| { | |
| "best_metric": 1.905003309249878, | |
| "best_model_checkpoint": "detr-r50-finetuned-mist1-gb-8ah-6l/checkpoint-5290", | |
| "epoch": 50.0, | |
| "eval_steps": 500, | |
| "global_step": 5750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.812173913043479e-06, | |
| "loss": 2.5222, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.2562553882598877, | |
| "eval_runtime": 5.7546, | |
| "eval_samples_per_second": 6.951, | |
| "eval_steps_per_second": 0.869, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.612173913043479e-06, | |
| "loss": 2.3827, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.2210755348205566, | |
| "eval_runtime": 5.7395, | |
| "eval_samples_per_second": 6.969, | |
| "eval_steps_per_second": 0.871, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.412173913043479e-06, | |
| "loss": 2.3441, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.2602248191833496, | |
| "eval_runtime": 5.7242, | |
| "eval_samples_per_second": 6.988, | |
| "eval_steps_per_second": 0.873, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 9.21217391304348e-06, | |
| "loss": 2.2896, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.2359230518341064, | |
| "eval_runtime": 5.684, | |
| "eval_samples_per_second": 7.037, | |
| "eval_steps_per_second": 0.88, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 9.013913043478261e-06, | |
| "loss": 2.2828, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.2430644035339355, | |
| "eval_runtime": 5.7198, | |
| "eval_samples_per_second": 6.993, | |
| "eval_steps_per_second": 0.874, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 8.813913043478261e-06, | |
| "loss": 2.2972, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.1629228591918945, | |
| "eval_runtime": 5.6906, | |
| "eval_samples_per_second": 7.029, | |
| "eval_steps_per_second": 0.879, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 8.615652173913043e-06, | |
| "loss": 2.3007, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.1544721126556396, | |
| "eval_runtime": 5.7423, | |
| "eval_samples_per_second": 6.966, | |
| "eval_steps_per_second": 0.871, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 8.417391304347827e-06, | |
| "loss": 2.2951, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.115345001220703, | |
| "eval_runtime": 5.7472, | |
| "eval_samples_per_second": 6.96, | |
| "eval_steps_per_second": 0.87, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 8.217391304347827e-06, | |
| "loss": 2.2595, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.1553213596343994, | |
| "eval_runtime": 5.6778, | |
| "eval_samples_per_second": 7.045, | |
| "eval_steps_per_second": 0.881, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 8.017391304347828e-06, | |
| "loss": 2.2327, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.205960750579834, | |
| "eval_runtime": 5.7224, | |
| "eval_samples_per_second": 6.99, | |
| "eval_steps_per_second": 0.874, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 7.817391304347826e-06, | |
| "loss": 2.2023, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 2.045210599899292, | |
| "eval_runtime": 5.6947, | |
| "eval_samples_per_second": 7.024, | |
| "eval_steps_per_second": 0.878, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 7.617391304347826e-06, | |
| "loss": 2.2117, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 2.087853193283081, | |
| "eval_runtime": 5.7626, | |
| "eval_samples_per_second": 6.941, | |
| "eval_steps_per_second": 0.868, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 7.417391304347827e-06, | |
| "loss": 2.1805, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 2.1812005043029785, | |
| "eval_runtime": 5.7549, | |
| "eval_samples_per_second": 6.951, | |
| "eval_steps_per_second": 0.869, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 7.217391304347827e-06, | |
| "loss": 2.1344, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 2.0991523265838623, | |
| "eval_runtime": 5.7805, | |
| "eval_samples_per_second": 6.92, | |
| "eval_steps_per_second": 0.865, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 7.017391304347827e-06, | |
| "loss": 2.1057, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 1.983435869216919, | |
| "eval_runtime": 5.7113, | |
| "eval_samples_per_second": 7.004, | |
| "eval_steps_per_second": 0.875, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 6.817391304347826e-06, | |
| "loss": 2.086, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.9609792232513428, | |
| "eval_runtime": 5.7575, | |
| "eval_samples_per_second": 6.947, | |
| "eval_steps_per_second": 0.868, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 6.617391304347827e-06, | |
| "loss": 2.0591, | |
| "step": 1955 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 2.100736141204834, | |
| "eval_runtime": 5.7633, | |
| "eval_samples_per_second": 6.94, | |
| "eval_steps_per_second": 0.868, | |
| "step": 1955 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 6.417391304347827e-06, | |
| "loss": 2.053, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 2.056126832962036, | |
| "eval_runtime": 5.7709, | |
| "eval_samples_per_second": 6.931, | |
| "eval_steps_per_second": 0.866, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 6.217391304347826e-06, | |
| "loss": 2.0387, | |
| "step": 2185 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 2.0596375465393066, | |
| "eval_runtime": 5.7884, | |
| "eval_samples_per_second": 6.91, | |
| "eval_steps_per_second": 0.864, | |
| "step": 2185 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 6.0173913043478264e-06, | |
| "loss": 2.0161, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.9885139465332031, | |
| "eval_runtime": 5.7465, | |
| "eval_samples_per_second": 6.961, | |
| "eval_steps_per_second": 0.87, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 5.817391304347827e-06, | |
| "loss": 2.0374, | |
| "step": 2415 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 2.0041000843048096, | |
| "eval_runtime": 5.7421, | |
| "eval_samples_per_second": 6.966, | |
| "eval_steps_per_second": 0.871, | |
| "step": 2415 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 5.617391304347827e-06, | |
| "loss": 2.0233, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 2.0102856159210205, | |
| "eval_runtime": 5.7047, | |
| "eval_samples_per_second": 7.012, | |
| "eval_steps_per_second": 0.876, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 5.417391304347826e-06, | |
| "loss": 2.0363, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 2.0540664196014404, | |
| "eval_runtime": 5.7156, | |
| "eval_samples_per_second": 6.998, | |
| "eval_steps_per_second": 0.875, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 5.2173913043478265e-06, | |
| "loss": 1.9837, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 1.9924190044403076, | |
| "eval_runtime": 5.6809, | |
| "eval_samples_per_second": 7.041, | |
| "eval_steps_per_second": 0.88, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 5.017391304347826e-06, | |
| "loss": 1.9943, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 2.0557620525360107, | |
| "eval_runtime": 5.7087, | |
| "eval_samples_per_second": 7.007, | |
| "eval_steps_per_second": 0.876, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 4.817391304347827e-06, | |
| "loss": 1.9846, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 1.9873688220977783, | |
| "eval_runtime": 5.6682, | |
| "eval_samples_per_second": 7.057, | |
| "eval_steps_per_second": 0.882, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 4.617391304347826e-06, | |
| "loss": 1.9601, | |
| "step": 3105 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 1.9554007053375244, | |
| "eval_runtime": 5.7979, | |
| "eval_samples_per_second": 6.899, | |
| "eval_steps_per_second": 0.862, | |
| "step": 3105 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 4.4173913043478265e-06, | |
| "loss": 1.9837, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 1.9988619089126587, | |
| "eval_runtime": 5.7796, | |
| "eval_samples_per_second": 6.921, | |
| "eval_steps_per_second": 0.865, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 4.217391304347827e-06, | |
| "loss": 1.9664, | |
| "step": 3335 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 1.9875919818878174, | |
| "eval_runtime": 5.7433, | |
| "eval_samples_per_second": 6.965, | |
| "eval_steps_per_second": 0.871, | |
| "step": 3335 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 4.017391304347826e-06, | |
| "loss": 1.966, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 1.9754610061645508, | |
| "eval_runtime": 5.8653, | |
| "eval_samples_per_second": 6.82, | |
| "eval_steps_per_second": 0.852, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "learning_rate": 3.819130434782609e-06, | |
| "loss": 1.9226, | |
| "step": 3565 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_loss": 1.9357328414916992, | |
| "eval_runtime": 5.765, | |
| "eval_samples_per_second": 6.938, | |
| "eval_steps_per_second": 0.867, | |
| "step": 3565 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 3.6191304347826088e-06, | |
| "loss": 1.9405, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 1.9239734411239624, | |
| "eval_runtime": 5.8194, | |
| "eval_samples_per_second": 6.874, | |
| "eval_steps_per_second": 0.859, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "learning_rate": 3.4191304347826086e-06, | |
| "loss": 1.9035, | |
| "step": 3795 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_loss": 1.9410585165023804, | |
| "eval_runtime": 5.8097, | |
| "eval_samples_per_second": 6.885, | |
| "eval_steps_per_second": 0.861, | |
| "step": 3795 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 3.219130434782609e-06, | |
| "loss": 1.8924, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 1.9291362762451172, | |
| "eval_runtime": 5.8014, | |
| "eval_samples_per_second": 6.895, | |
| "eval_steps_per_second": 0.862, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "learning_rate": 3.019130434782609e-06, | |
| "loss": 1.8801, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_loss": 1.9660656452178955, | |
| "eval_runtime": 5.7747, | |
| "eval_samples_per_second": 6.927, | |
| "eval_steps_per_second": 0.866, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 2.819130434782609e-06, | |
| "loss": 1.8698, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 1.9104881286621094, | |
| "eval_runtime": 5.7592, | |
| "eval_samples_per_second": 6.945, | |
| "eval_steps_per_second": 0.868, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "learning_rate": 2.619130434782609e-06, | |
| "loss": 1.8572, | |
| "step": 4255 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_loss": 1.944820761680603, | |
| "eval_runtime": 5.7796, | |
| "eval_samples_per_second": 6.921, | |
| "eval_steps_per_second": 0.865, | |
| "step": 4255 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 2.419130434782609e-06, | |
| "loss": 1.8756, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 1.9674819707870483, | |
| "eval_runtime": 5.7301, | |
| "eval_samples_per_second": 6.981, | |
| "eval_steps_per_second": 0.873, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "learning_rate": 2.219130434782609e-06, | |
| "loss": 1.8593, | |
| "step": 4485 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_loss": 1.9364864826202393, | |
| "eval_runtime": 5.8116, | |
| "eval_samples_per_second": 6.883, | |
| "eval_steps_per_second": 0.86, | |
| "step": 4485 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 2.019130434782609e-06, | |
| "loss": 1.8713, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 1.9382976293563843, | |
| "eval_runtime": 5.7132, | |
| "eval_samples_per_second": 7.001, | |
| "eval_steps_per_second": 0.875, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "learning_rate": 1.8191304347826088e-06, | |
| "loss": 1.8436, | |
| "step": 4715 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_loss": 1.967057466506958, | |
| "eval_runtime": 5.7284, | |
| "eval_samples_per_second": 6.983, | |
| "eval_steps_per_second": 0.873, | |
| "step": 4715 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "learning_rate": 1.6191304347826088e-06, | |
| "loss": 1.83, | |
| "step": 4830 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 1.9526548385620117, | |
| "eval_runtime": 5.6918, | |
| "eval_samples_per_second": 7.028, | |
| "eval_steps_per_second": 0.878, | |
| "step": 4830 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "learning_rate": 1.4191304347826089e-06, | |
| "loss": 1.857, | |
| "step": 4945 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_loss": 1.944758653640747, | |
| "eval_runtime": 5.7519, | |
| "eval_samples_per_second": 6.954, | |
| "eval_steps_per_second": 0.869, | |
| "step": 4945 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "learning_rate": 1.2191304347826089e-06, | |
| "loss": 1.8318, | |
| "step": 5060 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_loss": 1.9366220235824585, | |
| "eval_runtime": 5.7436, | |
| "eval_samples_per_second": 6.964, | |
| "eval_steps_per_second": 0.871, | |
| "step": 5060 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 1.0191304347826089e-06, | |
| "loss": 1.8177, | |
| "step": 5175 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_loss": 1.9388927221298218, | |
| "eval_runtime": 5.8021, | |
| "eval_samples_per_second": 6.894, | |
| "eval_steps_per_second": 0.862, | |
| "step": 5175 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "learning_rate": 8.191304347826088e-07, | |
| "loss": 1.8034, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_loss": 1.905003309249878, | |
| "eval_runtime": 5.7813, | |
| "eval_samples_per_second": 6.919, | |
| "eval_steps_per_second": 0.865, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "learning_rate": 6.191304347826088e-07, | |
| "loss": 1.8226, | |
| "step": 5405 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_loss": 1.9226171970367432, | |
| "eval_runtime": 5.8014, | |
| "eval_samples_per_second": 6.895, | |
| "eval_steps_per_second": 0.862, | |
| "step": 5405 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 4.1913043478260874e-07, | |
| "loss": 1.818, | |
| "step": 5520 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 1.9150111675262451, | |
| "eval_runtime": 5.7701, | |
| "eval_samples_per_second": 6.932, | |
| "eval_steps_per_second": 0.867, | |
| "step": 5520 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "learning_rate": 2.191304347826087e-07, | |
| "loss": 1.8148, | |
| "step": 5635 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_loss": 1.9168732166290283, | |
| "eval_runtime": 5.7338, | |
| "eval_samples_per_second": 6.976, | |
| "eval_steps_per_second": 0.872, | |
| "step": 5635 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 1.91304347826087e-08, | |
| "loss": 1.7984, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 1.9223819971084595, | |
| "eval_runtime": 5.7595, | |
| "eval_samples_per_second": 6.945, | |
| "eval_steps_per_second": 0.868, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 5750, | |
| "total_flos": 1.098949102848e+19, | |
| "train_loss": 2.026795845363451, | |
| "train_runtime": 4682.5898, | |
| "train_samples_per_second": 4.912, | |
| "train_steps_per_second": 1.228 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5750, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1.098949102848e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |