repo_name
stringlengths
8
38
pr_number
int64
3
47.1k
pr_title
stringlengths
8
175
pr_description
stringlengths
2
19.8k
author
null
date_created
stringlengths
25
25
date_merged
stringlengths
25
25
filepath
stringlengths
6
136
before_content
stringlengths
54
884k
after_content
stringlengths
56
884k
pr_author
stringlengths
3
21
previous_commit
stringlengths
40
40
pr_commit
stringlengths
40
40
comment
stringlengths
2
25.4k
comment_author
stringlengths
3
29
__index_level_0__
int64
0
5.1k
py-why/dowhy
882
Revise user guide entry for intrinsic causal influence
null
null
2023-02-24 21:36:44+00:00
2023-03-07 16:13:45+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/quantify_intrinsic_causal_influence.rst
Quantifying Intrinsic Causal Influence ====================================== By quantifying intrinsic causal influence, we answer the question: How strong is the causal influence of a source node to a target node that is not inherited from the parents of the source node? Naturally, descendants will have a z...
Quantifying Intrinsic Causal Influence ====================================== By quantifying intrinsic causal influence, we answer the question: How strong is the causal influence of a source node to a target node that is not inherited from the parents of the source node? Naturally, descendants will have a z...
bloebp
a1dcccbc805cb28aa4840fe9bce8338278632a50
12168ea7bd7a30d4c0d6501f69c7161ddb073845
We did not explain "simple linear relationship" before this text.
kailashbuki
84
py-why/dowhy
882
Revise user guide entry for intrinsic causal influence
null
null
2023-02-24 21:36:44+00:00
2023-03-07 16:13:45+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/quantify_intrinsic_causal_influence.rst
Quantifying Intrinsic Causal Influence ====================================== By quantifying intrinsic causal influence, we answer the question: How strong is the causal influence of a source node to a target node that is not inherited from the parents of the source node? Naturally, descendants will have a z...
Quantifying Intrinsic Causal Influence ====================================== By quantifying intrinsic causal influence, we answer the question: How strong is the causal influence of a source node to a target node that is not inherited from the parents of the source node? Naturally, descendants will have a z...
bloebp
a1dcccbc805cb28aa4840fe9bce8338278632a50
12168ea7bd7a30d4c0d6501f69c7161ddb073845
Nit [for consistency reason]: intrinsic causal influence
kailashbuki
85
py-why/dowhy
882
Revise user guide entry for intrinsic causal influence
null
null
2023-02-24 21:36:44+00:00
2023-03-07 16:13:45+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/quantify_intrinsic_causal_influence.rst
Quantifying Intrinsic Causal Influence ====================================== By quantifying intrinsic causal influence, we answer the question: How strong is the causal influence of a source node to a target node that is not inherited from the parents of the source node? Naturally, descendants will have a z...
Quantifying Intrinsic Causal Influence ====================================== By quantifying intrinsic causal influence, we answer the question: How strong is the causal influence of a source node to a target node that is not inherited from the parents of the source node? Naturally, descendants will have a z...
bloebp
a1dcccbc805cb28aa4840fe9bce8338278632a50
12168ea7bd7a30d4c0d6501f69c7161ddb073845
In that case, we would also need to introduce something like ``Y = max(0, X) + N``, since the following train will not leave earlier if the train is faster. I am not sure if we really need to make it that complicated here in the "how to use the method" section, since this is mostly to show how the API works? I only ...
bloebp
86
py-why/dowhy
882
Revise user guide entry for intrinsic causal influence
null
null
2023-02-24 21:36:44+00:00
2023-03-07 16:13:45+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/quantify_intrinsic_causal_influence.rst
Quantifying Intrinsic Causal Influence ====================================== By quantifying intrinsic causal influence, we answer the question: How strong is the causal influence of a source node to a target node that is not inherited from the parents of the source node? Naturally, descendants will have a z...
Quantifying Intrinsic Causal Influence ====================================== By quantifying intrinsic causal influence, we answer the question: How strong is the causal influence of a source node to a target node that is not inherited from the parents of the source node? Naturally, descendants will have a z...
bloebp
a1dcccbc805cb28aa4840fe9bce8338278632a50
12168ea7bd7a30d4c0d6501f69c7161ddb073845
Removed "simple" here. This is more referring to the ``Y = X + N`` setup, since users might think it is restricted to linear relationships.
bloebp
87
py-why/dowhy
882
Revise user guide entry for intrinsic causal influence
null
null
2023-02-24 21:36:44+00:00
2023-03-07 16:13:45+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/quantify_intrinsic_causal_influence.rst
Quantifying Intrinsic Causal Influence ====================================== By quantifying intrinsic causal influence, we answer the question: How strong is the causal influence of a source node to a target node that is not inherited from the parents of the source node? Naturally, descendants will have a z...
Quantifying Intrinsic Causal Influence ====================================== By quantifying intrinsic causal influence, we answer the question: How strong is the causal influence of a source node to a target node that is not inherited from the parents of the source node? Naturally, descendants will have a z...
bloebp
a1dcccbc805cb28aa4840fe9bce8338278632a50
12168ea7bd7a30d4c0d6501f69c7161ddb073845
Fixed
bloebp
88
py-why/dowhy
875
Revise attributing distributional changes user guide entry
null
null
2023-02-17 22:20:40+00:00
2023-03-07 16:14:00+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/attribute_distributional_changes.rst
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: What mechanism in my system changed between two sets of data? For example, in a distributed computing system, we want to know why an important system metric changed in a negative w...
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: **What mechanism in my system changed between two sets of data? Or in other words, which node in my data behaves differently?** Here we want to identify the node or nodes in the gr...
bloebp
12168ea7bd7a30d4c0d6501f69c7161ddb073845
5d449be765fbb04d77a35d2ace7978bfc6d90309
Is this really better? Note that we already use the word "mechanism" in our methods and language. So I believe it does make sense to refer to it here, no?
petergtz
89
py-why/dowhy
875
Revise attributing distributional changes user guide entry
null
null
2023-02-17 22:20:40+00:00
2023-03-07 16:14:00+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/attribute_distributional_changes.rst
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: What mechanism in my system changed between two sets of data? For example, in a distributed computing system, we want to know why an important system metric changed in a negative w...
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: **What mechanism in my system changed between two sets of data? Or in other words, which node in my data behaves differently?** Here we want to identify the node or nodes in the gr...
bloebp
12168ea7bd7a30d4c0d6501f69c7161ddb073845
5d449be765fbb04d77a35d2ace7978bfc6d90309
It would definitely make sense, but seeing it from a perspective of someone who just wants to "quickly" look through the functionalities, I think this formulation helps more in understanding what the purpose of the method is. Otherwise, one first needs to understand what we mean by a 'mechanism' etc.
bloebp
90
py-why/dowhy
875
Revise attributing distributional changes user guide entry
null
null
2023-02-17 22:20:40+00:00
2023-03-07 16:14:00+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/attribute_distributional_changes.rst
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: What mechanism in my system changed between two sets of data? For example, in a distributed computing system, we want to know why an important system metric changed in a negative w...
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: **What mechanism in my system changed between two sets of data? Or in other words, which node in my data behaves differently?** Here we want to identify the node or nodes in the gr...
bloebp
12168ea7bd7a30d4c0d6501f69c7161ddb073845
5d449be765fbb04d77a35d2ace7978bfc6d90309
Can we have both? One for continuity across the documentation and one for the quick reader? "What mechanism ....? Or in other words, which node in my data behaves differently?"
emrekiciman
91
py-why/dowhy
875
Revise attributing distributional changes user guide entry
null
null
2023-02-17 22:20:40+00:00
2023-03-07 16:14:00+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/attribute_distributional_changes.rst
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: What mechanism in my system changed between two sets of data? For example, in a distributed computing system, we want to know why an important system metric changed in a negative w...
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: **What mechanism in my system changed between two sets of data? Or in other words, which node in my data behaves differently?** Here we want to identify the node or nodes in the gr...
bloebp
12168ea7bd7a30d4c0d6501f69c7161ddb073845
5d449be765fbb04d77a35d2ace7978bfc6d90309
That's a good compromise, will change it.
bloebp
92
py-why/dowhy
875
Revise attributing distributional changes user guide entry
null
null
2023-02-17 22:20:40+00:00
2023-03-07 16:14:00+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/attribute_distributional_changes.rst
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: What mechanism in my system changed between two sets of data? For example, in a distributed computing system, we want to know why an important system metric changed in a negative w...
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: **What mechanism in my system changed between two sets of data? Or in other words, which node in my data behaves differently?** Here we want to identify the node or nodes in the gr...
bloebp
12168ea7bd7a30d4c0d6501f69c7161ddb073845
5d449be765fbb04d77a35d2ace7978bfc6d90309
uptick in latency of ___? application?
kailashbuki
93
py-why/dowhy
875
Revise attributing distributional changes user guide entry
null
null
2023-02-17 22:20:40+00:00
2023-03-07 16:14:00+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/attribute_distributional_changes.rst
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: What mechanism in my system changed between two sets of data? For example, in a distributed computing system, we want to know why an important system metric changed in a negative w...
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: **What mechanism in my system changed between two sets of data? Or in other words, which node in my data behaves differently?** Here we want to identify the node or nodes in the gr...
bloebp
12168ea7bd7a30d4c0d6501f69c7161ddb073845
5d449be765fbb04d77a35d2ace7978bfc6d90309
It would be more accurate to say "Estimate the conditional distributions from `old` data...".
kailashbuki
94
py-why/dowhy
875
Revise attributing distributional changes user guide entry
null
null
2023-02-17 22:20:40+00:00
2023-03-07 16:14:00+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/attribute_distributional_changes.rst
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: What mechanism in my system changed between two sets of data? For example, in a distributed computing system, we want to know why an important system metric changed in a negative w...
Attributing Distributional Changes ================================== When attributing distribution changes, we answer the question: **What mechanism in my system changed between two sets of data? Or in other words, which node in my data behaves differently?** Here we want to identify the node or nodes in the gr...
bloebp
12168ea7bd7a30d4c0d6501f69c7161ddb073845
5d449be765fbb04d77a35d2ace7978bfc6d90309
Replace old mechanisms with new one by one.
kailashbuki
95
py-why/dowhy
870
Revise gcm user guide entry for counterfactuals
null
null
2023-02-14 23:09:28+00:00
2023-03-07 16:14:22+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/computing_counterfactuals.rst
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome z for a variable Z where variable X was set to a value x. What would have happened to the value of Z, had I intervened on X to assign it a different value x'? As a concrete ...
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome :math:`z` for a variable :math:`Z` where variable :math:`X` was set to a value :math:`x`. What would have happened to the value of :math:`Z`, had I intervened on :math:`X` to...
bloebp
5d449be765fbb04d77a35d2ace7978bfc6d90309
d7b7cc65c5b6780fbc0a32ec4cd9f94e17878353
1. The data generation process can be made more realistic. Currently, there can be negative dosages too (absurd!). 2. Should we also explain what Y is?
kailashbuki
96
py-why/dowhy
870
Revise gcm user guide entry for counterfactuals
null
null
2023-02-14 23:09:28+00:00
2023-03-07 16:14:22+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/computing_counterfactuals.rst
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome z for a variable Z where variable X was set to a value x. What would have happened to the value of Z, had I intervened on X to assign it a different value x'? As a concrete ...
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome :math:`z` for a variable :math:`Z` where variable :math:`X` was set to a value :math:`x`. What would have happened to the value of :math:`Z`, had I intervened on :math:`X` to...
bloebp
5d449be765fbb04d77a35d2ace7978bfc6d90309
d7b7cc65c5b6780fbc0a32ec4cd9f94e17878353
I would replace "when intervening on ..." with "had we intervened on ...". Reads better. Clearly states we are contemplating about the past.
kailashbuki
97
py-why/dowhy
870
Revise gcm user guide entry for counterfactuals
null
null
2023-02-14 23:09:28+00:00
2023-03-07 16:14:22+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/computing_counterfactuals.rst
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome z for a variable Z where variable X was set to a value x. What would have happened to the value of Z, had I intervened on X to assign it a different value x'? As a concrete ...
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome :math:`z` for a variable :math:`Z` where variable :math:`X` was set to a value :math:`x`. What would have happened to the value of :math:`Z`, had I intervened on :math:`X` to...
bloebp
5d449be765fbb04d77a35d2ace7978bfc6d90309
d7b7cc65c5b6780fbc0a32ec4cd9f94e17878353
Invertibility leads to point counterfactuals. But we may also have distributional counterfactuals, where we estimate the posterior distribution of noise terms given the observational evidence. We should make the latter explicit here as most readers of Pearl are familiar with the notion of distributional counterfactuals...
kailashbuki
98
py-why/dowhy
870
Revise gcm user guide entry for counterfactuals
null
null
2023-02-14 23:09:28+00:00
2023-03-07 16:14:22+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/computing_counterfactuals.rst
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome z for a variable Z where variable X was set to a value x. What would have happened to the value of Z, had I intervened on X to assign it a different value x'? As a concrete ...
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome :math:`z` for a variable :math:`Z` where variable :math:`X` was set to a value :math:`x`. What would have happened to the value of :math:`Z`, had I intervened on :math:`X` to...
bloebp
5d449be765fbb04d77a35d2ace7978bfc6d90309
d7b7cc65c5b6780fbc0a32ec4cd9f94e17878353
I completely rewrote the example to keep it connected to the cholesterol example while being somewhat 'realistic'.
bloebp
99
py-why/dowhy
870
Revise gcm user guide entry for counterfactuals
null
null
2023-02-14 23:09:28+00:00
2023-03-07 16:14:22+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/computing_counterfactuals.rst
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome z for a variable Z where variable X was set to a value x. What would have happened to the value of Z, had I intervened on X to assign it a different value x'? As a concrete ...
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome :math:`z` for a variable :math:`Z` where variable :math:`X` was set to a value :math:`x`. What would have happened to the value of :math:`Z`, had I intervened on :math:`X` to...
bloebp
5d449be765fbb04d77a35d2ace7978bfc6d90309
d7b7cc65c5b6780fbc0a32ec4cd9f94e17878353
This sentence changed in the new version.
bloebp
100
py-why/dowhy
870
Revise gcm user guide entry for counterfactuals
null
null
2023-02-14 23:09:28+00:00
2023-03-07 16:14:22+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/computing_counterfactuals.rst
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome z for a variable Z where variable X was set to a value x. What would have happened to the value of Z, had I intervened on X to assign it a different value x'? As a concrete ...
Computing Counterfactuals ========================== By computing counterfactuals, we answer the question: I observed a certain outcome :math:`z` for a variable :math:`Z` where variable :math:`X` was set to a value :math:`x`. What would have happened to the value of :math:`Z`, had I intervened on :math:`X` to...
bloebp
5d449be765fbb04d77a35d2ace7978bfc6d90309
d7b7cc65c5b6780fbc0a32ec4cd9f94e17878353
Good point, I added a note at the bottom. Allowing counterfactual distributions (beyond point-wise) is anyway something high on the todo list.
bloebp
101
py-why/dowhy
867
Revise arrow strength user guide
Signed-off-by: Patrick Bloebaum <bloebp@amazon.com>
null
2023-02-13 21:29:38+00:00
2023-03-07 16:14:39+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/quantify_arrow_strength.rst
Quantifying Arrow Strength ================================= By quantifying the strength of an arrow, we answer the question: How strong is the causal influence from a cause to its direct effect? How to use it ^^^^^^^^^^^^^^ To see how the method works, let us generate some data. >>> import numpy as np, pandas...
Quantifying Arrow Strength ================================= By quantifying the strength of an arrow, we answer the question: **How strong is the causal influence from a cause to its direct effect?** While there are different definitions for measuring causal influences in the literature, DoWhy offers an implemen...
bloebp
d7b7cc65c5b6780fbc0a32ec4cd9f94e17878353
7ca6de528ac70ec8cb2f29d2e0389149cec0090a
`the library` -> `DoWhy`
petergtz
102
py-why/dowhy
867
Revise arrow strength user guide
Signed-off-by: Patrick Bloebaum <bloebp@amazon.com>
null
2023-02-13 21:29:38+00:00
2023-03-07 16:14:39+00:00
docs/source/user_guide/gcm_based_inference/answering_causal_questions/quantify_arrow_strength.rst
Quantifying Arrow Strength ================================= By quantifying the strength of an arrow, we answer the question: How strong is the causal influence from a cause to its direct effect? How to use it ^^^^^^^^^^^^^^ To see how the method works, let us generate some data. >>> import numpy as np, pandas...
Quantifying Arrow Strength ================================= By quantifying the strength of an arrow, we answer the question: **How strong is the causal influence from a cause to its direct effect?** While there are different definitions for measuring causal influences in the literature, DoWhy offers an implemen...
bloebp
d7b7cc65c5b6780fbc0a32ec4cd9f94e17878353
7ca6de528ac70ec8cb2f29d2e0389149cec0090a
Nit: I would leave out "as mentioned before".
kailashbuki
103
py-why/dowhy
846
Enhancement: warn about unobserved graph variables in `causal_model.identify_effect`.
Closes #810. Introduces a `UserWarning` that is emitted if there are any graph variables that are not contained in the observed data (`self._data`). Adds a unit test for this implementation. Currently, the return values of `self.get_common_causes()`, `self.get_instruments()` and `self.get_effect_modifiers()` are...
null
2023-02-04 14:44:46+00:00
2023-02-14 20:21:29+00:00
dowhy/causal_model.py
""" Module containing the main model class for the dowhy package. """ import logging from itertools import combinations from sympy import init_printing import dowhy.causal_estimators as causal_estimators import dowhy.causal_refuters as causal_refuters import dowhy.graph_learners as graph_learners import dowhy.utils....
""" Module containing the main model class for the dowhy package. """ import logging import typing import warnings from itertools import combinations from sympy import init_printing import dowhy.causal_estimators as causal_estimators import dowhy.causal_refuters as causal_refuters import dowhy.graph_learners as grap...
MFreidank
2a2b3f4f7d02f8157d4b2ea39588b305e048eca8
23214a9850544780e21f0afecb390446ceca48a2
From the wording, it's not clear whether this is an ok or not ok thing. To resolve this, I'd suggest giving a little more guidance. For larger graphs, we might also want to add counts too. Maybe something like: "The graph defines N variables. K were found in the dataset and will be analyzed as observed variables...
emrekiciman
104
py-why/dowhy
846
Enhancement: warn about unobserved graph variables in `causal_model.identify_effect`.
Closes #810. Introduces a `UserWarning` that is emitted if there are any graph variables that are not contained in the observed data (`self._data`). Adds a unit test for this implementation. Currently, the return values of `self.get_common_causes()`, `self.get_instruments()` and `self.get_effect_modifiers()` are...
null
2023-02-04 14:44:46+00:00
2023-02-14 20:21:29+00:00
dowhy/causal_model.py
""" Module containing the main model class for the dowhy package. """ import logging from itertools import combinations from sympy import init_printing import dowhy.causal_estimators as causal_estimators import dowhy.causal_refuters as causal_refuters import dowhy.graph_learners as graph_learners import dowhy.utils....
""" Module containing the main model class for the dowhy package. """ import logging import typing import warnings from itertools import combinations from sympy import init_printing import dowhy.causal_estimators as causal_estimators import dowhy.causal_refuters as causal_refuters import dowhy.graph_learners as grap...
MFreidank
2a2b3f4f7d02f8157d4b2ea39588b305e048eca8
23214a9850544780e21f0afecb390446ceca48a2
I think you could check all variables by calling causalgraph.py CausalGraph::get_all_nodes(...)
emrekiciman
105
py-why/dowhy
846
Enhancement: warn about unobserved graph variables in `causal_model.identify_effect`.
Closes #810. Introduces a `UserWarning` that is emitted if there are any graph variables that are not contained in the observed data (`self._data`). Adds a unit test for this implementation. Currently, the return values of `self.get_common_causes()`, `self.get_instruments()` and `self.get_effect_modifiers()` are...
null
2023-02-04 14:44:46+00:00
2023-02-14 20:21:29+00:00
dowhy/causal_model.py
""" Module containing the main model class for the dowhy package. """ import logging from itertools import combinations from sympy import init_printing import dowhy.causal_estimators as causal_estimators import dowhy.causal_refuters as causal_refuters import dowhy.graph_learners as graph_learners import dowhy.utils....
""" Module containing the main model class for the dowhy package. """ import logging import typing import warnings from itertools import combinations from sympy import init_printing import dowhy.causal_estimators as causal_estimators import dowhy.causal_refuters as causal_refuters import dowhy.graph_learners as grap...
MFreidank
2a2b3f4f7d02f8157d4b2ea39588b305e048eca8
23214a9850544780e21f0afecb390446ceca48a2
Regarding your comment about whether the warnings should be visible or not by default ... how about we have a short message that says "N variables are assumed unobserved because they are not in the dataset. Pass/set the ____ flag for additional details".
emrekiciman
106
py-why/dowhy
846
Enhancement: warn about unobserved graph variables in `causal_model.identify_effect`.
Closes #810. Introduces a `UserWarning` that is emitted if there are any graph variables that are not contained in the observed data (`self._data`). Adds a unit test for this implementation. Currently, the return values of `self.get_common_causes()`, `self.get_instruments()` and `self.get_effect_modifiers()` are...
null
2023-02-04 14:44:46+00:00
2023-02-14 20:21:29+00:00
dowhy/causal_model.py
""" Module containing the main model class for the dowhy package. """ import logging from itertools import combinations from sympy import init_printing import dowhy.causal_estimators as causal_estimators import dowhy.causal_refuters as causal_refuters import dowhy.graph_learners as graph_learners import dowhy.utils....
""" Module containing the main model class for the dowhy package. """ import logging import typing import warnings from itertools import combinations from sympy import init_printing import dowhy.causal_estimators as causal_estimators import dowhy.causal_refuters as causal_refuters import dowhy.graph_learners as grap...
MFreidank
2a2b3f4f7d02f8157d4b2ea39588b305e048eca8
23214a9850544780e21f0afecb390446ceca48a2
I've tried to adapt the logging to resolve your comments. There are now two levels to the output: * A high level warning, emitted using `warnings.warn` that points out the issue and suggests additional details are available when configuring `logging` levels. * A lower level, more detailed warning, emitted using `self...
MFreidank
107
py-why/dowhy
846
Enhancement: warn about unobserved graph variables in `causal_model.identify_effect`.
Closes #810. Introduces a `UserWarning` that is emitted if there are any graph variables that are not contained in the observed data (`self._data`). Adds a unit test for this implementation. Currently, the return values of `self.get_common_causes()`, `self.get_instruments()` and `self.get_effect_modifiers()` are...
null
2023-02-04 14:44:46+00:00
2023-02-14 20:21:29+00:00
dowhy/causal_model.py
""" Module containing the main model class for the dowhy package. """ import logging from itertools import combinations from sympy import init_printing import dowhy.causal_estimators as causal_estimators import dowhy.causal_refuters as causal_refuters import dowhy.graph_learners as graph_learners import dowhy.utils....
""" Module containing the main model class for the dowhy package. """ import logging import typing import warnings from itertools import combinations from sympy import init_printing import dowhy.causal_estimators as causal_estimators import dowhy.causal_refuters as causal_refuters import dowhy.graph_learners as grap...
MFreidank
2a2b3f4f7d02f8157d4b2ea39588b305e048eca8
23214a9850544780e21f0afecb390446ceca48a2
That's easier indeed, thanks for pointing this out. Resolved now.
MFreidank
108
py-why/dowhy
846
Enhancement: warn about unobserved graph variables in `causal_model.identify_effect`.
