| { | |
| "best_metric": 0.8043478260869565, | |
| "best_model_checkpoint": "beit-base-patch16-224-DMAE-5e-5\\checkpoint-98", | |
| "epoch": 34.285714285714285, | |
| "eval_steps": 500, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.381615161895752, | |
| "eval_runtime": 0.7932, | |
| "eval_samples_per_second": 57.995, | |
| "eval_steps_per_second": 3.782, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.3503844738006592, | |
| "eval_runtime": 0.8112, | |
| "eval_samples_per_second": 56.707, | |
| "eval_steps_per_second": 3.698, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.3654, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.307466745376587, | |
| "eval_runtime": 0.7822, | |
| "eval_samples_per_second": 58.809, | |
| "eval_steps_per_second": 3.835, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.244441032409668, | |
| "eval_runtime": 0.7842, | |
| "eval_samples_per_second": 58.66, | |
| "eval_steps_per_second": 3.826, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2088861465454102, | |
| "eval_runtime": 0.7852, | |
| "eval_samples_per_second": 58.586, | |
| "eval_steps_per_second": 3.821, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.2386, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.1710909605026245, | |
| "eval_runtime": 0.7777, | |
| "eval_samples_per_second": 59.15, | |
| "eval_steps_per_second": 3.858, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.14915132522583, | |
| "eval_runtime": 0.8062, | |
| "eval_samples_per_second": 57.059, | |
| "eval_steps_per_second": 3.721, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4782608695652174, | |
| "eval_loss": 1.1050454378128052, | |
| "eval_runtime": 0.7812, | |
| "eval_samples_per_second": 58.886, | |
| "eval_steps_per_second": 3.84, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.132, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.0928192138671875, | |
| "eval_runtime": 0.7982, | |
| "eval_samples_per_second": 57.631, | |
| "eval_steps_per_second": 3.759, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 1.0229545831680298, | |
| "eval_runtime": 0.8342, | |
| "eval_samples_per_second": 55.144, | |
| "eval_steps_per_second": 3.596, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 0.9864158034324646, | |
| "eval_runtime": 0.7922, | |
| "eval_samples_per_second": 58.067, | |
| "eval_steps_per_second": 3.787, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 1.015, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 1.005975604057312, | |
| "eval_runtime": 0.7792, | |
| "eval_samples_per_second": 59.037, | |
| "eval_steps_per_second": 3.85, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.8789530396461487, | |
| "eval_runtime": 1.0307, | |
| "eval_samples_per_second": 44.628, | |
| "eval_steps_per_second": 2.911, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.853344738483429, | |
| "eval_runtime": 0.9702, | |
| "eval_samples_per_second": 47.412, | |
| "eval_steps_per_second": 3.092, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.9217, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 14.86, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.8203216791152954, | |
| "eval_runtime": 0.8002, | |
| "eval_samples_per_second": 57.487, | |
| "eval_steps_per_second": 3.749, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.8151666522026062, | |
| "eval_runtime": 0.8012, | |
| "eval_samples_per_second": 57.414, | |
| "eval_steps_per_second": 3.744, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 16.86, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 0.8598169088363647, | |
| "eval_runtime": 0.7827, | |
| "eval_samples_per_second": 58.772, | |
| "eval_steps_per_second": 3.833, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 17.14, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.7989, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.7979407906532288, | |
| "eval_runtime": 0.7862, | |
| "eval_samples_per_second": 58.511, | |
| "eval_steps_per_second": 3.816, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 18.86, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.7554780840873718, | |
| "eval_runtime": 0.7872, | |
| "eval_samples_per_second": 58.437, | |
| "eval_steps_per_second": 3.811, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.7309, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7608695652173914, | |
| "eval_loss": 0.7019062042236328, | |
| "eval_runtime": 0.8452, | |
| "eval_samples_per_second": 54.425, | |
| "eval_steps_per_second": 3.549, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.86, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.7253907322883606, | |
| "eval_runtime": 0.8162, | |
| "eval_samples_per_second": 56.359, | |
| "eval_steps_per_second": 3.676, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.7267776131629944, | |
| "eval_runtime": 0.7792, | |
| "eval_samples_per_second": 59.037, | |
| "eval_steps_per_second": 3.85, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 22.86, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.6267, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 22.86, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.7095625400543213, | |
| "eval_runtime": 0.7832, | |
| "eval_samples_per_second": 58.735, | |
| "eval_steps_per_second": 3.831, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.7568075060844421, | |
| "eval_runtime": 0.7902, | |
| "eval_samples_per_second": 58.215, | |
| "eval_steps_per_second": 3.797, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 24.86, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 0.6565390825271606, | |
| "eval_runtime": 0.7832, | |
| "eval_samples_per_second": 58.735, | |
| "eval_steps_per_second": 3.831, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 25.71, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.564, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 0.703463077545166, | |
| "eval_runtime": 0.8392, | |
| "eval_samples_per_second": 54.815, | |
| "eval_steps_per_second": 3.575, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 26.86, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 0.7081196904182434, | |
| "eval_runtime": 0.7792, | |
| "eval_samples_per_second": 59.037, | |
| "eval_steps_per_second": 3.85, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8043478260869565, | |
| "eval_loss": 0.6235552430152893, | |
| "eval_runtime": 0.7842, | |
| "eval_samples_per_second": 58.66, | |
| "eval_steps_per_second": 3.826, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.5625, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 28.86, | |
| "eval_accuracy": 0.8043478260869565, | |
| "eval_loss": 0.6090804934501648, | |
| "eval_runtime": 0.7902, | |
| "eval_samples_per_second": 58.214, | |
| "eval_steps_per_second": 3.797, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.7608695652173914, | |
| "eval_loss": 0.652334988117218, | |
| "eval_runtime": 0.7812, | |
| "eval_samples_per_second": 58.886, | |
| "eval_steps_per_second": 3.84, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 30.86, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 0.6235449910163879, | |
| "eval_runtime": 0.7912, | |
| "eval_samples_per_second": 58.141, | |
| "eval_steps_per_second": 3.792, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 31.43, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.5185, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 0.5974029898643494, | |
| "eval_runtime": 0.7762, | |
| "eval_samples_per_second": 59.265, | |
| "eval_steps_per_second": 3.865, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 32.86, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 0.5943583846092224, | |
| "eval_runtime": 0.7887, | |
| "eval_samples_per_second": 58.324, | |
| "eval_steps_per_second": 3.804, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 0.6004733443260193, | |
| "eval_runtime": 0.7862, | |
| "eval_samples_per_second": 58.511, | |
| "eval_steps_per_second": 3.816, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 34.29, | |
| "learning_rate": 0.0, | |
| "loss": 0.5005, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 34.29, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 0.6005403399467468, | |
| "eval_runtime": 0.7812, | |
| "eval_samples_per_second": 58.886, | |
| "eval_steps_per_second": 3.84, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 34.29, | |
| "step": 120, | |
| "total_flos": 5.659246428298445e+17, | |
| "train_loss": 0.8312262058258056, | |
| "train_runtime": 279.5249, | |
| "train_samples_per_second": 30.48, | |
| "train_steps_per_second": 0.429 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 5.659246428298445e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |