SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2010.16262 · NeurIPS · 2020

Experimental design for MRI by greedy policy search

Max Welling, Tim Bakker

arXiv · PDF · Open in the Atlas

Code that ran

16 of 18 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; 3 honoured the contract we drafted, 3 violated it and 10 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/2010.16262.json · badge: /api/ran/2010.16262.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
Timsey/pg_mri canonical (classifier label)
classifier basis: census
16 ran · 2 no recorded run
FunctionStatusWhere it lives
acquire_rows_in_batch_parallel Ran
ran on a synthesized input
Timsey/pg_mri/src/policy_model/policy_model_utils.py
code served (permissive licence) · get_code("bb8cc619d6012476")
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 6, 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 torch.nn as nn

torch.manual_seed(0)

# Build small tensors with distinct dimensions for clarity
# k: shape (batch=2, 1, channels=4, rows=5, cols=3)
k = torch.randn(2, 1, 4, 5, 3)

# mask: shape (batch=2, 1, channels=4, rows=5, cols=3) or will be expanded
mask = torch.zeros(2, 1, 4, 5, 3)

# mk: shape (batch=2, 1, channels=4, rows=5, cols=3) or will be expanded
mk = torch.zeros(2, 1, 4, 5, 3)

# to_acquire: shape (batch=2, num_acquisitions=1) containing row indices
# Each element is a tensor with a single row index to acquire
to_acquire = torch.tensor([[1], [2]], dtype=torch.long)

ARGS = {
    'k': k,
    'mk': mk,
    'mask': mask,
    'to_acquire': to_acquire,
}
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 tuple, ndim 6, 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
apply_mask Ran
ran on a synthesized input
Timsey/pg_mri/src/helpers/transforms.py
code served (permissive licence) · get_code("92c6e74cc5959c54")
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. The image it ran in was 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 numpy as np

torch.manual_seed(0)

# Create a small k-space tensor
# Shape: (batch=2, height=4, width=3, complex=2)
data = torch.randn(2, 4, 3, 2)

# Create a mask function that takes a shape and seed and returns a mask tensor
def mask_func(shape, seed):
    """
    Simple mask function that creates a random binary mask.
    Args:
        shape: tuple of ints specifying the shape
        seed: random seed
    Returns:
        mask (torch.Tensor): binary mask with the given shape
    """
    if seed is not None:
        np.random.seed(seed)
    mask = torch.from_numpy(np.random.binomial(1, 0.5, shape).astype(np.float32))
    return mask

ARGS = {
    "data": data,
    "mask_func": mask_func,
    "seed": 42
}
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 — a tuple output has no array shape of its own; none is 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
build_policy_model Ran
ran on a synthesized input
Timsey/pg_mri/src/policy_model/policy_model_def.py
code served (permissive licence) · get_code("c495b239238c593a")
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 PolicyModel. The image it ran in was 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

torch.manual_seed(0)

class Args:
    def __init__(self):
        self.resolution = 4
        self.num_chans = 8
        self.num_layers = 2
        self.drop_prob = 0.1
        self.fc_size = 16
        self.device = 'cpu'

ARGS = {'args': Args()}
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 PolicyModel — a PolicyModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PolicyModel 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
build_reconstruction_model Ran
ran on a synthesized input
Timsey/pg_mri/src/reconstruction_model/reconstruction_model_def.py
code served (permissive licence) · get_code("31fbaa88a3b3f5d7")
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 UnetModel. The image it ran in was 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
from torch import nn

torch.manual_seed(0)

class ArgsConfig:
    def __init__(self, num_chans, num_pools, drop_prob, device):
        self.num_chans = num_chans
        self.num_pools = num_pools
        self.drop_prob = drop_prob
        self.device = device

ARGS = {
    'args': ArgsConfig(
        num_chans=32,
        num_pools=3,
        drop_prob=0.1,
        device='cpu'
    )
}
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 UnetModel — a UnetModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a UnetModel 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
compute_snr Ran
ran on a synthesized input
Timsey/pg_mri/src/compute_snr.py
code served (permissive licence) · get_code("2703e15dc7d86837")
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 numpy as np
import pickle
import tempfile
import os

# Create a simple args class with required attributes
class Args:
    def __init__(self):
        self.data_runs = 2

# Create temporary directory for pickle files
temp_dir = tempfile.mkdtemp()

# Generate mock gradient data
# Shape: (num_batches * data_runs, last_layer_size, second_to_last_layer_size + 1)
# Let's use: 4 total batches (2 runs × 2 batches per run), last_layer_size=3, second_to_last_layer_size+1=5
num_batches_per_run = 2
data_runs = 2
last_layer_size = 3
feature_size = 5

# Create weight gradients: (num_batches * data_runs, last_layer_size, feature_size - 1)
weight_grads_list = [
    np.random.randn(last_layer_size, feature_size - 1).astype(np.float32)
    for _ in range(num_batches_per_run * data_runs)
]

# Create bias gradients: (num_batches * data_runs, last_layer_size)
bias_grads_list = [
    np.random.randn(last_layer_size).astype(np.float32)
    for _ in range(num_batches_per_run * data_runs)
]

# Save to pickle files
weight_path = os.path.join(temp_dir, "weights.pkl")
bias_path = os.path.join(temp_dir, "bias.pkl")

with open(weight_path, 'wb') as f:
    pickle.dump(weight_grads_list, f)

with open(bias_path, 'wb') as f:
    pickle.dump(bias_grads_list, f)

# Create args object
args = Args()

ARGS = {
    "args": args,
    "weight_path": weight_path,
    "bias_path": bias_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
compute_ssim Ran
ran on a synthesized input
Timsey/pg_mri/src/helpers/torch_metrics.py
code served (permissive licence) · get_code("cfa8a628b40e30d9")
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. 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, 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 torch.nn.functional as F
from torch.autograd import Variable
from math import exp

torch.manual_seed(0)

# Create two small sample images
# Shape: (batch_size, channels, height, width)
img1 = torch.randn(2, 3, 4, 4)
img2 = torch.randn(2, 3, 4, 4)

# Ensure they are in the valid range [0, 1] for SSIM computation
img1 = torch.sigmoid(img1)
img2 = torch.sigmoid(img2)

ARGS = {
    "img1": img1,
    "img2": img2,
    "window_size": 11,
    "size_average": True,
    "data_range": 1.0
}
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
create_window Ran
ran, honoured the contract we drafted
Timsey/pg_mri/src/helpers/torch_metrics.py
code served (permissive licence) · get_code("6154e3744ece5728")
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 4, dtype family float, shape (3, 1, 5, 5) 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
from torch.autograd import Variable
from math import exp

torch.manual_seed(0)

def gaussian(window_size, sigma):
    gauss = torch.Tensor([exp(-(x - window_size // 2) ** 2 / float(2 * sigma ** 2)) for x in range(window_size)])
    return gauss / gauss.sum()

def create_window(window_size, channel):
    _1D_window = gaussian(window_size, 1.5).unsqueeze(1)
    _2D_window = _1D_window.mm(_1D_window.t()).float().unsqueeze(0).unsqueeze(0)
    window = Variable(_2D_window.expand(channel, 1, window_size, window_size).contiguous())
    return window

ARGS = {
    "window_size": 5,
    "channel": 3
}
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 4, dtype family float, shape (3, 1, 5, 5) — shape (3, 1, 5, 5) 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
evaluate_loss Ran
ran on a synthesized input
Timsey/pg_mri/src/train_reconstruction.py
code served (permissive licence) · get_code("47421e4b28120240")
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 torch.nn as nn
from torch.utils.tensorboard import SummaryWriter
import tempfile
import os

torch.manual_seed(0)

# Create a minimal reconstruction model
class ReconModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv2d(1, 1, kernel_size=3, padding=1)
    
    def forward(self, x):
        return self.conv(x)

# Create a minimal args object
class Args:
    def __init__(self):
        self.device = 'cpu'

# Create a minimal data loader (list of tuples with 9 elements each)
# The function unpacks: _, _, _, input, target, _, _, _, _ = data
data_loader = []
for i in range(2):
    # Create batch with 2 samples, each 4x3 spatial dimensions
    input_data = torch.randn(2, 4, 3)
    target_data = torch.randn(2, 4, 3)
    # Tuple of 9 elements as unpacked in the function
    data_tuple = (None, None, None, input_data, target_data, None, None, None, None)
    data_loader.append(data_tuple)

# Create a temporary directory for tensorboard writer
temp_dir = tempfile.mkdtemp()
writer = SummaryWriter(log_dir=temp_dir)

# Construct arguments
model = ReconModel()
args = Args()
epoch = 1

ARGS = {
    'args': args,
    'epoch': epoch,
    'model': model,
    'data_loader': data_loader,
    'writer': writer
}
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 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
gaussian Ran
ran, honoured the contract we drafted
Timsey/pg_mri/src/helpers/torch_metrics.py
code served (permissive licence) · get_code("c56b7ef16f309a45")
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) and returned Tensor, ndim 1, dtype family float, shape (5,) 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
from math import exp

torch.manual_seed(0)

ARGS = {
    "window_size": 5,
    "sigma": 1.0,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 (5,) — shape (5,) 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
load_json Ran
ran on a synthesized input
Timsey/pg_mri/src/helpers/utils.py
code served (permissive licence) · get_code("4ee4f4cc01e655ca")
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 dict. The image it ran in was 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 json
import tempfile
import os

# Create a temporary JSON file with some test data
temp_dir = tempfile.mkdtemp()
test_json_path = os.path.join(temp_dir, "test.json")

test_data = {
    "key1": "value1",
    "key2": [1, 2, 3],
    "key3": {"nested": "object"}
}

with open(test_json_path, 'w') as f:
    json.dump(test_data, f)

ARGS = {
    "path": test_json_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
mse Ran
ran, violated the contract we drafted
Timsey/pg_mri/src/reconstruction_model/reconstruction_model_utils.py
code served (permissive licence) · get_code("f8ee64e99d4c3f8e")
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 float64, ndim 0, 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

# Create small ground truth and prediction arrays
gt = np.array([[1.0, 2.0, 3.0, 4.0],
               [5.0, 6.0, 7.0, 8.0]])

pred = np.array([[1.1, 2.1, 2.9, 4.2],
                 [4.9, 6.1, 7.1, 7.8]])

ARGS = {
    "gt": gt,
    "pred": pred
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
nmse Ran
ran, violated the contract we drafted
Timsey/pg_mri/src/reconstruction_model/reconstruction_model_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from j-duan/VS-Net, is pointer only (licence: NONE) · get_code("d6ba4d694236aa92")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from j-duan/VS-Net, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

snr_from_grads Ran
ran on a synthesized input
Timsey/pg_mri/src/compute_snr.py
code served (permissive licence) · get_code("fffbeee23ca15eae")
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 numpy as np

class Args:
    def __init__(self, data_runs):
        self.data_runs = data_runs

# Create a small arguments object
data_runs = 2
args = Args(data_runs=data_runs)

# Create gradient array with shape (num_batches * data_runs, last_layer_size, second_to_last_layer_size + 1)
# Total batches across runs: 4 (2 runs × 2 batches per run)
# last_layer_size: 3
# second_to_last_layer_size + 1: 5
num_batches_per_run = 2
total_batches = num_batches_per_run * data_runs
last_layer_size = 3
second_to_last_layer_size_plus_1 = 5

grads = np.random.randn(total_batches, last_layer_size, second_to_last_layer_size_plus_1).astype(np.float32)

ARGS = {
    'args': args,
    'grads': grads
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
str2bool Ran
ran, violated the contract we drafted
Timsey/pg_mri/src/helpers/utils.py
code served (permissive licence) · get_code("1c31f800b71774a1")
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 bool, 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

def str2bool(v):
    if isinstance(v, bool):
        return v
    if v.lower() in ('yes', 'true', 't', 'y', '1'):
        return True
    elif v.lower() in ('no', 'false', 'f', 'n', '0'):
        return False
    else:
        raise ValueError('Boolean value expected.')

ARGS = {
    'v': '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 bool, 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
to_tensor Ran
ran, honoured the contract we drafted
Timsey/pg_mri/src/helpers/transforms.py
code served (permissive licence) · get_code("fb8b8d6173e8bede")
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 2, dtype family float, shape (2, 3) 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 numpy as np
import torch

torch.manual_seed(0)

# Test case 1: Real-valued numpy array
real_array = np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]], dtype=np.float32)

# Test case 2: Complex-valued numpy array
complex_array = np.array([[1.0 + 2.0j, 3.0 + 4.0j], [5.0 + 6.0j, 7.0 + 8.0j]], dtype=np.complex64)

# Create ARGS dict with both test cases
# The function takes a single 'data' parameter
ARGS = {
    "data": real_array
}
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 Tensor, ndim 2, dtype family float, shape (2, 3) — shape (2, 3) 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
train_epoch Ran
ran on a synthesized input
Timsey/pg_mri/src/train_reconstruction.py
code served (permissive licence) · get_code("99ec6e3ba8fb96b6")
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 torch.nn as nn
from torch.optim import Adam
from unittest.mock import MagicMock
import logging

torch.manual_seed(0)

# Setup logging
logging.basicConfig(level=logging.INFO)

# Create a minimal reconstruction model
class ReconModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv2d(1, 1, kernel_size=3, padding=1)
    
    def forward(self, x):
        return self.conv(x)

model = ReconModel()

# Create mock args
class Args:
    def __init__(self):
        self.device = 'cpu'
        self.report_interval = 1
        self.num_epochs = 2

args = Args()

# Create data loader with 9-tuples as expected by the function
# Each batch item: (_, _, _, input, target, _, _, _, _)
data_loader = []
batch_size = 2
for batch_idx in range(3):
    # input shape: (batch, height, width)
    # target shape: (batch, height, width)
    input_tensor = torch.randn(batch_size, 4, 4)
    target_tensor = torch.randn(batch_size, 4, 4)
    
    # Create 9-tuple with placeholders
    data_item = (
        None,  # placeholder 1
        None,  # placeholder 2
        None,  # placeholder 3
        input_tensor,   # input
        target_tensor,  # target
        None,  # placeholder 6
        None,  # placeholder 7
        None,  # placeholder 8
        None   # placeholder 9
    )
    data_loader.append(data_item)

# Create optimizer
optimizer = Adam(model.parameters(), lr=0.001)

# Create mock tensorboard writer
writer = MagicMock()
writer.add_scalar = MagicMock()

# Define epoch number
epoch = 0

ARGS = {
    'args': args,
    'epoch': epoch,
    'model': model,
    'data_loader': data_loader,
    'optimizer': optimizer,
    'writer': writer
}
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 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
fft2 Not yet run
unverified (harvested, no recorded run)
Timsey/pg_mri/src/helpers/transforms.py
code served (permissive licence) · get_code("9a768de7e61bd97d")
How it ran

no invocation recorded.

str2none Not yet run
unverified (harvested, no recorded run)
Timsey/pg_mri/src/helpers/utils.py
code served (permissive licence) · get_code("87e85996601fcb42")
How it ran

no invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 distinct functions with an identity, 7 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 7 later) and 11 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.

FunctionAlso present inEarliest-datedOthers named
gaussian 160 other papers
7 dated earlier, 153 later · 7 only as an assembly brick
1812.11941 (2018-12)
High Quality Monocular Depth Estimation via Transfer Learning
1904.01538 (2019-04)
Spatial Attentive Single-Image Deraining with a High Quality Real Rain Dataset
1907.11628 (2019-07)
Unsupervised Learning for Optical Flow Estimation Using Pyramid Convolution LSTM
and 157 more not named here
create_window 52 other papers
5 dated earlier, 47 later · 5 only as an assembly brick
1904.01538 (2019-04)
Spatial Attentive Single-Image Deraining with a High Quality Real Rain Dataset
1907.11628 (2019-07)
Unsupervised Learning for Optical Flow Estimation Using Pyramid Convolution LSTM
2002.11616 (2020-02)
Zooming Slow-Mo: Fast and Accurate One-Stage Space-Time Video Super-Resolution
and 49 more not named here
str2bool 5 other papers
5 later
2201.06648 (2022-01)
OmniPrint: A Configurable Printed Character Synthesizer
2207.12020 (2022-07)
Domain-invariant Feature Exploration for Domain Generalization
2305.02164 (2023-05)
Nonparametric Generative Modeling with Conditional Sliced-Wasserstein Flows
and 2 more not named here
nmse 3 other papers
2 dated earlier, 1 later
1907.10033 (2019-07)
VS-Net: Variable splitting network for accelerated parallel MRI reconstruction
1912.09278 (2019-12)
$Σ$-net: Systematic Evaluation of Iterative Deep Neural Networks for Fast Parallel MR Image Reconstruction
2604.12709 (2026-04)
Information-Theoretic Optimization for Task-Adapted Compressed Sensing Magnetic Resonance Imaging
mse 2 other papers
1 dated earlier, 1 later
1912.09278 (2019-12)
$Σ$-net: Systematic Evaluation of Iterative Deep Neural Networks for Fast Parallel MR Image Reconstruction
2604.12709 (2026-04)
Information-Theoretic Optimization for Task-Adapted Compressed Sensing Magnetic Resonance Imaging
load_json 1 other paper
1 later
2112.07133 (2021-12)
CLIP-Lite: Information Efficient Visual Representation Learning with Language Supervision
to_tensor 1 other paper
1 later
aaai_25475 (2023 (year only))
Deep Equilibrium Models for Snapshot Compressive Imaging

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

In today's clinical practice, magnetic resonance imaging (MRI) is routinely accelerated through subsampling of the associated Fourier domain. Currently, the construction of these subsampling strategies -known as experimental design -relies primarily on heuristics. We propose to learn experimental design strategies for accelerated MRI with policy gradient methods. Unexpectedly, our experiments show that a simple greedy approximation of the objective leads to solutions nearly on-par with the more general non-greedy approach. We offer a partial explanation for this phenomenon rooted in greater variance in the non-greedy objective's gradient estimates, and experimentally verify that this variance hampers non-greedy models in adapting their policies to individual MR images. We empirically show that this adaptivity is key to improving subsampling designs.

For agents

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

get_harvested_code_for_paper("2010.16262")
get_code_for_paper("2010.16262")
have("2010.16262")

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

curl https://syntology.ai/api/ran/2010.16262.json

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

[![Syntology run record](https://syntology.ai/api/ran/2010.16262.svg)](https://syntology.ai/paper/2010.16262)

Connect an agent — have() is free.