| { | |
| "best_metric": 0.7796974441892816, | |
| "best_model_checkpoint": "/home/hoan/projects/nsfw/models/efficientvit_l1.r224_in1k/run-hgfebf69/checkpoint-595", | |
| "epoch": 28.0, | |
| "eval_steps": 500, | |
| "global_step": 595, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.452508274337144e-07, | |
| "loss": 1.5355, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.3932926829268293, | |
| "eval_f1": 0.484929996783293, | |
| "eval_loss": 1.5276610851287842, | |
| "eval_precision": 0.6852684114434121, | |
| "eval_recall": 0.3932926829268293, | |
| "eval_runtime": 8.1255, | |
| "eval_samples_per_second": 444.032, | |
| "eval_steps_per_second": 3.569, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 2.905016548674288e-07, | |
| "loss": 1.5334, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 4.357524823011432e-07, | |
| "loss": 1.5242, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_accuracy": 0.4584257206208426, | |
| "eval_f1": 0.5440289476561175, | |
| "eval_loss": 1.5126081705093384, | |
| "eval_precision": 0.6975641567116644, | |
| "eval_recall": 0.4584257206208426, | |
| "eval_runtime": 8.3776, | |
| "eval_samples_per_second": 430.674, | |
| "eval_steps_per_second": 3.462, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 5.810033097348576e-07, | |
| "loss": 1.5104, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.5296563192904656, | |
| "eval_f1": 0.5944990700609136, | |
| "eval_loss": 1.4862231016159058, | |
| "eval_precision": 0.6890568361070714, | |
| "eval_recall": 0.5296563192904656, | |
| "eval_runtime": 8.1023, | |
| "eval_samples_per_second": 445.304, | |
| "eval_steps_per_second": 3.579, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.262541371685721e-07, | |
| "loss": 1.493, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 8.715049646022864e-07, | |
| "loss": 1.4698, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6513303769401331, | |
| "eval_f1": 0.6736289843952361, | |
| "eval_loss": 1.4428884983062744, | |
| "eval_precision": 0.7042447934146061, | |
| "eval_recall": 0.6513303769401331, | |
| "eval_runtime": 8.2686, | |
| "eval_samples_per_second": 436.352, | |
| "eval_steps_per_second": 3.507, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 1.0167557920360007e-06, | |
| "loss": 1.4457, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 1.1620066194697152e-06, | |
| "loss": 1.4126, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.7513858093126385, | |
| "eval_f1": 0.7236423084979209, | |
| "eval_loss": 1.3878988027572632, | |
| "eval_precision": 0.704223393234371, | |
| "eval_recall": 0.7513858093126385, | |
| "eval_runtime": 8.0477, | |
| "eval_samples_per_second": 448.328, | |
| "eval_steps_per_second": 3.604, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 1.3072574469034296e-06, | |
| "loss": 1.3745, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "eval_accuracy": 0.8009977827050998, | |
| "eval_f1": 0.7446621793568696, | |
| "eval_loss": 1.3229458332061768, | |
| "eval_precision": 0.7198332904237893, | |
| "eval_recall": 0.8009977827050998, | |
| "eval_runtime": 8.341, | |
| "eval_samples_per_second": 432.564, | |
| "eval_steps_per_second": 3.477, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 1.4075969248184277e-06, | |
| "loss": 1.3395, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 1.4038673596551303e-06, | |
| "loss": 1.2941, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_accuracy": 0.8245565410199557, | |
| "eval_f1": 0.7506711729341005, | |
| "eval_loss": 1.2522051334381104, | |
| "eval_precision": 0.7225360812946281, | |
| "eval_recall": 0.8245565410199557, | |
| "eval_runtime": 8.1304, | |
| "eval_samples_per_second": 443.768, | |
| "eval_steps_per_second": 3.567, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "learning_rate": 1.3955441156005382e-06, | |
| "loss": 1.2525, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 1.3826818162583345e-06, | |
| "loss": 1.2116, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8276053215077606, | |
| "eval_f1": 0.7512743652414022, | |
| "eval_loss": 1.1887985467910767, | |
| "eval_precision": 0.7288159081513773, | |
| "eval_recall": 0.8276053215077606, | |
| "eval_runtime": 8.187, | |
| "eval_samples_per_second": 440.697, | |
| "eval_steps_per_second": 3.542, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 1.3653648740447739e-06, | |
| "loss": 1.168, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.8284368070953437, | |
| "eval_f1": 0.7508179552480237, | |
| "eval_loss": 1.1201469898223877, | |
| "eval_precision": 0.6864978143631778, | |
| "eval_recall": 0.8284368070953437, | |
| "eval_runtime": 8.4146, | |
| "eval_samples_per_second": 428.78, | |
| "eval_steps_per_second": 3.446, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 1.343706936208727e-06, | |
| "loss": 1.1394, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 1.3178501389900054e-06, | |
| "loss": 1.1014, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "eval_accuracy": 0.8287139689578714, | |
| "eval_f1": 0.7513704434640557, | |
| "eval_loss": 1.0765742063522339, | |
| "eval_precision": 0.7699235349840203, | |
| "eval_recall": 0.8287139689578714, | |
| "eval_runtime": 8.0122, | |
| "eval_samples_per_second": 450.311, | |
| "eval_steps_per_second": 3.619, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 1.2879641748107618e-06, | |
| "loss": 1.0706, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "eval_accuracy": 0.8284368070953437, | |
| "eval_f1": 0.7507041432190336, | |
| "eval_loss": 1.0376129150390625, | |
| "eval_precision": 0.6863075433503277, | |
| "eval_recall": 0.8284368070953437, | |
| "eval_runtime": 8.5615, | |
| "eval_samples_per_second": 421.42, | |
| "eval_steps_per_second": 3.387, | |
| "step": 233 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 1.254245178621782e-06, | |
| "loss": 1.0439, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 1.2169144407123264e-06, | |
| "loss": 1.0194, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8284368070953437, | |
| "eval_f1": 0.7508179552480237, | |
| "eval_loss": 1.0062154531478882, | |
| "eval_precision": 0.6864978143631778, | |
| "eval_recall": 0.8284368070953437, | |
| "eval_runtime": 8.4048, | |
| "eval_samples_per_second": 429.28, | |
| "eval_steps_per_second": 3.45, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 1.1762169544310694e-06, | |
| "loss": 1.0048, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 1.132419808349121e-06, | |
| "loss": 0.9997, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.8284368070953437, | |
| "eval_f1": 0.7508179552480237, | |
| "eval_loss": 0.9836404919624329, | |
| "eval_precision": 0.6864978143631778, | |
| "eval_recall": 0.8284368070953437, | |
| "eval_runtime": 8.5535, | |
| "eval_samples_per_second": 421.814, | |
| "eval_steps_per_second": 3.39, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 13.46, | |
| "learning_rate": 1.0858104334170029e-06, | |
| "loss": 0.9776, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 13.98, | |
| "eval_accuracy": 0.8284368070953437, | |
| "eval_f1": 0.7507041432190336, | |
| "eval_loss": 0.9691689610481262, | |
| "eval_precision": 0.6863075433503277, | |
| "eval_recall": 0.8284368070953437, | |
| "eval_runtime": 7.9567, | |
| "eval_samples_per_second": 453.453, | |
| "eval_steps_per_second": 3.645, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 1.0366947166191075e-06, | |
| "loss": 0.9671, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 9.853949935052988e-07, | |
| "loss": 0.9587, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "eval_accuracy": 0.8292682926829268, | |
| "eval_f1": 0.7526922431410414, | |
| "eval_loss": 0.9562509059906006, | |
| "eval_precision": 0.7703043936882441, | |
| "eval_recall": 0.8292682926829268, | |
| "eval_runtime": 8.4186, | |
| "eval_samples_per_second": 428.574, | |
| "eval_steps_per_second": 3.445, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 15.29, | |
| "learning_rate": 9.32247932774231e-07, | |
| "loss": 0.9463, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 8.776023267913875e-07, | |
| "loss": 0.9492, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.83009977827051, | |
| "eval_f1": 0.7547519055308186, | |
| "eval_loss": 0.9446051120758057, | |
| "eval_precision": 0.7591494201844647, | |
| "eval_recall": 0.83009977827051, | |
| "eval_runtime": 8.2197, | |
| "eval_samples_per_second": 438.947, | |
| "eval_steps_per_second": 3.528, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 8.218168025421926e-07, | |
| "loss": 0.9437, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.83009977827051, | |
| "eval_f1": 0.7553598948352633, | |
| "eval_loss": 0.9370310306549072, | |
| "eval_precision": 0.7528714352407055, | |
| "eval_recall": 0.83009977827051, | |
| "eval_runtime": 8.2035, | |
| "eval_samples_per_second": 439.813, | |
| "eval_steps_per_second": 3.535, | |
| "step": 361 | |
| }, | |
| { | |
| "epoch": 17.13, | |
| "learning_rate": 7.652574680426876e-07, | |
| "loss": 0.9358, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 7.082955096538664e-07, | |
| "loss": 0.9363, | |
| "step": 377 | |
| }, | |
| { | |
| "epoch": 17.98, | |
| "eval_accuracy": 0.8317627494456763, | |
| "eval_f1": 0.758845429136837, | |
| "eval_loss": 0.9373316168785095, | |
| "eval_precision": 0.7519338830068296, | |
| "eval_recall": 0.8317627494456763, | |
| "eval_runtime": 8.2615, | |
| "eval_samples_per_second": 436.724, | |
| "eval_steps_per_second": 3.51, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 18.35, | |
| "learning_rate": 6.513047560679618e-07, | |
| "loss": 0.9316, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 18.96, | |
| "learning_rate": 5.94659224953694e-07, | |
| "loss": 0.9264, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 18.96, | |
| "eval_accuracy": 0.8323170731707317, | |
| "eval_f1": 0.7606463034975902, | |
| "eval_loss": 0.9292559027671814, | |
| "eval_precision": 0.7526638484809335, | |
| "eval_recall": 0.8323170731707317, | |
| "eval_runtime": 8.115, | |
| "eval_samples_per_second": 444.609, | |
| "eval_steps_per_second": 3.574, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "learning_rate": 5.387306683613124e-07, | |
| "loss": 0.9316, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8328713968957872, | |
| "eval_f1": 0.762008565582684, | |
| "eval_loss": 0.9269818067550659, | |
| "eval_precision": 0.7475944578147732, | |
| "eval_recall": 0.8328713968957872, | |
| "eval_runtime": 8.3826, | |
| "eval_samples_per_second": 430.415, | |
| "eval_steps_per_second": 3.46, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 20.19, | |
| "learning_rate": 4.838861329963851e-07, | |
| "loss": 0.92, | |
| "step": 429 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 4.304855513737253e-07, | |
| "loss": 0.9174, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "eval_accuracy": 0.8350886917960089, | |
| "eval_f1": 0.766268232245012, | |
| "eval_loss": 0.9251817464828491, | |
| "eval_precision": 0.7611039143133754, | |
| "eval_recall": 0.8350886917960089, | |
| "eval_runtime": 7.7339, | |
| "eval_samples_per_second": 466.52, | |
| "eval_steps_per_second": 3.75, | |
| "step": 446 | |
| }, | |
| { | |
| "epoch": 21.41, | |
| "learning_rate": 3.7887937966016937e-07, | |
| "loss": 0.9212, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 21.98, | |
| "eval_accuracy": 0.8364745011086474, | |
| "eval_f1": 0.7701721753563872, | |
| "eval_loss": 0.9216660261154175, | |
| "eval_precision": 0.7861397608879442, | |
| "eval_recall": 0.8364745011086474, | |
| "eval_runtime": 8.2364, | |
| "eval_samples_per_second": 438.057, | |
| "eval_steps_per_second": 3.521, | |
| "step": 467 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "learning_rate": 3.29406297708525e-07, | |
| "loss": 0.9222, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 2.8239098637685624e-07, | |
| "loss": 0.9142, | |
| "step": 481 | |
| }, | |
| { | |
| "epoch": 22.96, | |
| "eval_accuracy": 0.8389689578713969, | |
| "eval_f1": 0.7751934812972323, | |
| "eval_loss": 0.9198487997055054, | |
| "eval_precision": 0.7580660317760047, | |
| "eval_recall": 0.8389689578713969, | |
| "eval_runtime": 8.1907, | |
| "eval_samples_per_second": 440.497, | |
| "eval_steps_per_second": 3.541, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 23.25, | |
| "learning_rate": 2.3814199672006268e-07, | |
| "loss": 0.9239, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 23.86, | |
| "learning_rate": 1.9694972503777733e-07, | |
| "loss": 0.9178, | |
| "step": 507 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8356430155210643, | |
| "eval_f1": 0.7695758947830811, | |
| "eval_loss": 0.9179719686508179, | |
| "eval_precision": 0.7756305393423396, | |
| "eval_recall": 0.8356430155210643, | |
| "eval_runtime": 8.0754, | |
| "eval_samples_per_second": 446.788, | |
| "eval_steps_per_second": 3.591, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "learning_rate": 1.5908450706789677e-07, | |
| "loss": 0.9152, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "eval_accuracy": 0.8395232815964523, | |
| "eval_f1": 0.7765057889032959, | |
| "eval_loss": 0.9176101088523865, | |
| "eval_precision": 0.7933438992755063, | |
| "eval_recall": 0.8395232815964523, | |
| "eval_runtime": 8.0393, | |
| "eval_samples_per_second": 448.793, | |
| "eval_steps_per_second": 3.607, | |
| "step": 531 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "learning_rate": 1.2479484383312357e-07, | |
| "loss": 0.9204, | |
| "step": 533 | |
| }, | |
| { | |
| "epoch": 25.69, | |
| "learning_rate": 9.430577078390059e-08, | |
| "loss": 0.9096, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 25.98, | |
| "eval_accuracy": 0.8398004434589801, | |
| "eval_f1": 0.7763506725004051, | |
| "eval_loss": 0.9181774258613586, | |
| "eval_precision": 0.7954486998022142, | |
| "eval_recall": 0.8398004434589801, | |
| "eval_runtime": 8.1901, | |
| "eval_samples_per_second": 440.533, | |
| "eval_steps_per_second": 3.541, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 26.31, | |
| "learning_rate": 6.781738094068847e-08, | |
| "loss": 0.9157, | |
| "step": 559 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 4.550351172787507e-08, | |
| "loss": 0.9211, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 26.96, | |
| "eval_accuracy": 0.8364745011086474, | |
| "eval_f1": 0.7710155859925437, | |
| "eval_loss": 0.9194357991218567, | |
| "eval_precision": 0.7851756179427326, | |
| "eval_recall": 0.8364745011086474, | |
| "eval_runtime": 8.2806, | |
| "eval_samples_per_second": 435.717, | |
| "eval_steps_per_second": 3.502, | |
| "step": 573 | |
| }, | |
| { | |
| "epoch": 27.53, | |
| "learning_rate": 2.751060411733735e-08, | |
| "loss": 0.912, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_f1": 0.7796974441892816, | |
| "eval_loss": 0.9164049625396729, | |
| "eval_precision": 0.796046494839443, | |
| "eval_recall": 0.8409090909090909, | |
| "eval_runtime": 8.3541, | |
| "eval_samples_per_second": 431.882, | |
| "eval_steps_per_second": 3.471, | |
| "step": 595 | |
| } | |
| ], | |
| "logging_steps": 13, | |
| "max_steps": 630, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 0.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "label_smoothing_factor": 0.019838947875582387, | |
| "learning_rate": 1.4078157120498472e-06, | |
| "metric": "eval/loss", | |
| "weight_decay": 0.024972382732841077 | |
| } | |
| } | |