scitex 2.10.3__py3-none-any.whl → 2.13.0__py3-none-any.whl

scitex/stats/_mcp_tool_schemas.py

@@ -0,0 +1,384 @@
+#!/usr/bin/env python3
+# Timestamp: 2026-01-08
+# File: src/scitex/stats/_mcp_tool_schemas.py
+# ----------------------------------------
+
+"""Tool schemas for the scitex-stats MCP server."""
+
+from __future__ import annotations
+
+import mcp.types as types
+
+__all__ = ["get_tool_schemas"]
+
+
+def get_tool_schemas() -> list[types.Tool]:
+    """Return all tool schemas for the Stats MCP server."""
+    return [
+        # Test Recommendation
+        types.Tool(
+            name="recommend_tests",
+            description=(
+                "Recommend appropriate statistical tests based on data characteristics. "
+                "Returns ranked list of tests with rationale."
+            ),
+            inputSchema={
+                "type": "object",
+                "properties": {
+                    "n_groups": {
+                        "type": "integer",
+                        "description": "Number of groups to compare",
+                        "default": 2,
+                    },
+                    "sample_sizes": {
+                        "type": "array",
+                        "items": {"type": "integer"},
+                        "description": "Sample sizes for each group",
+                    },
+                    "outcome_type": {
+                        "type": "string",
+                        "description": "Type of outcome variable",
+                        "enum": ["continuous", "ordinal", "categorical", "binary"],
+                        "default": "continuous",
+                    },
+                    "design": {
+                        "type": "string",
+                        "description": "Study design",
+                        "enum": ["between", "within", "mixed"],
+                        "default": "between",
+                    },
+                    "paired": {
+                        "type": "boolean",
+                        "description": "Whether data is paired/matched",
+                        "default": False,
+                    },
+                    "has_control_group": {
+                        "type": "boolean",
+                        "description": "Whether there is a control group",
+                        "default": False,
+                    },
+                    "top_k": {
+                        "type": "integer",
+                        "description": "Number of recommendations to return",
+                        "default": 3,
+                    },
+                },
+            },
+        ),
+        # Run Statistical Test
+        types.Tool(
+            name="run_test",
+            description=(
+                "Execute a statistical test on provided data. "
+                "Returns test statistic, p-value, effect size, and confidence intervals."
+            ),
+            inputSchema={
+                "type": "object",
+                "properties": {
+                    "test_name": {
+                        "type": "string",
+                        "description": "Name of test to run",
+                        "enum": [
+                            "ttest_ind",
+                            "ttest_paired",
+                            "ttest_1samp",
+                            "brunner_munzel",
+                            "mannwhitneyu",
+                            "wilcoxon",
+                            "anova",
+                            "kruskal",
+                            "chi2",
+                            "fisher_exact",
+                            "pearson",
+                            "spearman",
+                            "kendall",
+                        ],
+                    },
+                    "data": {
+                        "type": "array",
+                        "items": {"type": "array", "items": {"type": "number"}},
+                        "description": "Data arrays for each group",
+                    },
+                    "alternative": {
+                        "type": "string",
+                        "description": "Alternative hypothesis",
+                        "enum": ["two-sided", "less", "greater"],
+                        "default": "two-sided",
+                    },
+                },
+                "required": ["test_name", "data"],
+            },
+        ),
+        # Format Results
+        types.Tool(
+            name="format_results",
+            description=(
+                "Format statistical results in journal style (APA, Nature, etc.)"
+            ),
+            inputSchema={
+                "type": "object",
+                "properties": {
+                    "test_name": {
+                        "type": "string",
+                        "description": "Name of the statistical test",
+                    },
+                    "statistic": {
+                        "type": "number",
+                        "description": "Test statistic value",
+                    },
+                    "p_value": {
+                        "type": "number",
+                        "description": "P-value",
+                    },
+                    "df": {
+                        "type": "number",
+                        "description": "Degrees of freedom (if applicable)",
+                    },
+                    "effect_size": {
+                        "type": "number",
+                        "description": "Effect size value",
+                    },
+                    "effect_size_name": {
+                        "type": "string",
+                        "description": "Name of effect size measure (d, r, eta2, etc.)",
+                    },
+                    "style": {
+                        "type": "string",
+                        "description": "Journal formatting style",
+                        "enum": ["apa", "nature", "science", "brief"],
+                        "default": "apa",
+                    },
+                    "ci_lower": {
+                        "type": "number",
+                        "description": "Lower bound of confidence interval",
+                    },
+                    "ci_upper": {
+                        "type": "number",
+                        "description": "Upper bound of confidence interval",
+                    },
+                },
+                "required": ["test_name", "statistic", "p_value"],
+            },
+        ),
+        # Power Analysis
+        types.Tool(
+            name="power_analysis",
+            description=(
+                "Calculate statistical power or required sample size. "
+                "Supports various test types."
+            ),
+            inputSchema={
+                "type": "object",
+                "properties": {
+                    "test_type": {
+                        "type": "string",
+                        "description": "Type of statistical test",
+                        "enum": ["ttest", "anova", "correlation", "chi2"],
+                        "default": "ttest",
+                    },
+                    "effect_size": {
+                        "type": "number",
+                        "description": "Expected effect size (Cohen's d, f, r, w)",
+                    },
+                    "alpha": {
+                        "type": "number",
+                        "description": "Significance level",
+                        "default": 0.05,
+                    },
+                    "power": {
+                        "type": "number",
+                        "description": "Desired statistical power (for sample size calculation)",
+                        "default": 0.8,
+                    },
+                    "n": {
+                        "type": "integer",
+                        "description": "Sample size (for power calculation)",
+                    },
+                    "n_groups": {
+                        "type": "integer",
+                        "description": "Number of groups (for ANOVA)",
+                        "default": 2,
+                    },
+                    "ratio": {
+                        "type": "number",
+                        "description": "Ratio of group sizes (n2/n1)",
+                        "default": 1.0,
+                    },
+                },
+            },
+        ),
+        # Correct P-values
+        types.Tool(
+            name="correct_pvalues",
+            description=(
+                "Apply multiple comparison correction to p-values. "
+                "Supports Bonferroni, FDR (Benjamini-Hochberg), Holm, and Sidak methods."
+            ),
+            inputSchema={
+                "type": "object",
+                "properties": {
+                    "pvalues": {
+                        "type": "array",
+                        "items": {"type": "number"},
+                        "description": "Array of p-values to correct",
+                    },
+                    "method": {
+                        "type": "string",
+                        "description": "Correction method",
+                        "enum": ["bonferroni", "fdr_bh", "fdr_by", "holm", "sidak"],
+                        "default": "fdr_bh",
+                    },
+                    "alpha": {
+                        "type": "number",
+                        "description": "Family-wise error rate",
+                        "default": 0.05,
+                    },
+                },
+                "required": ["pvalues"],
+            },
+        ),
+        # Descriptive Statistics
+        types.Tool(
+            name="describe",
+            description=(
+                "Calculate descriptive statistics for data. "
+                "Returns mean, std, median, quartiles, skewness, kurtosis."
+            ),
+            inputSchema={
+                "type": "object",
+                "properties": {
+                    "data": {
+                        "type": "array",
+                        "items": {"type": "number"},
+                        "description": "Data array to describe",
+                    },
+                    "percentiles": {
+                        "type": "array",
+                        "items": {"type": "number"},
+                        "description": "Percentiles to calculate (0-100)",
+                        "default": [25, 50, 75],
+                    },
+                },
+                "required": ["data"],
+            },
+        ),
+        # Effect Size Calculation
+        types.Tool(
+            name="effect_size",
+            description=(
+                "Calculate effect size between groups. "
+                "Returns Cohen's d, Hedges' g, or other appropriate measure."
+            ),
+            inputSchema={
+                "type": "object",
+                "properties": {
+                    "group1": {
+                        "type": "array",
+                        "items": {"type": "number"},
+                        "description": "First group data",
+                    },
+                    "group2": {
+                        "type": "array",
+                        "items": {"type": "number"},
+                        "description": "Second group data",
+                    },
+                    "measure": {
+                        "type": "string",
+                        "description": "Effect size measure",
+                        "enum": ["cohens_d", "hedges_g", "glass_delta", "cliffs_delta"],
+                        "default": "cohens_d",
+                    },
+                    "pooled": {
+                        "type": "boolean",
+                        "description": "Use pooled standard deviation",
+                        "default": True,
+                    },
+                },
+                "required": ["group1", "group2"],
+            },
+        ),
+        # Normality Test
+        types.Tool(
+            name="normality_test",
+            description=(
+                "Test whether data follows a normal distribution. "
+                "Returns test statistic and p-value."
+            ),
+            inputSchema={
+                "type": "object",
+                "properties": {
+                    "data": {
+                        "type": "array",
+                        "items": {"type": "number"},
+                        "description": "Data to test for normality",
+                    },
+                    "method": {
+                        "type": "string",
+                        "description": "Normality test method",
+                        "enum": ["shapiro", "dagostino", "anderson", "lilliefors"],
+                        "default": "shapiro",
+                    },
+                },
+                "required": ["data"],
+            },
+        ),
+        # Post-hoc Tests
+        types.Tool(
+            name="posthoc_test",
+            description=(
+                "Run post-hoc pairwise comparisons after significant ANOVA/Kruskal. "
+                "Supports Tukey HSD, Dunnett, Games-Howell, Dunn."
+            ),
+            inputSchema={
+                "type": "object",
+                "properties": {
+                    "groups": {
+                        "type": "array",
+                        "items": {"type": "array", "items": {"type": "number"}},
+                        "description": "Data arrays for each group",
+                    },
+                    "group_names": {
+                        "type": "array",
+                        "items": {"type": "string"},
+                        "description": "Names for each group",
+                    },
+                    "method": {
+                        "type": "string",
+                        "description": "Post-hoc method",
+                        "enum": ["tukey", "dunnett", "games_howell", "dunn"],
+                        "default": "tukey",
+                    },
+                    "control_group": {
+                        "type": "integer",
+                        "description": "Index of control group (for Dunnett)",
+                        "default": 0,
+                    },
+                },
+                "required": ["groups"],
+            },
+        ),
+        # P-value to Stars
+        types.Tool(
+            name="p_to_stars",
+            description=("Convert p-value to significance stars (*, **, ***, ns)"),
+            inputSchema={
+                "type": "object",
+                "properties": {
+                    "p_value": {
+                        "type": "number",
+                        "description": "P-value to convert",
+                    },
+                    "thresholds": {
+                        "type": "array",
+                        "items": {"type": "number"},
+                        "description": "Significance thresholds",
+                        "default": [0.001, 0.01, 0.05],
+                    },
+                },
+                "required": ["p_value"],
+            },
+        ),
+    ]
+
+
+# EOF

scitex/stats/_schema.py

@@ -7,7 +7,7 @@ Statistical Result Schema - DEPRECATED
 This module is deprecated. Import from scitex.schema._stats instead:
     from scitex.schema._stats import Position, StatStyling, StatPositioning, StatResult
 
-For the new simplified API, use scitex.fts._stats.
+For the new simplified API, use scitex.io.bundle._stats.
 """
 
 import warnings

scitex/stats/correct/_correct_bonferroni.py

@@ -536,7 +536,7 @@ def run_main():
 
     args = parse_args()
 
-    CONFIG, sys.stdout, sys.stderr, plt, CC, rng_manager = stx.session.start(
+    CONFIG, sys.stdout, sys.stderr, plt, CC, rng = stx.session.start(
         sys,
         plt,
         args=args,

scitex/stats/correct/_correct_fdr.py

@@ -392,7 +392,7 @@ def demo(verbose=False):
     """Demonstrate FDR correction."""
     import scitex as stx
 
-    # CONFIG, sys.stdout, sys.stderr, plt, CC, rng_manager
+    # CONFIG, sys.stdout, sys.stderr, plt, CC, rng
 
     logger.info("Demonstrating False Discovery Rate correction")
 

scitex/stats/correct/_correct_fdr_.py

@@ -603,7 +603,7 @@ def run_main():
 
     args = parse_args()
 
-    CONFIG, sys.stdout, sys.stderr, plt, CC, rng_manager = stx.session.start(
+    CONFIG, sys.stdout, sys.stderr, plt, CC, rng = stx.session.start(
         sys,
         plt,
         args=args,

scitex/stats/correct/_correct_holm.py

@@ -506,7 +506,7 @@ def run_main():
 
     args = parse_args()
 
-    CONFIG, sys.stdout, sys.stderr, plt, CC, rng_manager = stx.session.start(
+    CONFIG, sys.stdout, sys.stderr, plt, CC, rng = stx.session.start(
         sys,
         plt,
         args=args,

scitex/stats/correct/_correct_sidak.py

@@ -333,7 +333,7 @@ def main():
     )
     args = parser.parse_args([])
 
-    CONFIG, sys.stdout, sys.stderr, plt, CC, rng_manager = stx.session.start(
+    CONFIG, sys.stdout, sys.stderr, plt, CC, rng = stx.session.start(
         sys=sys,
         plt=plt,
         args=args,

scitex/stats/effect_sizes/_cliffs_delta.py

@@ -304,7 +304,7 @@ def run_main():
 
     args = parse_args()
 
-    CONFIG, sys.stdout, sys.stderr, plt, CC, rng_manager = stx.session.start(
+    CONFIG, sys.stdout, sys.stderr, plt, CC, rng = stx.session.start(
         sys,
         plt,
         args=args,

scitex/stats/effect_sizes/_cohens_d.py

@@ -317,7 +317,7 @@ def run_main():
 
     args = parse_args()
 
-    CONFIG, sys.stdout, sys.stderr, plt, CC, rng_manager = stx.session.start(
+    CONFIG, sys.stdout, sys.stderr, plt, CC, rng = stx.session.start(
         sys,
         plt,
         args=args,

scitex/stats/effect_sizes/_epsilon_squared.py

@@ -288,7 +288,7 @@ def run_main():
 
     args = parse_args()
 
-    CONFIG, sys.stdout, sys.stderr, plt, CC, rng_manager = stx.session.start(
+    CONFIG, sys.stdout, sys.stderr, plt, CC, rng = stx.session.start(
         sys,
         plt,
         args=args,

scitex/stats/effect_sizes/_eta_squared.py

@@ -274,7 +274,7 @@ def run_main():
 
     args = parse_args()
 
-    CONFIG, sys.stdout, sys.stderr, plt, CC, rng_manager = stx.session.start(
+    CONFIG, sys.stdout, sys.stderr, plt, CC, rng = stx.session.start(
         sys,
         plt,
         args=args,

scitex/stats/effect_sizes/_prob_superiority.py

@@ -269,7 +269,7 @@ def run_main():
 
     args = parse_args()
 
-    CONFIG, sys.stdout, sys.stderr, plt, CC, rng_manager = stx.session.start(
+    CONFIG, sys.stdout, sys.stderr, plt, CC, rng = stx.session.start(
         sys,
         plt,
         args=args,

scitex/stats/io/__init__.py

@@ -1,29 +1,28 @@
 #!/usr/bin/env python3
-# -*- coding: utf-8 -*-
 # Timestamp: "2025-12-13 (ywatanabe)"
 # File: /home/ywatanabe/proj/scitex-code/src/scitex/stats/io/__init__.py
 
 """
-I/O operations for scitex.stats - Statistical results bundles (.statsz).
+I/O operations for scitex.stats - Statistical results bundles (.stats).
 
 This module handles:
-    - .statsz bundle load/save operations
+    - .stats bundle load/save operations
     - Statistical comparison metadata
     - P-value and effect size validation
 """
 
 from ._bundle import (
-    validate_statsz_spec,
-    load_statsz_bundle,
-    save_statsz_bundle,
-    STATSZ_SCHEMA_SPEC,
+    STATS_SCHEMA_SPEC,
+    load_stats_bundle,
+    save_stats_bundle,
+    validate_stats_spec,
 )
 
 __all__ = [
-    "validate_statsz_spec",
-    "load_statsz_bundle",
-    "save_statsz_bundle",
-    "STATSZ_SCHEMA_SPEC",
+    "validate_stats_spec",
+    "load_stats_bundle",
+    "save_stats_bundle",
+    "STATS_SCHEMA_SPEC",
 ]
 
 # EOF

scitex/stats/io/_bundle.py

@@ -1,10 +1,9 @@
 #!/usr/bin/env python3
-# -*- coding: utf-8 -*-
 # Timestamp: "2025-12-13 (ywatanabe)"
 # File: /home/ywatanabe/proj/scitex-code/src/scitex/stats/io/_bundle.py
 
 """
-SciTeX .statsz Bundle I/O - Statistics-specific bundle operations.
+SciTeX .stats Bundle I/O - Statistics-specific bundle operations.
 
 Handles:
     - Statistical results specification validation
@@ -17,14 +16,14 @@ from pathlib import Path
 from typing import Any, Dict, List
 
 __all__ = [
-    "validate_statsz_spec",
-    "load_statsz_bundle",
-    "save_statsz_bundle",
-    "STATSZ_SCHEMA_SPEC",
+    "validate_stats_spec",
+    "load_stats_bundle",
+    "save_stats_bundle",
+    "STATS_SCHEMA_SPEC",
 ]
 
-# Schema specification for .statsz bundles
-STATSZ_SCHEMA_SPEC = {
+# Schema specification for .stats bundles
+STATS_SCHEMA_SPEC = {
     "name": "scitex.stats.stats",
     "version": "1.0.0",
     "required_fields": ["schema"],
@@ -32,8 +31,8 @@ STATSZ_SCHEMA_SPEC = {
 }
 
 
-def validate_statsz_spec(spec: Dict[str, Any]) -> List[str]:
-    """Validate .statsz-specific fields.
+def validate_stats_spec(spec: Dict[str, Any]) -> List[str]:
+    """Validate .stats-specific fields.
 
     Args:
         spec: The specification dictionary to validate.
@@ -91,8 +90,8 @@ def validate_statsz_spec(spec: Dict[str, Any]) -> List[str]:
     return errors
 
 
-def load_statsz_bundle(bundle_dir: Path) -> Dict[str, Any]:
-    """Load .statsz bundle contents from directory.
+def load_stats_bundle(bundle_dir: Path) -> Dict[str, Any]:
+    """Load .stats bundle contents from directory.
 
     Args:
         bundle_dir: Path to the bundle directory.
@@ -105,7 +104,7 @@ def load_statsz_bundle(bundle_dir: Path) -> Dict[str, Any]:
     # Load specification
     spec_file = bundle_dir / "stats.json"
     if spec_file.exists():
-        with open(spec_file, "r") as f:
+        with open(spec_file) as f:
             result["spec"] = json.load(f)
     else:
         result["spec"] = None
@@ -115,16 +114,17 @@ def load_statsz_bundle(bundle_dir: Path) -> Dict[str, Any]:
     if data_file.exists():
         try:
             import pandas as pd
+
             result["data"] = pd.read_csv(data_file)
         except ImportError:
-            with open(data_file, "r") as f:
+            with open(data_file) as f:
                 result["data"] = f.read()
 
     return result
 
 
-def save_statsz_bundle(data: Dict[str, Any], dir_path: Path) -> None:
-    """Save .statsz bundle contents to directory.
+def save_stats_bundle(data: Dict[str, Any], dir_path: Path) -> None:
+    """Save .stats bundle contents to directory.
 
     Args:
         data: Bundle data dictionary.