SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.01216 · 2021

Evidential Turing Processes

arXiv · PDF · Open in the Atlas

Code that ran

6 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 0 failures of our instrument; 2 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2106.01216.json · badge: /api/ran/2106.01216.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
ituvisionlab/EvidentialCalibration canonical (classifier label)
classifier basis: census
6 ran · 2 no recorded run
FunctionStatusWhere it lives
auroc Ran
ran on a synthesized input
ituvisionlab/EvidentialCalibration/scores.py
code served (permissive licence) · get_code("f48a597c2c8cc253")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from sklearn.metrics import roc_auc_score

torch.manual_seed(0)

# For auroc function, we need:
# - preds: predicted probabilities/scores (array-like, 1D or 2D)
# - target: ground truth binary labels (array-like, 1D)
# - minibatch: boolean flag (not used in function body, but listed as parameter)

# Create small sample data
batch_size = 2
num_samples = 4

# Generate binary targets
target = np.array([0, 1, 1, 0])

# Generate predictions (scores between 0 and 1)
preds = np.array([0.1, 0.9, 0.8, 0.2])

ARGS = {
    "preds": preds,
    "target": target,
    "minibatch": True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
err Ran
ran on a synthesized input
ituvisionlab/EvidentialCalibration/scores.py
code served (permissive licence) · get_code("71198529aba3a9fb")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function expects:
# - preds: a tensor that supports .argmax(1) and != comparison
# - target: a tensor that supports .argmax(1) and != comparison
# - minibatch: a boolean flag

# Create small tensors for preds and target
# preds should have shape (batch_size, num_classes) to support argmax(1)
# target should have shape (batch_size, num_classes) to support argmax(1)
preds = torch.randn(2, 4)  # batch_size=2, num_classes=4
target = torch.randn(2, 4)  # batch_size=2, num_classes=4

ARGS = {
    "preds": preds,
    "target": target,
    "minibatch": True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
nll Ran
ran on a synthesized input
ituvisionlab/EvidentialCalibration/scores.py
code served (permissive licence) · get_code("3d7e17b5dada4cfd")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch as th

th.manual_seed(0)

# Create small tensors for the function
# preds: probability predictions (batch_size=2, num_classes=4)
# Should be probabilities, so values in [0, 1]
preds = th.tensor([
    [0.1, 0.2, 0.3, 0.4],
    [0.25, 0.25, 0.25, 0.25]
], dtype=th.float32)

# target: one-hot encoded targets (batch_size=2, num_classes=4)
target = th.tensor([
    [0.0, 0.0, 1.0, 0.0],
    [1.0, 0.0, 0.0, 0.0]
], dtype=th.float32)

# minibatch: boolean flag
minibatch = True

ARGS = {
    "preds": preds,
    "target": target,
    "minibatch": minibatch
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
prepare_data Ran
ran on a synthesized input
ituvisionlab/EvidentialCalibration/dataprep.py
code served (permissive licence) · get_code("e7a7abab03a8b754")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from torch.utils import data
from torch.utils.data import DataLoader, TensorDataset

torch.manual_seed(0)

# Create minimal fake datasets to avoid downloads
class FakeFashionMNIST:
    def __init__(self, root, train=True, download=False, transform=None):
        self.data = torch.randn(100, 1, 28, 28)
        self.targets = torch.randint(0, 10, (100,))
        self.transform = transform
    
    def __len__(self):
        return len(self.data)
    
    def __getitem__(self, idx):
        img = self.data[idx]
        if self.transform:
            img = self.transform(img)
        return img, self.targets[idx]

class FakeMNIST:
    def __init__(self, root, train=True, download=False, transform=None):
        self.data = torch.randn(60, 1, 28, 28)
        self.targets = torch.randint(0, 10, (60,))
        self.transform = transform
    
    def __len__(self):
        return len(self.data)
    
    def __getitem__(self, idx):
        img = self.data[idx]
        if self.transform:
            img = self.transform(img)
        return img, self.targets[idx]

class FakeKMNIST:
    def __init__(self, root, train=False, download=False, transform=None):
        self.data = torch.randn(30, 1, 28, 28)
        self.targets = torch.randint(0, 10, (30,))
        self.transform = transform
    
    def __len__(self):
        return len(self.data)
    
    def __getitem__(self, idx):
        img = self.data[idx]
        if self.transform:
            img = self.transform(img)
        return img, self.targets[idx]

# Monkey patch the datasets module to use fake datasets
import torchvision.datasets as datasets_module
datasets_module.FashionMNIST = FakeFashionMNIST
datasets_module.MNIST = FakeMNIST
datasets_module.KMNIST = FakeKMNIST

ARGS = {
    'data_set': 'fashion',
    'batch_size': 2,
    'iscuda': False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
prepare_for_training_with_ood Ran
ran on a synthesized input
ituvisionlab/EvidentialCalibration/dataprep.py
code served (permissive licence) · get_code("3bab291577d514ab")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from torch.utils import data

class SimpleDataset(data.Dataset):
    def __init__(self, size):
        self.size = size
    
    def __len__(self):
        return self.size
    
    def __getitem__(self, idx):
        return np.array([idx, idx+1, idx+2, idx+3], dtype=np.float32)

train_dataset = SimpleDataset(10)
ood_dataset = SimpleDataset(15)

ARGS = {
    "train_dataset": train_dataset,
    "ood_dataset": ood_dataset,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 3, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
train Ran
ran on a synthesized input
ituvisionlab/EvidentialCalibration/script.py
code served (permissive licence) · get_code("d25a34aea14dc1bd")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned SGD. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import torch.optim as optim

torch.manual_seed(0)

# Define a minimal model that has a loss method
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def loss(self, data, target):
        output = self.linear(data)
        criterion = nn.CrossEntropyLoss()
        return criterion(output, target)

# Create model
model = MinimalModel()

# Create train_loader: list of (data, target) tuples
train_loader = [
    (torch.randn(2, 4), torch.randint(0, 3, (2,))),
    (torch.randn(2, 4), torch.randint(0, 3, (2,)))
]

# Create ood_loader: list of (ood_data, ood_target) tuples
ood_loader = [
    (torch.randn(2, 4), torch.randint(0, 3, (2,))),
    (torch.randn(2, 4), torch.randint(0, 3, (2,)))
]

# Create optimizer
opt = optim.SGD(model.parameters(), lr=0.01)

# Define ARGS dict with all parameters
ARGS = {
    'model': model,
    'train_loader': train_loader,
    'ood_loader': ood_loader,
    'epoch': 1,
    'opt': opt,
    'iscuda': False,
    'verbose': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SGD — a SGD output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SGD output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_results Not yet run
unverified (harvested, no recorded run)
ituvisionlab/EvidentialCalibration/utils.py
code served (permissive licence) · get_code("5f5462e36c39a5aa")
How it ran

no invocation recorded.

prepare_data_fashion Not yet run
unverified (harvested, no recorded run)
ituvisionlab/EvidentialCalibration/dataprep.py
code served (permissive licence) · get_code("f7d9198cfd460409")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 8 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

A probabilistic classifier with reliable predictive uncertainties i) fits successfully to the target domain data, ii) provides calibrated class probabilities in difficult regions of the target domain (e.g.\ class overlap), and iii) accurately identifies queries coming out of the target domain and rejects them. We introduce an original combination of Evidential Deep Learning, Neural Processes, and Neural Turing Machines capable of providing all three essential properties mentioned above for total uncertainty quantification. We observe our method on five classification tasks to be the only one that can excel all three aspects of total calibration with a single standalone predictor. Our unified solution delivers an implementation-friendly and compute efficient recipe for safety clearance and provides intellectual economy to an investigation of algorithmic roots of epistemic awareness in deep neural nets.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2106.01216")
get_code_for_paper("2106.01216")
have("2106.01216")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2106.01216.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2106.01216.svg)](https://syntology.ai/paper/2106.01216)

Connect an agent — have() is free.