mini-causal 0.1.3__py3-none-any.whl

mini_causal/__init__.py

@@ -0,0 +1 @@
+

mini_causal/model_selection.py

@@ -0,0 +1 @@
+from mini_causal.utils import treatment_control_selection

mini_causal/utils.py

@@ -0,0 +1,157 @@
+import numpy as np
+import pandas as pd
+from sklearn import clone
+
+def clone_model(model):
+    return clone(model)
+
+
+def treatment_control_selection(df, control_size=0.2, random_seed=None):
+    """
+    Randomly split dataframe into treatment and control groups.
+
+    Args:
+        df (pd.DataFrame):
+            the input dataframe
+
+        control_size (float):
+            Fraction assigned to control group
+
+        random_seed(int or None):
+            for reproducibilit
+
+    Returns:
+        treatment,control
+    """
+
+    if random_seed is not None:
+        np.random.seed(random_seed)
+
+    n_samples = len(df)
+
+    # Shuffle row indices
+    indices = np.random.permutation(n_samples)
+
+    # Split point
+    split_idx = int(n_samples * (1 - control_size))
+
+    # Indices
+    treat_idx = indices[:split_idx]
+    control_idx = indices[split_idx:]
+
+
+    treatment = df.iloc[treat_idx]
+    control = df.iloc[control_idx]
+
+    return treatment,control
+
+
+
+def make_treatment_control_df(treatment,control)->pd.DataFrame:
+    df=pd.concat([treatment,control])
+
+    return df.sort_index()
+
+
+def summary_stats(treatment:np.ndarray,control:np.ndarray)->pd.DataFrame:
+    n_treatment=len(treatment)
+    n_control=len(control)
+
+    mean_control_preds=float(np.mean(control))
+    mean_treatment_preds=float(np.mean(treatment))
+
+    variance_treatment=float(np.var(treatment,ddof=1))
+    variance_control=float(np.var(control,ddof=1))
+
+    return pd.DataFrame(
+        {
+            "type":["treatment","control"],
+            "n":[n_treatment,n_control],
+            "mean":[mean_treatment_preds,mean_control_preds],
+            "sample_variance":[variance_treatment,variance_control]
+        })
+    
+
+
+def causal_effects_stratum(X:pd.DataFrame,causal_effects:np.ndarray,strata_feature:str):
+    """
+    Args:
+        X (pd.DataFrame):
+            X values stored in a dataframe that will be used for the prediction.
+
+        causal_effects(np.ndarray):
+            the causal effects values
+        
+        strata_feature(str):
+           the name of the feature that will be used for strata
+
+    Returns:
+        pd.DataFrame:
+           - a dataframe with the following columns: strata_feature,mean,size and weight of the strata
+           -  the mean is based on the type arg i.e if preds ,then it will be the mean for the 
+           predictions, else if probs then it will be the mean for the predicted probabilities
+    """
+    N=len(X)
+    X["causal_effects"]=causal_effects
+        
+    results=X.groupby(strata_feature)["causal_effects"].agg(["mean","size"]).reset_index()
+    results["weight_rss"]= results["size"] * N
+
+    return results
+ 
+def stratum_weight_rss(X:pd.DataFrame,
+                       treatment_model,
+                       control_model,
+                       feature:str,
+                       strata_feature:str,
+                       type:str="preds")->pd.DataFrame:
+    """
+    The causal effect based on the stratum to validate the effect per stratum of the treatment.
+
+    Args:
+        X (pd.DataFrame):
+            X values stored in a dataframe that will be used for the prediction.
+
+        strata_feature (str):
+            the name of the feature that will be used for strata
+
+        treatment_model:
+            the trained treatment estimator
+
+        control_model:
+            the trained control estimator
+
+        type (str): {preds,probs}
+            return the stratum values based on the predicted probabilites or predicted values of the target
+
+
+    Returns:
+        pd.DataFrame:
+           - a dataframe with the following columns: strata_feature,mean,size and weight of the strata
+           -  the mean is based on the type arg i.e if preds ,then it will be the mean for the 
+           predictions, else if probs then it will be the mean for the predicted probabilities
+
+    """
+    X_control=X.drop([feature,strata_feature],axis=1)
+    X_treatment=X.drop(strata_feature,axis=1)
+
+    treatment_probs=treatment_model.predict_proba(X_treatment)[:,1]
+    control_probs=control_model.predict_proba(X_control)[:,1]
+
+    treatment_preds=treatment_model.predict(X_treatment)
+    control_preds=control_model.predict(X_control)
+    
+    causal_effects_probs=treatment_probs-control_probs
+    causal_effects=treatment_preds-control_preds
+
+    if type=="preds":
+        return causal_effects_stratum(X=X,causal_effects=causal_effects_probs,
+                                      strata_feature=strata_feature)
+    
+    elif type=="probs":
+        return causal_effects_stratum(X=X,causal_effects=causal_effects,
+                                      strata_feature=strata_feature)
+    
+    else:
+        raise ValueError("Type invalid")
+    

mini_causal-0.1.3.dist-info/METADATA

@@ -0,0 +1,21 @@
+Metadata-Version: 2.4
+Name: mini-causal
+Version: 0.1.3
+Summary: A mini package for causal inference
+Author-email: Masemene Matlakana Benny <bennymasemene46@gmail.com>
+Project-URL: Homepage, https://github.com/MasemeneMatlakanaBenny/Mini_Causal
+Project-URL: Issues, https://github.com/MasemeneMatlakana/Mini_Causal/issues
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.12
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: pandas>=3.0.2
+Requires-Dist: numpy>=2.4.4
+Requires-Dist: scipy>=1.17.1
+Dynamic: license-file
+
+# MiniCausal
+
+![logo](images/cropped_logo.jpeg)

mini_causal-0.1.3.dist-info/RECORD

@@ -0,0 +1,8 @@
+mini_causal/__init__.py,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
+mini_causal/model_selection.py,sha256=MHdg8SmFI_GUbUaawktnZNv86lFzyC_Z4DKQi8gCYTU,58
+mini_causal/utils.py,sha256=iObTWTvvsj_Ak8gNFrZ7UCpdWPwhBudL9BOSBlf42Xs,4686
+mini_causal-0.1.3.dist-info/licenses/LICENSE,sha256=inTwgDJnDsOeOMXHNRzKEsjfs9A3l2Ib70mJgz9RmhE,1079
+mini_causal-0.1.3.dist-info/METADATA,sha256=ml6-QQh1rm63frHwTfXwghMrfaq4FrHPjWWnYwDEDaU,710
+mini_causal-0.1.3.dist-info/WHEEL,sha256=aeYiig01lYGDzBgS8HxWXOg3uV61G9ijOsup-k9o1sk,91
+mini_causal-0.1.3.dist-info/top_level.txt,sha256=z_s2J0gL5M0LfBt99Z-i3uJaw08pbJd0zkhRdV-HNak,12
+mini_causal-0.1.3.dist-info/RECORD,,

mini_causal-0.1.3.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (82.0.1)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

mini_causal-0.1.3.dist-info/licenses/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 MasemeneMatlakanaBenny
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

mini_causal-0.1.3.dist-info/top_level.txt

@@ -0,0 +1 @@
+mini_causal