SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1506.05908 · 2015

Deep Knowledge Tracing

arXiv · PDF · Open in the Atlas

Code that ran

10 of 12 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, 1 violated it and 9 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/1506.05908.json · badge: /api/ran/1506.05908.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
qqhann/KnowledgeTracing community (classifier label)
classifier basis not recorded on the link
9 ran · 2 no recorded run
ZhijieXiong/pyedmine reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_bes Ran
ran, violated the contract we drafted
ZhijieXiong/pyedmine/edmine/evaluator/SequentialDLKTEvaluator.py
code served (permissive licence) · get_code("2316c44149b4355c")
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 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

torch.manual_seed(0)

# The function takes three lists of numeric values
# all_predict_score: predicted confidence scores (0-1)
# all_r: reference scores (0-1)
# all_correctness: binary correctness labels (0 or 1)

# Create small lists of values for each parameter
all_predict_score = [0.9, 0.3, 0.7, 0.2]
all_r = [0.85, 0.4, 0.75, 0.1]
all_correctness = [1, 0, 1, 0]

ARGS = {
    "all_predict_score": all_predict_score,
    "all_r": all_r,
    "all_correctness": all_correctness,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, 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
get_Seq2Seq Ran
ran on a synthesized input
qqhann/KnowledgeTracing/model/seq2seq.py
code served (permissive licence) · get_code("ee0926c66c8a2526")
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 Seq2Seq. 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)

# Define parameters for get_Seq2Seq
NUM_EMBEDDINGS = 100
ENC_EMB_DIM = 8
HID_DIM = 16
N_LAYERS = 2
ENC_DROPOUT = 0.5
OUTPUT_DIM = 250  # Must be > 2 + 124 + 124 = 250 based on forward slicing
DEC_EMB_DIM = 8
DEC_DROPOUT = 0.5
dev = 'cpu'

ARGS = {
    'NUM_EMBEDDIGNS': NUM_EMBEDDINGS,
    'ENC_EMB_DIM': ENC_EMB_DIM,
    'HID_DIM': HID_DIM,
    'N_LAYERS': N_LAYERS,
    'ENC_DROPOUT': ENC_DROPOUT,
    'OUTPUT_DIM': OUTPUT_DIM,
    'DEC_EMB_DIM': DEC_EMB_DIM,
    'DEC_DROPOUT': DEC_DROPOUT,
    'dev': dev,
}
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 Seq2Seq — a Seq2Seq output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Seq2Seq 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_exp_paths Ran
ran on a synthesized input
qqhann/KnowledgeTracing/src/path.py
code served (permissive licence) · get_code("1c9910ac2237898e")
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
from pathlib import Path

# Create a temporary project directory structure
temp_dir = tempfile.TemporaryDirectory()
projectdir = Path(temp_dir.name)

# Create the config directory structure
config_name = "test_config"
configdir = projectdir / 'config' / config_name
configdir.mkdir(parents=True, exist_ok=True)

# Create some dummy .json files in the config directory
(configdir / "exp1.json").touch()
(configdir / "exp2.json").touch()
(configdir / "exp3.json").touch()

ARGS = {
    "projectdir": projectdir,
    "config_name": config_name,
}
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 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_knowledge_concepts_dict Ran
ran on a synthesized input
qqhann/KnowledgeTracing/src/data.py
code served (permissive licence) · get_code("5b9a73137a87c6ea")
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

# Construct valid arguments for get_knowledge_concepts_dict
# The function expects:
# data: List[List[Tuple[int, int]]] - a list of sequences, where each sequence
#       is a list of tuples containing (question_id, answer_id)

# Create sample data with distinct question IDs
data = [
    [(1, 0), (2, 1), (3, 0)],
    [(2, 1), (4, 0), (1, 1)],
    [(5, 0), (3, 1)]
]

ARGS = {
    "data": data
}
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
get_logger Ran
ran on a synthesized input
qqhann/KnowledgeTracing/src/log.py
code served (permissive licence) · get_code("8137614e705c6de1")
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 Logger. 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 logging
import tempfile
import os

# Create a temporary file for logging output
temp_dir = tempfile.gettempdir()
log_filename = os.path.join(temp_dir, "test_logger.log")

ARGS = {
    "name": "test_logger",
    "filename": log_filename
}
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 Logger — a Logger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Logger 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_loss_batch_seq2seq Ran
ran on a synthesized input
qqhann/KnowledgeTracing/model/seq2seq.py
code served (permissive licence) · get_code("99f511bc70adf79d")
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 torch
import torch.nn as nn

torch.manual_seed(0)

# Build a minimal seq2seq model that matches the function's usage
class SimpleSeq2SeqModel(nn.Module):
    def __init__(self, input_size, hidden_size, output_size):
        super().__init__()
        self.hidden_size = hidden_size
        self.encoder = nn.LSTM(input_size, hidden_size, batch_first=False)
        self.decoder = nn.LSTM(input_size, hidden_size, batch_first=False)
        self.fc_out = nn.Linear(hidden_size, output_size)
        self.fc_prob = nn.Linear(hidden_size, output_size)
    
    def forward(self, input_src, input_trg):
        # input_src: (seq_len, batch_size, input_size)
        # input_trg: (seq_len, batch_size, input_size)
        
        # Encode
        enc_out, (h_n, c_n) = self.encoder(input_src)
        
        # Decode
        dec_out, _ = self.decoder(input_trg, (h_n, c_n))
        
        # Generate outputs
        out = self.fc_out(dec_out)  # (seq_len, batch_size, output_size)
        out_prob = self.fc_prob(dec_out)  # (seq_len, batch_size, output_size)
        
        return out, out_prob

# Create model
model = SimpleSeq2SeqModel(input_size=4, hidden_size=5, output_size=3)

# Create loss function
loss_func = nn.MSELoss()

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

# Create small tensors matching the expected dimensions
batch_size = 2
seq_len = 3
input_size = 4
output_size = 3

xs_src = torch.randn(batch_size, seq_len, input_size)  # (batch, seq_len, input_size)
xs_trg = torch.randn(batch_size, seq_len, input_size)  # (batch, seq_len, input_size)
ys = torch.randn(batch_size, seq_len, output_size)     # (batch, seq_len, output_size)
yq = torch.ones(batch_size, seq_len, output_size)      # (batch, seq_len, output_size)
ya = torch.randn(batch_size, seq_len)                  # (batch, seq_len)
yp = torch.randn(batch_size, seq_len, output_size)     # (batch, seq_len, output_size)

ARGS = {
    "extend_forward": 0,
    "ks_loss": None,
}
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 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
get_option_fallback Ran
ran on a synthesized input
qqhann/KnowledgeTracing/src/config.py
code served (permissive licence) · get_code("27420e93f2ee17d6")
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
from typing import Dict

torch.manual_seed(0)

# Test case 1: Simple flat merge
options_1 = {'a': 2, 'd': 3}
fallback_1 = {'a': 1, 'b': 2}

# Test case 2: Nested dictionaries
options_2 = {'a': 2, 'd': 3, 'c': {'e': 6, 'g': 7}}
fallback_2 = {'a': 1, 'b': 2, 'c': {'e': 4, 'f': 5}}

# Test case 3: Empty options
options_3 = {}
fallback_3 = {'x': 10, 'y': 20}

# Test case 4: Empty fallback
options_4 = {'p': 100, 'q': 200}
fallback_4 = {}

ARGS = {
    'options': options_2,
    'fallback': fallback_2,
    'hard': False,
    'depth': 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 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
load_json Ran
ran on a synthesized input
qqhann/KnowledgeTracing/src/path.py
code served (permissive licence) · get_code("7186e392cdb3794b")
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
from pathlib import Path

# Create a temporary JSON file with sample configuration
with tempfile.NamedTemporaryFile(mode='w', suffix='.json', delete=False) as f:
    config_dict = {
        "model": "test_model",
        "batch_size": 32,
        "learning_rate": 0.001,
        "epochs": 10,
        "layers": [64, 128, 256]
    }
    json.dump(config_dict, f)
    temp_json_path = Path(f.name)

ARGS = {
    "cfgjsonpath": temp_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
load_rep_cfg Ran
ran on a synthesized input
qqhann/KnowledgeTracing/src/path.py
code served (permissive licence) · get_code("13710178c3163331")
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
from pathlib import Path

# Create a temporary directory and a JSON file with the expected structure
temp_dir = tempfile.TemporaryDirectory()
temp_path = Path(temp_dir.name)

# Create a sample report JSON file with a 'config' key
report_data = {
    "config": {
        "model": "test_model",
        "learning_rate": 0.001,
        "batch_size": 32,
        "epochs": 10
    },
    "results": {
        "accuracy": 0.95
    }
}

report_json_path = temp_path / "report.json"
with open(report_json_path, 'w') as f:
    json.dump(report_data, f)

ARGS = {
    "reportpath": report_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
varible Ran
ran on a synthesized input
qqhann/KnowledgeTracing/model/dkvmn.py
code served (permissive licence) · get_code("efc39cac785550fe")
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, 4) 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)

# Create a small tensor
tensor = torch.randn(2, 4)

# Set device to CPU
device = 'cpu'

# Define ARGS mapping all parameters
ARGS = {
    'tensor': tensor,
    'device': device
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) 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_source Not yet run
unverified (harvested, no recorded run)
qqhann/KnowledgeTracing/src/data.py
code served (permissive licence) · get_code("cf4867fc8dd16528")
How it ran

no invocation recorded.

re_numbering_knowledge_concepts Not yet run
unverified (harvested, no recorded run)
qqhann/KnowledgeTracing/src/data.py
code served (permissive licence) · get_code("123b744c11729533")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 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 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.

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

Knowledge tracing---where a machine models the knowledge of a student as they interact with coursework---is a well established problem in computer supported education. Though effectively modeling student knowledge would have high educational impact, the task has many inherent challenges. In this paper we explore the utility of using Recurrent Neural Networks (RNNs) to model student learning. The RNN family of models have important advantages over previous methods in that they do not require the explicit encoding of human domain knowledge, and can capture more complex representations of student knowledge. Using neural networks results in substantial improvements in prediction performance on a range of knowledge tracing datasets. Moreover the learned model can be used for intelligent curriculum design and allows straightforward interpretation and discovery of structure in student tasks. These results suggest a promising new line of research for knowledge tracing and an exemplary application task for RNNs.

For agents

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

get_harvested_code_for_paper("1506.05908")
get_code_for_paper("1506.05908")
have("1506.05908")

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

curl https://syntology.ai/api/ran/1506.05908.json

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

[![Syntology run record](https://syntology.ai/api/ran/1506.05908.svg)](https://syntology.ai/paper/1506.05908)

Connect an agent — have() is free.