| { | |
| "best_metric": 0.9347826086956522, | |
| "best_model_checkpoint": "vit-base-patch16-224-finetuned-teeth_dataset/checkpoint-138", | |
| "epoch": 40.0, | |
| "eval_steps": 500, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.008695652173913044, | |
| "eval_loss": 4.653251647949219, | |
| "eval_runtime": 128.9785, | |
| "eval_samples_per_second": 3.566, | |
| "eval_steps_per_second": 0.116, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.006521739130434782, | |
| "eval_loss": 4.584750175476074, | |
| "eval_runtime": 9.4797, | |
| "eval_samples_per_second": 48.525, | |
| "eval_steps_per_second": 1.582, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 2.967961311340332, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 4.6048, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_accuracy": 0.030434782608695653, | |
| "eval_loss": 4.460752964019775, | |
| "eval_runtime": 9.6304, | |
| "eval_samples_per_second": 47.766, | |
| "eval_steps_per_second": 1.558, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.08478260869565217, | |
| "eval_loss": 4.285670757293701, | |
| "eval_runtime": 9.7655, | |
| "eval_samples_per_second": 47.104, | |
| "eval_steps_per_second": 1.536, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_accuracy": 0.11521739130434783, | |
| "eval_loss": 4.1469807624816895, | |
| "eval_runtime": 9.6545, | |
| "eval_samples_per_second": 47.646, | |
| "eval_steps_per_second": 1.554, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "grad_norm": 3.3595643043518066, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 4.2716, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "eval_accuracy": 0.20434782608695654, | |
| "eval_loss": 3.964137315750122, | |
| "eval_runtime": 9.7181, | |
| "eval_samples_per_second": 47.334, | |
| "eval_steps_per_second": 1.544, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "eval_accuracy": 0.31521739130434784, | |
| "eval_loss": 3.7704641819000244, | |
| "eval_runtime": 9.5221, | |
| "eval_samples_per_second": 48.309, | |
| "eval_steps_per_second": 1.575, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.556402683258057, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 3.7404, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.41956521739130437, | |
| "eval_loss": 3.5808911323547363, | |
| "eval_runtime": 9.3828, | |
| "eval_samples_per_second": 49.026, | |
| "eval_steps_per_second": 1.599, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "eval_accuracy": 0.45217391304347826, | |
| "eval_loss": 3.4765827655792236, | |
| "eval_runtime": 9.5003, | |
| "eval_samples_per_second": 48.419, | |
| "eval_steps_per_second": 1.579, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 9.87, | |
| "eval_accuracy": 0.508695652173913, | |
| "eval_loss": 3.2980847358703613, | |
| "eval_runtime": 9.6725, | |
| "eval_samples_per_second": 47.557, | |
| "eval_steps_per_second": 1.551, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 10.67, | |
| "grad_norm": 3.8709869384765625, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 3.1589, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "eval_accuracy": 0.6086956521739131, | |
| "eval_loss": 3.1131505966186523, | |
| "eval_runtime": 9.5498, | |
| "eval_samples_per_second": 48.169, | |
| "eval_steps_per_second": 1.571, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6695652173913044, | |
| "eval_loss": 2.949446201324463, | |
| "eval_runtime": 9.5423, | |
| "eval_samples_per_second": 48.206, | |
| "eval_steps_per_second": 1.572, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "eval_accuracy": 0.6782608695652174, | |
| "eval_loss": 2.836071729660034, | |
| "eval_runtime": 9.5423, | |
| "eval_samples_per_second": 48.207, | |
| "eval_steps_per_second": 1.572, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "grad_norm": 4.395231246948242, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 2.6384, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "eval_accuracy": 0.7347826086956522, | |
| "eval_loss": 2.6520774364471436, | |
| "eval_runtime": 9.5638, | |
| "eval_samples_per_second": 48.098, | |
| "eval_steps_per_second": 1.568, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 14.93, | |
| "eval_accuracy": 0.758695652173913, | |
| "eval_loss": 2.4943137168884277, | |
| "eval_runtime": 9.5047, | |
| "eval_samples_per_second": 48.397, | |
| "eval_steps_per_second": 1.578, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 4.303729057312012, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.1342, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7847826086956522, | |
| "eval_loss": 2.3421921730041504, | |
| "eval_runtime": 9.1584, | |
| "eval_samples_per_second": 50.227, | |
| "eval_steps_per_second": 1.638, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "eval_accuracy": 0.8108695652173913, | |
| "eval_loss": 2.2326693534851074, | |
| "eval_runtime": 9.6001, | |
| "eval_samples_per_second": 47.916, | |
| "eval_steps_per_second": 1.562, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 17.87, | |
| "eval_accuracy": 0.8260869565217391, | |
| "eval_loss": 2.083353281021118, | |
| "eval_runtime": 9.5492, | |
| "eval_samples_per_second": 48.172, | |
| "eval_steps_per_second": 1.571, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 18.67, | |
| "grad_norm": 3.915405750274658, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 1.714, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 18.93, | |
| "eval_accuracy": 0.8565217391304348, | |
| "eval_loss": 1.983383059501648, | |
| "eval_runtime": 9.4581, | |
| "eval_samples_per_second": 48.636, | |
| "eval_steps_per_second": 1.586, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8673913043478261, | |
| "eval_loss": 1.8931976556777954, | |
| "eval_runtime": 9.4385, | |
| "eval_samples_per_second": 48.737, | |
| "eval_steps_per_second": 1.589, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "eval_accuracy": 0.8586956521739131, | |
| "eval_loss": 1.861843228340149, | |
| "eval_runtime": 9.2697, | |
| "eval_samples_per_second": 49.624, | |
| "eval_steps_per_second": 1.618, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 21.33, | |
| "grad_norm": 3.428410768508911, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 1.4427, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 21.87, | |
| "eval_accuracy": 0.8891304347826087, | |
| "eval_loss": 1.6974326372146606, | |
| "eval_runtime": 9.5173, | |
| "eval_samples_per_second": 48.333, | |
| "eval_steps_per_second": 1.576, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 22.93, | |
| "eval_accuracy": 0.8891304347826087, | |
| "eval_loss": 1.6662733554840088, | |
| "eval_runtime": 9.3013, | |
| "eval_samples_per_second": 49.456, | |
| "eval_steps_per_second": 1.613, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 4.027013778686523, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 1.1858, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8847826086956522, | |
| "eval_loss": 1.6013683080673218, | |
| "eval_runtime": 9.5313, | |
| "eval_samples_per_second": 48.262, | |
| "eval_steps_per_second": 1.574, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 24.8, | |
| "eval_accuracy": 0.9043478260869565, | |
| "eval_loss": 1.5112110376358032, | |
| "eval_runtime": 9.5654, | |
| "eval_samples_per_second": 48.09, | |
| "eval_steps_per_second": 1.568, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 25.87, | |
| "eval_accuracy": 0.9108695652173913, | |
| "eval_loss": 1.473188042640686, | |
| "eval_runtime": 9.4856, | |
| "eval_samples_per_second": 48.495, | |
| "eval_steps_per_second": 1.581, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "grad_norm": 3.1277294158935547, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 1.0222, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 26.93, | |
| "eval_accuracy": 0.9065217391304348, | |
| "eval_loss": 1.430389404296875, | |
| "eval_runtime": 9.3124, | |
| "eval_samples_per_second": 49.396, | |
| "eval_steps_per_second": 1.611, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9130434782608695, | |
| "eval_loss": 1.3915013074874878, | |
| "eval_runtime": 9.4281, | |
| "eval_samples_per_second": 48.79, | |
| "eval_steps_per_second": 1.591, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "eval_accuracy": 0.9217391304347826, | |
| "eval_loss": 1.3509211540222168, | |
| "eval_runtime": 9.5983, | |
| "eval_samples_per_second": 47.925, | |
| "eval_steps_per_second": 1.563, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 29.33, | |
| "grad_norm": 2.7500855922698975, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.8306, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 29.87, | |
| "eval_accuracy": 0.9282608695652174, | |
| "eval_loss": 1.3053604364395142, | |
| "eval_runtime": 9.4585, | |
| "eval_samples_per_second": 48.633, | |
| "eval_steps_per_second": 1.586, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 30.93, | |
| "eval_accuracy": 0.9260869565217391, | |
| "eval_loss": 1.2869884967803955, | |
| "eval_runtime": 9.1724, | |
| "eval_samples_per_second": 50.151, | |
| "eval_steps_per_second": 1.635, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 2.7954471111297607, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.7391, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.9282608695652174, | |
| "eval_loss": 1.264487385749817, | |
| "eval_runtime": 9.6279, | |
| "eval_samples_per_second": 47.778, | |
| "eval_steps_per_second": 1.558, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 32.8, | |
| "eval_accuracy": 0.9260869565217391, | |
| "eval_loss": 1.2453793287277222, | |
| "eval_runtime": 9.2796, | |
| "eval_samples_per_second": 49.571, | |
| "eval_steps_per_second": 1.616, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 33.87, | |
| "eval_accuracy": 0.9282608695652174, | |
| "eval_loss": 1.2394675016403198, | |
| "eval_runtime": 9.5473, | |
| "eval_samples_per_second": 48.181, | |
| "eval_steps_per_second": 1.571, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 34.67, | |
| "grad_norm": 2.7077043056488037, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.6971, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 34.93, | |
| "eval_accuracy": 0.9304347826086956, | |
| "eval_loss": 1.207598090171814, | |
| "eval_runtime": 9.5844, | |
| "eval_samples_per_second": 47.994, | |
| "eval_steps_per_second": 1.565, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9326086956521739, | |
| "eval_loss": 1.1821191310882568, | |
| "eval_runtime": 9.4964, | |
| "eval_samples_per_second": 48.44, | |
| "eval_steps_per_second": 1.58, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 36.8, | |
| "eval_accuracy": 0.9347826086956522, | |
| "eval_loss": 1.1735903024673462, | |
| "eval_runtime": 9.4168, | |
| "eval_samples_per_second": 48.849, | |
| "eval_steps_per_second": 1.593, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 37.33, | |
| "grad_norm": 2.4264731407165527, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.6758, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 37.87, | |
| "eval_accuracy": 0.9326086956521739, | |
| "eval_loss": 1.1671130657196045, | |
| "eval_runtime": 9.6709, | |
| "eval_samples_per_second": 47.565, | |
| "eval_steps_per_second": 1.551, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 38.93, | |
| "eval_accuracy": 0.9347826086956522, | |
| "eval_loss": 1.1656177043914795, | |
| "eval_runtime": 9.2896, | |
| "eval_samples_per_second": 49.518, | |
| "eval_steps_per_second": 1.615, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 2.507667303085327, | |
| "learning_rate": 0.0, | |
| "loss": 0.6445, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.9347826086956522, | |
| "eval_loss": 1.1649013757705688, | |
| "eval_runtime": 9.7408, | |
| "eval_samples_per_second": 47.224, | |
| "eval_steps_per_second": 1.54, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "step": 150, | |
| "total_flos": 1.4270027608424448e+18, | |
| "train_loss": 1.9666850312550863, | |
| "train_runtime": 1497.2167, | |
| "train_samples_per_second": 15.362, | |
| "train_steps_per_second": 0.1 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1.4270027608424448e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |