194, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:07,906 - Iter: 8, Total Loss: 0.5964, loss_single: 0.5964, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:07,942 - Iter: 9, Total Loss: 0.5768, loss_single: 0.5768, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:07,978 - Iter: 10, Total Loss: 0.5580, loss_single: 0.5580, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,014 - Iter: 11, Total Loss: 0.5429, loss_single: 0.5429, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:08,050 - Iter: 12, Total Loss: 0.5288, loss_single: 0.5288, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:08,086 - Iter: 13, Total Loss: 0.5124, loss_single: 0.5124, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,121 - Iter: 14, Total Loss: 0.4992, loss_single: 0.4992, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,156 - Iter: 15, Total Loss: 0.4864, loss_single: 0.4864, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,192 - Iter: 16, Total Loss: 0.4715, loss_single: 0.4715, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,227 - Iter: 17, Total Loss: 0.4558, loss_single: 0.4558, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,263 - Iter: 18, Total Loss: 0.4396, loss_single: 0.4396, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,299 - Iter: 19, Total Loss: 0.4248, loss_single: 0.4248, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:08,334 - Iter: 20, Total Loss: 0.4129, loss_single: 0.4129, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,370 - Iter: 21, Total Loss: 0.4028, loss_single: 0.4028, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,406 - Iter: 22, Total Loss: 0.3942, loss_single: 0.3942, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,442 - Iter: 23, Total Loss: 0.3871, loss_single: 0.3871, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:08,478 - Iter: 24, Total Loss: 0.3813, loss_single: 0.3813, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:08,513 - Iter: 25, Total Loss: 0.3763, loss_single: 0.3763, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,548 - Iter: 26, Total Loss: 0.3726, loss_single: 0.3726, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,584 - Iter: 27, Total Loss: 0.3701, loss_single: 0.3701, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,619 - Iter: 28, Total Loss: 0.3679, loss_single: 0.3679, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,655 - Iter: 29, Total Loss: 0.3658, loss_single: 0.3658, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,691 - Iter: 30, Total Loss: 0.3639, loss_single: 0.3639, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:08,730 - Iter: 31, Total Loss: 0.3626, loss_single: 0.3626, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.039 sec
+2026-06-24 05:38:08,765 - Iter: 32, Total Loss: 0.3615, loss_single: 0.3615, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,801 - Iter: 33, Total Loss: 0.3604, loss_single: 0.3604, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,836 - Iter: 34, Total Loss: 0.3593, loss_single: 0.3593, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,871 - Iter: 35, Total Loss: 0.3584, loss_single: 0.3584, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,907 - Iter: 36, Total Loss: 0.3578, loss_single: 0.3578, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,942 - Iter: 37, Total Loss: 0.3571, loss_single: 0.3571, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:08,978 - Iter: 38, Total Loss: 0.3564, loss_single: 0.3564, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,014 - Iter: 39, Total Loss: 0.3560, loss_single: 0.3560, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,050 - Iter: 40, Total Loss: 0.3557, loss_single: 0.3557, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,097 - Iter: 41, Total Loss: 0.3553, loss_single: 0.3553, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,133 - Iter: 42, Total Loss: 0.3550, loss_single: 0.3550, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,168 - Iter: 43, Total Loss: 0.3547, loss_single: 0.3547, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,204 - Iter: 44, Total Loss: 0.3544, loss_single: 0.3544, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,240 - Iter: 45, Total Loss: 0.3542, loss_single: 0.3542, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,275 - Iter: 46, Total Loss: 0.3540, loss_single: 0.3540, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,310 - Iter: 47, Total Loss: 0.3537, loss_single: 0.3537, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,346 - Iter: 48, Total Loss: 0.3535, loss_single: 0.3535, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,381 - Iter: 49, Total Loss: 0.3533, loss_single: 0.3533, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,418 - Iter: 50, Total Loss: 0.3532, loss_single: 0.3532, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:09,453 - Iter: 51, Total Loss: 0.3530, loss_single: 0.3530, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,489 - Iter: 52, Total Loss: 0.3528, loss_single: 0.3528, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,525 - Iter: 53, Total Loss: 0.3527, loss_single: 0.3527, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,560 - Iter: 54, Total Loss: 0.3525, loss_single: 0.3525, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,596 - Iter: 55, Total Loss: 0.3524, loss_single: 0.3524, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,631 - Iter: 56, Total Loss: 0.3522, loss_single: 0.3522, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,667 - Iter: 57, Total Loss: 0.3520, loss_single: 0.3520, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,703 - Iter: 58, Total Loss: 0.3520, loss_single: 0.3520, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,738 - Iter: 59, Total Loss: 0.3519, loss_single: 0.3519, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,774 - Iter: 60, Total Loss: 0.3517, loss_single: 0.3517, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,809 - Iter: 61, Total Loss: 0.3517, loss_single: 0.3517, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,846 - Iter: 62, Total Loss: 0.3516, loss_single: 0.3516, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:09,881 - Iter: 63, Total Loss: 0.3515, loss_single: 0.3515, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,916 - Iter: 64, Total Loss: 0.3514, loss_single: 0.3514, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:09,952 - Iter: 65, Total Loss: 0.3513, loss_single: 0.3513, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:09,988 - Iter: 66, Total Loss: 0.3511, loss_single: 0.3511, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,023 - Iter: 67, Total Loss: 0.3511, loss_single: 0.3511, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,059 - Iter: 68, Total Loss: 0.3510, loss_single: 0.3510, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,095 - Iter: 69, Total Loss: 0.3509, loss_single: 0.3509, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,130 - Iter: 70, Total Loss: 0.3508, loss_single: 0.3508, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,166 - Iter: 71, Total Loss: 0.3507, loss_single: 0.3507, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:10,201 - Iter: 72, Total Loss: 0.3506, loss_single: 0.3506, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,237 - Iter: 73, Total Loss: 0.3505, loss_single: 0.3505, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,273 - Iter: 74, Total Loss: 0.3505, loss_single: 0.3505, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,309 - Iter: 75, Total Loss: 0.3504, loss_single: 0.3504, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,344 - Iter: 76, Total Loss: 0.3503, loss_single: 0.3503, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,380 - Iter: 77, Total Loss: 0.3502, loss_single: 0.3502, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:10,416 - Iter: 78, Total Loss: 0.3502, loss_single: 0.3502, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,451 - Iter: 79, Total Loss: 0.3501, loss_single: 0.3501, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,487 - Iter: 80, Total Loss: 0.3500, loss_single: 0.3500, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,522 - Iter: 81, Total Loss: 0.3499, loss_single: 0.3499, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,558 - Iter: 82, Total Loss: 0.3498, loss_single: 0.3498, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,594 - Iter: 83, Total Loss: 0.3498, loss_single: 0.3498, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:10,629 - Iter: 84, Total Loss: 0.3497, loss_single: 0.3497, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,665 - Iter: 85, Total Loss: 0.3496, loss_single: 0.3496, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,700 - Iter: 86, Total Loss: 0.3495, loss_single: 0.3495, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,736 - Iter: 87, Total Loss: 0.3495, loss_single: 0.3495, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,772 - Iter: 88, Total Loss: 0.3494, loss_single: 0.3494, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:10,808 - Iter: 89, Total Loss: 0.3493, loss_single: 0.3493, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,847 - Iter: 90, Total Loss: 0.3493, loss_single: 0.3493, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,883 - Iter: 91, Total Loss: 0.3492, loss_single: 0.3492, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,920 - Iter: 92, Total Loss: 0.3491, loss_single: 0.3491, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.037 sec