Closes #810. Introduces a `UserWarning` that is emitted if there are any graph variables that are not contained in the observed data (`self._data`). Adds a unit test for this implementation. Currently, the return values of `self.get_common_causes()`, `self.get_instruments()` and `self.get_effect_modifiers()` are...
null
2023-02-04 14:44:46+00:00
2023-02-14 20:21:29+00:00
tests/test_causal_model.py
import pandas as pd import pytest from pytest import mark from sklearn import linear_model import dowhy import dowhy.datasets from dowhy import CausalModel class TestCausalModel(object): @mark.parametrize( ["beta", "num_samples", "num_treatments"], [ (10, 100, 1), ], ) ...
import pandas as pd import pytest from pytest import mark from sklearn import linear_model import dowhy import dowhy.datasets from dowhy import CausalModel class TestCausalModel(object): @mark.parametrize( ["beta", "num_samples", "num_treatments"], [ (10, 100, 1), ], ) ...
MFreidank
2a2b3f4f7d02f8157d4b2ea39588b305e048eca8
23214a9850544780e21f0afecb390446ceca48a2
Were these whitespace edits intentional?
emrekiciman
109
py-why/dowhy
846
Enhancement: warn about unobserved graph variables in `causal_model.identify_effect`.
Closes #810. Introduces a `UserWarning` that is emitted if there are any graph variables that are not contained in the observed data (`self._data`). Adds a unit test for this implementation. Currently, the return values of `self.get_common_causes()`, `self.get_instruments()` and `self.get_effect_modifiers()` are...
null
2023-02-04 14:44:46+00:00
2023-02-14 20:21:29+00:00
tests/test_causal_model.py
import pandas as pd import pytest from pytest import mark from sklearn import linear_model import dowhy import dowhy.datasets from dowhy import CausalModel class TestCausalModel(object): @mark.parametrize( ["beta", "num_samples", "num_treatments"], [ (10, 100, 1), ], ) ...
import pandas as pd import pytest from pytest import mark from sklearn import linear_model import dowhy import dowhy.datasets from dowhy import CausalModel class TestCausalModel(object): @mark.parametrize( ["beta", "num_samples", "num_treatments"], [ (10, 100, 1), ], ) ...
MFreidank
2a2b3f4f7d02f8157d4b2ea39588b305e048eca8
23214a9850544780e21f0afecb390446ceca48a2
I believe this is causing the format check to fail in the build process.
emrekiciman
110
py-why/dowhy
846
Enhancement: warn about unobserved graph variables in `causal_model.identify_effect`.
Closes #810. Introduces a `UserWarning` that is emitted if there are any graph variables that are not contained in the observed data (`self._data`). Adds a unit test for this implementation. Currently, the return values of `self.get_common_causes()`, `self.get_instruments()` and `self.get_effect_modifiers()` are...
null
2023-02-04 14:44:46+00:00
2023-02-14 20:21:29+00:00
tests/test_causal_model.py
import pandas as pd import pytest from pytest import mark from sklearn import linear_model import dowhy import dowhy.datasets from dowhy import CausalModel class TestCausalModel(object): @mark.parametrize( ["beta", "num_samples", "num_treatments"], [ (10, 100, 1), ], ) ...
import pandas as pd import pytest from pytest import mark from sklearn import linear_model import dowhy import dowhy.datasets from dowhy import CausalModel class TestCausalModel(object): @mark.parametrize( ["beta", "num_samples", "num_treatments"], [ (10, 100, 1), ], ) ...
MFreidank
2a2b3f4f7d02f8157d4b2ea39588b305e048eca8
23214a9850544780e21f0afecb390446ceca48a2
No, they weren't intentional. Will look into fixing this, might have been caused by some editor auto-formatting.
MFreidank
111
py-why/dowhy
838
Rewriting the User Guide
Hi all, This is a first stab at unifying the user guide for effect estimation and gcm sub-package. This is an extremely early draft and will require a lot of iterations before we can put it into the right place. I'm already publishing it here to avoid going in the wrong direction and wasting time. To simplify iterat...
null
2023-01-27 15:39:34+00:00
2023-07-27 18:47:39+00:00
docs/source/user_guide/intro.rst
Introduction to DoWhy ===================== The need for causal inference ---------------------------------- Predictive models uncover patterns that connect the inputs and outcome in observed data. To intervene, however, we need to estimate the effect of changing an input from its current value, for which no data exi...
Introduction to DoWhy ===================== Much like machine learning libraries have done for prediction, DoWhy is a Python library that aims to spark causal thinking and analysis. DoWhy provides a wide variety of algorithms for effect estimation, prediction, quantification of causal influences, causal structure learn...
petergtz
ccc4af28684f251d457c740d502f1e3e0da327fd
0df1f4c5d2f737dd70bacfaf00058b1da1d5dbf9
Can also link the examples here. Maybe something along the line of "See also the example notebooks[Link] how to use them in a concrete problem."
bloebp
112
py-why/dowhy
838
Rewriting the User Guide
Hi all, This is a first stab at unifying the user guide for effect estimation and gcm sub-package. This is an extremely early draft and will require a lot of iterations before we can put it into the right place. I'm already publishing it here to avoid going in the wrong direction and wasting time. To simplify iterat...
null
2023-01-27 15:39:34+00:00
2023-07-27 18:47:39+00:00
docs/source/user_guide/intro.rst
Introduction to DoWhy ===================== The need for causal inference ---------------------------------- Predictive models uncover patterns that connect the inputs and outcome in observed data. To intervene, however, we need to estimate the effect of changing an input from its current value, for which no data exi...
Introduction to DoWhy ===================== Much like machine learning libraries have done for prediction, DoWhy is a Python library that aims to spark causal thinking and analysis. DoWhy provides a wide variety of algorithms for effect estimation, prediction, quantification of causal influences, causal structure learn...
petergtz
ccc4af28684f251d457c740d502f1e3e0da327fd
0df1f4c5d2f737dd70bacfaf00058b1da1d5dbf9
Suggestion: To perform tasks, DoWhy leverages two powerful frameworks, namely graphical causal models (GCM) and potential outcomes (PO), depending on the task at hand. What’s common to most tasks is that they require a causal graph, which is modeled after the problem domain. For that reason, this user guide starts wit...
bloebp
113
py-why/dowhy
838
Rewriting the User Guide
Hi all, This is a first stab at unifying the user guide for effect estimation and gcm sub-package. This is an extremely early draft and will require a lot of iterations before we can put it into the right place. I'm already publishing it here to avoid going in the wrong direction and wasting time. To simplify iterat...
null
2023-01-27 15:39:34+00:00
2023-07-27 18:47:39+00:00
docs/source/user_guide/intro.rst
Introduction to DoWhy ===================== The need for causal inference ---------------------------------- Predictive models uncover patterns that connect the inputs and outcome in observed data. To intervene, however, we need to estimate the effect of changing an input from its current value, for which no data exi...
Introduction to DoWhy ===================== Much like machine learning libraries have done for prediction, DoWhy is a Python library that aims to spark causal thinking and analysis. DoWhy provides a wide variety of algorithms for effect estimation, prediction, quantification of causal influences, causal structure learn...
petergtz
ccc4af28684f251d457c740d502f1e3e0da327fd
0df1f4c5d2f737dd70bacfaf00058b1da1d5dbf9
Comments on the new image: Its really great! Two things: - Can maybe more generally say “Quantifying causal contributions” or something along the line instead of specifically ICC, seeing that we also have direct arrow strength. - Would it make sense to add an optional validation/refutation step? Although this might ...
bloebp
114
py-why/dowhy
838
Rewriting the User Guide
Hi all, This is a first stab at unifying the user guide for effect estimation and gcm sub-package. This is an extremely early draft and will require a lot of iterations before we can put it into the right place. I'm already publishing it here to avoid going in the wrong direction and wasting time. To simplify iterat...
null
2023-01-27 15:39:34+00:00
2023-07-27 18:47:39+00:00
docs/source/user_guide/intro.rst
Introduction to DoWhy ===================== The need for causal inference ---------------------------------- Predictive models uncover patterns that connect the inputs and outcome in observed data. To intervene, however, we need to estimate the effect of changing an input from its current value, for which no data exi...
Introduction to DoWhy ===================== Much like machine learning libraries have done for prediction, DoWhy is a Python library that aims to spark causal thinking and analysis. DoWhy provides a wide variety of algorithms for effect estimation, prediction, quantification of causal influences, causal structure learn...
petergtz
ccc4af28684f251d457c740d502f1e3e0da327fd
0df1f4c5d2f737dd70bacfaf00058b1da1d5dbf9
> Can maybe more generally say “Quantifying causal contributions” or something along the line instead of specifically ICC, seeing that we also have direct arrow strength. Yes. The PowerPoint presentation (which acts as source for this png) is alongside everything else. Just open it, make the necessary changes and ta...
petergtz
115
py-why/dowhy
829
Pin Numpy version to 1.23.1
This makes it compatible with Numba version 0.56.4. The way the Numpy version was specified before, installs version 1.24.0 which is incompatible with Numba, see also https://github.com/numba/numba/pull/8691. Numba is a transient dependency coming in through `econml` and `sparse`. The resulting error looks like this...
null
2023-01-24 09:37:39+00:00
2023-01-26 16:19:34+00:00
pyproject.toml
[tool.poetry] name = "dowhy" # # 0.0.0 is standard placeholder for poetry-dynamic-versioning # any changes to this should not be checked in # version = "0.0.0" description = "DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions" authors = ["PyWhy Community <a...
[tool.poetry] name = "dowhy" # # 0.0.0 is standard placeholder for poetry-dynamic-versioning # any changes to this should not be checked in # version = "0.0.0" description = "DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions" authors = ["PyWhy Community <a...
petergtz
96a055ae4448ecddeb458d86b10588ea3295c4a3
ca340303e993fa4be1831f74afccf9161e19a9d8
I don't think the `==` is strictly necessary here. With semver you can just remove the `^` symbol to match the explicit version.
darthtrevino
116
py-why/dowhy
811
Enhancement/remove data from estimators
* Remove data, treatment and outcome from estimator object * estimate_effect now have data, treatment and outcome as parameters
null
2023-01-03 18:39:35+00:00
2023-01-20 11:06:39+00:00
dowhy/causal_estimator.py
import copy import logging from collections import namedtuple from typing import Dict, List, Optional, Union import numpy as np import pandas as pd import sympy as sp from sklearn.utils import resample import dowhy.interpreters as interpreters from dowhy.causal_identifier.identified_estimand import IdentifiedEstimand...
import copy import logging from collections import namedtuple from dataclasses import dataclass from typing import Any, Dict, List, Optional, Tuple, Union import numpy as np import pandas as pd import sympy as sp from sklearn.utils import resample import dowhy.interpreters as interpreters from dowhy.causal_identifier...
andresmor-ms
325cf4e245de3e55b85a42c5fefc36f6ef34db46
c3e1c75696a8297aa48977ea4c7b7e8dc50dc6aa
do we require treatment_name as a parameter here since it can be inferred from self? In general, I think it is okay for internal methods that start with _ to require this argument, but definitely user-facing methods can avoid it.
amit-sharma
117
py-why/dowhy
811
Enhancement/remove data from estimators
* Remove data, treatment and outcome from estimator object * estimate_effect now have data, treatment and outcome as parameters
null
2023-01-03 18:39:35+00:00
2023-01-20 11:06:39+00:00
dowhy/causal_estimator.py
import copy import logging from collections import namedtuple from typing import Dict, List, Optional, Union import numpy as np import pandas as pd import sympy as sp from sklearn.utils import resample import dowhy.interpreters as interpreters from dowhy.causal_identifier.identified_estimand import IdentifiedEstimand...
import copy import logging from collections import namedtuple from dataclasses import dataclass from typing import Any, Dict, List, Optional, Tuple, Union import numpy as np import pandas as pd import sympy as sp from sklearn.utils import resample import dowhy.interpreters as interpreters from dowhy.causal_identifier...
andresmor-ms
325cf4e245de3e55b85a42c5fefc36f6ef34db46
c3e1c75696a8297aa48977ea4c7b7e8dc50dc6aa
same comment, I suggest removing the treatment_name argument.
amit-sharma
118
py-why/dowhy
811
Enhancement/remove data from estimators
* Remove data, treatment and outcome from estimator object * estimate_effect now have data, treatment and outcome as parameters
null
2023-01-03 18:39:35+00:00
2023-01-20 11:06:39+00:00
dowhy/causal_estimators/econml.py
import inspect from importlib import import_module from typing import Any, Callable, List, Optional, Protocol, Union from warnings import warn import numpy as np import pandas as pd from numpy.distutils.misc_util import is_sequence from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.causal_i...
import inspect from importlib import import_module from typing import Any, Callable, List, Optional, Protocol, Union from warnings import warn import numpy as np import pandas as pd from numpy.distutils.misc_util import is_sequence from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.causal_i...
andresmor-ms
325cf4e245de3e55b85a42c5fefc36f6ef34db46
c3e1c75696a8297aa48977ea4c7b7e8dc50dc6aa
we might not need treatment_name
amit-sharma
119
py-why/dowhy
811
Enhancement/remove data from estimators
* Remove data, treatment and outcome from estimator object * estimate_effect now have data, treatment and outcome as parameters
null
2023-01-03 18:39:35+00:00
2023-01-20 11:06:39+00:00
dowhy/causal_estimators/generalized_linear_model_estimator.py
import itertools from typing import Any, List, Optional, Union import pandas as pd import statsmodels.api as sm from dowhy.causal_estimator import CausalEstimator from dowhy.causal_estimators.regression_estimator import RegressionEstimator from dowhy.causal_identifier import IdentifiedEstimand class GeneralizedLine...
import itertools from typing import Any, List, Optional, Union import pandas as pd import statsmodels.api as sm from dowhy.causal_estimator import CausalEstimator from dowhy.causal_estimators.regression_estimator import RegressionEstimator from dowhy.causal_identifier import IdentifiedEstimand class GeneralizedLine...
andresmor-ms
325cf4e245de3e55b85a42c5fefc36f6ef34db46
c3e1c75696a8297aa48977ea4c7b7e8dc50dc6aa
treatment and outcome names can be avoided for predict_fn and _build_model.
amit-sharma
120
py-why/dowhy
811
Enhancement/remove data from estimators
* Remove data, treatment and outcome from estimator object * estimate_effect now have data, treatment and outcome as parameters
null
2023-01-03 18:39:35+00:00
2023-01-20 11:06:39+00:00
dowhy/causal_estimators/instrumental_variable_estimator.py
from typing import Any, Dict, List, Optional, Union import numpy as np import pandas as pd import sympy as sp import sympy.stats as spstats from statsmodels.sandbox.regression.gmm import IV2SLS from dowhy.causal_estimator import CausalEstimate, CausalEstimator, RealizedEstimand from dowhy.causal_identifier import Ide...
from typing import Any, Dict, List, Optional, Union import numpy as np import pandas as pd import sympy as sp import sympy.stats as spstats from statsmodels.sandbox.regression.gmm import IV2SLS from dowhy.causal_estimator import CausalEstimate, CausalEstimator, RealizedEstimand from dowhy.causal_identifier import Ide...
andresmor-ms
325cf4e245de3e55b85a42c5fefc36f6ef34db46
c3e1c75696a8297aa48977ea4c7b7e8dc50dc6aa
can avoid treatment_name and outcome_name here too
amit-sharma
121
py-why/dowhy
811
Enhancement/remove data from estimators
* Remove data, treatment and outcome from estimator object * estimate_effect now have data, treatment and outcome as parameters
null
2023-01-03 18:39:35+00:00
2023-01-20 11:06:39+00:00
dowhy/causal_estimators/regression_estimator.py
from typing import Any, List, Optional, Union import numpy as np import pandas as pd import statsmodels.api as sm from dowhy.causal_estimator import CausalEstimate, CausalEstimator, IdentifiedEstimand class RegressionEstimator(CausalEstimator): """Compute effect of treatment using some regression function. ...
from typing import Any, List, Optional, Union import numpy as np import pandas as pd import statsmodels.api as sm from dowhy.causal_estimator import CausalEstimate, CausalEstimator, IdentifiedEstimand class RegressionEstimator(CausalEstimator): """Compute effect of treatment using some regression function. ...
andresmor-ms
325cf4e245de3e55b85a42c5fefc36f6ef34db46
c3e1c75696a8297aa48977ea4c7b7e8dc50dc6aa
treatment_name can be avoided here too as a parameter
amit-sharma
122
py-why/dowhy
811
Enhancement/remove data from estimators
* Remove data, treatment and outcome from estimator object * estimate_effect now have data, treatment and outcome as parameters
null
2023-01-03 18:39:35+00:00
2023-01-20 11:06:39+00:00
dowhy/causal_estimators/two_stage_regression_estimator.py
import copy from typing import Any, List, Optional, Type, Union import numpy as np import pandas as pd from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.causal_estimators.linear_regression_estimator import LinearRegressionEstimator from dowhy.causal_identifier import EstimandType, Identifi...
import copy from typing import Any, List, Optional, Type, Union import numpy as np import pandas as pd from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.causal_estimators.linear_regression_estimator import LinearRegressionEstimator from dowhy.causal_identifier import EstimandType, Identifi...
andresmor-ms
325cf4e245de3e55b85a42c5fefc36f6ef34db46
c3e1c75696a8297aa48977ea4c7b7e8dc50dc6aa
treatment name parameter can be avoided here too
amit-sharma
123
py-why/dowhy
789
Add Python 3.10 support
* Add Python version condition on `autogluon-tabular`. * Remove `autogluon-tabular` from dev-dependencies to add Python 3.10 to our builds. **Note**: `autogluon-tabular` has a dependency constraint on `scikit-learn` of `<1.2.0`. So we're able to advance `scikit-learn` past `1.0.2`, but not to the most recent versio...
null
2022-12-13 17:13:03+00:00
2022-12-15 14:13:24+00:00
pyproject.toml
[tool.poetry] name = "dowhy" # # 0.0.0 is standard placeholder for poetry-dynamic-versioning # any changes to this should not be checked in # version = "0.0.0" description = "DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions" authors = ["PyWhy Community <a...
[tool.poetry] name = "dowhy" # # 0.0.0 is standard placeholder for poetry-dynamic-versioning # any changes to this should not be checked in # version = "0.0.0" description = "DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions" authors = ["PyWhy Community <a...
darthtrevino
5bbac57b8104d99b03bd5e439a7c671bc7215303
7c6f56fb9023b023e694beba51656d826d9c1843
Is tensorflow still required? I thought this was only a dependency of autogluon.
petergtz
124
py-why/dowhy
789
Add Python 3.10 support
* Add Python version condition on `autogluon-tabular`. * Remove `autogluon-tabular` from dev-dependencies to add Python 3.10 to our builds. **Note**: `autogluon-tabular` has a dependency constraint on `scikit-learn` of `<1.2.0`. So we're able to advance `scikit-learn` past `1.0.2`, but not to the most recent versio...
null
2022-12-13 17:13:03+00:00
2022-12-15 14:13:24+00:00
tests/gcm/ml/test_autogluon.py
import tempfile import numpy as np from flaky import flaky from pytest import approx from sklearn.model_selection import train_test_split from dowhy.gcm.fcms import AdditiveNoiseModel, ClassifierFCM from dowhy.gcm.ml.autolguon import AutoGluonClassifier, AutoGluonRegressor from dowhy.gcm.util.general import shape_int...
import tempfile import numpy as np from flaky import flaky from pytest import approx, importorskip, mark from sklearn.model_selection import train_test_split from dowhy.gcm.fcms import AdditiveNoiseModel, ClassifierFCM autolguon = importorskip("dowhy.gcm.ml.autolguon") from dowhy.gcm.ml.autolguon import AutoGluonCla...
darthtrevino
5bbac57b8104d99b03bd5e439a7c671bc7215303
7c6f56fb9023b023e694beba51656d826d9c1843
I think for this there's a better way to do it. Reading https://docs.pytest.org/en/7.1.x/how-to/skipping.html#skipping-on-a-missing-import-dependency it should be possible to do: ```python autolguon = pytest.importorskip("dowhy.gcm.ml.autolguon") from dowhy.gcm.ml.autolguon import AutoGluonClassifier, AutoGluonReg...
petergtz
125
py-why/dowhy
789
Add Python 3.10 support
* Add Python version condition on `autogluon-tabular`. * Remove `autogluon-tabular` from dev-dependencies to add Python 3.10 to our builds. **Note**: `autogluon-tabular` has a dependency constraint on `scikit-learn` of `<1.2.0`. So we're able to advance `scikit-learn` past `1.0.2`, but not to the most recent versio...
null
2022-12-13 17:13:03+00:00
2022-12-15 14:13:24+00:00
tests/gcm/test_auto.py
import networkx as nx import numpy as np import pandas as pd from flaky import flaky from sklearn.ensemble import HistGradientBoostingClassifier, HistGradientBoostingRegressor from sklearn.linear_model import ElasticNetCV, LassoCV, LinearRegression, LogisticRegression, RidgeCV from sklearn.naive_bayes import GaussianNB...
import networkx as nx import numpy as np import pandas as pd from flaky import flaky from pytest import mark from sklearn.ensemble import HistGradientBoostingClassifier, HistGradientBoostingRegressor from sklearn.linear_model import ElasticNetCV, LassoCV, LinearRegression, LogisticRegression, RidgeCV from sklearn.naive...
darthtrevino
5bbac57b8104d99b03bd5e439a7c671bc7215303
7c6f56fb9023b023e694beba51656d826d9c1843
With the skip marker, is it really necessary to comment out the code? Is PyTest not entirely skipping this test? Assuming that, we can simply move the `AutoGluonClassifier` import into the test function itself. I just pushed a new commit trying this out.
petergtz
126
py-why/dowhy
789
Add Python 3.10 support
* Add Python version condition on `autogluon-tabular`. * Remove `autogluon-tabular` from dev-dependencies to add Python 3.10 to our builds. **Note**: `autogluon-tabular` has a dependency constraint on `scikit-learn` of `<1.2.0`. So we're able to advance `scikit-learn` past `1.0.2`, but not to the most recent versio...
null
2022-12-13 17:13:03+00:00
2022-12-15 14:13:24+00:00
tests/gcm/test_auto.py
import networkx as nx import numpy as np import pandas as pd from flaky import flaky from sklearn.ensemble import HistGradientBoostingClassifier, HistGradientBoostingRegressor from sklearn.linear_model import ElasticNetCV, LassoCV, LinearRegression, LogisticRegression, RidgeCV from sklearn.naive_bayes import GaussianNB...
import networkx as nx import numpy as np import pandas as pd from flaky import flaky from pytest import mark from sklearn.ensemble import HistGradientBoostingClassifier, HistGradientBoostingRegressor from sklearn.linear_model import ElasticNetCV, LassoCV, LinearRegression, LogisticRegression, RidgeCV from sklearn.naive...
darthtrevino
5bbac57b8104d99b03bd5e439a7c671bc7215303
7c6f56fb9023b023e694beba51656d826d9c1843
I think the reasoning was that the import would fail, but let's see what it does
darthtrevino
127
py-why/dowhy
789
Add Python 3.10 support
* Add Python version condition on `autogluon-tabular`. * Remove `autogluon-tabular` from dev-dependencies to add Python 3.10 to our builds. **Note**: `autogluon-tabular` has a dependency constraint on `scikit-learn` of `<1.2.0`. So we're able to advance `scikit-learn` past `1.0.2`, but not to the most recent versio...
null
2022-12-13 17:13:03+00:00
2022-12-15 14:13:24+00:00
tests/gcm/test_auto.py
import networkx as nx import numpy as np import pandas as pd from flaky import flaky from sklearn.ensemble import HistGradientBoostingClassifier, HistGradientBoostingRegressor from sklearn.linear_model import ElasticNetCV, LassoCV, LinearRegression, LogisticRegression, RidgeCV from sklearn.naive_bayes import GaussianNB...
import networkx as nx import numpy as np import pandas as pd from flaky import flaky from pytest import mark from sklearn.ensemble import HistGradientBoostingClassifier, HistGradientBoostingRegressor from sklearn.linear_model import ElasticNetCV, LassoCV, LinearRegression, LogisticRegression, RidgeCV from sklearn.naive...
darthtrevino
5bbac57b8104d99b03bd5e439a7c671bc7215303
7c6f56fb9023b023e694beba51656d826d9c1843
LGTM!
darthtrevino
128
py-why/dowhy
786
Make autogluon optional
It looks like since this has not been listed in the 'extra' group, it's not marked as optional. Right now, a `pip install dowhy` pulls in autogluon and pytorch and makes this a pretty large installation. I believe this change should fix this. Signed-off-by: Peter Goetz <pego@amazon.com>
null
2022-12-09 22:26:14+00:00
2022-12-12 09:07:24+00:00
pyproject.toml
[tool.poetry] name = "dowhy" # # 0.0.0 is standard placeholder for poetry-dynamic-versioning # any changes to this should not be checked in # version = "0.0.0" description = "DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions" authors = ["PyWhy Community <a...
[tool.poetry] name = "dowhy" # # 0.0.0 is standard placeholder for poetry-dynamic-versioning # any changes to this should not be checked in # version = "0.0.0" description = "DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions" authors = ["PyWhy Community <a...
petergtz
6ebfc4464bd5c41880f25299df25c3c1cd05b20c
b5ae49a256217f1cae5a45033e8e6bc84599e508
I think the package name is autogluon.tabular.
bloebp
129
py-why/dowhy
786
Make autogluon optional
It looks like since this has not been listed in the 'extra' group, it's not marked as optional. Right now, a `pip install dowhy` pulls in autogluon and pytorch and makes this a pretty large installation. I believe this change should fix this. Signed-off-by: Peter Goetz <pego@amazon.com>
null
2022-12-09 22:26:14+00:00
2022-12-12 09:07:24+00:00
pyproject.toml
[tool.poetry] name = "dowhy" # # 0.0.0 is standard placeholder for poetry-dynamic-versioning # any changes to this should not be checked in # version = "0.0.0" description = "DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions" authors = ["PyWhy Community <a...
[tool.poetry] name = "dowhy" # # 0.0.0 is standard placeholder for poetry-dynamic-versioning # any changes to this should not be checked in # version = "0.0.0" description = "DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions" authors = ["PyWhy Community <a...
petergtz
6ebfc4464bd5c41880f25299df25c3c1cd05b20c
b5ae49a256217f1cae5a45033e8e6bc84599e508
They both work. According to https://peps.python.org/pep-0503/#normalized-names, a dot gets replaced by a dash. Note also that we use `autogluon-tabular` already in other places in this project file. Also tested this locally. Both names work.
petergtz
130
py-why/dowhy
778
[Version Bump v0.9] Release notes and updated docstrings
This PR includes the release notes for v0.9 Signed-off-by: Amit Sharma <amit_sharma@live.com>
null
2022-12-04 10:24:21+00:00
2022-12-06 05:37:44+00:00
docs/source/code_repo.rst
Release notes ============= DoWhy is hosted on GitHub. You can browse the code in a html-friendly format `here <https://github.com/Microsoft/dowhy>`_. v0.7.1: Added Graph refuter. Support for dagitty graphs and external estimators --------------------------------------------------------------------------------------...
Release notes ============= DoWhy is hosted on GitHub. You can browse the code in a html-friendly format `here <https://github.com/Microsoft/dowhy>`_. v0.9: New functional API (preview), faster refutations, and better independence tests for GCMs -----------------------------------------------------------------------...
amit-sharma
649b3c32a11ddff86e932c24b3551161ba5f5747
9e938f333ec5436697a4f48a7ebb0e63a39a0303
This should also list: ``` * New API for unit change attribution (@kailashbuki) * New quality option `BEST` for auto-assignment of causal mechanisms, which uses the optional auto-ML library AutoGluon (@bloebp) ```
petergtz
131
py-why/dowhy
778
[Version Bump v0.9] Release notes and updated docstrings
This PR includes the release notes for v0.9 Signed-off-by: Amit Sharma <amit_sharma@live.com>
null
2022-12-04 10:24:21+00:00
2022-12-06 05:37:44+00:00
docs/source/code_repo.rst
Release notes ============= DoWhy is hosted on GitHub. You can browse the code in a html-friendly format `here <https://github.com/Microsoft/dowhy>`_. v0.7.1: Added Graph refuter. Support for dagitty graphs and external estimators --------------------------------------------------------------------------------------...
Release notes ============= DoWhy is hosted on GitHub. You can browse the code in a html-friendly format `here <https://github.com/Microsoft/dowhy>`_. v0.9: New functional API (preview), faster refutations, and better independence tests for GCMs -----------------------------------------------------------------------...
amit-sharma
649b3c32a11ddff86e932c24b3551161ba5f5747
9e938f333ec5436697a4f48a7ebb0e63a39a0303
> Also, you can now browse docs separately for different versions of DoWhy This was already available in the previous version, so you can probably remove that. How about: ``` * New PyData theme for documentation with new homepage, Getting started guide, revised User Guide and examples page (@petergtz) ``` ...
petergtz
132
py-why/dowhy
778
[Version Bump v0.9] Release notes and updated docstrings
This PR includes the release notes for v0.9 Signed-off-by: Amit Sharma <amit_sharma@live.com>
null
2022-12-04 10:24:21+00:00
2022-12-06 05:37:44+00:00
docs/source/code_repo.rst
Release notes ============= DoWhy is hosted on GitHub. You can browse the code in a html-friendly format `here <https://github.com/Microsoft/dowhy>`_. v0.7.1: Added Graph refuter. Support for dagitty graphs and external estimators --------------------------------------------------------------------------------------...
Release notes ============= DoWhy is hosted on GitHub. You can browse the code in a html-friendly format `here <https://github.com/Microsoft/dowhy>`_. v0.9: New functional API (preview), faster refutations, and better independence tests for GCMs -----------------------------------------------------------------------...
amit-sharma
649b3c32a11ddff86e932c24b3551161ba5f5747
9e938f333ec5436697a4f48a7ebb0e63a39a0303
Got it, making those changes.
amit-sharma
133
py-why/dowhy
778
[Version Bump v0.9] Release notes and updated docstrings
This PR includes the release notes for v0.9 Signed-off-by: Amit Sharma <amit_sharma@live.com>
null
2022-12-04 10:24:21+00:00
2022-12-06 05:37:44+00:00
docs/source/dowhy.rst
API reference ============= Subpackages ----------- .. toctree:: :maxdepth: 4 dowhy.api dowhy.causal_estimators dowhy.causal_identifiers dowhy.causal_refuters dowhy.data_transformers dowhy.do_samplers dowhy.gcm dowhy.graph_learners dowhy.interpreters dowhy.utils Submodules ---------...
API reference ============= Subpackages ----------- .. toctree:: :maxdepth: 4 dowhy.api dowhy.causal_estimators dowhy.causal_identifier dowhy.causal_refuters dowhy.data_transformers dowhy.do_samplers dowhy.gcm dowhy.graph_learners dowhy.interpreters dowhy.utils Submodules ----------...
amit-sharma
649b3c32a11ddff86e932c24b3551161ba5f5747
9e938f333ec5436697a4f48a7ebb0e63a39a0303
Did you change this on purpose? Typically, this part is called the "API Reference", so I would assume users would look for that when they search for documentation on DoWhy's API.
petergtz
134
py-why/dowhy
778
[Version Bump v0.9] Release notes and updated docstrings
This PR includes the release notes for v0.9 Signed-off-by: Amit Sharma <amit_sharma@live.com>
null
2022-12-04 10:24:21+00:00
2022-12-06 05:37:44+00:00
docs/source/dowhy.rst
API reference ============= Subpackages ----------- .. toctree:: :maxdepth: 4 dowhy.api dowhy.causal_estimators dowhy.causal_identifiers dowhy.causal_refuters dowhy.data_transformers dowhy.do_samplers dowhy.gcm dowhy.graph_learners dowhy.interpreters dowhy.utils Submodules ---------...
API reference ============= Subpackages ----------- .. toctree:: :maxdepth: 4 dowhy.api dowhy.causal_estimators dowhy.causal_identifier dowhy.causal_refuters dowhy.data_transformers dowhy.do_samplers dowhy.gcm dowhy.graph_learners dowhy.interpreters dowhy.utils Submodules ----------...
amit-sharma
649b3c32a11ddff86e932c24b3551161ba5f5747
9e938f333ec5436697a4f48a7ebb0e63a39a0303
No, let me revert it back.
amit-sharma
135
py-why/dowhy
768
An attempt at a DCO-compliant version of multivalue treatment PR
Signed-off-by: Egor Kraev egor.kraev@wise.com
null
2022-11-21 08:19:21+00:00
2022-11-24 06:17:03+00:00
dowhy/causal_estimators/econml.py
import inspect from importlib import import_module import econml import numpy as np import pandas as pd from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wrapper class for estimators from the EconML library. For a li...
import inspect from importlib import import_module from typing import Callable import numpy as np import pandas as pd from numpy.distutils.misc_util import is_sequence from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wra...
EgorKraevTransferwise
f83a276393f7f89e8c7991e5750c2f1095127e04
6cf8366c48b5f010952c14f615585a41cd37c3bf
what does tt stand for? true treatment?
amit-sharma
136
py-why/dowhy
768
An attempt at a DCO-compliant version of multivalue treatment PR
Signed-off-by: Egor Kraev egor.kraev@wise.com
null
2022-11-21 08:19:21+00:00
2022-11-24 06:17:03+00:00
dowhy/causal_estimators/econml.py
import inspect from importlib import import_module import econml import numpy as np import pandas as pd from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wrapper class for estimators from the EconML library. For a li...
import inspect from importlib import import_module from typing import Callable import numpy as np import pandas as pd from numpy.distutils.misc_util import is_sequence from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wra...
EgorKraevTransferwise
f83a276393f7f89e8c7991e5750c2f1095127e04
6cf8366c48b5f010952c14f615585a41cd37c3bf
I did not understand what this loop does? An inline comment will help.
amit-sharma
137
py-why/dowhy
768
An attempt at a DCO-compliant version of multivalue treatment PR
Signed-off-by: Egor Kraev egor.kraev@wise.com
null
2022-11-21 08:19:21+00:00
2022-11-24 06:17:03+00:00
dowhy/causal_estimators/econml.py
import inspect from importlib import import_module import econml import numpy as np import pandas as pd from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wrapper class for estimators from the EconML library. For a li...
import inspect from importlib import import_module from typing import Callable import numpy as np import pandas as pd from numpy.distutils.misc_util import is_sequence from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wra...
EgorKraevTransferwise
f83a276393f7f89e8c7991e5750c2f1095127e04
6cf8366c48b5f010952c14f615585a41cd37c3bf
will it possible to add a docstring for these methods? It will be good to add a docstring to the the methods we expect people to access. In the past, we were not so consistent, but this will be a good practice.
amit-sharma
138
py-why/dowhy
768
An attempt at a DCO-compliant version of multivalue treatment PR
Signed-off-by: Egor Kraev egor.kraev@wise.com
null
2022-11-21 08:19:21+00:00
2022-11-24 06:17:03+00:00
dowhy/causal_estimators/econml.py
import inspect from importlib import import_module import econml import numpy as np import pandas as pd from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wrapper class for estimators from the EconML library. For a li...
import inspect from importlib import import_module from typing import Callable import numpy as np import pandas as pd from numpy.distutils.misc_util import is_sequence from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wra...
EgorKraevTransferwise
f83a276393f7f89e8c7991e5750c2f1095127e04
6cf8366c48b5f010952c14f615585a41cd37c3bf
Adding a line in docstring to clarify
EgorKraevTransferwise
139
py-why/dowhy
768
An attempt at a DCO-compliant version of multivalue treatment PR
Signed-off-by: Egor Kraev egor.kraev@wise.com
null
2022-11-21 08:19:21+00:00
2022-11-24 06:17:03+00:00
dowhy/causal_estimators/econml.py
import inspect from importlib import import_module import econml import numpy as np import pandas as pd from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wrapper class for estimators from the EconML library. For a li...
import inspect from importlib import import_module from typing import Callable import numpy as np import pandas as pd from numpy.distutils.misc_util import is_sequence from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wra...
EgorKraevTransferwise
f83a276393f7f89e8c7991e5750c2f1095127e04
6cf8366c48b5f010952c14f615585a41cd37c3bf
Adding
EgorKraevTransferwise
140
py-why/dowhy
768
An attempt at a DCO-compliant version of multivalue treatment PR
Signed-off-by: Egor Kraev egor.kraev@wise.com
null
2022-11-21 08:19:21+00:00
2022-11-24 06:17:03+00:00
dowhy/causal_estimators/econml.py
import inspect from importlib import import_module import econml import numpy as np import pandas as pd from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wrapper class for estimators from the EconML library. For a li...
import inspect from importlib import import_module from typing import Callable import numpy as np import pandas as pd from numpy.distutils.misc_util import is_sequence from dowhy.causal_estimator import CausalEstimate, CausalEstimator from dowhy.utils.api import parse_state class Econml(CausalEstimator): """Wra...
EgorKraevTransferwise
f83a276393f7f89e8c7991e5750c2f1095127e04
6cf8366c48b5f010952c14f615585a41cd37c3bf
Adding
EgorKraevTransferwise
141
py-why/dowhy
766
Adding autogluon as optional dependency to gcm module
This extends the auto model assignment by the 'BEST' parameter, which returns an autogluon model (i.e., an auto ML model).
null
2022-11-18 00:33:18+00:00
2022-11-23 21:05:13+00:00
poetry.lock
[[package]] name = "absl-py" version = "1.3.0" description = "Abseil Python Common Libraries, see https://github.com/abseil/abseil-py." category = "dev" optional = false python-versions = ">=3.6" [[package]] name = "alabaster" version = "0.7.12" description = "A configurable sidebar-enabled Sphinx theme" category = "m...
[[package]] name = "absl-py" version = "1.3.0" description = "Abseil Python Common Libraries, see https://github.com/abseil/abseil-py." category = "dev" optional = false python-versions = ">=3.6" [[package]] name = "alabaster" version = "0.7.12" description = "A configurable sidebar-enabled Sphinx theme" category = "m...
bloebp
0a03cb10baddae6d6ca94d4d38fa074f5391b426
f83a276393f7f89e8c7991e5750c2f1095127e04
Please make sure you're on the latest poetry version. I believe latest uses underscores, not dashes. Let's try and keep noise in the lock as low as possible.
petergtz
142
py-why/dowhy
766
Adding autogluon as optional dependency to gcm module
This extends the auto model assignment by the 'BEST' parameter, which returns an autogluon model (i.e., an auto ML model).
null
2022-11-18 00:33:18+00:00
2022-11-23 21:05:13+00:00
poetry.lock
[[package]] name = "absl-py" version = "1.3.0" description = "Abseil Python Common Libraries, see https://github.com/abseil/abseil-py." category = "dev" optional = false python-versions = ">=3.6" [[package]] name = "alabaster" version = "0.7.12" description = "A configurable sidebar-enabled Sphinx theme" category = "m...
[[package]] name = "absl-py" version = "1.3.0" description = "Abseil Python Common Libraries, see https://github.com/abseil/abseil-py." category = "dev" optional = false python-versions = ">=3.6" [[package]] name = "alabaster" version = "0.7.12" description = "A configurable sidebar-enabled Sphinx theme" category = "m...
bloebp
0a03cb10baddae6d6ca94d4d38fa074f5391b426
f83a276393f7f89e8c7991e5750c2f1095127e04
Sorted via Discord. Looks like the latest version of Poetry (1.2.2) actually does use dashes. Hopefully this will not be an issue anymore in the future.
petergtz
143
py-why/dowhy
764
Add convenience function to plot attribution scores
We're using function in so many places in notebooks, similar to the graph plot utility that I think it's worth providing this as a convenience function.
null
2022-11-15 22:35:03+00:00
2022-11-16 14:11:16+00:00
dowhy/gcm/util/plotting.py
import logging from typing import Any, Dict, List, Optional, Tuple import networkx as nx import pandas as pd from matplotlib import pyplot from networkx.drawing import nx_pydot _logger = logging.getLogger(__name__) def plot( causal_graph: nx.Graph, causal_strengths: Optional[Dict[Tuple[Any, Any], float]] = ...
import logging from typing import Any, Dict, List, Optional, Tuple import networkx as nx import numpy as np import pandas as pd from matplotlib import pyplot from networkx.drawing import nx_pydot _logger = logging.getLogger(__name__) def plot( causal_graph: nx.Graph, causal_strengths: Optional[Dict[Tuple[An...
petergtz
7568d9528b7108830821e8cc729c881523a58919
a067703ca430cb0a571ac31c9b90d21af8d0543b
It would be nice to have the xlabel as well. But not a must have.
kailashbuki
144
py-why/dowhy
764
Add convenience function to plot attribution scores
We're using function in so many places in notebooks, similar to the graph plot utility that I think it's worth providing this as a convenience function.
null
2022-11-15 22:35:03+00:00
2022-11-16 14:11:16+00:00
dowhy/gcm/util/plotting.py
import logging from typing import Any, Dict, List, Optional, Tuple import networkx as nx import pandas as pd from matplotlib import pyplot from networkx.drawing import nx_pydot _logger = logging.getLogger(__name__) def plot( causal_graph: nx.Graph, causal_strengths: Optional[Dict[Tuple[Any, Any], float]] = ...
import logging from typing import Any, Dict, List, Optional, Tuple import networkx as nx import numpy as np import pandas as pd from matplotlib import pyplot from networkx.drawing import nx_pydot _logger = logging.getLogger(__name__) def plot( causal_graph: nx.Graph, causal_strengths: Optional[Dict[Tuple[An...
petergtz
7568d9528b7108830821e8cc729c881523a58919
a067703ca430cb0a571ac31c9b90d21af8d0543b
Optionally: Can add something like `**kwargs_bar_plot`, which are passed into the `pyplot.bar` call.
bloebp
145
py-why/dowhy
764
Add convenience function to plot attribution scores
We're using function in so many places in notebooks, similar to the graph plot utility that I think it's worth providing this as a convenience function.
null
2022-11-15 22:35:03+00:00
2022-11-16 14:11:16+00:00
dowhy/gcm/util/plotting.py
import logging from typing import Any, Dict, List, Optional, Tuple import networkx as nx import pandas as pd from matplotlib import pyplot from networkx.drawing import nx_pydot _logger = logging.getLogger(__name__) def plot( causal_graph: nx.Graph, causal_strengths: Optional[Dict[Tuple[Any, Any], float]] = ...
import logging from typing import Any, Dict, List, Optional, Tuple import networkx as nx import numpy as np import pandas as pd from matplotlib import pyplot from networkx.drawing import nx_pydot _logger = logging.getLogger(__name__) def plot( causal_graph: nx.Graph, causal_strengths: Optional[Dict[Tuple[An...
petergtz
7568d9528b7108830821e8cc729c881523a58919
a067703ca430cb0a571ac31c9b90d21af8d0543b
I'll add them in a follow-up PR.
petergtz
146
py-why/dowhy
764
Add convenience function to plot attribution scores
We're using function in so many places in notebooks, similar to the graph plot utility that I think it's worth providing this as a convenience function.
null
2022-11-15 22:35:03+00:00
2022-11-16 14:11:16+00:00
dowhy/gcm/util/plotting.py
import logging from typing import Any, Dict, List, Optional, Tuple import networkx as nx import pandas as pd from matplotlib import pyplot from networkx.drawing import nx_pydot _logger = logging.getLogger(__name__) def plot( causal_graph: nx.Graph, causal_strengths: Optional[Dict[Tuple[Any, Any], float]] = ...
import logging from typing import Any, Dict, List, Optional, Tuple import networkx as nx import numpy as np import pandas as pd from matplotlib import pyplot from networkx.drawing import nx_pydot _logger = logging.getLogger(__name__) def plot( causal_graph: nx.Graph, causal_strengths: Optional[Dict[Tuple[An...
petergtz
7568d9528b7108830821e8cc729c881523a58919
a067703ca430cb0a571ac31c9b90d21af8d0543b
Fair. I'd still post-pone this into a separate PR.
petergtz
147
py-why/dowhy
762
Add APIs for attributing unit change without a mechanism change
- Added two new APIs for attributing unit-level output changes when the mechanism does not change. Signed-off-by: Kailash <111277+kailashbuki@users.noreply.github.com>
null
2022-11-14 09:40:52+00:00
2022-11-14 21:51:34+00:00
dowhy/gcm/unit_change.py
"""This module provides the APIs for attributing the change in the output value of a deterministic mechanism for a statistical unit. """ from abc import abstractmethod from typing import List, Optional import numpy as np import pandas as pd from sklearn.linear_model._base import LinearModel from sklearn.utils.validat...
"""This module provides the APIs for attributing the change in the output value of a deterministic mechanism for a statistical unit. """ from abc import abstractmethod from typing import List, Optional import numpy as np import pandas as pd from sklearn.linear_model._base import LinearModel from sklearn.utils.validat...
kailashbuki
675cbc444cc422b007ba72fdf3303b328b123c48
ac047066326d0de3646f8461c0c5c4dd9826287c
What about using ```if isinstance(mechanism, LinearPredictionModel)``` and then have only one method ```unit_change_input_only```? Same applies to the existing method. To further simplify it (i.e., avoid having multiple different methods), you could also consider adding a parameter ```analyze_inputs_only``` to the o...
bloebp
148
py-why/dowhy
762
Add APIs for attributing unit change without a mechanism change
- Added two new APIs for attributing unit-level output changes when the mechanism does not change. Signed-off-by: Kailash <111277+kailashbuki@users.noreply.github.com>
null
2022-11-14 09:40:52+00:00
2022-11-14 21:51:34+00:00
dowhy/gcm/unit_change.py
"""This module provides the APIs for attributing the change in the output value of a deterministic mechanism for a statistical unit. """ from abc import abstractmethod from typing import List, Optional import numpy as np import pandas as pd from sklearn.linear_model._base import LinearModel from sklearn.utils.validat...
"""This module provides the APIs for attributing the change in the output value of a deterministic mechanism for a statistical unit. """ from abc import abstractmethod from typing import List, Optional import numpy as np import pandas as pd from sklearn.linear_model._base import LinearModel from sklearn.utils.validat...
kailashbuki
675cbc444cc422b007ba72fdf3303b328b123c48
ac047066326d0de3646f8461c0c5c4dd9826287c
Under a single method, a user still has to select an appropriate wrapper (`SklearnLinearRegressionModel` or `SklearnRegressionModel`) for sklearn models, and pass its object. At this point, just calling a more explicit API is better, no? It is also more transparent. A bit more theoretical argument: With `isinstance` pa...
kailashbuki
149
py-why/dowhy
762
Add APIs for attributing unit change without a mechanism change
- Added two new APIs for attributing unit-level output changes when the mechanism does not change. Signed-off-by: Kailash <111277+kailashbuki@users.noreply.github.com>
null
2022-11-14 09:40:52+00:00
2022-11-14 21:51:34+00:00
dowhy/gcm/unit_change.py
"""This module provides the APIs for attributing the change in the output value of a deterministic mechanism for a statistical unit. """ from abc import abstractmethod from typing import List, Optional import numpy as np import pandas as pd from sklearn.linear_model._base import LinearModel from sklearn.utils.validat...
"""This module provides the APIs for attributing the change in the output value of a deterministic mechanism for a statistical unit. """ from abc import abstractmethod from typing import List, Optional import numpy as np import pandas as pd from sklearn.linear_model._base import LinearModel from sklearn.utils.validat...
kailashbuki
675cbc444cc422b007ba72fdf3303b328b123c48
ac047066326d0de3646f8461c0c5c4dd9826287c
> Under a single method, a user still has to select an appropriate wrapper (SklearnLinearRegressionModel or SklearnRegressionModel) for sklearn models, and pass its object That's right, one need to check if the model ```has_coefficients``` actually, since this is what you need in the linear solution. I see the conce...
bloebp
150
py-why/dowhy
762
Add APIs for attributing unit change without a mechanism change
- Added two new APIs for attributing unit-level output changes when the mechanism does not change. Signed-off-by: Kailash <111277+kailashbuki@users.noreply.github.com>
null
2022-11-14 09:40:52+00:00
2022-11-14 21:51:34+00:00
dowhy/gcm/unit_change.py
"""This module provides the APIs for attributing the change in the output value of a deterministic mechanism for a statistical unit. """ from abc import abstractmethod from typing import List, Optional import numpy as np import pandas as pd from sklearn.linear_model._base import LinearModel from sklearn.utils.validat...
"""This module provides the APIs for attributing the change in the output value of a deterministic mechanism for a statistical unit. """ from abc import abstractmethod from typing import List, Optional import numpy as np import pandas as pd from sklearn.linear_model._base import LinearModel from sklearn.utils.validat...
kailashbuki
675cbc444cc422b007ba72fdf3303b328b123c48
ac047066326d0de3646f8461c0c5c4dd9826287c
Force pushed a single helper API for all those four APIs.
kailashbuki
151
py-why/dowhy
762
Add APIs for attributing unit change without a mechanism change
- Added two new APIs for attributing unit-level output changes when the mechanism does not change. Signed-off-by: Kailash <111277+kailashbuki@users.noreply.github.com>
null
2022-11-14 09:40:52+00:00
2022-11-14 21:51:34+00:00
tests/gcm/test_unit_change.py
import numpy as np import pandas as pd import pytest from flaky import flaky from sklearn.ensemble import RandomForestRegressor as RFR from sklearn.exceptions import NotFittedError from sklearn.linear_model import LinearRegression from dowhy.gcm.ml.regression import SklearnRegressionModel from dowhy.gcm.unit_change im...
import numpy as np import pandas as pd import pytest from flaky import flaky from sklearn.ensemble import RandomForestRegressor as RFR from sklearn.exceptions import NotFittedError from sklearn.linear_model import LinearRegression from dowhy.gcm.ml.regression import SklearnRegressionModel from dowhy.gcm.unit_change im...
kailashbuki
675cbc444cc422b007ba72fdf3303b328b123c48
ac047066326d0de3646f8461c0c5c4dd9826287c
It looks like you don't really need `to_numpy()` as `assert_array_almost_equal` does this for you.
petergtz
152
py-why/dowhy
762
Add APIs for attributing unit change without a mechanism change
- Added two new APIs for attributing unit-level output changes when the mechanism does not change. Signed-off-by: Kailash <111277+kailashbuki@users.noreply.github.com>
null
2022-11-14 09:40:52+00:00
2022-11-14 21:51:34+00:00
tests/gcm/test_unit_change.py
import numpy as np import pandas as pd import pytest from flaky import flaky from sklearn.ensemble import RandomForestRegressor as RFR from sklearn.exceptions import NotFittedError from sklearn.linear_model import LinearRegression from dowhy.gcm.ml.regression import SklearnRegressionModel from dowhy.gcm.unit_change im...
import numpy as np import pandas as pd import pytest from flaky import flaky from sklearn.ensemble import RandomForestRegressor as RFR from sklearn.exceptions import NotFittedError from sklearn.linear_model import LinearRegression from dowhy.gcm.ml.regression import SklearnRegressionModel from dowhy.gcm.unit_change im...
kailashbuki
675cbc444cc422b007ba72fdf3303b328b123c48
ac047066326d0de3646f8461c0c5c4dd9826287c
Something I know now. Thank you. Force pushing the fix.
kailashbuki
153
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/example_notebooks/DoWhy-The Causal Story Behind Hotel Booking Cancellations.ipynb
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Exploring Causes of Hotel Booking Cancellations" ] }, { "attachments": { "Screenshot%20from%202020-09-29%2019-08-50.png": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABcUAAAMoCAYAAAAOXYhzAAAAinpUWHRSYXcgcHJvZmlsZSB0eXBl...
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Exploring Causes of Hotel Booking Cancellations" ] }, { "attachments": { "Screenshot%20from%202020-09-29%2019-08-50.png": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABcUAAAMoCAYAAAAOXYhzAAAAinpUWHRSYXcgcHJvZmlsZSB0eXBl...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
Good thought on rewording the title here. To keep the focus on the application, how about "Exploring Causes of Hotel Booking Cancellations"
amit-sharma
154
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/example_notebooks/DoWhy-The Causal Story Behind Hotel Booking Cancellations.ipynb
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Exploring Causes of Hotel Booking Cancellations" ] }, { "attachments": { "Screenshot%20from%202020-09-29%2019-08-50.png": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABcUAAAMoCAYAAAAOXYhzAAAAinpUWHRSYXcgcHJvZmlsZSB0eXBl...
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Exploring Causes of Hotel Booking Cancellations" ] }, { "attachments": { "Screenshot%20from%202020-09-29%2019-08-50.png": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABcUAAAMoCAYAAAAOXYhzAAAAinpUWHRSYXcgcHJvZmlsZSB0eXBl...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
Done
petergtz
155
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/example_notebooks/dowhy_simple_example.ipynb
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Basic Example for Calculating the Causal Effect\n", "This is a quick introduction to the DoWhy causal inference library.\n", "We will load in a sample dataset and estimate the causal effect of a (pre-specified) treatment vari...
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Basic Example for Calculating the Causal Effect\n", "This is a quick introduction to the DoWhy causal inference library.\n", "We will load in a sample dataset and estimate the causal effect of a (pre-specified) treatment vari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
we could just say, "Causal Effect".
amit-sharma
156
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/example_notebooks/dowhy_simple_example.ipynb
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Basic Example for Calculating the Causal Effect\n", "This is a quick introduction to the DoWhy causal inference library.\n", "We will load in a sample dataset and estimate the causal effect of a (pre-specified) treatment vari...
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Basic Example for Calculating the Causal Effect\n", "This is a quick introduction to the DoWhy causal inference library.\n", "We will load in a sample dataset and estimate the causal effect of a (pre-specified) treatment vari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
Done
petergtz
157
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/example_notebooks/nb_index.rst
Example notebooks ================= These examples are also available on `GitHub <https://github .com/py-why/dowhy/tree/main/docs/source/example_notebooks>`_. You can `run them locally <https://docs.jupyter .org/en/latest/running.html>`_ after cloning `DoWhy <https://github.com/py-why/dowhy>`_ and `installing Jupyter ...
Example notebooks ================= These examples are also available on `GitHub <https://github .com/py-why/dowhy/tree/main/docs/source/example_notebooks>`_. You can `run them locally <https://docs.jupyter .org/en/latest/running.html>`_ after cloning `DoWhy <https://github.com/py-why/dowhy>`_ and `installing Jupyter ...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
nice thought to add the level. for the second bullet (API), given that the APIs will likely merge in the future, how about calling it a "Task"? We can think of effect estimation and Intervention via GCM as the two tasks here. Edit: seeing the later examples, "Root Cause Analysis" may also fit nicely as a "Task" tha...
amit-sharma
158
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/example_notebooks/nb_index.rst
Example notebooks ================= These examples are also available on `GitHub <https://github .com/py-why/dowhy/tree/main/docs/source/example_notebooks>`_. You can `run them locally <https://docs.jupyter .org/en/latest/running.html>`_ after cloning `DoWhy <https://github.com/py-why/dowhy>`_ and `installing Jupyter ...
Example notebooks ================= These examples are also available on `GitHub <https://github .com/py-why/dowhy/tree/main/docs/source/example_notebooks>`_. You can `run them locally <https://docs.jupyter .org/en/latest/running.html>`_ after cloning `DoWhy <https://github.com/py-why/dowhy>`_ and `installing Jupyter ...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
this one is "advanced"
amit-sharma
159
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/example_notebooks/nb_index.rst
Example notebooks ================= These examples are also available on `GitHub <https://github .com/py-why/dowhy/tree/main/docs/source/example_notebooks>`_. You can `run them locally <https://docs.jupyter .org/en/latest/running.html>`_ after cloning `DoWhy <https://github.com/py-why/dowhy>`_ and `installing Jupyter ...
Example notebooks ================= These examples are also available on `GitHub <https://github .com/py-why/dowhy/tree/main/docs/source/example_notebooks>`_. You can `run them locally <https://docs.jupyter .org/en/latest/running.html>`_ after cloning `DoWhy <https://github.com/py-why/dowhy>`_ and `installing Jupyter ...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
the conditional effects notebook (dowhy-conditional-treatment-effects) is a key one that introduces users to compute the conditional causal effect. Given that many people use doWhy to estimate the conditional effect, one option is to include it in the "introductory examples" section. We can mark it as Level: Beginner ...
amit-sharma
160
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/example_notebooks/nb_index.rst
Example notebooks ================= These examples are also available on `GitHub <https://github .com/py-why/dowhy/tree/main/docs/source/example_notebooks>`_. You can `run them locally <https://docs.jupyter .org/en/latest/running.html>`_ after cloning `DoWhy <https://github.com/py-why/dowhy>`_ and `installing Jupyter ...
Example notebooks ================= These examples are also available on `GitHub <https://github .com/py-why/dowhy/tree/main/docs/source/example_notebooks>`_. You can `run them locally <https://docs.jupyter .org/en/latest/running.html>`_ after cloning `DoWhy <https://github.com/py-why/dowhy>`_ and `installing Jupyter ...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
You're right. Originally, my thought was that there is the DoSampler API, GCM, and DoWhy's original "standard API", but we should make this forward-looking and we really want to focus on the tasks that users want to accomplish. 👍
petergtz
161
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/example_notebooks/nb_index.rst
Example notebooks ================= These examples are also available on `GitHub <https://github .com/py-why/dowhy/tree/main/docs/source/example_notebooks>`_. You can `run them locally <https://docs.jupyter .org/en/latest/running.html>`_ after cloning `DoWhy <https://github.com/py-why/dowhy>`_ and `installing Jupyter ...
Example notebooks ================= These examples are also available on `GitHub <https://github .com/py-why/dowhy/tree/main/docs/source/example_notebooks>`_. You can `run them locally <https://docs.jupyter .org/en/latest/running.html>`_ after cloning `DoWhy <https://github.com/py-why/dowhy>`_ and `installing Jupyter ...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
Added
petergtz
162
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
cool title 👍
amit-sharma
163
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
typo: model.
amit-sharma
164
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
we also use causal graphs for effect identification step. Since the two frameworks are mentioned above, I guess we can focus on an end-user's perspective here. Maybe rephrase it to, "For effect inference, DoWhy offers a 4-step recipe consisting of modeling a causal graph, identification, estimation, and refutation. ...
amit-sharma
165
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
haha, this might be too aggressive? How about, To understand what these four steps mean (and why we need four steps), the best place to learn more is the user's guide ... chapter. Alternatively, you can dive into the code and explore basic features in the doc notebook."
amit-sharma
166
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
given the new structure, should the further resources go into the user guide section where we talk about causality?
amit-sharma
167
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
at this point, it will also be good to highlight the fact that we interface with EconML for conditional effects, since people have found that integration useful. We can add a sentence like, "For estimation of conditional effects, you can also use methods from EconML using the same API, refer to this [notebook](dowhy-...
amit-sharma
168
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
This actually not a new formulation, but an existing one: - https://www.pywhy.org/dowhy/v0.8/user_guide/causality_intro.html?highlight=estimation - https://www.pywhy.org/dowhy/main/getting_started/index.html#next-steps I do get your point though. Alternative suggestion: Kailash and I have already started working r...
petergtz
169
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
Yes, maybe. Have also already thought about the best place for this section. If it's okay, I would also like to post-pone this change until we have reworked the User Guide. Hopefully, that will give us a better understanding where this will fit.
petergtz
170
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
Don e
petergtz
171
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
Sounds good. We can postpone this.
amit-sharma
172
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/getting_started/index.rst
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
Getting Started =============== Installation ^^^^^^^^^^^^ The simplest installation is through `pip <https://pypi.org/project/dowhy/>`__ or conda: .. tab-set-code:: .. code-block:: pip pip install dowhy .. code-block:: conda conda install -c conda-forge dowhy Further installation scenari...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
Sounds good!
amit-sharma
173
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/index.rst
DoWhy documentation =================== .. toctree:: :maxdepth: 3 :hidden: :glob: getting_started/index User Guide <user_guide/index> Examples <example_notebooks/nb_index> dowhy Contributing <contributing> code_repo **Date**: |today| **Version**: |version| **Related resources**:...
DoWhy documentation =================== .. toctree:: :maxdepth: 3 :hidden: :glob: getting_started/index User Guide <user_guide/index> Examples <example_notebooks/nb_index> dowhy Contributing <contributing> code_repo **Date**: |today| **Version**: |version| **Related resources**:...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
love this new sentence +1
amit-sharma
174
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/index.rst
DoWhy documentation =================== .. toctree:: :maxdepth: 3 :hidden: :glob: getting_started/index User Guide <user_guide/index> Examples <example_notebooks/nb_index> dowhy Contributing <contributing> code_repo **Date**: |today| **Version**: |version| **Related resources**:...
DoWhy documentation =================== .. toctree:: :maxdepth: 3 :hidden: :glob: getting_started/index User Guide <user_guide/index> Examples <example_notebooks/nb_index> dowhy Contributing <contributing> code_repo **Date**: |today| **Version**: |version| **Related resources**:...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
Some of these are action links (check the source, raise an issue), so it may be worth putting them first. Source Repo | Issues and Ideas | Join the community (Discord) | PyWhy organization | DoWhy on PyPI
amit-sharma
175
py-why/dowhy
760
Restructure documentation
The result of this can be seen at https://petergtz.github.io/dowhy/main - Add logo to docs and README - Move 4-step process documentation from docs landing page to effect inference user guide - Provide documentation overview on docs starting page. - Move citation page from docs landing page to user guide - Intro...
null
2022-11-11 14:23:31+00:00
2022-11-17 13:13:21+00:00
docs/source/index.rst
DoWhy documentation =================== .. toctree:: :maxdepth: 3 :hidden: :glob: getting_started/index User Guide <user_guide/index> Examples <example_notebooks/nb_index> dowhy Contributing <contributing> code_repo **Date**: |today| **Version**: |version| **Related resources**:...
DoWhy documentation =================== .. toctree:: :maxdepth: 3 :hidden: :glob: getting_started/index User Guide <user_guide/index> Examples <example_notebooks/nb_index> dowhy Contributing <contributing> code_repo **Date**: |today| **Version**: |version| **Related resources**:...
petergtz
30c102e358258b103fec8cd3ce42de9cf2051a50
de8969a5c100e5186f7892f16073129b6d0cffc6
Done
petergtz
176
py-why/dowhy
746
Functional api/causal estimators
* Introduce `fit()` method to estimators. * Refactor constructors to avoid using `*args` and `**kwargs` and have more explicit parameters. * Refactor refuters and other parts of the code to use `fit()` and modify arguments to `estimate_effect()`
null
2022-11-04 16:15:39+00:00
2022-12-03 17:07:53+00:00
docs/source/example_notebooks/dowhy-conditional-treatment-effects.ipynb
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Conditional Average Treatment Effects (CATE) with DoWhy and EconML\n", "\n", "This is an experimental feature where we use [EconML](https://github.com/microsoft/econml) methods from DoWhy. Using EconML allows CATE estimation ...
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Conditional Average Treatment Effects (CATE) with DoWhy and EconML\n", "\n", "This is an experimental feature where we use [EconML](https://github.com/microsoft/econml) methods from DoWhy. Using EconML allows CATE estimation ...
andresmor-ms
11c4e0dafd6e824eb81ad14262457d954ae61468
affe0952f4aba6845247355c171565510c2c1673
obtaining a build error. can you change the import of deepiv too? "from econml.iv.nnet import DeepIV" CellExecutionError in example_notebooks/dowhy-conditional-treatment-effects.ipynb: [147](https://github.com/py-why/dowhy/actions/runs/3568922214/jobs/5998324083#step:6:148)------------------ [148](https://gith...
amit-sharma
177
py-why/dowhy
746
Functional api/causal estimators
* Introduce `fit()` method to estimators. * Refactor constructors to avoid using `*args` and `**kwargs` and have more explicit parameters. * Refactor refuters and other parts of the code to use `fit()` and modify arguments to `estimate_effect()`
null
2022-11-04 16:15:39+00:00
2022-12-03 17:07:53+00:00
docs/source/example_notebooks/dowhy-conditional-treatment-effects.ipynb
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Conditional Average Treatment Effects (CATE) with DoWhy and EconML\n", "\n", "This is an experimental feature where we use [EconML](https://github.com/microsoft/econml) methods from DoWhy. Using EconML allows CATE estimation ...
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Conditional Average Treatment Effects (CATE) with DoWhy and EconML\n", "\n", "This is an experimental feature where we use [EconML](https://github.com/microsoft/econml) methods from DoWhy. Using EconML allows CATE estimation ...
andresmor-ms
11c4e0dafd6e824eb81ad14262457d954ae61468
affe0952f4aba6845247355c171565510c2c1673
Changed import.
andresmor-ms
178
py-why/dowhy
746
Functional api/causal estimators
* Introduce `fit()` method to estimators. * Refactor constructors to avoid using `*args` and `**kwargs` and have more explicit parameters. * Refactor refuters and other parts of the code to use `fit()` and modify arguments to `estimate_effect()`
null
2022-11-04 16:15:39+00:00
2022-12-03 17:07:53+00:00
docs/source/example_notebooks/dowhy-conditional-treatment-effects.ipynb
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Conditional Average Treatment Effects (CATE) with DoWhy and EconML\n", "\n", "This is an experimental feature where we use [EconML](https://github.com/microsoft/econml) methods from DoWhy. Using EconML allows CATE estimation ...
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Conditional Average Treatment Effects (CATE) with DoWhy and EconML\n", "\n", "This is an experimental feature where we use [EconML](https://github.com/microsoft/econml) methods from DoWhy. Using EconML allows CATE estimation ...
andresmor-ms
11c4e0dafd6e824eb81ad14262457d954ae61468
affe0952f4aba6845247355c171565510c2c1673
This is not needed, only need to specify the correct method_name parameter
andresmor-ms
179
py-why/dowhy
746
Functional api/causal estimators
* Introduce `fit()` method to estimators. * Refactor constructors to avoid using `*args` and `**kwargs` and have more explicit parameters. * Refactor refuters and other parts of the code to use `fit()` and modify arguments to `estimate_effect()`
null
2022-11-04 16:15:39+00:00
2022-12-03 17:07:53+00:00
docs/source/example_notebooks/dowhy_functional_api.ipynb
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Functional API Preview\n", "\n", "This notebook is part of a set of notebooks that provides a preview of the proposed functional API for dowhy. For details on the new API for DoWhy, check out https://github.com/py-why/dowhy/w...
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Functional API Preview\n", "\n", "This notebook is part of a set of notebooks that provides a preview of the proposed functional API for dowhy. For details on the new API for DoWhy, check out https://github.com/py-why/dowhy/w...
andresmor-ms
11c4e0dafd6e824eb81ad14262457d954ae61468
affe0952f4aba6845247355c171565510c2c1673
can you also add an example where estimate_effect is called again for a different (test) dataset?
amit-sharma
180
py-why/dowhy
746
Functional api/causal estimators
* Introduce `fit()` method to estimators. * Refactor constructors to avoid using `*args` and `**kwargs` and have more explicit parameters. * Refactor refuters and other parts of the code to use `fit()` and modify arguments to `estimate_effect()`
null
2022-11-04 16:15:39+00:00
2022-12-03 17:07:53+00:00
docs/source/example_notebooks/dowhy_functional_api.ipynb
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Functional API Preview\n", "\n", "This notebook is part of a set of notebooks that provides a preview of the proposed functional API for dowhy. For details on the new API for DoWhy, check out https://github.com/py-why/dowhy/w...
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Functional API Preview\n", "\n", "This notebook is part of a set of notebooks that provides a preview of the proposed functional API for dowhy. For details on the new API for DoWhy, check out https://github.com/py-why/dowhy/w...
andresmor-ms
11c4e0dafd6e824eb81ad14262457d954ae61468
affe0952f4aba6845247355c171565510c2c1673
You mean to call `estimate_effect()` without fitting and providing a new data? I'm not sure how that would work, would it be replacing the self._data without the other changes in the `fit()`? Let me know how you expect it to work to be sure I implement it the correct way.
andresmor-ms
181
py-why/dowhy
746
Functional api/causal estimators
* Introduce `fit()` method to estimators. * Refactor constructors to avoid using `*args` and `**kwargs` and have more explicit parameters. * Refactor refuters and other parts of the code to use `fit()` and modify arguments to `estimate_effect()`
null
2022-11-04 16:15:39+00:00
2022-12-03 17:07:53+00:00
docs/source/example_notebooks/dowhy_functional_api.ipynb
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Functional API Preview\n", "\n", "This notebook is part of a set of notebooks that provides a preview of the proposed functional API for dowhy. For details on the new API for DoWhy, check out https://github.com/py-why/dowhy/w...
{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Functional API Preview\n", "\n", "This notebook is part of a set of notebooks that provides a preview of the proposed functional API for dowhy. For details on the new API for DoWhy, check out https://github.com/py-why/dowhy/w...
andresmor-ms
11c4e0dafd6e824eb81ad14262457d954ae61468
affe0952f4aba6845247355c171565510c2c1673
no, I meant that you fit an estimator once, and then call estimate_effect twice for two different datasets. Just to show that we can fit once, and then call estimate_effect multiple times.
amit-sharma
182
py-why/dowhy
746
Functional api/causal estimators
* Introduce `fit()` method to estimators. * Refactor constructors to avoid using `*args` and `**kwargs` and have more explicit parameters. * Refactor refuters and other parts of the code to use `fit()` and modify arguments to `estimate_effect()`
null
2022-11-04 16:15:39+00:00
2022-12-03 17:07:53+00:00
docs/source/example_notebooks/sensitivity_analysis_nonparametric_estimators.ipynb
{ "cells": [ { "cell_type": "markdown", "id": "0bbaacaa", "metadata": {}, "source": [ "# Sensitivity analysis for non-parametric causal estimators\n", "Sensitivity analysis helps us study how robust an estimated effect is when the assumption of no unobserved confounding is violated. That is, how ...
{ "cells": [ { "cell_type": "markdown", "id": "0bbaacaa", "metadata": {}, "source": [ "# Sensitivity analysis for non-parametric causal estimators\n", "Sensitivity analysis helps us study how robust an estimated effect is when the assumption of no unobserved confounding is violated. That is, how ...
andresmor-ms
11c4e0dafd6e824eb81ad14262457d954ae61468
affe0952f4aba6845247355c171565510c2c1673
if there's an easy to remove the vscode hash from every notebook commit, that will be nice. I suspect this hash will change with every commit. If not, it is fairly minor and we can ignore.
amit-sharma
183