| { | |
| "best_metric": 0.9130434782608695, | |
| "best_model_checkpoint": "vit-base-patch16-224-ve-U11-b-24\\checkpoint-117", | |
| "epoch": 22.153846153846153, | |
| "eval_steps": 500, | |
| "global_step": 144, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.5434782608695652, | |
| "eval_loss": 1.3798073530197144, | |
| "eval_runtime": 0.9211, | |
| "eval_samples_per_second": 49.941, | |
| "eval_steps_per_second": 2.171, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 5.419117647058824e-05, | |
| "loss": 1.3792, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.3090617656707764, | |
| "eval_runtime": 0.8387, | |
| "eval_samples_per_second": 54.85, | |
| "eval_steps_per_second": 2.385, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 1.2226966619491577, | |
| "eval_runtime": 0.7418, | |
| "eval_samples_per_second": 62.012, | |
| "eval_steps_per_second": 2.696, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 5.0147058823529414e-05, | |
| "loss": 1.2783, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6086956521739131, | |
| "eval_loss": 1.1262723207473755, | |
| "eval_runtime": 0.7431, | |
| "eval_samples_per_second": 61.9, | |
| "eval_steps_per_second": 2.691, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 4.610294117647059e-05, | |
| "loss": 1.1226, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.0465526580810547, | |
| "eval_runtime": 0.7448, | |
| "eval_samples_per_second": 61.765, | |
| "eval_steps_per_second": 2.685, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 0.9854069352149963, | |
| "eval_runtime": 0.7677, | |
| "eval_samples_per_second": 59.917, | |
| "eval_steps_per_second": 2.605, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 4.205882352941176e-05, | |
| "loss": 0.9881, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.9302965402603149, | |
| "eval_runtime": 0.7343, | |
| "eval_samples_per_second": 62.642, | |
| "eval_steps_per_second": 2.724, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 3.801470588235294e-05, | |
| "loss": 0.8707, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 0.8805552124977112, | |
| "eval_runtime": 0.7752, | |
| "eval_samples_per_second": 59.337, | |
| "eval_steps_per_second": 2.58, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 0.8234447240829468, | |
| "eval_runtime": 0.726, | |
| "eval_samples_per_second": 63.36, | |
| "eval_steps_per_second": 2.755, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "learning_rate": 3.397058823529412e-05, | |
| "loss": 0.7604, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8260869565217391, | |
| "eval_loss": 0.7159415483474731, | |
| "eval_runtime": 0.7512, | |
| "eval_samples_per_second": 61.236, | |
| "eval_steps_per_second": 2.662, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 2.992647058823529e-05, | |
| "loss": 0.6452, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "eval_accuracy": 0.8478260869565217, | |
| "eval_loss": 0.6929341554641724, | |
| "eval_runtime": 0.7972, | |
| "eval_samples_per_second": 57.705, | |
| "eval_steps_per_second": 2.509, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8695652173913043, | |
| "eval_loss": 0.6490562558174133, | |
| "eval_runtime": 0.7845, | |
| "eval_samples_per_second": 58.637, | |
| "eval_steps_per_second": 2.549, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 2.5882352941176472e-05, | |
| "loss": 0.5576, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 12.92, | |
| "eval_accuracy": 0.8478260869565217, | |
| "eval_loss": 0.5923656225204468, | |
| "eval_runtime": 0.7527, | |
| "eval_samples_per_second": 61.115, | |
| "eval_steps_per_second": 2.657, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 13.85, | |
| "learning_rate": 2.1838235294117645e-05, | |
| "loss": 0.4708, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8478260869565217, | |
| "eval_loss": 0.5550963878631592, | |
| "eval_runtime": 0.8057, | |
| "eval_samples_per_second": 57.091, | |
| "eval_steps_per_second": 2.482, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 14.92, | |
| "eval_accuracy": 0.8043478260869565, | |
| "eval_loss": 0.6354050636291504, | |
| "eval_runtime": 0.8771, | |
| "eval_samples_per_second": 52.446, | |
| "eval_steps_per_second": 2.28, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 1.7794117647058825e-05, | |
| "loss": 0.422, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8695652173913043, | |
| "eval_loss": 0.5129868388175964, | |
| "eval_runtime": 0.7568, | |
| "eval_samples_per_second": 60.785, | |
| "eval_steps_per_second": 2.643, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 16.92, | |
| "learning_rate": 1.375e-05, | |
| "loss": 0.3546, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 16.92, | |
| "eval_accuracy": 0.8695652173913043, | |
| "eval_loss": 0.5302236676216125, | |
| "eval_runtime": 0.7453, | |
| "eval_samples_per_second": 61.721, | |
| "eval_steps_per_second": 2.684, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9130434782608695, | |
| "eval_loss": 0.4435981810092926, | |
| "eval_runtime": 0.7717, | |
| "eval_samples_per_second": 59.607, | |
| "eval_steps_per_second": 2.592, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 18.46, | |
| "learning_rate": 9.705882352941177e-06, | |
| "loss": 0.3353, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 18.92, | |
| "eval_accuracy": 0.8260869565217391, | |
| "eval_loss": 0.5620782375335693, | |
| "eval_runtime": 0.7372, | |
| "eval_samples_per_second": 62.402, | |
| "eval_steps_per_second": 2.713, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 5.661764705882353e-06, | |
| "loss": 0.3106, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8695652173913043, | |
| "eval_loss": 0.4911867082118988, | |
| "eval_runtime": 0.7695, | |
| "eval_samples_per_second": 59.776, | |
| "eval_steps_per_second": 2.599, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.92, | |
| "eval_accuracy": 0.8913043478260869, | |
| "eval_loss": 0.4746726453304291, | |
| "eval_runtime": 0.8093, | |
| "eval_samples_per_second": 56.842, | |
| "eval_steps_per_second": 2.471, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "learning_rate": 1.6176470588235295e-06, | |
| "loss": 0.312, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8913043478260869, | |
| "eval_loss": 0.46030691266059875, | |
| "eval_runtime": 0.7427, | |
| "eval_samples_per_second": 61.935, | |
| "eval_steps_per_second": 2.693, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 22.15, | |
| "eval_accuracy": 0.8913043478260869, | |
| "eval_loss": 0.4597944915294647, | |
| "eval_runtime": 0.7386, | |
| "eval_samples_per_second": 62.279, | |
| "eval_steps_per_second": 2.708, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 22.15, | |
| "step": 144, | |
| "total_flos": 1.4061948496538665e+18, | |
| "train_loss": 0.689174536201689, | |
| "train_runtime": 307.4489, | |
| "train_samples_per_second": 63.933, | |
| "train_steps_per_second": 0.468 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 144, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 24, | |
| "save_steps": 500, | |
| "total_flos": 1.4061948496538665e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |