Estimate refutation throws error: PicklingError: logger cannot be pickled

[arainboldt]

Estimate refutation throws the following error:

PicklingError: logger cannot be pickled

from dowhy.causal_model import CausalModel
import networkx as nx
import pandas as pd
import numpy as np

edge_list = [
            
        ('a', 'b'),
        ('a', 'c'),
        ('c', 'd'),
        ('b', 'd'),
    ]

expert_G = nx.DiGraph()
for edge in edge_list:
    expert_G.add_edge(edge[0],edge[1])


dot_graph = to_pydot(expert_G)

data = pd.DataFrame(
    np.random.binomial(1, np.random.uniform(0,1,size=4), size=(100,4)),
    columns=list('abcd')
)

model = CausalModel(
    data=data,
    treatment='b',
    outcome='d',
    graph=str(dot_graph)
)

import statsmodels

identified_estimand = model.identify_effect(proceed_when_unidentifiable=True)
identified_estimand.identifier.logger = None
estimate = model.estimate_effect(identified_estimand,
                                              method_name="backdoor.generalized_linear_model",
                                              method_params={
                                                  'glm_family':statsmodels.api.families.Binomial()
                                              },
                                              target_units="ate")


print(estimate.interpret('textual_effect_interpreter'))

refutel = model.refute_estimate(identified_estimand, estimate, "random_common_cause")

This can also include a verbatim copy of outputs, or screenshots.

Expected behavior
A clear and concise description of what you expected to happen.

Version information:

• DoWhy version 0.9.1

Full Traceback

---------------------------------------------------------------------------
PicklingError                             Traceback (most recent call last)
Cell In[16], line 41
     31 estimate = model.estimate_effect(identified_estimand,
     32                                               method_name="backdoor.generalized_linear_model",
     33                                               method_params={
     34                                                   'glm_family':statsmodels.api.families.Binomial()
     35                                               },
     36                                               target_units="ate")
     39 print(estimate.interpret('textual_effect_interpreter'))
---> 41 refutel = model.refute_estimate(identified_estimand, estimate, "random_common_cause")

File [~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/dowhy/causal_model.py:417](https://file+.vscode-resource.vscode-cdn.net/home/***/intell/repos/dev/causal/ml/notebooks/~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/dowhy/causal_model.py:417), in CausalModel.refute_estimate(self, estimand, estimate, method_name, show_progress_bar, **kwargs)
    414     refuter_class = causal_refuters.get_class_object(method_name)
    416 refuter = refuter_class(self._data, identified_estimand=estimand, estimate=estimate, **kwargs)
--> 417 res = refuter.refute_estimate(show_progress_bar)
    418 return res

File [~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/dowhy/causal_refuters/random_common_cause.py:45](https://file+.vscode-resource.vscode-cdn.net/home/***/intell/repos/dev/causal/ml/notebooks/~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/dowhy/causal_refuters/random_common_cause.py:45), in RandomCommonCause.refute_estimate(self, show_progress_bar)
     44 def refute_estimate(self, show_progress_bar=False):
---> 45     refute = refute_random_common_cause(
     46         self._data,
     47         self._target_estimand,
     48         self._estimate,
     49         self._num_simulations,
     50         self._random_state,
     51         show_progress_bar,
     52         self._n_jobs,
     53         self._verbose,
     54     )
     55     refute.add_refuter(self)
     56     return refute

File [~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/dowhy/causal_refuters/random_common_cause.py:120](https://file+.vscode-resource.vscode-cdn.net/home/***/intell/repos/dev/causal/ml/notebooks/~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/dowhy/causal_refuters/random_common_cause.py:120), in refute_random_common_cause(data, target_estimand, estimate, num_simulations, random_state, show_progress_bar, n_jobs, verbose, **_)
    117     random_state = np.random.RandomState(seed=random_state)
    119 # Run refutation in parallel
--> 120 sample_estimates = Parallel(n_jobs=n_jobs, verbose=verbose)(
    121     delayed(_refute_once)(data, identified_estimand, estimate, random_state)
    122     for _ in tqdm(
    123         range(num_simulations),
    124         colour=CausalRefuter.PROGRESS_BAR_COLOR,
    125         disable=not show_progress_bar,
    126         desc="Refuting Estimates: ",
    127     )
    128 )
    129 sample_estimates = np.array(sample_estimates)
    131 refute = CausalRefutation(
    132     estimate.value, np.mean(sample_estimates), refutation_type="Refute: Add a random common cause"
    133 )

File [~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/parallel.py:1085](https://file+.vscode-resource.vscode-cdn.net/home/***/intell/repos/dev/causal/ml/notebooks/~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/parallel.py:1085), in Parallel.__call__(self, iterable)
   1076 try:
   1077     # Only set self._iterating to True if at least a batch
   1078     # was dispatched. In particular this covers the edge
   (...)
   1082     # was very quick and its callback already dispatched all the
   1083     # remaining jobs.
   1084     self._iterating = False
-> 1085     if self.dispatch_one_batch(iterator):
   1086         self._iterating = self._original_iterator is not None
   1088     while self.dispatch_one_batch(iterator):

File [~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/parallel.py:901](https://file+.vscode-resource.vscode-cdn.net/home/***/intell/repos/dev/causal/ml/notebooks/~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/parallel.py:901), in Parallel.dispatch_one_batch(self, iterator)
    899     return False
    900 else:
--> 901     self._dispatch(tasks)
    902     return True

File [~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/parallel.py:819](https://file+.vscode-resource.vscode-cdn.net/home/***/intell/repos/dev/causal/ml/notebooks/~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/parallel.py:819), in Parallel._dispatch(self, batch)
    817 with self._lock:
    818     job_idx = len(self._jobs)
--> 819     job = self._backend.apply_async(batch, callback=cb)
    820     # A job can complete so quickly than its callback is
    821     # called before we get here, causing self._jobs to
    822     # grow. To ensure correct results ordering, .insert is
    823     # used (rather than .append) in the following line
    824     self._jobs.insert(job_idx, job)

File [~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/_parallel_backends.py:208](https://file+.vscode-resource.vscode-cdn.net/home/***/intell/repos/dev/causal/ml/notebooks/~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/_parallel_backends.py:208), in SequentialBackend.apply_async(self, func, callback)
    206 def apply_async(self, func, callback=None):
    207     """Schedule a func to be run"""
--> 208     result = ImmediateResult(func)
    209     if callback:
    210         callback(result)

File [~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/_parallel_backends.py:597](https://file+.vscode-resource.vscode-cdn.net/home/***/intell/repos/dev/causal/ml/notebooks/~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/_parallel_backends.py:597), in ImmediateResult.__init__(self, batch)
    594 def __init__(self, batch):
    595     # Don't delay the application, to avoid keeping the input
    596     # arguments in memory
--> 597     self.results = batch()

File [~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/parallel.py:288](https://file+.vscode-resource.vscode-cdn.net/home/***/intell/repos/dev/causal/ml/notebooks/~/.cache/pypoetry/virtualenvs/causal-JHqzxB11-py3.8/lib/python3.8/site-packages/joblib/parallel.py:288), in BatchedCalls.__call__(self)
    284 def __call__(self):
    285     # Set the default nested backend to self._backend but do not set the
    286     # change the default number of processes to -1
...
   1739     import pickle
-> 1740     raise pickle.PicklingError('logger cannot be pickled')
   1741 return getLogger, (self.name,)

PicklingError: logger cannot be pickled

[amit-sharma]

When I run this code, I'm unable to reproduce the error. I had to create my own data since encode_data is not available and your graph does not include the approval node. Can you provide a self-contained example to debug this error?

[arainboldt]

My bad. Sorry for the incomplete example. I'll update this.

[arainboldt]

@amit-sharma thanks for the notification. I've updated the example, should work now.