| { | |
| "best_metric": 0.9411764705882353, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-fish/checkpoint-21", | |
| "epoch": 60.0, | |
| "eval_steps": 500, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.29411764705882354, | |
| "eval_loss": 1.803544521331787, | |
| "eval_runtime": 0.4219, | |
| "eval_samples_per_second": 40.297, | |
| "eval_steps_per_second": 2.37, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.29411764705882354, | |
| "eval_loss": 1.786075234413147, | |
| "eval_runtime": 0.425, | |
| "eval_samples_per_second": 39.998, | |
| "eval_steps_per_second": 2.353, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_accuracy": 0.29411764705882354, | |
| "eval_loss": 1.7553783655166626, | |
| "eval_runtime": 0.409, | |
| "eval_samples_per_second": 41.562, | |
| "eval_steps_per_second": 2.445, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.35294117647058826, | |
| "eval_loss": 1.6953703165054321, | |
| "eval_runtime": 0.4264, | |
| "eval_samples_per_second": 39.865, | |
| "eval_steps_per_second": 2.345, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_accuracy": 0.4117647058823529, | |
| "eval_loss": 1.678010106086731, | |
| "eval_runtime": 0.4097, | |
| "eval_samples_per_second": 41.492, | |
| "eval_steps_per_second": 2.441, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "eval_accuracy": 0.4117647058823529, | |
| "eval_loss": 1.65355384349823, | |
| "eval_runtime": 0.3958, | |
| "eval_samples_per_second": 42.953, | |
| "eval_steps_per_second": 2.527, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_accuracy": 0.4117647058823529, | |
| "eval_loss": 1.6221718788146973, | |
| "eval_runtime": 0.4077, | |
| "eval_samples_per_second": 41.701, | |
| "eval_steps_per_second": 2.453, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 8.47060489654541, | |
| "learning_rate": 4.850746268656717e-05, | |
| "loss": 1.6467, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5294117647058824, | |
| "eval_loss": 1.4681596755981445, | |
| "eval_runtime": 0.4106, | |
| "eval_samples_per_second": 41.405, | |
| "eval_steps_per_second": 2.436, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "eval_accuracy": 0.5294117647058824, | |
| "eval_loss": 1.3260958194732666, | |
| "eval_runtime": 0.4111, | |
| "eval_samples_per_second": 41.351, | |
| "eval_steps_per_second": 2.432, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "eval_accuracy": 0.5294117647058824, | |
| "eval_loss": 1.1887730360031128, | |
| "eval_runtime": 0.4034, | |
| "eval_samples_per_second": 42.144, | |
| "eval_steps_per_second": 2.479, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "eval_accuracy": 0.5294117647058824, | |
| "eval_loss": 1.0433154106140137, | |
| "eval_runtime": 0.4207, | |
| "eval_samples_per_second": 40.405, | |
| "eval_steps_per_second": 2.377, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5882352941176471, | |
| "eval_loss": 0.8211753368377686, | |
| "eval_runtime": 0.4003, | |
| "eval_samples_per_second": 42.464, | |
| "eval_steps_per_second": 2.498, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "eval_accuracy": 0.7058823529411765, | |
| "eval_loss": 0.7240034341812134, | |
| "eval_runtime": 0.4119, | |
| "eval_samples_per_second": 41.269, | |
| "eval_steps_per_second": 2.428, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "eval_accuracy": 0.8235294117647058, | |
| "eval_loss": 0.6390114426612854, | |
| "eval_runtime": 0.4132, | |
| "eval_samples_per_second": 41.145, | |
| "eval_steps_per_second": 2.42, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.5593705177307129, | |
| "eval_runtime": 0.413, | |
| "eval_samples_per_second": 41.162, | |
| "eval_steps_per_second": 2.421, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.233682155609131, | |
| "learning_rate": 4.104477611940299e-05, | |
| "loss": 0.782, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8235294117647058, | |
| "eval_loss": 0.46473824977874756, | |
| "eval_runtime": 0.3967, | |
| "eval_samples_per_second": 42.853, | |
| "eval_steps_per_second": 2.521, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.42642152309417725, | |
| "eval_runtime": 0.42, | |
| "eval_samples_per_second": 40.479, | |
| "eval_steps_per_second": 2.381, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.3983331024646759, | |
| "eval_runtime": 0.4396, | |
| "eval_samples_per_second": 38.669, | |
| "eval_steps_per_second": 2.275, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.3759661912918091, | |
| "eval_runtime": 0.4039, | |
| "eval_samples_per_second": 42.09, | |
| "eval_steps_per_second": 2.476, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.3751123547554016, | |
| "eval_runtime": 0.4128, | |
| "eval_samples_per_second": 41.182, | |
| "eval_steps_per_second": 2.422, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.35528773069381714, | |
| "eval_runtime": 0.3998, | |
| "eval_samples_per_second": 42.522, | |
