SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1901.07441 · 2019

PadChest: A large chest x-ray image dataset with multi-label annotated reports

arXiv · PDF · Open in the Atlas

Code that ran

11 of 16 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; 1 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 4 failures of our instrument; 5 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/1901.07441.json · badge: /api/ran/1901.07441.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
auriml/Rx-thorax-automatic-captioning canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
rajpurkarlab/chexzero community (classifier label)
classifier basis not recorded on the link
9 ran · 5 no recorded run
FunctionStatusWhere it lives
basic_clean Ran
ran; our contract draft was wrong, not the code
rajpurkarlab/chexzero/simple_tokenizer.py
pointer only (licence: MIT); the stored copy get_code serves, from ltpwy/MSCI, is pointer only (licence: NONE) · get_code("98f385d847636a3e")
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 ltpwy/MSCI, 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.

evaluate_model Ran
ran on a synthesized input
rajpurkarlab/chexzero/metrics.py
code served (permissive licence) · get_code("5dba521f400b96c2")
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 torch.nn as nn
import os
import tempfile
from pathlib import Path

torch.manual_seed(0)

# Create a minimal mock for run_zero_shot since it's called in evaluate_model
def run_zero_shot(cxr_labels, cxr_pair_templates, model_path, cxr_filepath=None, 
                  final_label_path=None, alt_labels_dict=None, softmax_eval=False,
                  context_length=77, pretrained=False, use_bootstrap=False, cutlabels=False):
    """Mock implementation of run_zero_shot"""
    # Return dummy results
    cxr_results = {"accuracy": 0.85, "labels": cxr_labels}
    y_pred = torch.tensor([0, 1, 0, 1], dtype=torch.long)
    return cxr_results, y_pred

# Patch the function into the module namespace
import sys
current_module = sys.modules[__name__]
current_module.run_zero_shot = run_zero_shot

# Create temporary directories for file paths
temp_dir = tempfile.mkdtemp()
X_dir_path = os.path.join(temp_dir, "images")
y_dir_path = os.path.join(temp_dir, "labels.csv")
model_path_str = os.path.join(temp_dir, "model.pt")

# Create the directories
os.makedirs(X_dir_path, exist_ok=True)

# Create dummy files
Path(y_dir_path).touch()
Path(model_path_str).touch()

# Define CXR labels
cxr_labels = ["pneumonia", "fracture", "opacity", "edema"]

# Define optional alternative labels dictionary
alt_labels_dict = {
    "pneumonia": ["pneumonia", "lung infection"],
    "fracture": ["fracture", "bone break"]
}

ARGS = {
    "X_dir": X_dir_path,
    "y_dir": y_dir_path,
    "model_path": model_path_str,
    "cxr_labels": cxr_labels,
    "alt_labels_dict": alt_labels_dict
}
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 — 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
extract_filenames Ran
ran on a synthesized input
rajpurkarlab/chexzero/preprocess_padchest.py
code served (permissive licence) · get_code("201f97cd8887fb7f")
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 list, 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 tempfile
import os

# Create a temporary txt file with sample image filenames
# The file has 2 unnecessary lines at top and bottom
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_path = temp_file.name

# Write content with 2 lines at top and bottom to skip
content = """header_line_1
header_line_2
image1.jpg some other info
image2.png some other info
image3.jpg some other info
footer_line_1
footer_line_2"""

temp_file.write(content)
temp_file.close()

ARGS = {
    "txt_path": temp_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list 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
get_best_alt_labels Ran
ran on a synthesized input
rajpurkarlab/chexzero/metrics.py
code served (permissive licence) · get_code("4487d0b083693aeb")
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 pandas as pd

# Analyze the function to understand what it needs:
# - res_df: a DataFrame where columns are named like "path_name_something"
#   and it accesses res_df[col][0] to get mean_auc values
# - cxr_labels: something that supports "in" operator to check if path_name exists

# Create a minimal res_df with a few columns
res_df = pd.DataFrame({
    'pathology_auc': [0.75],
    'finding_auc': [0.82],
    'disease_auc': [0.68],
    'pathology_alt': [0.80],
    'finding_alt': [0.79]
})

# Create cxr_labels as a set or list that contains some of the path names
cxr_labels = {'pathology', 'finding', 'disease'}

ARGS = {
    'res_df': res_df,
    'cxr_labels': cxr_labels
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — 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
get_pairs Ran
ran; our contract draft was wrong, not the code
rajpurkarlab/chexzero/simple_tokenizer.py
code served (permissive licence) · get_code("d919ae32e5e4e616")
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 set. 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

torch.manual_seed(0)

# The function get_pairs takes a word parameter
# A word is represented as a tuple of symbols (variable-length strings)
# The function iterates through the word and creates pairs of consecutive symbols

word = ("hello", "world", "test")

ARGS = {
    "word": word
}
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 set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set 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_unique_labels Ran
ran on a synthesized input
rajpurkarlab/chexzero/preprocess_padchest.py
code served (permissive licence) · get_code("1eca4821b33e6694")
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 list, 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 pandas as pd

# Create a sample DataFrame with labels in string representation format
labels_df = pd.DataFrame({
    'Labels': [
        "['label1', 'label2', 'label3']",
        "['LABEL1', 'label4']",
        "['label2', 'label5']",
        "invalid_format",
        "['label3', 'LABEL4']"
    ]
})

ARGS = {
    'labels_df': labels_df,
    'column': 'Labels'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list 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
labels2set Ran
ran; our contract draft was wrong, not the code
auriml/Rx-thorax-automatic-captioning/multilabel_text_classifier.py
pointer only (licence: NONE) · get_code("02dbfbb304ff1c8b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, auriml/Rx-thorax-automatic-captioning 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.

load_data Ran
ran on a synthesized input
rajpurkarlab/chexzero/data_process.py
code served (permissive licence) · get_code("2ab162b21ee3aeda")
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 DataFrame. 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 pandas as pd
import io

# Create a minimal CSV file in memory
csv_data = """col1,col2,col3
1,2,3
4,5,6
7,8,9"""

# Create a temporary filepath by writing to a StringIO object
# Since we can't do actual file I/O, we'll create a real temp file
import tempfile
import os

temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.csv')
temp_file.write(csv_data)
temp_file.close()

filepath = temp_file.name

ARGS = {
    "filepath": filepath
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 and dtype family int 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
string2numeric_hash Ran
ran, honoured the contract we drafted
auriml/Rx-thorax-automatic-captioning/multilabel_text_classifier.py
pointer only (licence: NONE) · get_code("83daed126452d499")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, auriml/Rx-thorax-automatic-captioning 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.

train_batch Ran
ran on a synthesized input
rajpurkarlab/chexzero/run_train.py
code served (permissive licence) · get_code("8789fc3097e78c26")
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
import torch.nn as nn

torch.manual_seed(0)

# Define a minimal model that matches how train_batch uses it
class CLIPLikeModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear_img = nn.Linear(4, 2)
        self.linear_txt = nn.Linear(4, 2)
    
    def forward(self, images, texts):
        logits_per_image = self.linear_img(images)
        logits_per_text = self.linear_txt(texts)
        return logits_per_image, logits_per_text

# Create model, device, criterion, and optimizer
device = torch.device("cpu")
model = CLIPLikeModel()
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)

# Create tiny batch tensors (batch_size=2, feature_dim=4)
batch_size = 2
images = torch.randn(batch_size, 4)
texts = torch.randn(batch_size, 4)

ARGS = {
    "images": images,
    "texts": texts,
    "model": model,
    "device": device,
    "criterion": criterion,
    "optimizer": optimizer,
}
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
whitespace_clean Ran
ran; our contract draft was wrong, not the code
rajpurkarlab/chexzero/simple_tokenizer.py
pointer only (licence: MIT); the stored copy get_code serves, from ltpwy/MSCI, is pointer only (licence: NONE) · get_code("9542161e9640b858")
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 ltpwy/MSCI, 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.

build_model Not yet run
unverified (harvested, no recorded run)
rajpurkarlab/chexzero/model.py
code served (permissive licence) · get_code("ccfcbefe529dfd77")
How it ran

no invocation recorded.

f1_mcc_bootstrap Not yet run
unverified (harvested, no recorded run)
rajpurkarlab/chexzero/metrics.py
code served (permissive licence) · get_code("8f76b210cb2b9935")
How it ran

no invocation recorded.

get_cxr_paths_list Not yet run
unverified (harvested, no recorded run)
rajpurkarlab/chexzero/data_process.py
code served (permissive licence) · get_code("ed79f26b800fd59c")
How it ran

no invocation recorded.

preprocess Not yet run
unverified (harvested, no recorded run)
rajpurkarlab/chexzero/data_process.py
code served (permissive licence) · get_code("bbb2005719f3fcd1")
How it ran

no invocation recorded.

preprocess_data Not yet run
unverified (harvested, no recorded run)
rajpurkarlab/chexzero/preprocess_padchest.py
code served (permissive licence) · get_code("7c2197559c73f7c3")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 4 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 4 later) and 12 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
get_pairs 258 other papers
3 dated earlier, 254 later, 1 undated · 4 only as an assembly brick
1802.06474 (2018-02)
A Closed-form Solution to Photorealistic Image Stylization
1803.01128 (2018-03)
Seq2Sick: Evaluating the Robustness of Sequence-to-Sequence Models with Adversarial Examples
1809.00125 (2018-09)
Simple Fusion: Return of the Language Model
and 255 more not named here
whitespace_clean 231 other papers
230 later, 1 undated · 4 only as an assembly brick
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2108.10904 (2021-08)
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
2109.08472 (2021-09)
ActionCLIP: A New Paradigm for Video Action Recognition
and 228 more not named here
basic_clean 218 other papers
217 later, 1 undated · 4 only as an assembly brick
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2108.10904 (2021-08)
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
2109.08472 (2021-09)
ActionCLIP: A New Paradigm for Video Action Recognition
and 215 more not named here
build_model 119 other papers
1 dated earlier, 116 later, 2 undated
1809.11096 (2018-09)
Large Scale GAN Training for High Fidelity Natural Image Synthesis
2104.14294 (2021-04)
Emerging Properties in Self-Supervised Vision Transformers
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
and 116 more not named here

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

We present a labeled large-scale, high resolution chest x-ray dataset for the automated exploration of medical images along with their associated reports. This dataset includes more than 160,000 images obtained from 67,000 patients that were interpreted and reported by radiologists at Hospital San Juan Hospital (Spain) from 2009 to 2017, covering six different position views and additional information on image acquisition and patient demography. The reports were labeled with 174 different radiographic findings, 19 differential diagnoses and 104 anatomic locations organized as a hierarchical taxonomy and mapped onto standard Unified Medical Language System (UMLS) terminology. Of these reports, 27% were manually annotated by trained physicians and the remaining set was labeled using a supervised method based on a recurrent neural network with attention mechanisms. The labels generated were then validated in an independent test set achieving a 0.93 Micro-F1 score. To the best of our knowledge, this is one of the largest public chest x-ray database suitable for training supervised models concerning radiographs, and the first to contain radiographic reports in Spanish. The PadChest dataset can be downloaded from http://bimcv.cipf.es/bimcv-projects/padchest/.

For agents

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

get_harvested_code_for_paper("1901.07441")
get_code_for_paper("1901.07441")
have("1901.07441")

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

curl https://syntology.ai/api/ran/1901.07441.json

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

[![Syntology run record](https://syntology.ai/api/ran/1901.07441.svg)](https://syntology.ai/paper/1901.07441)

Connect an agent — have() is free.