SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.01753 · NeurIPS · 2022

HYPRO: A Hybridly Normalized Probabilistic Model for Long-Horizon Prediction of Event Sequences

Hongyuan Mei, Siqiao Xue, James Zhang, Xiaoming Shi

arXiv · PDF · Open in the Atlas

Code that ran

18 of 24 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (6 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 17 have no contract verdict; 1 failure of our instrument; 6 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/2210.01753.json · badge: /api/ran/2210.01753.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
alipay/hypro_tpp canonical (classifier label)
classifier basis: plan_row
7 ran · 0 no recorded run
hongyuanmei/neural-hawkes-particle-smoothing canonical (classifier label)
classifier basis not recorded on the link
3 ran · 2 no recorded run
ant-research/hypro_tpp community
Papers with Code archive
5 ran · 0 no recorded run
ivan-chai/hotpp-benchmark community (classifier label)
classifier basis not recorded on the link
2 ran · 4 no recorded run
yangalan123/anhp-andtt unrelated (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
BaseModel Ran
ran on a synthesized input · constructed an object; no result was computed
ant-research/hypro_tpp/hypro_tpp/models/xfmr_nhp_fast.py
code served (permissive licence) · get_code("bfffdf8e1209b24c")
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, BaseModel: 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
from torch import nn

torch.manual_seed(0)

# Create a minimal model_config dict with all required keys
model_config = {
    'num_event_types_no_pad': 10,
    'num_event_types_pad': 11,
    'event_pad_index': 10,
    'hidden_dim': 4,
    'num_steps_integral_loss': 20,
    'add_bos': False,
    'name': 'BaseModel'
}

ARGS = {
    'model_config': model_config
}
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 BaseModel — 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 BaseModel 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
EncoderLayer Ran
ran on a synthesized input · constructed an object; no result was computed
ant-research/hypro_tpp/hypro_tpp/models/xfmr_nhp_fast.py
code served (permissive licence) · get_code("ad8afbb3cfd7019b")
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, EncoderLayer: 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)

# Build a minimal self-attention module that matches how it's called
class MinimalSelfAttention(nn.Module):
    def __init__(self, d_model):
        super(MinimalSelfAttention, self).__init__()
        self.d_model = d_model
    
    def forward(self, query, key, value, mask):
        # Called as self.self_attn(x, x, x, mask)
        # Just return something of the right shape
        batch_size = query.shape[0]
        seq_len = query.shape[1]
        return torch.zeros(batch_size, seq_len, self.d_model)

# Build a minimal feed-forward module
class MinimalFeedForward(nn.Module):
    def __init__(self, d_model):
        super(MinimalFeedForward, self).__init__()
        self.linear = nn.Linear(d_model, d_model)
    
    def forward(self, x):
        return self.linear(x)

# Set up dimensions
d_model = 4
batch_size = 2
seq_len = 3

# Create instances
self_attn = MinimalSelfAttention(d_model)
feed_forward = MinimalFeedForward(d_model)
dropout = 0.1
use_residual = True

ARGS = {
    'd_model': d_model,
    'self_attn': self_attn,
    'feed_forward': feed_forward,
    'use_residual': use_residual,
    'dropout': dropout
}
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 EncoderLayer — 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 EncoderLayer 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
MultiHeadAttention Ran
ran on a synthesized input · constructed an object; no result was computed
ant-research/hypro_tpp/hypro_tpp/models/xfmr_nhp_fast.py
code served (permissive licence) · get_code("8f782f28d3d8e855")
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, MultiHeadAttention: 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 math

torch.manual_seed(0)

# Create small tensors for input
batch_size = 2
seq_len = 3
d_input = 4
n_head = 2
d_model = 4

# Create sample query, key, value tensors
query = torch.randn(batch_size, seq_len, d_input)
key = torch.randn(batch_size, seq_len, d_input)
value = torch.randn(batch_size, seq_len, d_input)

# Create optional mask tensor (shape: batch_size x seq_len x seq_len)
# mask > 0 means positions to mask out
mask = torch.zeros(batch_size, seq_len, seq_len)

# Create optional dropout layer
dropout_layer = nn.Dropout(p=0.1)

ARGS = {
    "n_head": n_head,
    "d_input": d_input,
    "d_model": d_model,
    "dropout": 0.1,
    "output_linear": False
}
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 MultiHeadAttention — 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 MultiHeadAttention 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
SublayerConnection Ran
ran on a synthesized input · constructed an object; no result was computed
ant-research/hypro_tpp/hypro_tpp/models/xfmr_nhp_fast.py
code served (permissive licence) · get_code("d68aa2c5bcfcfed4")
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, SublayerConnection: 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)

# SublayerConnection.__init__ takes: self, d_model, dropout
# SublayerConnection.forward takes: self, x, sublayer
# 
# The forward method:
# - calls self.norm(x) where self.norm is nn.LayerNorm(d_model)
#   so x must be a tensor with last dimension == d_model
# - calls sublayer(self.norm(x)) where sublayer is a callable
#   that takes a tensor and returns a tensor of same shape
# - adds x to the result, so shapes must match
# - applies dropout

# Create a minimal sublayer module that takes input and returns same shape
class DummySublayer(nn.Module):
    def forward(self, x):
        return x

# Set up parameters
d_model = 4
dropout = 0.1

# Create a batch of inputs: (batch_size=2, seq_len=3, d_model=4)
x = torch.randn(2, 3, 4)

# Create the sublayer
sublayer = DummySublayer()

ARGS = {
    "d_model": d_model,
    "dropout": dropout,
}
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 SublayerConnection — 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 SublayerConnection 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
XFMRNHPFast Ran
ran on a synthesized input · constructed an object; no result was computed
ant-research/hypro_tpp/hypro_tpp/models/xfmr_nhp_fast.py
code served (permissive licence) · get_code("4a7ec9134a46cc23")
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, XFMRNHPFast: 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 math

torch.manual_seed(0)

# Build the model_config dict with all required parameters for XFMRNHPFast
model_config = {
    'name': 'XFMRNHPFast',
    'num_event_types_no_pad': 10,
    'num_event_types_pad': 11,
    'event_pad_index': 10,
    'hidden_dim': 64,
    'time_emb_dim': 16,
    'use_ln': True,
    'num_layers': 2,
    'num_heads': 2,
    'sharing_param_layer': True,
    'dropout': 0.1,
    'mc_num_sample_per_step': 5,
    'num_steps_integral_loss': 20,
    'add_bos': False,
}

# Create a minimal batch for the forward pass
# Assuming batch contains event sequences and times
batch = {
    'event_seq': torch.randint(0, 10, (2, 8)),  # batch_size=2, seq_len=8
    'time_seq': torch.randn(2, 8),               # batch_size=2, seq_len=8
    'length': torch.tensor([8, 7]),              # actual lengths for each sequence
}

ARGS = {
    'model_config': model_config,
}
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 XFMRNHPFast — 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 XFMRNHPFast 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
XFMRNHPFast Ran
ran on a synthesized input · constructed an object; no result was computed
yangalan123/anhp-andtt/anhp/model/xfmr_nhp_fast.py
code served (permissive licence) · get_code("583f1f767c810249")
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, XFMRNHPFast: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import math

torch.manual_seed(0)

# Create a minimal dataset stand-in
class Dataset:
    def __init__(self):
        self.num_types = 5
        self.pad_index = 0
        self.event_num = 3

# Create ARGS dictionary with all required parameters for XFMRNHPFast
ARGS = {
    'dataset': Dataset(),
    'd_model': 4,
    'n_layers': 2,
    'n_head': 2,
    'dropout': 0.1,
    'd_time': 4,
    'd_inner': 8,
    'use_norm': False,
    'sharing_param_layer': False
}
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 XFMRNHPFast — 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 XFMRNHPFast 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
attention Ran
ran; our contract draft was wrong, not the code
alipay/hypro_tpp/hypro_tpp/models/layers.py
code served (permissive licence) · get_code("8abbfdabbdf5e708")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) 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 math

torch.manual_seed(0)

# Attention typically works with shape (batch, num_heads, seq_len, d_k)
# We'll use:
# - batch_size = 2
# - num_heads = 3
# - seq_len = 5
# - d_k = 4

batch_size = 2
num_heads = 3
seq_len = 5
d_k = 4

query = torch.randn(batch_size, num_heads, seq_len, d_k)
key = torch.randn(batch_size, num_heads, seq_len, d_k)
value = torch.randn(batch_size, num_heads, seq_len, d_k)

# Mask should be broadcastable to scores shape (batch, num_heads, seq_len, seq_len)
# Typically used for padding or causal masking
# mask > 0 will be filled with large negative value
mask = torch.zeros(batch_size, num_heads, seq_len, seq_len)
# Set some positions to 1 to test masking
mask[0, 0, 0, 3:] = 1
mask[1, 1, 2, 4] = 1

# Dropout is a nn.Module with forward method
dropout = torch.nn.Dropout(p=0.1)

ARGS = {
    'query': query,
    'key': key,
    'value': value,
    'mask': mask,
    'dropout': dropout
}
Setup block drafted by
a language model, Claude Sonnet 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 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
colored Ran
ran on a synthesized input
hongyuanmei/neural-hawkes-particle-smoothing/mp_manager/core/color_text.py
code served (permissive licence) · get_code("1a34b8d0a489e850")
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 str. 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 os

ARGS = {
    'text': 'Hello, World!',
    'color': 'red',
    'on_color': 'on_grey',
    'attrs': ['bold', 'underline']
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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_delta Ran
ran on a synthesized input
ivan-chai/hotpp-benchmark/hotpp/losses/common.py
code served (permissive licence) · get_code("d51e14e958836775")
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

torch.manual_seed(0)

# Create small tensor inputs: batch_size=2, sequence_length=5, feature_dim=4
inputs = torch.randn(2, 5, 4)

# Create a mask: batch_size=2, sequence_length=5
mask = torch.ones(2, 5, dtype=torch.bool)

# Test with delta='last' (default behavior)
delta = 'last'

# Optional smoothing parameter
smoothing = 0.1

# Max delta clipping value
max_delta = 2.0

# Whether to exclude out-of-horizon deltas from mask
exclude_out_of_horizon = False

ARGS = {
    'inputs': inputs,
    'mask': mask,
    'delta': delta,
    'smoothing': smoothing,
    'max_delta': max_delta,
    'exclude_out_of_horizon': exclude_out_of_horizon,
}
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
create_dataloader Ran
ran on a synthesized input
alipay/hypro_tpp/hypro_tpp/preprocess/data_loader.py
code served (permissive licence) · get_code("2148f6ec1007f87b")
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 DataLoader. 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
from torch.utils.data import Dataset, DataLoader

torch.manual_seed(0)

class SimpleDataset(Dataset):
    def __init__(self):
        self.data = torch.randn(10, 4)
        self.labels = torch.randint(0, 2, (10,))
    
    def __len__(self):
        return len(self.data)
    
    def __getitem__(self, idx):
        return self.data[idx], self.labels[idx]
    
    def collate_fn(self, batch):
        x = torch.stack([item[0] for item in batch])
        y = torch.stack([torch.tensor(item[1]) for item in batch])
        return x, y

dataset = SimpleDataset()

ARGS = {
    'dataset': dataset,
    'batch_size': 2,
    'shuffle': 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 DataLoader — a DataLoader output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataLoader 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
find_alignment_mc Ran
ran on a synthesized input
alipay/hypro_tpp/hypro_tpp/utils/metrics.py
code served (permissive licence) · get_code("58ff88767659c8e2")
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. 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 numpy as np

# Create small sequences representing timestamps
seq1 = np.array([0.0, 1.0, 2.0, 3.0], dtype=np.float32)
seq2 = np.array([0.5, 1.5, 2.5], dtype=np.float32)

# Create deletion costs array - multiple cost values to test
del_cost = np.array([0.1, 0.5, 1.0], dtype=np.float32)

# Transportation cost per unit length
trans_cost = 0.2

ARGS = {
    "seq1": seq1,
    "seq2": seq2,
    "del_cost": del_cost,
    "trans_cost": trans_cost,
}
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 list — a list 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
get_horizon_lengths Ran
ran on a synthesized input
ivan-chai/hotpp-benchmark/hotpp/info.py
code served (permissive licence) · get_code("0d758078e8864fdf")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 1, dtype family int, shape (8,) 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)

# Create a minimal dataset class that satisfies the function's requirements
class MockDataset:
    def __init__(self):
        self.timestamps_field = 'timestamps'
        self.data = [
            {'timestamps': torch.tensor([1.0, 2.0, 3.0, 4.0])},
            {'timestamps': torch.tensor([0.5, 1.5, 2.5, 3.5])}
        ]
    
    def __iter__(self):
        return iter(self.data)

# Construct the dataset
dataset = MockDataset()

# Define horizon as a small float value
horizon = 1.5

# Create ARGS dictionary with all required parameters
ARGS = {
    'dataset': dataset,
    'horizon': horizon
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family int, shape (8,) — shape (8,) 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_config Ran
ran on a synthesized input
alipay/hypro_tpp/hypro_tpp/utils/log.py
code served (permissive licence) · get_code("08a43a55e023ab89")
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 os
import tempfile
import yaml

# Create temporary directory with config files
temp_dir = tempfile.mkdtemp()

# Create the main config file
config_data = {
    'data': {
        'data_name': 'dataset_a'
    },
    'model': {
        'layers': 2
    }
}

config_file_path = os.path.join(temp_dir, 'config.yaml')
with open(config_file_path, 'w') as f:
    yaml.dump(config_data, f)

# Create the data_spec.yaml file in the same directory
data_spec_data = {
    'dataset_a': {
        'input_dim': 10,
        'output_dim': 5
    }
}

data_spec_file_path = os.path.join(temp_dir, 'data_spec.yaml')
with open(data_spec_file_path, 'w') as f:
    yaml.dump(data_spec_data, f)

ARGS = {
    'config_filename': config_file_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — 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
makeHiddenForLambda Ran
ran on a synthesized input
hongyuanmei/neural-hawkes-particle-smoothing/nhps/models/utils.py
code served (permissive licence) · get_code("0de796a888ce97a7")
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 3, dtype family float, shape (2, 3, 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)

# The function expects:
# - states: a tuple of 4 tensors (c, cb, d, o)
# - dtime: a tensor
# All tensors need to be compatible for broadcasting after unsqueezing

# Create small tensors with distinct dimensions
# batch_size=2, feature_dim=4
batch_size = 2
feature_dim = 4
time_steps = 3

c = torch.randn(batch_size, feature_dim)
cb = torch.randn(batch_size, feature_dim)
d = torch.randn(batch_size, feature_dim)
o = torch.randn(batch_size, feature_dim)

states = (c, cb, d, o)

# dtime should be broadcastable with the other tensors after unsqueezing
# After unsqueeze(dim=2), it will have shape (batch_size, 1, 2)
# so we need it to have shape (batch_size, time_steps)
dtime = torch.randn(batch_size, time_steps)

ARGS = {
    "states": states,
    "dtime": dtime
}
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 3, dtype family float, shape (2, 3, 4) — shape (2, 3, 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
make_config_string Ran
ran on a synthesized input
alipay/hypro_tpp/hypro_tpp/utils/misc.py
code served (permissive licence) · get_code("3a664d327d5f2e80")
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 str. 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)

# Analyzing the function:
# - It takes a `config` dict and `max_num_key` int
# - It iterates through config.items()
# - It checks if k == 'name'
# - It accesses v and converts it to string
# - Returns a string with the last underscore removed

ARGS = {
    'config': {
        'name': 'model_v1',
        'learning_rate': 0.001,
        'batch_size': 32,
        'epochs': 10,
        'optimizer': 'adam'
    },
    'max_num_key': 4
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
run Ran
ran on a synthesized input
hongyuanmei/neural-hawkes-particle-smoothing/mp_manager/core/run_func.py
code served (permissive licence) · get_code("56bd9a8fb615a7cc")
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 torch.nn as nn

torch.manual_seed(0)

def sample_task():
    """A simple callable task that returns a result."""
    return {"status": "completed", "value": 42}

ARGS = {
    "task": sample_task
}
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
time_rmse_np Ran
ran on a synthesized input
alipay/hypro_tpp/hypro_tpp/utils/metrics.py
code served (permissive licence) · get_code("371695e02ece447c")
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

# Construct valid arguments for time_rmse_np
preds = {
    'dtimes': np.array([1.0, 2.0, 3.0, 4.0])
}

labels = {
    'dtimes': np.array([1.1, 2.1, 2.9, 3.95])
}

ARGS = {
    'preds': preds,
    'labels': labels
}
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
type_acc_np Ran
ran on a synthesized input
alipay/hypro_tpp/hypro_tpp/utils/metrics.py
code served (permissive licence) · get_code("d72eb69e38a09eba")
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

np.random.seed(0)

# Create predictions and labels as numpy arrays
# The function compares them element-wise and computes mean accuracy
preds = np.array([0, 1, 2, 1, 0, 3])
labels = np.array([0, 1, 1, 1, 0, 3])

ARGS = {
    "preds": preds,
    "labels": labels
}
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
device_run Not yet run
unverified (harvested, no recorded run)
hongyuanmei/neural-hawkes-particle-smoothing/mp_manager/core/run_func.py
code served (permissive licence) · get_code("a84d93385888f3b2")
How it ran

no invocation recorded.

eval_embeddings Not yet run
unverified (harvested, no recorded run)
ivan-chai/hotpp-benchmark/hotpp/eval_downstream.py
code served (permissive licence) · get_code("aeccc4362da3db37")
How it ran

no invocation recorded.

evaluate Not yet run
unverified (harvested, no recorded run)
ivan-chai/hotpp-benchmark/hotpp/evaluate.py
code served (permissive licence) · get_code("183f293f3497c300")
How it ran

no invocation recorded.

makeHiddenForBound Not yet run
unverified (harvested, no recorded run)
hongyuanmei/neural-hawkes-particle-smoothing/nhps/models/utils.py
code served (permissive licence) · get_code("502423075a8fcbb0")
How it ran

no invocation recorded.

parse_result Not yet run
unverified (harvested, no recorded run)
ivan-chai/hotpp-benchmark/hotpp/eval_downstream.py
code served (permissive licence) · get_code("7893975b324c3d57")
How it ran

no invocation recorded.

test Not yet run
unverified (harvested, no recorded run)
ivan-chai/hotpp-benchmark/hotpp/evaluate.py
code served (permissive licence) · get_code("d35314d71442b416")
How it ran

no invocation recorded.

Also present in other papers' code

Of 24 samples lifted from this paper's linked repositories, 24 carry a function identity and 0 do not. Of the 24 distinct functions with an identity, 7 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 0 the same month, 2 later) and 17 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
attention
brick of this paper's assembly: helper
4 other papers
1 dated earlier, 3 later
2002.09291 (2020-02)
Transformer Hawkes Process
2305.16646 (2023-05)
Language Models Can Improve Event Prediction by Few-Shot Abductive Reasoning
2307.08097 (2023-07)
EasyTPP: Towards Open Benchmarking Temporal Point Processes
and 1 more not named here
SublayerConnection
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2310.04993 (2023-10)
Prompt-augmented Temporal Point Process for Streaming Event Sequence
colored 1 other paper
1 dated earlier
1905.05570 (2019-05)
Imputing Missing Events in Continuous-Time Event Streams
device_run 1 other paper
1 dated earlier
1905.05570 (2019-05)
Imputing Missing Events in Continuous-Time Event Streams
makeHiddenForBound 1 other paper
1 dated earlier
1905.05570 (2019-05)
Imputing Missing Events in Continuous-Time Event Streams
makeHiddenForLambda 1 other paper
1 dated earlier
1905.05570 (2019-05)
Imputing Missing Events in Continuous-Time Event Streams
run 1 other paper
1 dated earlier
1905.05570 (2019-05)
Imputing Missing Events in Continuous-Time Event Streams

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 this paper, we tackle the important yet under-investigated problem of making long-horizon prediction of event sequences. Existing state-of-the-art models do not perform well at this task due to their autoregressive structure. We propose HYPRO, a hybridly normalized probabilistic model that naturally fits this task: its first part is an autoregressive base model that learns to propose predictions; its second part is an energy function that learns to reweight the proposals such that more realistic predictions end up with higher probabilities. We also propose efficient training and inference algorithms for this model. Experiments on multiple real-world datasets demonstrate that our proposed HYPRO model can significantly outperform previous models at making long-horizon predictions of future events. We also conduct a range of ablation studies to investigate the effectiveness of each component of our proposed methods.

For agents

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

get_harvested_code_for_paper("2210.01753")
get_code_for_paper("2210.01753")
have("2210.01753")

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

curl https://syntology.ai/api/ran/2210.01753.json

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

[![Syntology run record](https://syntology.ai/api/ran/2210.01753.svg)](https://syntology.ai/paper/2210.01753)

Connect an agent — have() is free.