SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2301.08249 · AAAI · 2023

Causal Conditional Hidden Markov Model for Multimodal Traffic Prediction

Yu Zhao, Pan Deng, Junting Liu, Xiaofeng Jia, Mulan Wang

arXiv · PDF · Open in the Atlas

Code that ran

11 of 13 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (8 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 9 have no contract verdict; 2 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/2301.08249.json · badge: /api/ran/2301.08249.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
EternityZY/CCHMM community
Papers with Code archive
11 ran · 2 no recorded run
FunctionStatusWhere it lives
Attention Ran
ran on a synthesized input · constructed an object; no result was computed
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("597fc4a17fd92e8b")
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, Attention: 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)

ARGS = {
    'hidden_dim': 4,
    'hidden_num': 2,
    'bias': 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 Attention — 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 Attention 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
GraphGRU Ran
ran on a synthesized input · constructed an object; no result was computed
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("d0a1ad0fefeef10e")
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, GraphGRU: 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 numpy as np

torch.manual_seed(0)

# Define small tensor dimensions
batch_size = 2
node_num = 3
in_dim = 4
hidden_dim = 5
gcn_depth = 2

# Create normalized adjacency matrix
# Shape: (node_num, node_num) for graph convolution
norm_adj = torch.eye(node_num) / node_num
norm_adj = norm_adj.unsqueeze(0).expand(batch_size, -1, -1)  # (batch, node, node)

# Create input tensor
# Shape: (batch_size, node_num, in_dim)
inputs = torch.randn(batch_size, node_num, in_dim)

# Construct ARGS dictionary with all GraphGRU __init__ parameters
ARGS = {
    'in_dim': in_dim,
    'hidden_dim': hidden_dim,
    'norm_adj': norm_adj,
    'gcn_depth': gcn_depth,
    'dropout_type': 'zoneout',
    'dropout_prob': 0.3,
    'alpha': 0.3,
}
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 GraphGRU — 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 GraphGRU 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
NonlinearTransforms Ran
ran on a synthesized input · constructed an object; no result was computed
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("c35c78412ee2bf5a")
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, NonlinearTransforms: 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)

# Define a simple activation function
activation = nn.ReLU()

# Parameters for NonlinearTransforms
in_dim = 4
out_dim = 8

ARGS = {
    "in_dim": in_dim,
    "out_dim": out_dim,
    "activation": activation
}
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 NonlinearTransforms — 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 NonlinearTransforms 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
PosteriorNet Ran
ran on a synthesized input · constructed an object; no result was computed
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("44ec0108f5988c87")
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, PosteriorNet: 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)

# Build a minimal normalized adjacency matrix for the graph
num_nodes = 4
norm_adj = torch.eye(num_nodes)
norm_adj = norm_adj / norm_adj.sum(dim=1, keepdim=True)

# Build a minimal SCM model (unused in the snippet but required parameter)
class MinimalSCM(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(10, 10)
    
    def forward(self, x):
        return self.linear(x)

scm_model = MinimalSCM()

# Define ARGS dictionary with all PosteriorNet parameters
ARGS = {
    'tpos_dim': 3,
    'POI_dim': 5,
    'weather_dim': 4,
    'road_dim': None,
    'input_channels': [2, 2, 2, 1],
    'output_channels': [2, 2, 2, 1],
    'hidden_num': 4,
    'hidden_dim': 64,
    'norm_adj': norm_adj,
    'gcn_depth': 2,
    'SCM_model': scm_model,
    'Posterior_type': 'GCN',
    'dropout_prob': 0.3,
    'var_type': False,
    'use_reparameterize': True,
    'mu_type': False,
    'activation': nn.ReLU(inplace=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 PosteriorNet — 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 PosteriorNet 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
PriorNet Ran
ran on a synthesized input · constructed an object; no result was computed
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("d80214b291d04a83")
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, PriorNet: 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
import numpy as np

torch.manual_seed(0)

# Build normalized adjacency matrix for graph operations
num_nodes = 5
norm_adj = torch.randn(num_nodes, num_nodes)
norm_adj = norm_adj / (norm_adj.sum(dim=1, keepdim=True) + 1e-8)

# Build a minimal SCM model (can be None or a simple module)
class DummySCM(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(10, 10)
    
    def forward(self, x):
        return self.linear(x)

scm_model = DummySCM()

ARGS = {
    'tpos_dim': 4,
    'POI_dim': 3,
    'weather_dim': 2,
    'hidden_num': 4,
    'hidden_dim': 8,
    'gcn_depth': 2,
    'norm_adj': norm_adj,
    'SCM_model': scm_model,
    'Prior_type': 'GraphGRU',
    'var_type': False,
    'dropout_prob': 0.3,
    'mu_type': False,
    'activation': nn.ReLU(inplace=True)
}
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 PriorNet — 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 PriorNet 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
SCM Ran
ran on a synthesized input · constructed an object; no result was computed
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("90f0422e78153086")
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, SCM: 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)

# Build a small nonlinear activation
activation = nn.ReLU(inplace=True)

# Parameters for SCM
in_dim = 4
hidden_dim = 3
hidden_num = 2
scm_type = 'nonlinear'
nonlinear_activation = activation

ARGS = {
    'in_dim': in_dim,
    'hidden_dim': hidden_dim,
    'hidden_num': hidden_num,
    'scm_type': scm_type,
    'nonlinear_activation': nonlinear_activation,
}
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 SCM — 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 SCM 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
_h_A Ran
ran; our contract draft was wrong, not the code
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("d2c010b419f09240")
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)

# Construct arguments for _h_A function
# _h_A takes parameters: A (a matrix/tensor) and m (an integer)

# A should be a square matrix that will be used in matrix operations
# The function computes A*A, then passes it to matrix_poly
# matrix_poly expects a square matrix of size (d, d) and an integer d
A = torch.randn(4, 4)

# m is an integer parameter used as the dimension argument to matrix_poly
m = 4

ARGS = {
    "A": A,
    "m": m
}
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
gconv Ran
ran on a synthesized input
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("7d7bf352b9643fa4")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from torch import nn

INIT_ARGS = {}

ARGS = {
    'A': torch.randn(3, 3),  # hw
    'x': torch.randn(2, 3, 4)  # bwc
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
linear Ran
ran on a synthesized input · constructed an object; no result was computed
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("8c01dd40f3a8b2b6")
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, linear: 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)

ARGS = {
    'c_in': 4,
    'c_out': 3,
    'bias': 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 linear — 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 linear 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
matrix_poly Ran
ran; our contract draft was wrong, not the code
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("b09b6a0ef2ebc9e0")
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 (4, 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
import torch.nn as nn

torch.manual_seed(0)

# The function expects:
# - matrix: a torch tensor (used with .device, torch.div, torch.matrix_power)
# - d: an integer (used in torch.eye(d), torch.div(matrix, d), torch.matrix_power(x, d))

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

# Create an integer dimension parameter
d = 4

ARGS = {
    "matrix": matrix,
    "d": d
}
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 (4, 4) — shape (4, 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
mixpropGCN Ran
ran on a synthesized input · constructed an object; no result was computed
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("bdde914c03aee222")
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, mixpropGCN: 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
import torch.nn.functional as F

torch.manual_seed(0)

# Create a small normalized adjacency matrix
# Shape: (width, width) = (4, 4) for graph convolution
norm_adj = torch.randn(4, 4)
norm_adj = norm_adj / (norm_adj.abs().sum(dim=1, keepdim=True) + 1e-8)

# Define parameters for mixpropGCN
in_dim = 3
out_dim = 5
gdep = 2
dropout_prob = 0.1
alpha = 0.3

ARGS = {
    'in_dim': in_dim,
    'out_dim': out_dim,
    'gdep': gdep,
    'dropout_prob': dropout_prob,
    'alpha': alpha,
    'norm_adj': norm_adj,
}
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 mixpropGCN — 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 mixpropGCN 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
CCausalHMM Not yet run
unverified (harvested, no recorded run)
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("57c9c2eab908962d")
How it ran

no invocation recorded.

kl_normal_log Not yet run
unverified (harvested, no recorded run)
EternityZY/CCHMM/models/CausalHMM.py
code served (permissive licence) · get_code("371e0870ae77ecbd")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 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.

FunctionAlso present inEarliest-datedOthers named
matrix_poly
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2106.07635 (2021-06)
Variational Causal Networks: Approximate Bayesian Inference over Causal Structures

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

Multimodal traffic flow can reflect the health of the transportation system, and its prediction is crucial to urban traffic management. Recent works overemphasize spatio-temporal correlations of traffic flow, ignoring the physical concepts that lead to the generation of observations and their causal relationship. Spatio-temporal correlations are considered unstable under the influence of different conditions, and spurious correlations may exist in observations. In this paper, we analyze the physical concepts affecting the generation of multimode traffic flow from the perspective of the observation generation principle and propose a Causal Conditional Hidden Markov Model (CCHMM) to predict multimodal traffic flow. In the latent variables inference stage, a posterior network disentangles the causal representations of the concepts of interest from conditional information and observations, and a causal propagation module mines their causal relationship. In the data generation stage, a prior network samples the causal latent variables from the prior distribution and feeds them into the generator to generate multimodal traffic flow. We use a mutually supervised training method for the prior and posterior to enhance the identifiability of the model. Experiments on real-world datasets show that CCHMM can effectively disentangle causal representations of concepts of interest and identify causality, and accurately predict multimodal traffic flow.

For agents

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

get_harvested_code_for_paper("2301.08249")
get_code_for_paper("2301.08249")
have("2301.08249")

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

curl https://syntology.ai/api/ran/2301.08249.json

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

[![Syntology run record](https://syntology.ai/api/ran/2301.08249.svg)](https://syntology.ai/paper/2301.08249)

Connect an agent — have() is free.