update doc

docs/api/data/pyhealth.data.Patient.rst

@@ -1,7 +1,9 @@
 pyhealth.data.Patient
 =========================
 
-Another basic data structure in the package. A Patient is a collection of Visit for the current patients. It contains all necessary attributes of a patient, such as ethnicity, mortality status, gender, etc. It can support various healthcare tasks.
+Another basic data structure in the package. A Patient is a collection of Visit for
+the current patients. It contains all necessary attributes of a patient, such as
+ethnicity, mortality status, gender, etc. It can support various healthcare tasks.
 
 .. autoclass:: pyhealth.data.Patient
  :members:

docs/api/data/pyhealth.data.Visit.rst

@@ -1,7 +1,10 @@
 pyhealth.data.Visit
 =========================
 
-Another basic data structure in the package. A Visit is a single encounter in hispital. It is a container a sequence of Event for each information aspect, such as diagnosis or medications. It also contains other necessary attributes for supporting healthcare tasks, such as the date of the visit.
+Another basic data structure in the package. A Visit is a single encounter in
+hospital. It is a container a sequence of Event for each information aspect,
+such as diagnosis or medications. It also contains other necessary attributes
+for supporting healthcare tasks, such as the date of the visit.
 
 .. autoclass:: pyhealth.data.Visit
  :members:

docs/api/datasets.rst

@@ -10,4 +10,5 @@ Datasets
  datasets/pyhealth.datasets.eICUDataset
  datasets/pyhealth.datasets.OMOPDataset
  datasets/pyhealth.datasets.splitter
+ datasets/pyhealth.datasets.utils
 

docs/api/datasets/pyhealth.datasets.utils.rst

@@ -0,0 +1,13 @@
+pyhealth.datasets.splitter
+===================================
+
+Several utility functions.
+
+.. automodule:: pyhealth.datasets.utils
+ :members:
+ :undoc-members:
+ :show-inheritance:
+
+
+
+

docs/api/metrics.rst

@@ -10,4 +10,4 @@ tasks, such as drug drug interaction (DDI) rate.
 
  metrics/pyhealth.metrics.multiclass
  metrics/pyhealth.metrics.multilabel
- metrics/pyhealth.metrics.multiclass_avg_patient
+ metrics/pyhealth.metrics.binary

docs/api/metrics/pyhealth.metrics.binary.rst

@@ -0,0 +1,6 @@
+pyhealth.metrics.multiclass
+===================================
+
+.. currentmodule:: pyhealth.metrics.binary
+
+.. autofunction:: binary_metrics_fn

docs/api/metrics/pyhealth.metrics.multiclass.rst

@@ -3,13 +3,5 @@
 
 .. currentmodule:: pyhealth.metrics.multiclass
 
-.. autofunction:: accuracy_score
-.. autofunction:: precision_score
-.. autofunction:: recall_score
-.. autofunction:: f1_score
-.. autofunction:: roc_auc_score
-.. autofunction:: average_precision_score
-.. autofunction:: jaccard_score
-.. autofunction:: cohen_kappa_score
-.. autofunction:: r2_score
-.. autofunction:: confusion_matrix
+.. autofunction:: multiclass_metrics_fn
+

docs/api/metrics/pyhealth.metrics.multilabel.rst

@@ -3,13 +3,4 @@
 
 .. currentmodule:: pyhealth.metrics.multilabel
 
-.. autofunction:: accuracy_multilabel
-.. autofunction:: precision_multilabel
-.. autofunction:: recall_multilabel
-.. autofunction:: f1_multilabel
-.. autofunction:: roc_auc_multilabel
-.. autofunction:: pr_auc_multilabel
-.. autofunction:: jaccard_multilabel
-.. autofunction:: cohen_kappa_multilabel
-.. autofunction:: r2_score_multilabel
-.. autofunction:: ddi_rate_score
+.. autofunction:: multilabel_metrics_fn

pyhealth/datasets/base_dataset.py

@@ -49,8 +49,8 @@ class BaseDataset(ABC, Dataset):
  code_mapping: a dictionary containing the code mapping information.
  The key is a str of the source code vocabulary and the value is of
  two formats:
- (1) a str of the target code vocabulary;
- (2) a tuple with two elements. The first element is a str of the
+ - a str of the target code vocabulary;
+ - a tuple with two elements. The first element is a str of the
  target code vocabulary and the second element is a dict with
  keys "source_kwargs" or "target_kwargs" and values of the
  corresponding kwargs for the `CrossMap.map()` method.

pyhealth/models/classicml.py

@@ -11,14 +11,17 @@
 
 class ClassicML:
  """Call classical ML models
- 
+
+ Note:
+ The classic ML models are under development. Please use with caution!
+
  Args:
  dataset: the dataset object
  tables: a list of table names to be used
  target: the target table name
  classifier: the classifier object from sklearn
  mode: the mode of the model, can be "multilabel", "binary", "multiclass"
- 
+
  Examples:
  >>> from pyhealth.datasets import OMOPDataset
  >>> dataset = OMOPDataset(
@@ -27,7 +30,7 @@ class ClassicML:
  ... ) # load dataset
  >>> from pyhealth.tasks import mortality_prediction_omop_fn
  >>> dataset.set_task(mortality_prediction_omop_fn) # set task
- 
+
  >>> from pyhealth.models import ClassicML
  >>> from sklearn.ensemble import RandomForestClassifier as RF
  >>> model = ClassicML(
@@ -37,7 +40,6 @@ class ClassicML:
  ... mode="binary",
  ... classifier=RF(max_depth=6, max_features="sqrt", n_jobs=-1, n_estimators=50),
  ... )
- 
  """
 
  def __init__(

pyhealth/tasks/drug_recommendation.py

@@ -5,7 +5,7 @@ def drug_recommendation_mimic3_fn(patient: Patient):
  """Processes a single patient for the drug recommendation task.
 
  Drug recommendation aims at recommending a set of drugs given the patient health
-  history (e.g., conditions and procedures).
+ history (e.g., conditions and procedures).
 
  Args:
  patient: a Patient object
@@ -25,7 +25,6 @@ def drug_recommendation_mimic3_fn(patient: Patient):
  >>> dataset.set_task(drug_recommendation_mimic3_fn) # set task
  >>> dataset.samples[0] # exampe of an training sample
  [{'visit_id': '130744', 'patient_id': '103', 'conditions': [['42', '109', '19', '122', '98', '663', '58', '51']], 'procedures': [['1']], 'label': [['2', '3', '4']]}]
- 
  """
  samples = []
  for i in range(len(patient)):
@@ -72,7 +71,7 @@ def drug_recommendation_mimic4_fn(patient: Patient):
  """Processes a single patient for the drug recommendation task.
 
  Drug recommendation aims at recommending a set of drugs given the patient health
-  history (e.g., conditions and procedures).
+ history (e.g., conditions and procedures).
 
  Args:
  patient: a Patient object
@@ -81,7 +80,7 @@ def drug_recommendation_mimic4_fn(patient: Patient):
  samples: a list of samples, each sample is a dict with patient_id, visit_id,
  and other task-specific attributes as key
  
- Examples:
+  Examples:
  >>> from pyhealth.datasets import MIMIC4Dataset
  >>> mimic4_ds = MIMIC4Dataset(
  ... root="/srv/local/data/physionet.org/files/mimiciv/2.0/hosp",
@@ -92,7 +91,6 @@ def drug_recommendation_mimic4_fn(patient: Patient):
  >>> dataset.set_task(drug_recommendation_mimic4_fn) # set task
  >>> dataset.samples[0] # exampe of an training sample
  [{'visit_id': '130744', 'patient_id': '103', 'conditions': [['42', '109', '19', '122', '98', '663', '58', '51']], 'procedures': [['1']], 'label': [['2', '3', '4']]}]
- 
  """
  samples = []
  for i in range(len(patient)):
@@ -139,7 +137,7 @@ def drug_recommendation_eicu_fn(patient: Patient):
  """Processes a single patient for the drug recommendation task.
 
  Drug recommendation aims at recommending a set of drugs given the patient health
-  history (e.g., conditions and procedures).
+ history (e.g., conditions and procedures).
 
  Args:
  patient: a Patient object
@@ -148,7 +146,7 @@ def drug_recommendation_eicu_fn(patient: Patient):
  samples: a list of samples, each sample is a dict with patient_id, visit_id,
  and other task-specific attributes as key
  
- Examples:
+  Examples:
  >>> from pyhealth.datasets import eICUDataset
  >>> eicu_ds = eICUDataset(
  ... root="/srv/local/data/physionet.org/files/eicu-crd/2.0",
@@ -160,7 +158,6 @@ def drug_recommendation_eicu_fn(patient: Patient):
  >>> dataset.set_task(drug_recommendation_eicu_fn) # set task
  >>> dataset.samples[0] # exampe of an training sample
  [{'visit_id': '130744', 'patient_id': '103', 'conditions': [['42', '109', '98', '663', '58', '51']], 'procedures': [['1']], 'label': [['2', '3', '4']]}]
- 
  """
  samples = []
  for i in range(len(patient)):
@@ -205,7 +202,7 @@ def drug_recommendation_omop_fn(patient: Patient):
  """Processes a single patient for the drug recommendation task.
 
  Drug recommendation aims at recommending a set of drugs given the patient health
-  history (e.g., conditions and procedures).
+ history (e.g., conditions and procedures).
 
  Args:
  patient: a Patient object
@@ -225,7 +222,6 @@ def drug_recommendation_omop_fn(patient: Patient):
  >>> dataset.set_task(drug_recommendation_eicu_fn) # set task
  >>> dataset.samples[0] # exampe of an training sample
  [{'visit_id': '130744', 'patient_id': '103', 'conditions': [['42', '109', '98', '663', '58', '51'], ['98', '663', '58', '51']], 'procedures': [['1'], ['2', '3']], 'label': [['2', '3', '4'], ['0', '1', '4', '5']]}]
- 
  """
 
  samples = []

pyhealth/tasks/length_of_stay_prediction.py

@@ -44,7 +44,7 @@ def length_of_stay_prediction_mimic3_fn(patient: Patient):
 
  Note that we define the task as a multi-class classification task.
  
- Examples:
+  Examples:
  >>> from pyhealth.datasets import MIMIC3Dataset
  >>> mimic3_ds = MIMIC3Dataset(
  ... root="/srv/local/data/physionet.org/files/mimiciii/1.4",
@@ -55,7 +55,6 @@ def length_of_stay_prediction_mimic3_fn(patient: Patient):
  >>> dataset.set_task(length_of_stay_prediction_mimic3_fn) # set task
  >>> dataset.samples[0] # exampe of an training sample
  [{'visit_id': '130744', 'patient_id': '103', 'conditions': [['42', '109', '19', '122', '98', '663', '58', '51']], 'procedures': [['1']], 'label': 4}]
- 
  """
  samples = []
 
@@ -102,7 +101,7 @@ def length_of_stay_prediction_mimic4_fn(patient: Patient):
 
  Note that we define the task as a multi-class classification task.
  
- Examples:
+  Examples:
  >>> from pyhealth.datasets import MIMIC4Dataset
  >>> mimic4_ds = MIMIC4Dataset(
  ... root="/srv/local/data/physionet.org/files/mimiciv/2.0/hosp",
@@ -113,8 +112,6 @@ def length_of_stay_prediction_mimic4_fn(patient: Patient):
  >>> dataset.set_task(length_of_stay_prediction_mimic4_fn) # set task
  >>> dataset.samples[0] # exampe of an training sample
  [{'visit_id': '130744', 'patient_id': '103', 'conditions': [['42', '109', '19', '122', '98', '663', '58', '51']], 'procedures': [['1']], 'label': 2}]
- 
- 
  """
  samples = []
 
@@ -161,7 +158,7 @@ def length_of_stay_prediction_eicu_fn(patient: Patient):
 
  Note that we define the task as a multi-class classification task.
  
- Examples:
+  Examples:
  >>> from pyhealth.datasets import eICUDataset
  >>> eicu_ds = eICUDataset(
  ... root="/srv/local/data/physionet.org/files/eicu-crd/2.0",
@@ -173,8 +170,6 @@ def length_of_stay_prediction_eicu_fn(patient: Patient):
  >>> dataset.set_task(length_of_stay_prediction_eicu_fn) # set task
  >>> dataset.samples[0] # exampe of an training sample
  [{'visit_id': '130744', 'patient_id': '103', 'conditions': [['42', '109', '98', '663', '58', '51']], 'procedures': [['1']], 'label': 5}]
- 
- 
  """
  samples = []
 
@@ -232,7 +227,6 @@ def length_of_stay_prediction_omop_fn(patient: Patient):
  >>> dataset.set_task(length_of_stay_prediction_eicu_fn) # set task
  >>> dataset.samples[0] # exampe of an training sample
  [{'visit_id': '130744', 'patient_id': '103', 'conditions': [['42', '109', '98', '663', '58', '51']], 'procedures': [['1']], 'label': 7}]
- 
  """
  samples = []