hita's picture
model 6 output
b035940
{
"best_metric": 0.6572880744934082,
"best_model_checkpoint": "segformer-b0-finetuned-segments-sidewalk-outputs/checkpoint-40",
"epoch": 10.0,
"global_step": 40,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.25,
"learning_rate": 5.9250000000000004e-05,
"loss": 1.4918,
"step": 1
},
{
"epoch": 0.5,
"learning_rate": 5.85e-05,
"loss": 1.2963,
"step": 2
},
{
"epoch": 0.75,
"learning_rate": 5.775e-05,
"loss": 1.311,
"step": 3
},
{
"epoch": 1.0,
"learning_rate": 5.6999999999999996e-05,
"loss": 1.2207,
"step": 4
},
{
"epoch": 1.25,
"learning_rate": 5.625e-05,
"loss": 1.2096,
"step": 5
},
{
"epoch": 1.5,
"learning_rate": 5.550000000000001e-05,
"loss": 1.1354,
"step": 6
},
{
"epoch": 1.75,
"learning_rate": 5.475e-05,
"loss": 1.0495,
"step": 7
},
{
"epoch": 2.0,
"learning_rate": 5.4000000000000005e-05,
"loss": 1.1637,
"step": 8
},
{
"epoch": 2.25,
"learning_rate": 5.325e-05,
"loss": 0.894,
"step": 9
},
{
"epoch": 2.5,
"learning_rate": 5.25e-05,
"loss": 0.9463,
"step": 10
},
{
"epoch": 2.75,
"learning_rate": 5.1750000000000004e-05,
"loss": 1.1739,
"step": 11
},
{
"epoch": 3.0,
"learning_rate": 5.1e-05,
"loss": 0.8987,
"step": 12
},
{
"epoch": 3.25,
"learning_rate": 5.025e-05,
"loss": 0.9905,
"step": 13
},
{
"epoch": 3.5,
"learning_rate": 4.95e-05,
"loss": 0.7884,
"step": 14
},
{
"epoch": 3.75,
"learning_rate": 4.875e-05,
"loss": 0.7571,
"step": 15
},
{
"epoch": 4.0,
"learning_rate": 4.8e-05,
"loss": 0.8714,
"step": 16
},
{
"epoch": 4.25,
"learning_rate": 4.7249999999999997e-05,
"loss": 0.6458,
"step": 17
},
{
"epoch": 4.5,
"learning_rate": 4.6500000000000005e-05,
"loss": 0.9995,
"step": 18
},
{
"epoch": 4.75,
"learning_rate": 4.575e-05,
"loss": 0.7726,
"step": 19
},
{
"epoch": 5.0,
"learning_rate": 4.5e-05,
"loss": 0.8034,
"step": 20
},
{
"epoch": 5.0,
"eval_loss": 1.0316002368927002,
"eval_mean_accuracy": 0.7020655253953606,
"eval_mean_iou": 0.364862905521422,
"eval_overall_accuracy": 0.6520662124792448,
"eval_per_category_accuracy": [
NaN,
0.41182481182481184,
0.7103931261770244,
0.9839786381842457
],
"eval_per_category_iou": [
0.0,
0.41003696685408964,
0.6273174914575999,
0.42209716377399836
],
"eval_runtime": 2.5163,
"eval_samples_per_second": 0.795,
"eval_steps_per_second": 0.397,
"step": 20
},
{
"epoch": 5.25,
"learning_rate": 4.4250000000000005e-05,
"loss": 0.814,
"step": 21
},
{
"epoch": 5.5,
"learning_rate": 4.35e-05,
"loss": 0.5981,
"step": 22
},
{
"epoch": 5.75,
"learning_rate": 4.275e-05,
"loss": 0.6839,
"step": 23
},
{
"epoch": 6.0,
"learning_rate": 4.2e-05,
"loss": 0.6713,
"step": 24
},
{
"epoch": 6.25,
"learning_rate": 4.125e-05,
"loss": 0.7819,
"step": 25
},
{
"epoch": 6.5,
"learning_rate": 4.05e-05,
"loss": 0.6704,
"step": 26
},
{
"epoch": 6.75,
"learning_rate": 3.975e-05,
"loss": 0.6488,
"step": 27
},
{
"epoch": 7.0,
"learning_rate": 3.9e-05,
"loss": 0.6917,
"step": 28
},
{
"epoch": 7.25,
"learning_rate": 3.825e-05,
"loss": 0.5342,
"step": 29
},
{
"epoch": 7.5,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.9604,
"step": 30
},
{
"epoch": 7.75,
"learning_rate": 3.6750000000000006e-05,
"loss": 0.682,
"step": 31
},
{
"epoch": 8.0,
"learning_rate": 3.6e-05,
"loss": 0.4576,
"step": 32
},
{
"epoch": 8.25,
"learning_rate": 3.525e-05,
"loss": 0.5357,
"step": 33
},
{
"epoch": 8.5,
"learning_rate": 3.45e-05,
"loss": 0.5737,
"step": 34
},
{
"epoch": 8.75,
"learning_rate": 3.375e-05,
"loss": 0.5618,
"step": 35
},
{
"epoch": 9.0,
"learning_rate": 3.3e-05,
"loss": 0.5151,
"step": 36
},
{
"epoch": 9.25,
"learning_rate": 3.225e-05,
"loss": 0.6138,
"step": 37
},
{
"epoch": 9.5,
"learning_rate": 3.15e-05,
"loss": 0.7417,
"step": 38
},
{
"epoch": 9.75,
"learning_rate": 3.0749999999999995e-05,
"loss": 0.5539,
"step": 39
},
{
"epoch": 10.0,
"learning_rate": 3e-05,
"loss": 0.5346,
"step": 40
},
{
"epoch": 10.0,
"eval_loss": 0.6572880744934082,
"eval_mean_accuracy": 0.8304500516094228,
"eval_mean_iou": 0.512457610913599,
"eval_overall_accuracy": 0.8077099766100484,
"eval_per_category_accuracy": [
NaN,
0.7121329121329122,
0.8221295315442562,
0.9570877111511
],
"eval_per_category_iou": [
0.0,
0.7026242881329845,
0.725572305974316,
0.6216338495470954
],
"eval_runtime": 2.5198,
"eval_samples_per_second": 0.794,
"eval_steps_per_second": 0.397,
"step": 40
}
],
"max_steps": 80,
"num_train_epochs": 20,
"total_flos": 1402430470225920.0,
"trial_name": null,
"trial_params": null
}