pySEQTarget 0.10.0__tar.gz → 0.10.1__tar.gz

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pySEQTarget
-Version: 0.10.0
+Version: 0.10.1
 Summary: Sequentially Nested Target Trial Emulation
 Author-email: Ryan O'Dea <ryan.odea@psi.ch>, Alejandro Szmulewicz <aszmulewicz@hsph.harvard.edu>, Tom Palmer <tom.palmer@bristol.ac.uk>, Miguel Hernan <mhernan@hsph.harvard.edu>
 Maintainer-email: Ryan O'Dea <ryan.odea@psi.ch>
@@ -68,8 +68,9 @@ From the user side, this amounts to creating a dataclass, `SEQopts`, and then fe
 ```python
 import polars as pl
 from pySEQTarget import SEQuential, SEQopts
+from pySEQTarget.data import load_data
 
-data = pl.from_pandas(SEQdata)
+data = load_data("SEQdata")
 options = SEQopts(km_curves = True)
 
 # Initiate the class
@@ -77,17 +78,18 @@ model = SEQuential(data,
                    id_col = "ID",
                    time_col = "time",
                    eligible_col = "eligible",
+                   treatment_col = "tx_init",
+                   outcome_col = "outcome",
                    time_varying_cols = ["N", "L", "P"],
                    fixed_cols = ["sex"],
                    method = "ITT",
-                   options = options)
+                   parameters = options)
 model.expand()  # Construct the nested structure
 model.bootstrap(bootstrap_nboot = 20) # Run 20 bootstrap samples
 model.fit() # Fit the model
 model.survival() # Create survival curves
 model.plot() # Create and show a plot of the survival curves
 model.collect() # Collection of important information
-
 ```
 
 ## Assumptions
@@ -95,4 +97,3 @@ There are several key assumptions in this package -
 1. User provided `time_col` begins at 0 per unique `id_col`, we also assume this column contains only integers and continues by 1 for every time step, e.g. (0, 1, 2, 3, 4, ...) is allowed and (0, 1, 2, 2.5, ...) or (0, 1, 4, 5) are not
     1. Provided `time_col` entries may be out of order at intake as a sort is enforced at expansion.
 2. `eligible_col` and elements of `excused_colnames` are once 1, only 1 (with respect to `time_col`) flag variables.
-

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/README.md

@@ -31,8 +31,9 @@ From the user side, this amounts to creating a dataclass, `SEQopts`, and then fe
 ```python
 import polars as pl
 from pySEQTarget import SEQuential, SEQopts
+from pySEQTarget.data import load_data
 
-data = pl.from_pandas(SEQdata)
+data = load_data("SEQdata")
 options = SEQopts(km_curves = True)
 
 # Initiate the class
@@ -40,17 +41,18 @@ model = SEQuential(data,
                    id_col = "ID",
                    time_col = "time",
                    eligible_col = "eligible",
+                   treatment_col = "tx_init",
+                   outcome_col = "outcome",
                    time_varying_cols = ["N", "L", "P"],
                    fixed_cols = ["sex"],
                    method = "ITT",
-                   options = options)
+                   parameters = options)
 model.expand()  # Construct the nested structure
 model.bootstrap(bootstrap_nboot = 20) # Run 20 bootstrap samples
 model.fit() # Fit the model
 model.survival() # Create survival curves
 model.plot() # Create and show a plot of the survival curves
 model.collect() # Collection of important information
-
 ```
 
 ## Assumptions
@@ -58,4 +60,3 @@ There are several key assumptions in this package -
 1. User provided `time_col` begins at 0 per unique `id_col`, we also assume this column contains only integers and continues by 1 for every time step, e.g. (0, 1, 2, 3, 4, ...) is allowed and (0, 1, 2, 2.5, ...) or (0, 1, 4, 5) are not
     1. Provided `time_col` entries may be out of order at intake as a sort is enforced at expansion.
 2. `eligible_col` and elements of `excused_colnames` are once 1, only 1 (with respect to `time_col`) flag variables.
-

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/pySEQTarget/SEQuential.py

@@ -7,9 +7,10 @@ from typing import List, Literal, Optional
 import numpy as np
 import polars as pl
 
-from .analysis import (_calculate_hazard, _calculate_survival, _outcome_fit,
-                       _pred_risk, _risk_estimates, _subgroup_fit)
-from .error import _datachecker, _param_checker
+from .analysis import (_calculate_hazard, _calculate_survival, _clamp,
+                       _outcome_fit, _pred_risk, _risk_estimates,
+                       _subgroup_fit)
+from .error import _data_checker, _param_checker
 from .expansion import _binder, _diagnostics, _dynamic, _random_selection
 from .helpers import _col_string, _format_time, bootstrap_loop
 from .initialization import (_cense_denominator, _cense_numerator,
@@ -101,7 +102,7 @@ class SEQuential:
                     self.cense_denominator = _cense_denominator(self)
 
         _param_checker(self)
-        _datachecker(self)
+        _data_checker(self)
 
     def expand(self) -> None:
         """
@@ -190,7 +191,6 @@ class SEQuential:
             )
             id_counts = Counter(sampled_IDs)
             self._boot_samples.append(id_counts)
-        return self
 
     @bootstrap_loop
     def fit(self) -> None:
@@ -266,7 +266,7 @@ class SEQuential:
 
         risk_data = _pred_risk(self)
         surv_data = _calculate_survival(self, risk_data)
-        self.km_data = pl.concat([risk_data, surv_data])
+        self.km_data = _clamp(pl.concat([risk_data, surv_data]))
         self.risk_estimates = _risk_estimates(self)
 
         end = time.perf_counter()

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/pySEQTarget/analysis/__init__.py

@@ -3,6 +3,7 @@ from ._outcome_fit import _outcome_fit as _outcome_fit
 from ._risk_estimates import _risk_estimates as _risk_estimates
 from ._subgroup_fit import _subgroup_fit as _subgroup_fit
 from ._survival_pred import _calculate_survival as _calculate_survival
+from ._survival_pred import _clamp as _clamp
 from ._survival_pred import \
     _get_outcome_predictions as _get_outcome_predictions
 from ._survival_pred import _pred_risk as _pred_risk

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/pySEQTarget/analysis/_survival_pred.py

@@ -370,3 +370,11 @@ def _calculate_survival(self, risk_data):
             [(1 - pl.col("pred")).alias("pred"), pl.lit("survival").alias("estimate")]
         )
     return surv
+
+
+def _clamp(data):
+    """Clamp prediction and CI columns to [0, 1] bounds."""
+    cols = ["pred", "LCI", "UCI"]
+    exists = [c for c in cols if c in data.columns]
+
+    return data.with_columns([pl.col(col).clip(0.0, 1.0) for col in exists])

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/pySEQTarget/error/__init__.py

@@ -1,2 +1,2 @@
-from ._datachecker import _datachecker as _datachecker
+from ._data_checker import _data_checker as _data_checker
 from ._param_checker import _param_checker as _param_checker

pyseqtarget-0.10.0/pySEQTarget/error/_datachecker.py → pyseqtarget-0.10.1/pySEQTarget/error/_data_checker.py

@@ -1,7 +1,7 @@
 import polars as pl
 
 
-def _datachecker(self):
+def _data_checker(self):
     check = self.data.group_by(self.id_col).agg(
         [pl.len().alias("row_count"), pl.col(self.time_col).max().alias("max_time")]
     )

pyseqtarget-0.10.1/pySEQTarget/expansion/_selection.py

@@ -0,0 +1,44 @@
+import polars as pl
+
+
+def _random_selection(self):
+    """
+    Handles the case where random selection is applied for data from
+    the __mapper -> __binder -> optionally __dynamic pipeline
+    """
+    UIDs = (
+        self.DT.select(
+            [self.id_col, "trial", f"{self.treatment_col}{self.indicator_baseline}"]
+        )
+        .with_columns(
+            (
+                pl.col(self.id_col).cast(pl.Utf8) + "_" + pl.col("trial").cast(pl.Utf8)
+            ).alias("trialID")
+        )
+        .filter(
+            pl.col(f"{self.treatment_col}{self.indicator_baseline}")
+            == self.treatment_level[0]
+        )
+        .unique("trialID")
+        .get_column("trialID")
+        .to_list()
+    )
+
+    NIDs = len(UIDs)
+    sample = self._rng.choice(
+        UIDs, size=int(self.selection_sample * NIDs), replace=False
+    )
+
+    self.DT = (
+        self.DT.with_columns(
+            (
+                pl.col(self.id_col).cast(pl.Utf8) + "_" + pl.col("trial").cast(pl.Utf8)
+            ).alias("trialID")
+        )
+        .filter(
+            pl.col("trialID").is_in(sample)
+            | pl.col(f"{self.treatment_col}{self.indicator_baseline}")
+            != self.treatment_level[0]
+        )
+        .drop("trialID")
+    )

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/pySEQTarget/helpers/_bootstrap.py

@@ -58,13 +58,14 @@ def bootstrap_loop(method):
         start = time.perf_counter()
 
         results = []
+        original_DT = self.DT
+
         full = method(self, *args, **kwargs)
         results.append(full)
 
         if getattr(self, "bootstrap_nboot") > 0 and getattr(
             self, "_boot_samples", None
         ):
-            original_DT = self.DT
             nboot = self.bootstrap_nboot
             ncores = self.ncores
             seed = getattr(self, "seed", None)

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/pySEQTarget/weighting/_weight_fit.py

@@ -26,7 +26,7 @@ def _fit_LTFU(self, WDT):
         WDT,
         "cense_colname",
         [self.cense_numerator, self.cense_denominator],
-        ["cense_numerator", "cense_denominator"],
+        ["cense_numerator_model", "cense_denominator_model"],
         "cense_eligible_colname",
     )
 
@@ -39,7 +39,7 @@ def _fit_visit(self, WDT):
         WDT,
         "visit_colname",
         [self.cense_numerator, self.cense_denominator],
-        ["visit_numerator", "visit_denominator"],
+        ["visit_numerator_model", "visit_denominator_model"],
     )
 
 

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/pySEQTarget/weighting/_weight_pred.py

@@ -150,8 +150,8 @@ def _weight_predict(self, WDT):
                     .alias("numerator")
                 )
     if self.cense_colname is not None:
-        p_num = _predict_model(self, self.cense_numerator, WDT).flatten()
-        p_denom = _predict_model(self, self.cense_denominator, WDT).flatten()
+        p_num = _predict_model(self, self.cense_numerator_model, WDT).flatten()
+        p_denom = _predict_model(self, self.cense_denominator_model, WDT).flatten()
         WDT = WDT.with_columns(
             [
                 pl.Series("cense_numerator", p_num),
@@ -164,8 +164,8 @@ def _weight_predict(self, WDT):
         WDT = WDT.with_columns(pl.lit(1.0).alias("_cense"))
 
     if self.visit_colname is not None:
-        p_num = _predict_model(self, self.visit_numerator, WDT).flatten()
-        p_denom = _predict_model(self, self.visit_denominator, WDT).flatten()
+        p_num = _predict_model(self, self.visit_numerator_model, WDT).flatten()
+        p_denom = _predict_model(self, self.visit_denominator_model, WDT).flatten()
 
         WDT = WDT.with_columns(
             [

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/pySEQTarget.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pySEQTarget
-Version: 0.10.0
+Version: 0.10.1
 Summary: Sequentially Nested Target Trial Emulation
 Author-email: Ryan O'Dea <ryan.odea@psi.ch>, Alejandro Szmulewicz <aszmulewicz@hsph.harvard.edu>, Tom Palmer <tom.palmer@bristol.ac.uk>, Miguel Hernan <mhernan@hsph.harvard.edu>
 Maintainer-email: Ryan O'Dea <ryan.odea@psi.ch>
@@ -68,8 +68,9 @@ From the user side, this amounts to creating a dataclass, `SEQopts`, and then fe
 ```python
 import polars as pl
 from pySEQTarget import SEQuential, SEQopts
+from pySEQTarget.data import load_data
 
-data = pl.from_pandas(SEQdata)
+data = load_data("SEQdata")
 options = SEQopts(km_curves = True)
 
 # Initiate the class
@@ -77,17 +78,18 @@ model = SEQuential(data,
                    id_col = "ID",
                    time_col = "time",
                    eligible_col = "eligible",
+                   treatment_col = "tx_init",
+                   outcome_col = "outcome",
                    time_varying_cols = ["N", "L", "P"],
                    fixed_cols = ["sex"],
                    method = "ITT",
-                   options = options)
+                   parameters = options)
 model.expand()  # Construct the nested structure
 model.bootstrap(bootstrap_nboot = 20) # Run 20 bootstrap samples
 model.fit() # Fit the model
 model.survival() # Create survival curves
 model.plot() # Create and show a plot of the survival curves
 model.collect() # Collection of important information
-
 ```
 
 ## Assumptions
@@ -95,4 +97,3 @@ There are several key assumptions in this package -
 1. User provided `time_col` begins at 0 per unique `id_col`, we also assume this column contains only integers and continues by 1 for every time step, e.g. (0, 1, 2, 3, 4, ...) is allowed and (0, 1, 2, 2.5, ...) or (0, 1, 4, 5) are not
     1. Provided `time_col` entries may be out of order at intake as a sort is enforced at expansion.
 2. `eligible_col` and elements of `excused_colnames` are once 1, only 1 (with respect to `time_col`) flag variables.
-

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/pySEQTarget.egg-info/SOURCES.txt

@@ -18,7 +18,7 @@ pySEQTarget/analysis/_subgroup_fit.py
 pySEQTarget/analysis/_survival_pred.py
 pySEQTarget/data/__init__.py
 pySEQTarget/error/__init__.py
-pySEQTarget/error/_datachecker.py
+pySEQTarget/error/_data_checker.py
 pySEQTarget/error/_param_checker.py
 pySEQTarget/expansion/__init__.py
 pySEQTarget/expansion/_binder.py

{pyseqtarget-0.10.0 → pyseqtarget-0.10.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "pySEQTarget"
-version = "0.10.0"
+version = "0.10.1"
 description = "Sequentially Nested Target Trial Emulation"
 readme = "README.md"
 license = {text = "MIT"}

pyseqtarget-0.10.0/pySEQTarget/expansion/_selection.py

@@ -1,31 +0,0 @@
-import polars as pl
-
-
-def _random_selection(self):
-    """
-    Handles the case where random selection is applied for data from
-    the __mapper -> __binder -> optionally __dynamic pipeline
-    """
-    UIDs = (
-        self.DT.select(
-            [self.id_col, "trial", f"{self.treatment_col}{self.indicator_baseline}"]
-        )
-        .with_columns((pl.col(self.id_col) + "_" + pl.col("trial")).alias("trialID"))
-        .filter(pl.col(f"{self.treatment_col}{self.indicator_baseline}") == 0)
-        .unique("trialID")
-        .to_series()
-        .to_list()
-    )
-
-    NIDs = len(UIDs)
-    sample = self._rng.choice(
-        UIDs, size=int(self.selection_sample * NIDs), replace=False
-    )
-
-    self.DT = (
-        self.DT.with_columns(
-            (pl.col(self.id_col) + "_" + pl.col("trial")).alias("trialID")
-        )
-        .filter(pl.col("trialID").is_in(sample))
-        .drop("trialID")
-    )