| { | |
| "best_metric": 2.7267191410064697, | |
| "best_model_checkpoint": "detr-resnet50-cd45rb-subdataset/checkpoint-2420", | |
| "epoch": 10.0, | |
| "global_step": 2420, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9e-05, | |
| "loss": 3.774, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.962869644165039, | |
| "eval_runtime": 11.2116, | |
| "eval_samples_per_second": 8.919, | |
| "eval_steps_per_second": 1.16, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8e-05, | |
| "loss": 3.6768, | |
| "step": 484 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.8228960037231445, | |
| "eval_runtime": 11.2793, | |
| "eval_samples_per_second": 8.866, | |
| "eval_steps_per_second": 1.153, | |
| "step": 484 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7e-05, | |
| "loss": 3.6554, | |
| "step": 726 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.8254785537719727, | |
| "eval_runtime": 11.3653, | |
| "eval_samples_per_second": 8.799, | |
| "eval_steps_per_second": 1.144, | |
| "step": 726 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 6e-05, | |
| "loss": 3.5964, | |
| "step": 968 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.8284406661987305, | |
| "eval_runtime": 11.2727, | |
| "eval_samples_per_second": 8.871, | |
| "eval_steps_per_second": 1.153, | |
| "step": 968 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 5e-05, | |
| "loss": 3.5668, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.750312089920044, | |
| "eval_runtime": 20.6727, | |
| "eval_samples_per_second": 4.837, | |
| "eval_steps_per_second": 0.629, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4e-05, | |
| "loss": 3.5733, | |
| "step": 1452 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.755269765853882, | |
| "eval_runtime": 26.9669, | |
| "eval_samples_per_second": 3.708, | |
| "eval_steps_per_second": 0.482, | |
| "step": 1452 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3e-05, | |
| "loss": 3.5197, | |
| "step": 1694 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.7466766834259033, | |
| "eval_runtime": 11.2293, | |
| "eval_samples_per_second": 8.905, | |
| "eval_steps_per_second": 1.158, | |
| "step": 1694 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2e-05, | |
| "loss": 3.5313, | |
| "step": 1936 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.752133846282959, | |
| "eval_runtime": 12.0479, | |
| "eval_samples_per_second": 8.3, | |
| "eval_steps_per_second": 1.079, | |
| "step": 1936 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 1e-05, | |
| "loss": 3.4862, | |
| "step": 2178 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.7274692058563232, | |
| "eval_runtime": 12.3145, | |
| "eval_samples_per_second": 8.121, | |
| "eval_steps_per_second": 1.056, | |
| "step": 2178 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0, | |
| "loss": 3.503, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.7267191410064697, | |
| "eval_runtime": 11.8778, | |
| "eval_samples_per_second": 8.419, | |
| "eval_steps_per_second": 1.094, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2420, | |
| "total_flos": 4.6164408869376e+18, | |
| "train_loss": 3.588286495996901, | |
| "train_runtime": 26239.9252, | |
| "train_samples_per_second": 0.368, | |
| "train_steps_per_second": 0.092 | |
| } | |
| ], | |
| "max_steps": 2420, | |
| "num_train_epochs": 10, | |
| "total_flos": 4.6164408869376e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |