| { | |
| "best_metric": 0.4536682665348053, | |
| "best_model_checkpoint": "SavedModels/Cal_ViT-large-patch16-224_A.ipynb/checkpoint-504", | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 3780, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7936507936507936, | |
| "grad_norm": 7.900213241577148, | |
| "learning_rate": 0.000292063492063492, | |
| "loss": 0.9853, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.744, | |
| "eval_f1": 0.7302269434370317, | |
| "eval_loss": 0.7107657194137573, | |
| "eval_precision": 0.7712189868380369, | |
| "eval_recall": 0.7286540520411489, | |
| "eval_runtime": 12.0784, | |
| "eval_samples_per_second": 20.698, | |
| "eval_steps_per_second": 2.649, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 1.5873015873015874, | |
| "grad_norm": 8.41816234588623, | |
| "learning_rate": 0.0002841269841269841, | |
| "loss": 0.6452, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.82, | |
| "eval_f1": 0.8043829742202345, | |
| "eval_loss": 0.4917103350162506, | |
| "eval_precision": 0.8345023770555686, | |
| "eval_recall": 0.8097303681981102, | |
| "eval_runtime": 11.4769, | |
| "eval_samples_per_second": 21.783, | |
| "eval_steps_per_second": 2.788, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 16.99386215209961, | |
| "learning_rate": 0.00027619047619047615, | |
| "loss": 0.3935, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.816, | |
| "eval_f1": 0.7985758096573877, | |
| "eval_loss": 0.4719848334789276, | |
| "eval_precision": 0.8189957264957265, | |
| "eval_recall": 0.8055637015314435, | |
| "eval_runtime": 11.5276, | |
| "eval_samples_per_second": 21.687, | |
| "eval_steps_per_second": 2.776, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 3.1746031746031744, | |
| "grad_norm": 15.368912696838379, | |
| "learning_rate": 0.00026825396825396825, | |
| "loss": 0.2416, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.9682539682539684, | |
| "grad_norm": 19.9014949798584, | |
| "learning_rate": 0.0002603174603174603, | |
| "loss": 0.1334, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.844, | |
| "eval_f1": 0.8364854215918044, | |
| "eval_loss": 0.4536682665348053, | |
| "eval_precision": 0.8404496818970504, | |
| "eval_recall": 0.8392659312014151, | |
| "eval_runtime": 11.5071, | |
| "eval_samples_per_second": 21.726, | |
| "eval_steps_per_second": 2.781, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 20.753414154052734, | |
| "learning_rate": 0.0002523809523809524, | |
| "loss": 0.0575, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.836, | |
| "eval_f1": 0.8300810566073022, | |
| "eval_loss": 0.520072877407074, | |
| "eval_precision": 0.8326529759666744, | |
| "eval_recall": 0.8296722990271377, | |
| "eval_runtime": 11.5252, | |
| "eval_samples_per_second": 21.692, | |
| "eval_steps_per_second": 2.777, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 5.555555555555555, | |
| "grad_norm": 16.8087100982666, | |
| "learning_rate": 0.00024444444444444443, | |
| "loss": 0.1109, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.832, | |
| "eval_f1": 0.8266870743093288, | |
| "eval_loss": 0.6697499752044678, | |
| "eval_precision": 0.8294946162169833, | |
| "eval_recall": 0.8305852301820044, | |
| "eval_runtime": 11.4108, | |
| "eval_samples_per_second": 21.909, | |
| "eval_steps_per_second": 2.804, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 6.349206349206349, | |
| "grad_norm": 14.420797348022461, | |
| "learning_rate": 0.0002365079365079365, | |
| "loss": 0.0983, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.84, | |
| "eval_f1": 0.837053931393554, | |
| "eval_loss": 0.5976704359054565, | |
| "eval_precision": 0.8379383751169328, | |
| "eval_recall": 0.8369804031094353, | |
| "eval_runtime": 11.3574, | |
| "eval_samples_per_second": 22.012, | |
| "eval_steps_per_second": 2.818, | |
| "step": 882 | |
| }, | |
| { | |
| "epoch": 7.142857142857143, | |
| "grad_norm": 18.421493530273438, | |
| "learning_rate": 0.00022857142857142854, | |
| "loss": 0.0694, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.936507936507937, | |
| "grad_norm": 16.508926391601562, | |
| "learning_rate": 0.0002206349206349206, | |
| "loss": 0.0739, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.824, | |
| "eval_f1": 0.8220051553384886, | |
| "eval_loss": 0.5506980419158936, | |
| "eval_precision": 0.8243932700223792, | |
| "eval_recall": 0.8214786109947401, | |
| "eval_runtime": 11.3601, | |
| "eval_samples_per_second": 22.007, | |
| "eval_steps_per_second": 2.817, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 8.73015873015873, | |
| "grad_norm": 16.224796295166016, | |
| "learning_rate": 0.00021269841269841268, | |
| "loss": 0.0395, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.82, | |
| "eval_f1": 0.8177316510649844, | |
| "eval_loss": 0.6578373312950134, | |
| "eval_precision": 0.8200073051100985, | |
| "eval_recall": 0.8173119443280733, | |
| "eval_runtime": 11.3675, | |
| "eval_samples_per_second": 21.992, | |
| "eval_steps_per_second": 2.815, | |
| "step": 1134 | |
| }, | |
| { | |
| "epoch": 9.523809523809524, | |
| "grad_norm": 17.235092163085938, | |
| "learning_rate": 0.00020476190476190475, | |
| "loss": 0.0416, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.82, | |
| "eval_f1": 0.8114134932173007, | |
| "eval_loss": 0.7492188811302185, | |
| "eval_precision": 0.8144965098919531, | |
| "eval_recall": 0.8123562817111204, | |
| "eval_runtime": 11.3856, | |
| "eval_samples_per_second": 21.958, | |
| "eval_steps_per_second": 2.811, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 10.317460317460318, | |
| "grad_norm": 14.80374526977539, | |
| "learning_rate": 0.0001968253968253968, | |
| "loss": 0.0332, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.828, | |
| "eval_f1": 0.8224526625354294, | |
| "eval_loss": 0.8895044326782227, | |
| "eval_precision": 0.8306511283758787, | |
| "eval_recall": 0.8204987664665083, | |
| "eval_runtime": 11.3762, | |
| "eval_samples_per_second": 21.976, | |
| "eval_steps_per_second": 2.813, | |
| "step": 1386 | |
| }, | |
| { | |
| "epoch": 11.11111111111111, | |
| "grad_norm": 19.538997650146484, | |
| "learning_rate": 0.00018888888888888888, | |
| "loss": 0.0184, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 11.904761904761905, | |
| "grad_norm": 18.353506088256836, | |
| "learning_rate": 0.00018095238095238093, | |
| "loss": 0.0324, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.812, | |
| "eval_f1": 0.8088087500673765, | |
| "eval_loss": 0.762988269329071, | |
| "eval_precision": 0.809127072437966, | |
| "eval_recall": 0.8106590094493319, | |
| "eval_runtime": 11.385, | |
| "eval_samples_per_second": 21.959, | |
| "eval_steps_per_second": 2.811, | |
| "step": 1512 | |
| }, | |
| { | |
| "epoch": 12.698412698412698, | |
| "grad_norm": 18.43031883239746, | |
| "learning_rate": 0.000173015873015873, | |
| "loss": 0.0335, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.836, | |
| "eval_f1": 0.8294732588375853, | |
| "eval_loss": 0.5692726969718933, | |
| "eval_precision": 0.8316624761497172, | |
| "eval_recall": 0.831677372806405, | |
| "eval_runtime": 11.4395, | |
| "eval_samples_per_second": 21.854, | |
| "eval_steps_per_second": 2.797, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 13.492063492063492, | |
| "grad_norm": 17.18678092956543, | |
| "learning_rate": 0.00016507936507936506, | |
| "loss": 0.0245, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.86, | |
| "eval_f1": 0.8567973521009923, | |
| "eval_loss": 0.6419259309768677, | |
| "eval_precision": 0.8568399743801923, | |
| "eval_recall": 0.8570689615044453, | |
| "eval_runtime": 11.436, | |
| "eval_samples_per_second": 21.861, | |
| "eval_steps_per_second": 2.798, | |
| "step": 1764 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 15.67866325378418, | |
| "learning_rate": 0.00015714285714285713, | |
| "loss": 0.0189, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.86, | |
| "eval_f1": 0.8584421340648372, | |
| "eval_loss": 0.7554742693901062, | |
| "eval_precision": 0.8589289889820719, | |
| "eval_recall": 0.858138411767444, | |
| "eval_runtime": 11.4276, | |
| "eval_samples_per_second": 21.877, | |
| "eval_steps_per_second": 2.8, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 15.079365079365079, | |
| "grad_norm": 20.463829040527344, | |
| "learning_rate": 0.00014920634920634917, | |
| "loss": 0.0099, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 15.873015873015873, | |
| "grad_norm": 19.685022354125977, | |
| "learning_rate": 0.00014126984126984124, | |
| "loss": 0.0109, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.848, | |
| "eval_f1": 0.8440413675707793, | |
| "eval_loss": 0.7789789438247681, | |
| "eval_precision": 0.84466028545071, | |
| "eval_recall": 0.8448267234557557, | |
| "eval_runtime": 11.3821, | |
| "eval_samples_per_second": 21.964, | |
| "eval_steps_per_second": 2.811, | |
| "step": 2016 | |
| }, | |
| { | |
| "epoch": 16.666666666666668, | |
| "grad_norm": 19.673765182495117, | |
| "learning_rate": 0.0001333333333333333, | |
| "loss": 0.0083, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.848, | |
| "eval_f1": 0.8440413675707793, | |
| "eval_loss": 0.7907350063323975, | |
| "eval_precision": 0.84466028545071, | |
| "eval_recall": 0.8448267234557557, | |
| "eval_runtime": 11.4578, | |
| "eval_samples_per_second": 21.819, | |
| "eval_steps_per_second": 2.793, | |
| "step": 2142 | |
| }, | |
| { | |
| "epoch": 17.46031746031746, | |
| "grad_norm": 17.990713119506836, | |
| "learning_rate": 0.00012539682539682538, | |
| "loss": 0.008, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.856, | |
| "eval_f1": 0.8525987278543515, | |
| "eval_loss": 0.8481489419937134, | |
| "eval_precision": 0.8536676046910646, | |
| "eval_recall": 0.8527399571754412, | |
| "eval_runtime": 11.4478, | |
| "eval_samples_per_second": 21.838, | |
| "eval_steps_per_second": 2.795, | |
| "step": 2268 | |
| }, | |
| { | |
| "epoch": 18.253968253968253, | |
| "grad_norm": 20.667980194091797, | |
| "learning_rate": 0.00011746031746031744, | |
| "loss": 0.0121, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.852, | |
| "eval_f1": 0.8488564789790646, | |
| "eval_loss": 0.8253383040428162, | |
| "eval_precision": 0.8494105075271848, | |
| "eval_recall": 0.8491557277847601, | |
| "eval_runtime": 11.4707, | |
| "eval_samples_per_second": 21.795, | |
| "eval_steps_per_second": 2.79, | |
| "step": 2394 | |
| }, | |
| { | |
| "epoch": 19.047619047619047, | |
| "grad_norm": 19.05530548095703, | |
| "learning_rate": 0.0001095238095238095, | |
| "loss": 0.0095, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 19.841269841269842, | |
| "grad_norm": 19.204843521118164, | |
| "learning_rate": 0.00010158730158730157, | |
| "loss": 0.0097, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.856, | |
| "eval_f1": 0.8531258452442141, | |
| "eval_loss": 0.829548180103302, | |
| "eval_precision": 0.8533592253010878, | |
| "eval_recall": 0.8534847321137643, | |
| "eval_runtime": 11.4107, | |
| "eval_samples_per_second": 21.909, | |
| "eval_steps_per_second": 2.804, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 20.634920634920636, | |
| "grad_norm": 20.698205947875977, | |
| "learning_rate": 9.365079365079364e-05, | |
| "loss": 0.0085, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.852, | |
| "eval_f1": 0.8488564789790646, | |
| "eval_loss": 0.8458440899848938, | |
| "eval_precision": 0.8494105075271848, | |
| "eval_recall": 0.8491557277847601, | |
| "eval_runtime": 11.4413, | |
| "eval_samples_per_second": 21.851, | |
| "eval_steps_per_second": 2.797, | |
| "step": 2646 | |
| }, | |
| { | |
| "epoch": 21.428571428571427, | |
| "grad_norm": 19.178499221801758, | |
| "learning_rate": 8.57142857142857e-05, | |
| "loss": 0.0085, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.856, | |
| "eval_f1": 0.8531258452442141, | |
| "eval_loss": 0.8456778526306152, | |
| "eval_precision": 0.8533592253010878, | |
| "eval_recall": 0.8534847321137643, | |
| "eval_runtime": 11.4396, | |
| "eval_samples_per_second": 21.854, | |
| "eval_steps_per_second": 2.797, | |
| "step": 2772 | |
| }, | |
| { | |
| "epoch": 22.22222222222222, | |
| "grad_norm": 19.464889526367188, | |
| "learning_rate": 7.777777777777777e-05, | |
| "loss": 0.0091, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.852, | |
| "eval_f1": 0.8488564789790646, | |
| "eval_loss": 0.8525780439376831, | |
| "eval_precision": 0.8494105075271848, | |
| "eval_recall": 0.8491557277847601, | |
| "eval_runtime": 11.4614, | |
| "eval_samples_per_second": 21.812, | |
| "eval_steps_per_second": 2.792, | |
| "step": 2898 | |
| }, | |
| { | |
| "epoch": 23.015873015873016, | |
| "grad_norm": 18.641748428344727, | |
| "learning_rate": 6.984126984126984e-05, | |
| "loss": 0.0088, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 23.80952380952381, | |
| "grad_norm": 16.954683303833008, | |
| "learning_rate": 6.190476190476189e-05, | |
| "loss": 0.0087, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.852, | |
| "eval_f1": 0.849117063374952, | |
| "eval_loss": 0.8523342609405518, | |
| "eval_precision": 0.8492288344390057, | |
| "eval_recall": 0.8493180654470978, | |
| "eval_runtime": 11.4777, | |
| "eval_samples_per_second": 21.781, | |
| "eval_steps_per_second": 2.788, | |
| "step": 3024 | |
| }, | |
| { | |
| "epoch": 24.603174603174605, | |
| "grad_norm": 17.625598907470703, | |
| "learning_rate": 5.396825396825396e-05, | |
| "loss": 0.0232, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.856, | |
| "eval_f1": 0.8503092627516886, | |
| "eval_loss": 0.8092365860939026, | |
| "eval_precision": 0.8546918325243449, | |
| "eval_recall": 0.8510880696364568, | |
| "eval_runtime": 11.4882, | |
| "eval_samples_per_second": 21.761, | |
| "eval_steps_per_second": 2.785, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 25.396825396825395, | |
| "grad_norm": 17.685829162597656, | |
| "learning_rate": 4.603174603174602e-05, | |
| "loss": 0.0162, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.844, | |
| "eval_f1": 0.8397000162501418, | |
| "eval_loss": 0.7603457570075989, | |
| "eval_precision": 0.8405141515263873, | |
| "eval_recall": 0.8410801564027371, | |
| "eval_runtime": 11.4597, | |
| "eval_samples_per_second": 21.816, | |
| "eval_steps_per_second": 2.792, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 26.19047619047619, | |
| "grad_norm": 18.523897171020508, | |
| "learning_rate": 3.809523809523809e-05, | |
| "loss": 0.0112, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 26.984126984126984, | |
| "grad_norm": 20.7585506439209, | |
| "learning_rate": 3.0158730158730154e-05, | |
| "loss": 0.0112, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.848, | |
| "eval_f1": 0.8432066944548815, | |
| "eval_loss": 0.7882433533668518, | |
| "eval_precision": 0.8447052075757308, | |
| "eval_recall": 0.8440819485174323, | |
| "eval_runtime": 11.4572, | |
| "eval_samples_per_second": 21.82, | |
| "eval_steps_per_second": 2.793, | |
| "step": 3402 | |
| }, | |
| { | |
| "epoch": 27.77777777777778, | |
| "grad_norm": 19.374753952026367, | |
| "learning_rate": 2.222222222222222e-05, | |
| "loss": 0.0073, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.844, | |
| "eval_f1": 0.8392035860120967, | |
| "eval_loss": 0.8009940981864929, | |
| "eval_precision": 0.8402476780185758, | |
| "eval_recall": 0.8399152818507657, | |
| "eval_runtime": 11.4891, | |
| "eval_samples_per_second": 21.76, | |
| "eval_steps_per_second": 2.785, | |
| "step": 3528 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 18.367250442504883, | |
| "learning_rate": 1.4285714285714284e-05, | |
| "loss": 0.0103, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.848, | |
| "eval_f1": 0.8445914212496674, | |
| "eval_loss": 0.7996464967727661, | |
| "eval_precision": 0.8447769509983977, | |
| "eval_recall": 0.8449890611180934, | |
| "eval_runtime": 11.5367, | |
| "eval_samples_per_second": 21.67, | |
| "eval_steps_per_second": 2.774, | |
| "step": 3654 | |
| }, | |
| { | |
| "epoch": 29.365079365079364, | |
| "grad_norm": 19.641939163208008, | |
| "learning_rate": 6.349206349206348e-06, | |
| "loss": 0.0086, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.848, | |
| "eval_f1": 0.8445914212496674, | |
| "eval_loss": 0.8016173839569092, | |
| "eval_precision": 0.8447769509983977, | |
| "eval_recall": 0.8449890611180934, | |
| "eval_runtime": 11.4935, | |
| "eval_samples_per_second": 21.752, | |
| "eval_steps_per_second": 2.784, | |
| "step": 3780 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3780, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.6444324389874913e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |