SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1504.08083 · 2015

Fast R-CNN

Ross Girshick

arXiv · PDF · Open in the Atlas

Code that ran

5 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 3 failures of our instrument; 3 no recorded run), measured 2026-11-21; 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/1504.08083.json · badge: /api/ran/1504.08083.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
bmstu-iu8-g1-2019-project/road-signs-recognition community
Papers with Code archive
2 ran · 1 no recorded run
zjZSTU/Fast-R-CNN community
Papers with Code archive
2 ran · 0 no recorded run
serengil/retinaface community
Papers with Code archive
1 ran · 0 no recorded run
mrnabati/RRPN community
Papers with Code archive
0 ran · 1 no recorded run
rh01/fast-rnn community
Papers with Code archive
0 ran · 1 no recorded run
FunctionStatusWhere it lives
ROI_Pool Ran
ran on a synthesized input · constructed an object; no result was computed
zjZSTU/Fast-R-CNN/py/models/vgg16_roi.py
code served (permissive licence) · get_code("26f153de51efe17c")
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 an instance of its own class, ROI_Pool: the run constructed the object and did not call it, so no result was computed. 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)

# ROI_Pool.__init__ expects:
# - size: a tuple/list of length 2 representing (height, width)

# ROI_Pool.forward expects:
# - feature_maps: a 4D tensor of shape (N, C, H, W)

# Construct the arguments
size = (8, 6)  # (height, width) - distinct dimensions for probing

# Create a small 4D feature map tensor: (batch=2, channels=4, height=16, width=12)
feature_maps = torch.randn(2, 4, 16, 12)

ARGS = {
    'size': size
}
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 ROI_Pool — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ROI_Pool 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
VGG16_RoI Ran
ran on a synthesized input · constructed an object; no result was computed
zjZSTU/Fast-R-CNN/py/models/vgg16_roi.py
code served (permissive licence) · get_code("fe710a62ca902ab4")
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 an instance of its own class, VGG16_RoI: the run constructed the object and did not call it, so no result was computed. 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 torchvision.models as models
import types as _types

torch.manual_seed(0)

class ROI_Pool(nn.Module):
    def __init__(self, size):
        super(ROI_Pool, self).__init__()
        assert len(size) == 2, 'size参数输入(长, 宽)'
        pool_func = nn.AdaptiveMaxPool2d
        self.roi_pool = pool_func(size)

    def forward(self, feature_maps):
        assert feature_maps.dim() == 4, 'Expected 4D input of (N, C, H, W)'
        return self.roi_pool(feature_maps)

roi_pool = _types.SimpleNamespace(**{n: globals()[n] for n in ['ROI_Pool'] if n in globals()})

class VGG16_RoI(nn.Module):
    def __init__(self, num_classes=1000, init_weights=True):
        super(VGG16_RoI, self).__init__()
        feature_list = [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 'M', 512, 512, 512, 'M', 512, 512, 512]
        self.features = models.vgg.make_layers(feature_list)
        self.roipool = roi_pool.ROI_Pool((7, 7))
        self.classifier = nn.Sequential(nn.Linear(512 * 7 * 7, 4096), nn.ReLU(True), nn.Dropout(), nn.Linear(4096, 4096), nn.ReLU(True), nn.Dropout())
        self.softmax = nn.Linear(4096, num_classes + 1)
        self.bbox = nn.Linear(4096, num_classes * 4)
        if init_weights:
            self._initialize_weights()

    def forward(self, x):
        x = self.features(x)
        x = self.roipool(x)
        x = torch.flatten(x, 1)
        x = self.classifier(x)
        classify = self.softmax(x)
        regression = self.bbox(x)
        return (classify, regression)

    def _initialize_weights(self):
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
                if m.bias is not None:
                    nn.init.constant_(m.bias, 0)
            elif isinstance(m, nn.BatchNorm2d):
                nn.init.constant_(m.weight, 1)
                nn.init.constant_(m.bias, 0)
            elif isinstance(m, nn.Linear):
                nn.init.normal_(m.weight, 0, 0.01)
                nn.init.constant_(m.bias, 0)

ARGS = {
    'num_classes': 10,
    'init_weights': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type VGG16_RoI — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a VGG16_RoI 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
class_indices Ran
ran; our contract draft was wrong, not the code
bmstu-iu8-g1-2019-project/road-signs-recognition/sources/detector/detector.py
code served (permissive licence) · get_code("4f0b88a82728e9d8")
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 torch
import numpy as np

torch.manual_seed(0)

# The function expects classes_arr to be array-like with len() and indexing support
# It maps each element to its index position
classes_arr = ['class_a', 'class_b', 'class_c', 'class_d']

ARGS = {
    'classes_arr': classes_arr
}
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 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
cpu_nms Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
serengil/retinaface/retinaface/commons/postprocess.py
code served (permissive licence) · get_code("03e0b25aa5a59339")
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 list, ndim 1, dtype family int; 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 sample detection boxes with format [x1, y1, x2, y2, score]
# We'll create a small set of detections with some overlapping boxes
dets = np.array([
    [10.0, 10.0, 50.0, 50.0, 0.9],   # High confidence box
    [15.0, 15.0, 55.0, 55.0, 0.8],   # Overlapping box with lower confidence
    [100.0, 100.0, 150.0, 150.0, 0.85],  # Non-overlapping box
    [105.0, 105.0, 155.0, 155.0, 0.7],   # Overlapping with previous
], dtype=np.float32)

threshold = 0.5

ARGS = {
    "dets": dets,
    "threshold": threshold,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — 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 int — 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
on_index_wrapper Ran
ran; our contract draft was wrong, not the code
bmstu-iu8-g1-2019-project/road-signs-recognition/sources/detector/detector.py
code served (permissive licence) · get_code("370b7a5120a43be6")
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 function. 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 tensorflow as tf
import keras.backend as K
import numpy as np

# Define a simple metric function that takes true and pred labels
def simple_metric(true_labels, pred_labels):
    """A simple metric function that returns mean squared error"""
    return tf.reduce_mean(tf.square(true_labels - pred_labels))

# Setup the arguments
classes_arr = np.array(['cat', 'dog', 'bird'])
class_indices_dict = {'cat': 0, 'dog': 1, 'bird': 2}
background_label = 'cat'
function = simple_metric

ARGS = {
    'function': function,
    'class_indices_dict': class_indices_dict,
    'background_label': background_label,
    'name': 'test_metric'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
add_fast_rcnn_outputs Not yet run
unverified (harvested, no recorded run)
mrnabati/RRPN/detectron/detectron/modeling/fast_rcnn_heads.py
code served (permissive licence) · get_code("42384bd3a48004e8")
How it ran

no invocation recorded.

create_detector_model Not yet run
unverified (harvested, no recorded run)
bmstu-iu8-g1-2019-project/road-signs-recognition/sources/detector/detector.py
code served (permissive licence) · get_code("af5421ea89189612")
How it ran

no invocation recorded.

prepare_roidb Not yet run
unverified (harvested, no recorded run)
rh01/fast-rnn/lib/roi_data_layer/roidb.py
pointer only (licence: NOASSERTION) · get_code("432429bede769b4d")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, rh01/fast-rnn (NOASSERTION). 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. 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 2026-11-21.

Repositories linked to this paper

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

Abstract

This paper proposes a Fast Region-based Convolutional Network method (Fast R-CNN) for object detection. Fast R-CNN builds on previous work to efficiently classify object proposals using deep convolutional networks. Compared to previous work, Fast R-CNN employs several innovations to improve training and testing speed while also increasing detection accuracy. Fast R-CNN trains the very deep VGG16 network 9× faster than R-CNN, is 213× faster at test-time, and achieves a higher mAP on PASCAL VOC 2012. Compared to SPPnet, Fast R-CNN trains VGG16 3× faster, tests 10× faster, and is more accurate. Fast R-CNN is implemented in Python and C++ (using Caffe) and is available under the open-source MIT License at https: //github.com/rbgirshick/fast-rcnn.

For agents

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

get_harvested_code_for_paper("1504.08083")
get_code_for_paper("1504.08083")
have("1504.08083")

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

curl https://syntology.ai/api/ran/1504.08083.json

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

[![Syntology run record](https://syntology.ai/api/ran/1504.08083.svg)](https://syntology.ai/paper/1504.08083)

Connect an agent — have() is free.