+2026-06-24 05:38:10,956 - Iter: 93, Total Loss: 0.3491, loss_single: 0.3491, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:10,992 - Iter: 94, Total Loss: 0.3490, loss_single: 0.3490, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:11,028 - Iter: 95, Total Loss: 0.3489, loss_single: 0.3489, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:11,064 - Iter: 96, Total Loss: 0.3489, loss_single: 0.3489, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:11,099 - Iter: 97, Total Loss: 0.3488, loss_single: 0.3488, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:11,135 - Iter: 98, Total Loss: 0.3487, loss_single: 0.3487, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:11,171 - Iter: 99, Total Loss: 0.3487, loss_single: 0.3487, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.035 sec
+2026-06-24 05:38:11,207 - Iter: 100, Total Loss: 0.3486, loss_single: 0.3486, loss_poissn: 0.0000, loss_pacbed: 0.0000, loss_sparse: 0.0000, loss_simlar: 0.0000, in 0.036 sec
+2026-06-24 05:38:11,319 - Saving summary figures for iter 100
+2026-06-24 05:38:15,016 - ### Finished 100 iterations, averaged iter_t = 0.039293 with std = 0.040 ###
+2026-06-24 05:38:15,016 -  
+2026-06-24 05:38:15,016 - ### The PtyRADSolver is finished in 9.370 sec ###
+2026-06-24 05:38:15,016 -  

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000016/c30_small__simulation_data1__sample_000016_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-644_C12_-38.8_phi12_109_C21_-35.9_phi21_346_C23_643_phi23_271_C30_1.46e+06_C32_4.66e+07_phi56_303/model_iter0100.hdf5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:219677351b1c561e9a8c36c83653a11c319331ecb9cc160367db4a83b74e42f9
+size 1616040

ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000016/c30_small__simulation_data1__sample_000016_full_N256_dp128_flipT000_random32_p6_1obj_1slice_plr1e-4_oalr5e-4_oplr5e-4_slr5e-4_ozblur1.0_oathr0.96_sng1.0_C10_-644_C12_-38.8_phi12_109_C21_-35.9_phi21_346_C23_643_phi23_271_C30_1.46e+06_C32_4.66e+07_phi56_303/simulation_data1__sample_000016.yaml

@@ -0,0 +1,140 @@
+init_params:
+  probe_kv: 300.0
+  probe_conv_angle: 10.52404499053955
+  probe_aberrations:
+    C10: -644.499755859375
+    C12: -38.8441276550293
+    C21: -35.941925048828125
+    C23: 643.4893798828125
+    C30: 1462981.375
+    C32: 4877.9560546875
+    C34: 393.0309143066406
+    C41: -405787.28125
+    C43: -96929.4609375
+    C45: 479755.1875
+    C50: 332991232.0
+    C52: -666431.125
+    C54: 6416499.5
+    C56: 16632105.0
+    phi12: 109.3821122744934
+    phi21: 346.3588108490766
+    phi23: 270.92390671189463
+    phi32: 147.2187716346686
+    phi34: 258.0738620149994
+    phi41: 352.4983586109743
+    phi43: 47.072581174744435
+    phi45: 124.5918712164121
+    phi52: 256.2613210545748
+    phi54: 101.01286291842332
+    phi56: 302.96935112765357
+  meas_Npix: 128
+  pos_N_scan_slow: 16
+  pos_N_scan_fast: 16
+  pos_scan_step_size: 0.41889888048171997
+  probe_pmode_max: 6
+  obj_Nlayer: 1
+  obj_slice_thickness: 20.0
+  meas_permute: null
+  meas_reshape:
+  - 256
+  - 128
+  - 128
+  meas_flipT:
+  - 0
+  - 0
+  - 0
+  meas_crop: null
+  meas_pad: null
+  meas_resample: null
+  pos_scan_affine: null
+  meas_params:
+    path: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_inputs/c30_small/simulation_data1__sample_000016_measurement.h5
+    key: measurement
+  random_seed: 20261860
+  obj_omode_max: 1
+  meas_calibration:
+    mode: kMax
+    value: 2.5
+model_params:
+  detector_blur_std: null
+  update_params:
+    obja:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    objp:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+    probe:
+      start_iter: 1
+      lr: 0.0001
+      end_iter: null
+    probe_pos_shifts:
+      start_iter: 1
+      lr: 0.0005
+      end_iter: null
+loss_params:
+  loss_single:
+    state: true
+    weight: 1.0
+    dp_pow: 0.5
+  loss_poissn:
+    state: false
+    weight: 1.0
+    dp_pow: 1.0
+    eps: 1.0e-06
+  loss_sparse:
+    state: false
+    weight: 0.1
+    ln_order: 1
+constraint_params:
+  ortho_pmode:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  fix_probe_int:
+    start_iter: 1
+    step: 1
+    end_iter: null
+  obj_rblur:
+    start_iter: null
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 0.4
+  obj_zblur:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    obj_type: both
+    kernel_size: 5
+    std: 1
+  obja_thresh:
+    start_iter: 1
+    step: 1
+    end_iter: null
+    relax: 0
+    thresh:
+    - 0.96
+    - 1.04
+  objp_postiv:
+    start_iter: null
+    step: 1
+    end_iter: null
+    relax: 0
+recon_params:
+  NITER: 100
+  BATCH_SIZE:
+    size: 32
+    grad_accumulation: 1
+  SAVE_ITERS: 100
+  output_dir: /gpfs/scratch/ailab/ai4physic/ptyrad_workspace/ptyrad_eval_test_100samples_seed20260624/ptyrad_output/c30_small/simulation_data1__sample_000016
+  recon_dir_affixes:
+  - default
+  prefix: c30_small__simulation_data1__sample_000016
+  postfix: ''
+  compiler_configs:
+    enable: false
+  prefix_time: false