verbose

README.md

@@ -23,15 +23,9 @@ Demo input files are provided in `demo_files/` to illustrate basic use.
 
 ## 🛠 Installation
 
-### Install from PyPI
-
-```bash
-pip install ecoffitter
-```
-
 ### Install from source
 
-Assuming in project directory:
+Assuming in project directory (after git cloning):
 
 ```bash
 pip install -e .
@@ -127,8 +121,6 @@ If `--outfile` ends in `.txt`, the tool writes:
 - ECOFF estimate
 - fitted mean & variance (per component)
 - mixture weights (if multiple distributions)
-- percentiles
-- likelihood values
 
 ### 2. PDF Report
 
@@ -137,8 +129,8 @@ If `--outfile` ends in `.pdf`, the tool writes:
 - histogram of observed MICs
 - fitted distribution curve(s)
 - ECOFF location marker
-- table of model parameters
-- censoring diagnostics
+- fitted mean & variance (per component)
+- mixture weights (if multiple distributions)
 
 ## 🚀 Command-Line Usage
 
@@ -147,7 +139,7 @@ Once installed, you can call the CLI.
 Example using demo files:
 
 ```bash
-ecoff-fitter   --input demo_files/input.txt   --params demo_files/params.txt   --outfile demo_files/output.txt
+ecoff-fitter   --input demo_files/censored.txt   --params demo_files/params.txt   --outfile demo_files/output.txt
 ```
 
 Instead of using a parameter file, you can also specify parameters directly.

demo_files/output.txt

@@ -1,6 +1,9 @@
-ECOFF: 0.29
-99th percentile: 0.29
-97.5th percentile: 0.22
-95th percentile: 0.17
-μ₁: 0.049303392476164935, σ₁: 2.1439340245815863
-μ₂: 14.283983662174911, σ₂: 1.354070842559145
+  ECOFF: 0.3963
+  log scale: -1.3354
+
+  Component 1:
+    μ = 0.0835
+    σ (folds) = 1.9534
+  Component 2:
+    μ = 3.8329
+    σ (folds) = 1.4984

gui.py

@@ -55,6 +55,16 @@ def __init__(self, root):
         self.percentile_entry.insert(0, "99")
         self.percentile_entry.pack(side="left", padx=10)
 
+        self.verbose_var = tk.BooleanVar(value=False)
+        verbose_frame = tk.Frame(root)
+        verbose_frame.pack(pady=5, anchor="w")
+        tk.Checkbutton(
+            verbose_frame,
+            text="model diagnostics",
+            variable=self.verbose_var,
+        ).pack(side="left")
+
+
         # OUTPUT FILE
         output_frame = tk.Frame(root)
         output_frame.pack(pady=5, anchor="w")
@@ -125,6 +135,7 @@ def run_ecoff(self):
         tails = int(self.tails_entry.get()) if self.tails_entry.get().strip() else None
         percentile = float(self.percentile_entry.get())
         outfile = self.output_entry.get()
+        verbose = self.verbose_var.get()
 
         if not input_file:
             messagebox.showerror("Error", "You must select an input MIC data file.")
@@ -163,16 +174,15 @@ def run_ecoff(self):
             text = "ECOFF RESULTS\n=====================================\n\n"
 
             global_report = GenerateReport.from_fitter(global_fitter, global_result)
+
             if len(individual_results) > 1:
-
-                text += global_report.to_text("GLOBAL FIT")
+                text += global_report.to_text("GLOBAL FIT", verbose=verbose)
                 text += "\nINDIVIDUAL FITS:\n-------------------------------------\n"
 
-
             # Individual fits
             for name, (fitter, result) in individual_results.items():
                 rep = GenerateReport.from_fitter(fitter, result)
-                text += rep.to_text(label=name)
+                text += rep.to_text(label=name, verbose=verbose)
 
 
             if outfile:

src/ecoff_fitter/cli.py

@@ -117,13 +117,13 @@ def main(argv: Optional[List[str]] = None) -> None:
         global_report = GenerateReport.from_fitter(global_fitter, global_result)
 
         if len(individual_results) > 1:
-            text += global_report.to_text("GLOBAL FIT")
+            text += global_report.to_text("GLOBAL FIT", verbose=args.verbose)
             text += "\nINDIVIDUAL FITS:\n-------------------------------------\n"
 
         # Individual fits
         for name, (fitter, result) in individual_results.items():
             rep = GenerateReport.from_fitter(fitter, result)
-            text += rep.to_text(label=name)       
+            text += rep.to_text(label=name, verbose=args.verbose)       
 
         if args.outfile:
             validate_output_path(args.outfile)

src/ecoff_fitter/core.py

@@ -326,3 +326,29 @@ def compute_ecoff(self, percentile: float) -> Tuple[Any, ...]:
             mus_sigmas.extend([mus[k], sigmas[k]])
 
         return (ecoff, z_percentile, *mus_sigmas)
+
+    def model_summary(self) -> dict[str, Any]:
+        K = self.distributions
+        n = self.obj_df.observations.sum()
+        wt_idx = int(np.argmin(self.mus_))
+
+        summary = {
+            "model_family": "interval-censored log-normal",
+            "model_type": "mixture" if K > 1 else "single",
+            "components": K,
+            "wild_type_component": wt_idx + 1,
+            "dilution_factor": self.dilution_factor,
+            "boundary_support": self.boundary_support,
+            "n_observations": n,
+            "log_likelihood": self.loglike_,
+            "converged": self.converged_,
+            "n_iter": self.n_iter_,
+            "pis": self.pis_,
+        }
+
+        # Information criteria (if identifiable)
+        k_params = 2 * K + (K - 1)
+        summary["aic"] = 2 * k_params - 2 * self.loglike_
+        summary["bic"] = np.log(n) * k_params - 2 * self.loglike_
+
+        return summary

src/ecoff_fitter/report.py

@@ -108,10 +108,10 @@ def to_text(self, label: str | None = None, verbose: bool = False) -> str:
             lines.append(sigma_line)
 
         # Verbose model details
-        if verbose and self.model is not None:
+        if verbose:
             lines.append("")
             lines.append("--- Model details ---")
-            lines.append(str(self.model))
+            lines.extend(self._format_model_summary())
 
         return "\n".join(lines) + "\n"
 
@@ -175,6 +175,28 @@ def _make_pdf(self, title: Optional[str] = None) -> Figure:
 
         fig.tight_layout(rect=(0, 0, 1, 0.95))
         return fig
+
+    def _format_model_summary(self) -> list[str]:
+        """
+        Format a structured model summary from the fitter, if available.
+        """
+        if not hasattr(self.fitter, "model_summary"):
+            return ["  (model summary unavailable)"]
+
+        summary = self.fitter.model_summary()
+        lines: list[str] = []
+
+        for key, value in summary.items():
+            if value is None:
+                continue
+
+            # Format arrays nicely
+            if isinstance(value, np.ndarray):
+                value = ", ".join(f"{v:.4f}" for v in value)
+
+            lines.append(f"  {key.replace('_', ' ')}: {value}")
+
+        return lines
 
 
 class CombinedReport:
@@ -212,13 +234,13 @@ def write_out(self) -> None:
             lines.append(f"\n===== INDIVIDUAL FIT: {name} =====")
             lines.append(report.to_text(label=name))
 
-        # Join and write file
         text = "\n".join(lines)
 
         with open(self.outfile, "w") as f:
             f.write(text)
 
         print(f"\nCombined text report saved to: {self.outfile}")
+
 
     def save_pdf(self) -> None:
         from matplotlib.backends.backend_pdf import PdfPages
@@ -239,3 +261,5 @@ def save_pdf(self) -> None:
         print(f"Combined PDF saved to {self.outfile}")
 
         print(f"Combined PDF saved to {self.outfile}")
+
+

tests/test_ecoff_fitter.py

@@ -236,3 +236,39 @@ def fit(self, **kwargs):
     assert fitter.converged_
     assert len(fitter.mus_) == 2
     assert np.isfinite(fitter.mus_[0])
+
+def test_model_summary_basic(simple_data):
+    """
+    model_summary() should return a structured, self-consistent summary
+    after fitting.
+    """
+    fitter = ECOFFitter(simple_data, distributions=2)
+    fitter.fit()
+
+    summary = fitter.model_summary()
+
+    # basic structure
+    assert isinstance(summary, dict)
+
+    # required keys
+    required_keys = {
+        "model_family",
+        "model_type",
+        "components",
+        "wild_type_component",
+        "n_observations",
+        "log_likelihood",
+        "converged",
+        "pis",
+        "aic",
+        "bic",
+    }
+    assert required_keys.issubset(summary.keys())
+
+    # internal consistency
+    assert summary["components"] == fitter.distributions
+    assert summary["n_observations"] == fitter.obj_df.observations.sum()
+    assert np.isclose(np.sum(summary["pis"]), 1.0)
+    assert summary["converged"] is True
+    assert summary["bic"] >= summary["aic"]
+

tests/test_report.py

@@ -2,7 +2,7 @@
 import sys
 import pytest
 from unittest.mock import MagicMock, patch, call, ANY
-
+import numpy as np
 import ecoff_fitter.cli as cli
 import ecoff_fitter.report as report
 
@@ -205,15 +205,29 @@ def test_generate_report_to_text_two_dist_verbose():
     fitter.dilution_factor = 2
     fitter.mus_ = [1, 2]
     fitter.sigmas_ = [0.2, 0.5]
-    fitter.model_ = "FAKE_MODEL_DETAILS"
 
-    r = cli.GenerateReport(fitter=fitter, ecoff=4.0, z=(10, 8, 6))
+    fitter.model_summary.return_value = {
+        "model_family": "interval-censored log-normal",
+        "model_type": "mixture",
+        "components": 2,
+        "wild_type_component": 1,
+        "n_observations": 100,
+        "log_likelihood": -123.45,
+        "converged": True,
+        "pis": [0.7, 0.3],
+        "aic": 260.9,
+        "bic": 270.1,
+    }
+
+    r = report.GenerateReport(fitter=fitter, ecoff=4.0, z=(10, 8, 6))
     text = r.to_text(verbose=True)
 
     assert "Component 1" in text
     assert "Component 2" in text
     assert "--- Model details ---" in text
-    assert "FAKE_MODEL_DETAILS" in text
+    assert "model family: interval-censored log-normal" in text
+    assert "components: 2" in text
+
 
 def test_combined_report_write_out(tmp_path):
     # CombinedReport now writes to its outfile
@@ -235,8 +249,7 @@ def test_combined_report_write_out(tmp_path):
         individual_reports={"A": report_A, "B": report_B},
     )
 
-    # Act
-    combined.write_out()   # NO ARGUMENT NOW
+    combined.write_out()  
 
     text = outfile.read_text()
 
@@ -253,3 +266,46 @@ def test_combined_report_write_out(tmp_path):
     report_A.to_text.assert_called_with(label="A")
     report_B.to_text.assert_called_with(label="B")
 
+
+def test_generate_report_format_model_summary():
+    """
+    _format_model_summary() should format key/value pairs from
+    fitter.model_summary() into readable text lines.
+    """
+    fitter = MagicMock()
+    fitter.model_summary.return_value = {
+        "model_family": "interval-censored log-normal",
+        "model_type": "mixture",
+        "components": 2,
+        "wild_type_component": 1,
+        "n_observations": 100,
+        "log_likelihood": -123.45,
+        "converged": True,
+        "pis": np.array([0.7, 0.3]),
+        "aic": 260.9,
+        "bic": 270.1,
+    }
+
+    r = report.GenerateReport(
+        fitter=fitter,
+        ecoff=4.0,
+        z=(10, 8, 6),
+    )
+
+    lines = r._format_model_summary()
+
+    # basic structure
+    assert isinstance(lines, list)
+    assert all(isinstance(line, str) for line in lines)
+
+    # content checks (formatted, not raw)
+    assert any("model family: interval-censored log-normal" in line for line in lines)
+    assert any("model type: mixture" in line for line in lines)
+    assert any("components: 2" in line for line in lines)
+    assert any("wild type component: 1" in line for line in lines)
+    assert any("log likelihood: -123.45" in line for line in lines)
+    assert any("converged: True" in line for line in lines)
+    assert any("pis: 0.7000, 0.3000" in line for line in lines)
+    assert any("aic: 260.9" in line for line in lines)
+    assert any("bic: 270.1" in line for line in lines)
+