| { | |
| "best_metric": 0.7608695652173914, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-da-4e-5\\checkpoint-356", | |
| "epoch": 38.26086956521739, | |
| "eval_steps": 500, | |
| "global_step": 440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 1.446, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 1.627474069595337, | |
| "eval_runtime": 0.8182, | |
| "eval_samples_per_second": 56.222, | |
| "eval_steps_per_second": 3.667, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 1.4497, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 1.5550435781478882, | |
| "eval_runtime": 0.6081, | |
| "eval_samples_per_second": 75.641, | |
| "eval_steps_per_second": 4.933, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 1.388, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.32608695652173914, | |
| "eval_loss": 1.3769100904464722, | |
| "eval_runtime": 0.6232, | |
| "eval_samples_per_second": 73.817, | |
| "eval_steps_per_second": 4.814, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 1.2755, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.41304347826086957, | |
| "eval_loss": 1.2483216524124146, | |
| "eval_runtime": 0.6131, | |
| "eval_samples_per_second": 75.024, | |
| "eval_steps_per_second": 4.893, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 3.93939393939394e-05, | |
| "loss": 1.1574, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.1545071601867676, | |
| "eval_runtime": 0.6112, | |
| "eval_samples_per_second": 75.266, | |
| "eval_steps_per_second": 4.909, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 3.838383838383839e-05, | |
| "loss": 1.0826, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0428757667541504, | |
| "eval_runtime": 0.6126, | |
| "eval_samples_per_second": 75.084, | |
| "eval_steps_per_second": 4.897, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 3.7373737373737376e-05, | |
| "loss": 0.9798, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.9124, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_accuracy": 0.5652173913043478, | |
| "eval_loss": 0.9318356513977051, | |
| "eval_runtime": 0.6231, | |
| "eval_samples_per_second": 73.82, | |
| "eval_steps_per_second": 4.814, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 3.535353535353536e-05, | |
| "loss": 0.8228, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5217391304347826, | |
| "eval_loss": 1.0361946821212769, | |
| "eval_runtime": 0.6136, | |
| "eval_samples_per_second": 74.962, | |
| "eval_steps_per_second": 4.889, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 3.434343434343435e-05, | |
| "loss": 0.733, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 0.9698829650878906, | |
| "eval_runtime": 0.6176, | |
| "eval_samples_per_second": 74.477, | |
| "eval_steps_per_second": 4.857, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.7086, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 0.8268952369689941, | |
| "eval_runtime": 0.6242, | |
| "eval_samples_per_second": 73.698, | |
| "eval_steps_per_second": 4.806, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "learning_rate": 3.232323232323232e-05, | |
| "loss": 0.6459, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.8167576789855957, | |
| "eval_runtime": 0.7592, | |
| "eval_samples_per_second": 60.591, | |
| "eval_steps_per_second": 3.952, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 11.3, | |
| "learning_rate": 3.131313131313132e-05, | |
| "loss": 0.5793, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6086956521739131, | |
| "eval_loss": 1.0780104398727417, | |
| "eval_runtime": 0.6607, | |
| "eval_samples_per_second": 69.628, | |
| "eval_steps_per_second": 4.541, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 12.17, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 0.5904, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 1.0166438817977905, | |
| "eval_runtime": 0.6302, | |
| "eval_samples_per_second": 72.998, | |
| "eval_steps_per_second": 4.761, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 2.9292929292929297e-05, | |
| "loss": 0.5866, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 13.91, | |
| "learning_rate": 2.8282828282828285e-05, | |
| "loss": 0.5155, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6304347826086957, | |
| "eval_loss": 0.8489349484443665, | |
| "eval_runtime": 0.6912, | |
| "eval_samples_per_second": 66.555, | |
| "eval_steps_per_second": 4.341, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.4693, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 0.8454135656356812, | |
| "eval_runtime": 0.7302, | |
| "eval_samples_per_second": 63.0, | |
| "eval_steps_per_second": 4.109, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "learning_rate": 2.6262626262626265e-05, | |
| "loss": 0.4928, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.8160586357116699, | |
| "eval_runtime": 0.6577, | |
| "eval_samples_per_second": 69.944, | |
| "eval_steps_per_second": 4.562, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 2.5252525252525253e-05, | |
| "loss": 0.4763, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 16.96, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.7665790319442749, | |
| "eval_runtime": 0.7467, | |
| "eval_samples_per_second": 61.606, | |
| "eval_steps_per_second": 4.018, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "learning_rate": 2.4242424242424244e-05, | |
| "loss": 0.4354, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.8828268647193909, | |
| "eval_runtime": 1.2933, | |
| "eval_samples_per_second": 35.568, | |
| "eval_steps_per_second": 2.32, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 18.26, | |
| "learning_rate": 2.3232323232323232e-05, | |
| "loss": 0.3661, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 18.96, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.8781685829162598, | |
| "eval_runtime": 0.6101, | |
| "eval_samples_per_second": 75.393, | |
| "eval_steps_per_second": 4.917, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "learning_rate": 2.2222222222222227e-05, | |
| "loss": 0.3951, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2.121212121212121e-05, | |
| "loss": 0.3652, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.9417612552642822, | |
| "eval_runtime": 0.6257, | |
| "eval_samples_per_second": 73.523, | |
| "eval_steps_per_second": 4.795, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 20.87, | |
| "learning_rate": 2.0202020202020206e-05, | |
| "loss": 0.3733, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.8962993025779724, | |
| "eval_runtime": 0.6461, | |
| "eval_samples_per_second": 71.191, | |
| "eval_steps_per_second": 4.643, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "learning_rate": 1.9191919191919194e-05, | |
| "loss": 0.3473, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.9052736759185791, | |
| "eval_runtime": 0.6161, | |
| "eval_samples_per_second": 74.659, | |
| "eval_steps_per_second": 4.869, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 22.61, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.2988, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 22.96, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 0.8318140506744385, | |
| "eval_runtime": 0.6121, | |
| "eval_samples_per_second": 75.147, | |
| "eval_steps_per_second": 4.901, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 23.48, | |
| "learning_rate": 1.7171717171717173e-05, | |
| "loss": 0.349, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6086956521739131, | |
| "eval_loss": 1.1128798723220825, | |
| "eval_runtime": 0.6252, | |
| "eval_samples_per_second": 73.582, | |
| "eval_steps_per_second": 4.799, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 24.35, | |
| "learning_rate": 1.616161616161616e-05, | |
| "loss": 0.2963, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 24.96, | |
| "eval_accuracy": 0.6304347826086957, | |
| "eval_loss": 1.055749535560608, | |
| "eval_runtime": 0.6631, | |
| "eval_samples_per_second": 69.366, | |
| "eval_steps_per_second": 4.524, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 25.22, | |
| "learning_rate": 1.5151515151515153e-05, | |
| "loss": 0.3025, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 0.956657886505127, | |
| "eval_runtime": 0.6232, | |
| "eval_samples_per_second": 73.817, | |
| "eval_steps_per_second": 4.814, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 26.09, | |
| "learning_rate": 1.4141414141414143e-05, | |
| "loss": 0.3428, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 26.96, | |
| "learning_rate": 1.3131313131313132e-05, | |
| "loss": 0.2676, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 26.96, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 1.0130687952041626, | |
| "eval_runtime": 0.6477, | |
| "eval_samples_per_second": 71.026, | |
| "eval_steps_per_second": 4.632, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 27.83, | |
| "learning_rate": 1.2121212121212122e-05, | |
| "loss": 0.2848, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.9575807452201843, | |
| "eval_runtime": 1.1023, | |
| "eval_samples_per_second": 41.732, | |
| "eval_steps_per_second": 2.722, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 28.7, | |
| "learning_rate": 1.1111111111111113e-05, | |
| "loss": 0.2757, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 28.96, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.9820958971977234, | |
| "eval_runtime": 0.6177, | |
| "eval_samples_per_second": 74.476, | |
| "eval_steps_per_second": 4.857, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 29.57, | |
| "learning_rate": 1.0101010101010103e-05, | |
| "loss": 0.2564, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.0166358947753906, | |
| "eval_runtime": 0.6301, | |
| "eval_samples_per_second": 73.0, | |
| "eval_steps_per_second": 4.761, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 30.43, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.2635, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 30.96, | |
| "eval_accuracy": 0.7608695652173914, | |
| "eval_loss": 0.9664152264595032, | |
| "eval_runtime": 0.6201, | |
| "eval_samples_per_second": 74.177, | |
| "eval_steps_per_second": 4.838, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 31.3, | |
| "learning_rate": 8.08080808080808e-06, | |
| "loss": 0.2413, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 0.9894470572471619, | |
| "eval_runtime": 0.6182, | |
| "eval_samples_per_second": 74.414, | |
| "eval_steps_per_second": 4.853, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 32.17, | |
| "learning_rate": 7.070707070707071e-06, | |
| "loss": 0.2321, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 32.96, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 1.0272178649902344, | |
| "eval_runtime": 0.6182, | |
| "eval_samples_per_second": 74.414, | |
| "eval_steps_per_second": 4.853, | |
| "step": 379 | |
| }, | |
| { | |
| "epoch": 33.04, | |
| "learning_rate": 6.060606060606061e-06, | |
| "loss": 0.2467, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 33.91, | |
| "learning_rate": 5.0505050505050515e-06, | |
| "loss": 0.2517, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.0312250852584839, | |
| "eval_runtime": 0.6231, | |
| "eval_samples_per_second": 73.82, | |
| "eval_steps_per_second": 4.814, | |
| "step": 391 | |
| }, | |
| { | |
| "epoch": 34.78, | |
| "learning_rate": 4.04040404040404e-06, | |
| "loss": 0.2161, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 34.96, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.0433132648468018, | |
| "eval_runtime": 0.6132, | |
| "eval_samples_per_second": 75.021, | |
| "eval_steps_per_second": 4.893, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 35.65, | |
| "learning_rate": 3.0303030303030305e-06, | |
| "loss": 0.2304, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.0157586336135864, | |
| "eval_runtime": 0.6177, | |
| "eval_samples_per_second": 74.476, | |
| "eval_steps_per_second": 4.857, | |
| "step": 414 | |
| }, | |
| { | |
| "epoch": 36.52, | |
| "learning_rate": 2.02020202020202e-06, | |
| "loss": 0.2194, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 36.96, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 1.0119597911834717, | |
| "eval_runtime": 0.6327, | |
| "eval_samples_per_second": 72.71, | |
| "eval_steps_per_second": 4.742, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 37.39, | |
| "learning_rate": 1.01010101010101e-06, | |
| "loss": 0.2395, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 1.0152524709701538, | |
| "eval_runtime": 0.7022, | |
| "eval_samples_per_second": 65.512, | |
| "eval_steps_per_second": 4.273, | |
| "step": 437 | |
| }, | |
| { | |
| "epoch": 38.26, | |
| "learning_rate": 0.0, | |
| "loss": 0.2199, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 38.26, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 1.0148721933364868, | |
| "eval_runtime": 0.6527, | |
| "eval_samples_per_second": 70.481, | |
| "eval_steps_per_second": 4.597, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 38.26, | |
| "step": 440, | |
| "total_flos": 8.989085534729011e+17, | |
| "train_loss": 0.5370159680193121, | |
| "train_runtime": 595.9621, | |
| "train_samples_per_second": 48.459, | |
| "train_steps_per_second": 0.738 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 8.989085534729011e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |