| { | |
| "best_metric": 0.6521739130434783, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-da2-4e-5\\checkpoint-414", | |
| "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-07, | |
| "loss": 6.6265, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 7.94173526763916, | |
| "eval_runtime": 0.6297, | |
| "eval_samples_per_second": 73.053, | |
| "eval_steps_per_second": 4.764, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 1.8181818181818183e-06, | |
| "loss": 6.1679, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 7.933855056762695, | |
| "eval_runtime": 0.6717, | |
| "eval_samples_per_second": 68.487, | |
| "eval_steps_per_second": 4.467, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 2.7272727272727272e-06, | |
| "loss": 6.8561, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 7.914877891540527, | |
| "eval_runtime": 0.6852, | |
| "eval_samples_per_second": 67.137, | |
| "eval_steps_per_second": 4.379, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 3.6363636363636366e-06, | |
| "loss": 6.399, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 7.862863063812256, | |
| "eval_runtime": 0.6527, | |
| "eval_samples_per_second": 70.477, | |
| "eval_steps_per_second": 4.596, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 6.7961, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 7.740795612335205, | |
| "eval_runtime": 0.8582, | |
| "eval_samples_per_second": 53.601, | |
| "eval_steps_per_second": 3.496, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 6.5019, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 7.4855546951293945, | |
| "eval_runtime": 0.6767, | |
| "eval_samples_per_second": 67.98, | |
| "eval_steps_per_second": 4.433, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 6.363636363636364e-06, | |
| "loss": 6.259, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 7.272727272727273e-06, | |
| "loss": 6.085, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 7.138808727264404, | |
| "eval_runtime": 0.6872, | |
| "eval_samples_per_second": 66.943, | |
| "eval_steps_per_second": 4.366, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 8.181818181818183e-06, | |
| "loss": 6.014, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 6.7123517990112305, | |
| "eval_runtime": 0.6777, | |
| "eval_samples_per_second": 67.88, | |
| "eval_steps_per_second": 4.427, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 5.6147, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 6.244385242462158, | |
| "eval_runtime": 0.6992, | |
| "eval_samples_per_second": 65.794, | |
| "eval_steps_per_second": 4.291, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 1e-05, | |
| "loss": 5.2206, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 5.69525146484375, | |
| "eval_runtime": 0.6151, | |
| "eval_samples_per_second": 74.78, | |
| "eval_steps_per_second": 4.877, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 4.9027, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 5.129998683929443, | |
| "eval_runtime": 0.6845, | |
| "eval_samples_per_second": 67.204, | |
| "eval_steps_per_second": 4.383, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 11.3, | |
| "learning_rate": 1.181818181818182e-05, | |
| "loss": 4.4278, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 4.462917327880859, | |
| "eval_runtime": 0.6892, | |
| "eval_samples_per_second": 66.743, | |
| "eval_steps_per_second": 4.353, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 12.17, | |
| "learning_rate": 1.2727272727272728e-05, | |
| "loss": 4.0841, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 3.8089237213134766, | |
| "eval_runtime": 0.7022, | |
| "eval_samples_per_second": 65.507, | |
| "eval_steps_per_second": 4.272, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 1.3636363636363637e-05, | |
| "loss": 3.6313, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 13.91, | |
| "learning_rate": 1.4545454545454546e-05, | |
| "loss": 3.0004, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 3.089322328567505, | |
| "eval_runtime": 0.7019, | |
| "eval_samples_per_second": 65.54, | |
| "eval_steps_per_second": 4.274, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 1.5454545454545454e-05, | |
| "loss": 2.5654, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 2.4630162715911865, | |
| "eval_runtime": 0.6591, | |
| "eval_samples_per_second": 69.793, | |
| "eval_steps_per_second": 4.552, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "learning_rate": 1.6363636363636366e-05, | |
| "loss": 2.094, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 1.8911186456680298, | |
| "eval_runtime": 0.6292, | |
| "eval_samples_per_second": 73.114, | |
| "eval_steps_per_second": 4.768, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 1.7272727272727274e-05, | |
| "loss": 1.8348, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 16.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 1.5454320907592773, | |
| "eval_runtime": 0.6571, | |
| "eval_samples_per_second": 70.0, | |
| "eval_steps_per_second": 4.565, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 1.5383, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 1.412402629852295, | |
| "eval_runtime": 0.6671, | |
| "eval_samples_per_second": 68.95, | |
| "eval_steps_per_second": 4.497, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 18.26, | |
| "learning_rate": 1.9090909090909094e-05, | |
| "loss": 1.4068, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 18.96, | |
| "eval_accuracy": 0.10869565217391304, | |
| "eval_loss": 1.4348046779632568, | |
| "eval_runtime": 0.6922, | |
| "eval_samples_per_second": 66.457, | |
| "eval_steps_per_second": 4.334, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4019, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2.090909090909091e-05, | |
| "loss": 1.379, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.15217391304347827, | |
| "eval_loss": 1.4207279682159424, | |
| "eval_runtime": 0.7142, | |
| "eval_samples_per_second": 64.407, | |
| "eval_steps_per_second": 4.2, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 20.87, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 1.3552, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "eval_accuracy": 0.34782608695652173, | |
| "eval_loss": 1.3785516023635864, | |
| "eval_runtime": 0.6411, | |
| "eval_samples_per_second": 71.747, | |
| "eval_steps_per_second": 4.679, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "learning_rate": 2.2727272727272733e-05, | |
| "loss": 1.262, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.4782608695652174, | |
| "eval_loss": 1.3579806089401245, | |
| "eval_runtime": 0.7197, | |
| "eval_samples_per_second": 63.915, | |
| "eval_steps_per_second": 4.168, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 22.61, | |
| "learning_rate": 2.363636363636364e-05, | |
| "loss": 1.2235, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 22.96, | |
| "eval_accuracy": 0.3695652173913043, | |
| "eval_loss": 1.4412944316864014, | |
| "eval_runtime": 0.7497, | |
| "eval_samples_per_second": 61.359, | |
| "eval_steps_per_second": 4.002, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 23.48, | |
| "learning_rate": 2.454545454545455e-05, | |
| "loss": 1.2065, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.1956521739130435, | |
| "eval_loss": 1.301337718963623, | |
| "eval_runtime": 0.7562, | |
| "eval_samples_per_second": 60.832, | |
| "eval_steps_per_second": 3.967, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 24.35, | |
| "learning_rate": 2.5454545454545457e-05, | |
| "loss": 1.2196, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 24.96, | |
| "eval_accuracy": 0.2608695652173913, | |
| "eval_loss": 1.3489079475402832, | |
| "eval_runtime": 0.6542, | |
| "eval_samples_per_second": 70.32, | |
| "eval_steps_per_second": 4.586, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 25.22, | |
| "learning_rate": 2.6363636363636365e-05, | |
| "loss": 1.1815, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.30434782608695654, | |
| "eval_loss": 1.315794587135315, | |
| "eval_runtime": 0.8187, | |
| "eval_samples_per_second": 56.187, | |
| "eval_steps_per_second": 3.664, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 26.09, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 1.0654, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 26.96, | |
| "learning_rate": 2.8181818181818185e-05, | |
| "loss": 1.0469, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 26.96, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2959192991256714, | |
| "eval_runtime": 0.6638, | |
| "eval_samples_per_second": 69.298, | |
| "eval_steps_per_second": 4.519, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 27.83, | |
| "learning_rate": 2.9090909090909093e-05, | |
| "loss": 0.9854, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5217391304347826, | |
| "eval_loss": 1.4711965322494507, | |
| "eval_runtime": 0.6942, | |
| "eval_samples_per_second": 66.268, | |
| "eval_steps_per_second": 4.322, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 28.7, | |
| "learning_rate": 3.0000000000000004e-05, | |
| "loss": 0.9443, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 28.96, | |
| "eval_accuracy": 0.32608695652173914, | |
| "eval_loss": 1.2060942649841309, | |
| "eval_runtime": 0.6631, | |
| "eval_samples_per_second": 69.366, | |
| "eval_steps_per_second": 4.524, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 29.57, | |
| "learning_rate": 3.090909090909091e-05, | |
| "loss": 0.9168, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.43478260869565216, | |
| "eval_loss": 1.0654621124267578, | |
| "eval_runtime": 0.6752, | |
| "eval_samples_per_second": 68.131, | |
| "eval_steps_per_second": 4.443, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 30.43, | |
| "learning_rate": 3.181818181818182e-05, | |
| "loss": 0.8138, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 30.96, | |
| "eval_accuracy": 0.5434782608695652, | |
| "eval_loss": 1.098128318786621, | |
| "eval_runtime": 0.6792, | |
| "eval_samples_per_second": 67.731, | |
| "eval_steps_per_second": 4.417, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 31.3, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.8066, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.6304347826086957, | |
| "eval_loss": 1.002252221107483, | |
| "eval_runtime": 0.7928, | |
| "eval_samples_per_second": 58.019, | |
| "eval_steps_per_second": 3.784, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 32.17, | |
| "learning_rate": 3.363636363636364e-05, | |
| "loss": 0.8035, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 32.96, | |
| "eval_accuracy": 0.5652173913043478, | |
| "eval_loss": 0.9529276490211487, | |
| "eval_runtime": 0.8117, | |
| "eval_samples_per_second": 56.672, | |
| "eval_steps_per_second": 3.696, | |
| "step": 379 | |
| }, | |
| { | |
| "epoch": 33.04, | |
| "learning_rate": 3.454545454545455e-05, | |
| "loss": 0.7924, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 33.91, | |
| "learning_rate": 3.545454545454546e-05, | |
| "loss": 0.7693, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0103802680969238, | |
| "eval_runtime": 0.7052, | |
| "eval_samples_per_second": 65.228, | |
| "eval_steps_per_second": 4.254, | |
| "step": 391 | |
| }, | |
| { | |
| "epoch": 34.78, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.6773, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 34.96, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0684396028518677, | |
| "eval_runtime": 0.7032, | |
| "eval_samples_per_second": 65.417, | |
| "eval_steps_per_second": 4.266, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 35.65, | |
| "learning_rate": 3.7272727272727276e-05, | |
| "loss": 0.6726, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 0.8252102136611938, | |
| "eval_runtime": 0.6251, | |
| "eval_samples_per_second": 73.584, | |
| "eval_steps_per_second": 4.799, | |
| "step": 414 | |
| }, | |
| { | |
| "epoch": 36.52, | |
| "learning_rate": 3.818181818181819e-05, | |
| "loss": 0.5859, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 36.96, | |
| "eval_accuracy": 0.6304347826086957, | |
| "eval_loss": 0.8157598972320557, | |
| "eval_runtime": 0.7572, | |
| "eval_samples_per_second": 60.753, | |
| "eval_steps_per_second": 3.962, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 37.39, | |
| "learning_rate": 3.909090909090909e-05, | |
| "loss": 0.6204, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.6086956521739131, | |
| "eval_loss": 0.8518310189247131, | |
| "eval_runtime": 0.8152, | |
| "eval_samples_per_second": 56.426, | |
| "eval_steps_per_second": 3.68, | |
| "step": 437 | |
| }, | |
| { | |
| "epoch": 38.26, | |
| "learning_rate": 0.0, | |
| "loss": 0.5921, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 38.26, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 0.8756792545318604, | |
| "eval_runtime": 0.7577, | |
| "eval_samples_per_second": 60.709, | |
| "eval_steps_per_second": 3.959, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 38.26, | |
| "step": 440, | |
| "total_flos": 8.976721800462336e+17, | |
| "train_loss": 2.735191390731118, | |
| "train_runtime": 613.0869, | |
| "train_samples_per_second": 47.041, | |
| "train_steps_per_second": 0.718 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 8.976721800462336e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |