{
"arch": "x -> Linear(88,H) -> relu -> Linear(H,H) -> relu -> Linear(H,38)",
"batch": 4096,
"data_contested_rows": 704695,
"data_git_sha": "8450e19e47981504a7af9b5dea9d4cede1b71d7a",
"data_rows": 1026211,
"data_seed": 1,
"dtype": "f32-le",
"epochs": 150,
"feature_version": 3,
"features_len": 88,
"git_sha": "8450e19e47981504a7af9b5dea9d4cede1b71d7a",
"hidden": 256,
"lr": 0.001,
"param_floats": 98342,
"param_order": [
"l1.weight",
"l1.bias",
"l2.weight",
"l2.bias",
"l3.weight",
"l3.bias"
],
"param_shapes": {
"l1.bias": [
256
],
"l1.weight": [
256,
88
],
"l2.bias": [
256
],
"l2.weight": [
256,
256
],
"l3.bias": [
38
],
"l3.weight": [
38,
256
]
},
"softmax_len": 38,
"teacher": "american()",
"train_rows": 923590,
"trainer": "pons-trainer 0.1.0",
"val_ce": 0.2369789481163025,
"val_rows": 102621,
"val_top1_constructive": 0.9636085033416748,
"val_top1_contested": 0.9476953744888306,
"val_top1_overall": 0.9526510238647461,
"wd": 0.01
}