Test Marginal Distribution

Setting

config = {
    "ignore": ["order_item"],
    "train": {
        "epochs": 500,
        "lr": 0.01,
        "samples": None,
        "naive_dequantization": True,
    },
    "finetune": {
        "activate": False,
        "samples": 1000,
        "epochs": 50,
        "lr": 0.01,
    },
    "eval": {
        "activate": True,
        "samples": 1000,
    },
    "partitions": 5,
}

model_config = EinetConfig(
    num_channels=1,
    num_leaves=5,
    depth=2,
    num_sums=5,
    num_repetitions=5,
    num_classes=1,
    leaf_type=Normal,
    layer_type='einsum',
    dropout=0.0
)

Training

Generating evaluation queries...
Preprocessing data...
Using naive dequantization...
Training model...

Evaluating model...
GMQ: 2.1544432608222936;
50%: 2.1131607190719963;
90%: 2.390177847321101;
95%: 3.5446638521435294;
MAX: 4.285541153729266
Evaluating model...
GMQ: 2.1544432608222936;
50%: 2.1131607190719963;
90%: 2.390177847321101;
95%: 3.5446638521435294;
MAX: 4.285541153729266

Results

Column 0

Column 1

Column 2

Column 3

Column 4

Column 5

Column 6