| "eval_steps_per_second": 2.501, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 21.6, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.31613630056381226, | |
| "eval_runtime": 0.4018, | |
| "eval_samples_per_second": 42.307, | |
| "eval_steps_per_second": 2.489, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.270626962184906, | |
| "eval_runtime": 0.3978, | |
| "eval_samples_per_second": 42.736, | |
| "eval_steps_per_second": 2.514, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 7.092751979827881, | |
| "learning_rate": 3.358208955223881e-05, | |
| "loss": 0.3228, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.21002241969108582, | |
| "eval_runtime": 0.423, | |
| "eval_samples_per_second": 40.194, | |
| "eval_steps_per_second": 2.364, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 24.8, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.18847431242465973, | |
| "eval_runtime": 0.3992, | |
| "eval_samples_per_second": 42.583, | |
| "eval_steps_per_second": 2.505, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 25.6, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.17268314957618713, | |
| "eval_runtime": 0.3893, | |
| "eval_samples_per_second": 43.663, | |
| "eval_steps_per_second": 2.568, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 26.4, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.18181481957435608, | |
| "eval_runtime": 0.4253, | |
| "eval_samples_per_second": 39.969, | |
| "eval_steps_per_second": 2.351, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.1958554983139038, | |
| "eval_runtime": 0.4047, | |
| "eval_samples_per_second": 42.006, | |
| "eval_steps_per_second": 2.471, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.18888713419437408, | |
| "eval_runtime": 0.4156, | |
| "eval_samples_per_second": 40.908, | |
| "eval_steps_per_second": 2.406, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 29.6, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.1994781792163849, | |
| "eval_runtime": 0.4146, | |
| "eval_samples_per_second": 41.008, | |
| "eval_steps_per_second": 2.412, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 30.4, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.20932523906230927, | |
| "eval_runtime": 0.4033, | |
| "eval_samples_per_second": 42.151, | |
| "eval_steps_per_second": 2.479, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 5.438349723815918, | |
| "learning_rate": 2.6119402985074626e-05, | |
| "loss": 0.2375, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.18688271939754486, | |
| "eval_runtime": 0.4352, | |
| "eval_samples_per_second": 39.059, | |
| "eval_steps_per_second": 2.298, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 32.8, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.16475379467010498, | |
| "eval_runtime": 0.4369, | |
| "eval_samples_per_second": 38.912, | |
| "eval_steps_per_second": 2.289, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 33.6, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.15764710307121277, | |
| "eval_runtime": 0.4055, | |
| "eval_samples_per_second": 41.923, | |
| "eval_steps_per_second": 2.466, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 34.4, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.17089737951755524, | |
| "eval_runtime": 0.4229, | |
| "eval_samples_per_second": 40.197, | |
| "eval_steps_per_second": 2.365, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.1716768890619278, | |
| "eval_runtime": 0.4195, | |
| "eval_samples_per_second": 40.52, | |
| "eval_steps_per_second": 2.384, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 36.8, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.178259015083313, | |
| "eval_runtime": 0.4058, | |
| "eval_samples_per_second": 41.888, | |
| "eval_steps_per_second": 2.464, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 37.6, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.19927644729614258, | |
| "eval_runtime": 0.418, | |
| "eval_samples_per_second": 40.668, | |
| "eval_steps_per_second": 2.392, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 38.4, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.20850598812103271, | |
| "eval_runtime": 0.4378, | |
| "eval_samples_per_second": 38.829, | |
| "eval_steps_per_second": 2.284, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 6.9101080894470215, | |
| "learning_rate": 1.865671641791045e-05, | |
| "loss": 0.1897, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.2028307318687439, | |
| "eval_runtime": 0.3953, | |
| "eval_samples_per_second": 43.001, | |
| "eval_steps_per_second": 2.529, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 40.8, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.1704062819480896, | |
| "eval_runtime": 0.3961, | |
| "eval_samples_per_second": 42.921, | |
| "eval_steps_per_second": 2.525, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 41.6, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.15204329788684845, | |
| "eval_runtime": 0.4075, | |
| "eval_samples_per_second": 41.72, | |
| "eval_steps_per_second": 2.454, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 42.4, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.13252082467079163, | |
| "eval_runtime": 0.4039, | |
| "eval_samples_per_second": 42.088, | |
| "eval_steps_per_second": 2.476, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.14505437016487122, | |
| "eval_runtime": 0.4043, | |
| "eval_samples_per_second": 42.048, | |
| "eval_steps_per_second": 2.473, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 44.8, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.16641516983509064, | |
| "eval_runtime": 0.4361, | |
| "eval_samples_per_second": 38.986, | |
| "eval_steps_per_second": 2.293, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 45.6, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.19269916415214539, | |
| "eval_runtime": 0.4099, | |
| "eval_samples_per_second": 41.472, | |
| "eval_steps_per_second": 2.44, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 46.4, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.22021445631980896, | |
| "eval_runtime": 0.4008, | |
| "eval_samples_per_second": 42.416, | |
| "eval_steps_per_second": 2.495, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 5.389186382293701, | |
| "learning_rate": 1.119402985074627e-05, | |
| "loss": 0.1676, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.25689512491226196, | |
| "eval_runtime": 0.4197, | |
| "eval_samples_per_second": 40.504, | |
| "eval_steps_per_second": 2.383, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 48.8, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.27476173639297485, | |
| "eval_runtime": 0.4298, | |
| "eval_samples_per_second": 39.551, | |
| "eval_steps_per_second": 2.327, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 49.6, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.26124510169029236, | |
| "eval_runtime": 0.3885, | |
| "eval_samples_per_second": 43.756, | |
| "eval_steps_per_second": 2.574, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 50.4, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.24144531786441803, | |
| "eval_runtime": 0.4017, | |
| "eval_samples_per_second": 42.318, | |
| "eval_steps_per_second": 2.489, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.18416434526443481, | |
| "eval_runtime": 0.3993, | |
| "eval_samples_per_second": 42.577, | |
| "eval_steps_per_second": 2.505, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 52.8, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.15971186757087708, | |
| "eval_runtime": 0.4143, | |
| "eval_samples_per_second": 41.033, | |
| "eval_steps_per_second": 2.414, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 53.6, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.1446918547153473, | |
| "eval_runtime": 0.3964, | |
| "eval_samples_per_second": 42.886, | |
| "eval_steps_per_second": 2.523, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 54.4, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.13591152429580688, | |
| "eval_runtime": 0.392, | |
| "eval_samples_per_second": 43.363, | |
| "eval_steps_per_second": 2.551, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "grad_norm": 13.033480644226074, | |
| "learning_rate": 3.7313432835820893e-06, | |
| "loss": 0.1452, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.13674524426460266, | |
| "eval_runtime": 0.4048, | |
| "eval_samples_per_second": 41.997, | |
| "eval_steps_per_second": 2.47, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 56.8, | |
| "eval_accuracy": 0.9411764705882353, | |
| "eval_loss": 0.1401834785938263, | |
| "eval_runtime": 0.4238, | |
| "eval_samples_per_second": 40.11, | |
| "eval_steps_per_second": 2.359, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 57.6, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.14624573290348053, | |
| "eval_runtime": 0.4222, | |
| "eval_samples_per_second": 40.266, | |
| "eval_steps_per_second": 2.369, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 58.4, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.1515001356601715, | |
| "eval_runtime": 0.4044, | |
| "eval_samples_per_second": 42.039, | |
| "eval_steps_per_second": 2.473, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.1584557294845581, | |
| "eval_runtime": 0.3957, | |
| "eval_samples_per_second": 42.96, | |
| "eval_steps_per_second": 2.527, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "step": 75, | |
| "total_flos": 2.1926045190131712e+17, | |
| "train_loss": 0.4769280139605204, | |
| "train_runtime": 264.8978, | |
| "train_samples_per_second": 41.62, | |
| "train_steps_per_second": 0.283 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 75, | |
| "save_steps": 500, | |
| "total_flos": 2.1926045190131712e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |