RSATSIModel 2.5.0__tar.gz

rsatsimodel-2.5.0/LICENSE

@@ -0,0 +1,3 @@
+MIT License
+
+Copyright (c) 2026 Dr Md Galal Uddin

rsatsimodel-2.5.0/MANIFEST.in

@@ -0,0 +1,3 @@
+include README.md
+include LICENSE
+recursive-include examples *

rsatsimodel-2.5.0/PKG-INFO

@@ -0,0 +1,159 @@
+Metadata-Version: 2.4
+Name: RSATSIModel
+Version: 2.5.0
+Summary: Python package for computing ATSI and RS-ATSI using CHL, SAL, and Sentinel-3 OLCI-style data.
+Author-email: Dr Md Galal Uddin <jalaluddinbd1987@gmail.com>
+License: MIT
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: pandas>=1.5.0
+Requires-Dist: numpy>=1.23.0
+Requires-Dist: matplotlib>=3.6.0
+Requires-Dist: openpyxl>=3.1.0
+Requires-Dist: scikit-learn>=1.2.0
+Dynamic: license-file
+
+# REMOTE SENSING (RS)-DRIVEN ATSIModel
+
+## Assessment of Trophic Status Index (ATSI) and RS-ATSI Python Package
+
+**ATSIModel** is an open-source Python package for computing:
+1. **ATSI** from in-situ **chlorophyll-a (CHL)** and **salinity (SAL)** data.
+2. **RS-ATSI** from **Sentinel-3 OLCI style remote-sensing inputs**, with CHL prediction, ATSI derivation, and a structured validation framework.
+
+The package is designed for **transitional, estuarine, and coastal waters**, and supports both standard ATSI computation and remote-sensing based ATSI workflows.
+
+## Author
+
+**Dr Md Galal Uddin**  
+School of Engineering, University of Galway, Ireland  
+Email: **jalaluddinbd1987@gmail.com**
+
+### Research Profiles
+- **Google Scholar**: https://scholar.google.com/citations?user=g6xaAOkAAAAJ&hl=en
+
+## Scientific Background
+
+ATSI is implemented as a **CHL-driven, salinity-conditioned trophic scoring framework**.  
+This package extends that framework into a **remote-sensing compatible RS-ATSI workflow**.
+
+## Standard ATSI workflow
+
+Required columns:
+
+| Column | Description | Unit |
+|---|---|---|
+| CHL | Chlorophyll-a concentration | µg/L |
+| SAL | Salinity | PSU |
+
+Example:
+
+```python
+from atsimodel.core import run_atsi
+df = run_atsi("examples/sample_atsi_input.xlsx")
+print(df.head())
+```
+
+Export:
+
+```python
+from atsimodel.utils import export_single
+export_single(df, out_base="outputs/ATSI_results", formats=("xlsx", "csv", "txt", "json"))
+```
+
+## RS-ATSI workflow
+
+Recommended input columns:
+- `CHL`
+- `SAL`
+- `Rhow_1`, `Rhow_2`, `Rhow_3`, `Rhow_4`, `Rhow_5`, `Rhow_6`, `Rhow_7`, `Rhow_8`, `Rhow_9`, `Rhow_10`, `Rhow_11`
+
+Recommended optional columns:
+- `SITE`
+- `SEASON`
+- `ZONE`
+- `SPLIT`
+
+Example RS table:
+
+| SITE | SEASON | ZONE | SPLIT | CHL | SAL | Rhow_1 | Rhow_2 | Rhow_3 | Rhow_4 | Rhow_5 | Rhow_6 | Rhow_7 | Rhow_8 | Rhow_9 | Rhow_10 | Rhow_11 |
+|---|---|---|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
+| S1 | Winter | Upper | TRAIN | 8.0 | 31.0 | 0.011 | 0.012 | 0.013 | 0.014 | 0.015 | 0.016 | 0.017 | 0.018 | 0.019 | 0.020 | 0.021 |
+
+Example:
+
+```python
+from atsimodel.rs_core import run_rs_atsi_pipeline
+
+results = run_rs_atsi_pipeline(
+    input_file="examples/sample_rs_atsi_input.xlsx",
+    output_dir="outputs/RS_ATSI_demo"
+)
+print(results["chl_validation_overall"])
+print(results["atsi_validation_overall"])
+```
+
+## Validation design
+
+### 1. CHL validation
+- R²
+- RMSE
+- MAE
+- Bias
+- train / test / independent validation
+- site-wise validation
+
+### 2. ATSI validation
+- RS-ATSI vs in-situ ATSI
+- confusion matrix by trophic class
+- agreement by station and season
+
+### 3. Spatial validation
+- hotspot consistency
+- estuarine gradient realism
+- upper / middle / lower estuary comparison
+
+## Exported outputs
+
+The package exports:
+- Excel workbook
+- CSV files
+- TXT files
+- JSON files
+
+Typical output tables include:
+- `predictions_all`
+- `chl_validation_overall`
+- `chl_validation_by_site_test`
+- `atsi_validation_overall`
+- `atsi_validation_by_site_test`
+- `atsi_confusion_matrix_test`
+- `station_season_agreement_test`
+- `spatial_zone_summary_test`
+
+## Scientific notes
+
+- ATSI remains **CHL-driven**
+- SAL is used to determine **threshold context**
+- ATSI values are clipped to **0–100**
+- Current class translation is operational:
+  - Unpolluted
+  - Moderate
+  - Eutrophic
+  - Hypertrophic
+
+## Limitations
+
+This release does not yet include:
+- uncertainty quantification
+- prediction intervals
+- raster ingestion
+- geospatial map production
+- SHAP explainability
+
+## Contact
+
+**Dr Md Galal Uddin**  
+School of Engineering, University of Galway, Ireland  
+Email: **jalaluddinbd1987@gmail.com**

rsatsimodel-2.5.0/README.md

@@ -0,0 +1,143 @@
+# REMOTE SENSING (RS)-DRIVEN ATSIModel
+
+## Assessment of Trophic Status Index (ATSI) and RS-ATSI Python Package
+
+**ATSIModel** is an open-source Python package for computing:
+1. **ATSI** from in-situ **chlorophyll-a (CHL)** and **salinity (SAL)** data.
+2. **RS-ATSI** from **Sentinel-3 OLCI style remote-sensing inputs**, with CHL prediction, ATSI derivation, and a structured validation framework.
+
+The package is designed for **transitional, estuarine, and coastal waters**, and supports both standard ATSI computation and remote-sensing based ATSI workflows.
+
+## Author
+
+**Dr Md Galal Uddin**  
+School of Engineering, University of Galway, Ireland  
+Email: **jalaluddinbd1987@gmail.com**
+
+### Research Profiles
+- **Google Scholar**: https://scholar.google.com/citations?user=g6xaAOkAAAAJ&hl=en
+
+## Scientific Background
+
+ATSI is implemented as a **CHL-driven, salinity-conditioned trophic scoring framework**.  
+This package extends that framework into a **remote-sensing compatible RS-ATSI workflow**.
+
+## Standard ATSI workflow
+
+Required columns:
+
+| Column | Description | Unit |
+|---|---|---|
+| CHL | Chlorophyll-a concentration | µg/L |
+| SAL | Salinity | PSU |
+
+Example:
+
+```python
+from atsimodel.core import run_atsi
+df = run_atsi("examples/sample_atsi_input.xlsx")
+print(df.head())
+```
+
+Export:
+
+```python
+from atsimodel.utils import export_single
+export_single(df, out_base="outputs/ATSI_results", formats=("xlsx", "csv", "txt", "json"))
+```
+
+## RS-ATSI workflow
+
+Recommended input columns:
+- `CHL`
+- `SAL`
+- `Rhow_1`, `Rhow_2`, `Rhow_3`, `Rhow_4`, `Rhow_5`, `Rhow_6`, `Rhow_7`, `Rhow_8`, `Rhow_9`, `Rhow_10`, `Rhow_11`
+
+Recommended optional columns:
+- `SITE`
+- `SEASON`
+- `ZONE`
+- `SPLIT`
+
+Example RS table:
+
+| SITE | SEASON | ZONE | SPLIT | CHL | SAL | Rhow_1 | Rhow_2 | Rhow_3 | Rhow_4 | Rhow_5 | Rhow_6 | Rhow_7 | Rhow_8 | Rhow_9 | Rhow_10 | Rhow_11 |
+|---|---|---|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
+| S1 | Winter | Upper | TRAIN | 8.0 | 31.0 | 0.011 | 0.012 | 0.013 | 0.014 | 0.015 | 0.016 | 0.017 | 0.018 | 0.019 | 0.020 | 0.021 |
+
+Example:
+
+```python
+from atsimodel.rs_core import run_rs_atsi_pipeline
+
+results = run_rs_atsi_pipeline(
+    input_file="examples/sample_rs_atsi_input.xlsx",
+    output_dir="outputs/RS_ATSI_demo"
+)
+print(results["chl_validation_overall"])
+print(results["atsi_validation_overall"])
+```
+
+## Validation design
+
+### 1. CHL validation
+- R²
+- RMSE
+- MAE
+- Bias
+- train / test / independent validation
+- site-wise validation
+
+### 2. ATSI validation
+- RS-ATSI vs in-situ ATSI
+- confusion matrix by trophic class
+- agreement by station and season
+
+### 3. Spatial validation
+- hotspot consistency
+- estuarine gradient realism
+- upper / middle / lower estuary comparison
+
+## Exported outputs
+
+The package exports:
+- Excel workbook
+- CSV files
+- TXT files
+- JSON files
+
+Typical output tables include:
+- `predictions_all`
+- `chl_validation_overall`
+- `chl_validation_by_site_test`
+- `atsi_validation_overall`
+- `atsi_validation_by_site_test`
+- `atsi_confusion_matrix_test`
+- `station_season_agreement_test`
+- `spatial_zone_summary_test`
+
+## Scientific notes
+
+- ATSI remains **CHL-driven**
+- SAL is used to determine **threshold context**
+- ATSI values are clipped to **0–100**
+- Current class translation is operational:
+  - Unpolluted
+  - Moderate
+  - Eutrophic
+  - Hypertrophic
+
+## Limitations
+
+This release does not yet include:
+- uncertainty quantification
+- prediction intervals
+- raster ingestion
+- geospatial map production
+- SHAP explainability
+
+## Contact
+
+**Dr Md Galal Uddin**  
+School of Engineering, University of Galway, Ireland  
+Email: **jalaluddinbd1987@gmail.com**

rsatsimodel-2.5.0/RSATSIModel.egg-info/PKG-INFO

@@ -0,0 +1,159 @@
+Metadata-Version: 2.4
+Name: RSATSIModel
+Version: 2.5.0
+Summary: Python package for computing ATSI and RS-ATSI using CHL, SAL, and Sentinel-3 OLCI-style data.
+Author-email: Dr Md Galal Uddin <jalaluddinbd1987@gmail.com>
+License: MIT
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: pandas>=1.5.0
+Requires-Dist: numpy>=1.23.0
+Requires-Dist: matplotlib>=3.6.0
+Requires-Dist: openpyxl>=3.1.0
+Requires-Dist: scikit-learn>=1.2.0
+Dynamic: license-file
+
+# REMOTE SENSING (RS)-DRIVEN ATSIModel
+
+## Assessment of Trophic Status Index (ATSI) and RS-ATSI Python Package
+
+**ATSIModel** is an open-source Python package for computing:
+1. **ATSI** from in-situ **chlorophyll-a (CHL)** and **salinity (SAL)** data.
+2. **RS-ATSI** from **Sentinel-3 OLCI style remote-sensing inputs**, with CHL prediction, ATSI derivation, and a structured validation framework.
+
+The package is designed for **transitional, estuarine, and coastal waters**, and supports both standard ATSI computation and remote-sensing based ATSI workflows.
+
+## Author
+
+**Dr Md Galal Uddin**  
+School of Engineering, University of Galway, Ireland  
+Email: **jalaluddinbd1987@gmail.com**
+
+### Research Profiles
+- **Google Scholar**: https://scholar.google.com/citations?user=g6xaAOkAAAAJ&hl=en
+
+## Scientific Background
+
+ATSI is implemented as a **CHL-driven, salinity-conditioned trophic scoring framework**.  
+This package extends that framework into a **remote-sensing compatible RS-ATSI workflow**.
+
+## Standard ATSI workflow
+
+Required columns:
+
+| Column | Description | Unit |
+|---|---|---|
+| CHL | Chlorophyll-a concentration | µg/L |
+| SAL | Salinity | PSU |
+
+Example:
+
+```python
+from atsimodel.core import run_atsi
+df = run_atsi("examples/sample_atsi_input.xlsx")
+print(df.head())
+```
+
+Export:
+
+```python
+from atsimodel.utils import export_single
+export_single(df, out_base="outputs/ATSI_results", formats=("xlsx", "csv", "txt", "json"))
+```
+
+## RS-ATSI workflow
+
+Recommended input columns:
+- `CHL`
+- `SAL`
+- `Rhow_1`, `Rhow_2`, `Rhow_3`, `Rhow_4`, `Rhow_5`, `Rhow_6`, `Rhow_7`, `Rhow_8`, `Rhow_9`, `Rhow_10`, `Rhow_11`
+
+Recommended optional columns:
+- `SITE`
+- `SEASON`
+- `ZONE`
+- `SPLIT`
+
+Example RS table:
+
+| SITE | SEASON | ZONE | SPLIT | CHL | SAL | Rhow_1 | Rhow_2 | Rhow_3 | Rhow_4 | Rhow_5 | Rhow_6 | Rhow_7 | Rhow_8 | Rhow_9 | Rhow_10 | Rhow_11 |
+|---|---|---|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
+| S1 | Winter | Upper | TRAIN | 8.0 | 31.0 | 0.011 | 0.012 | 0.013 | 0.014 | 0.015 | 0.016 | 0.017 | 0.018 | 0.019 | 0.020 | 0.021 |
+
+Example:
+
+```python
+from atsimodel.rs_core import run_rs_atsi_pipeline
+
+results = run_rs_atsi_pipeline(
+    input_file="examples/sample_rs_atsi_input.xlsx",
+    output_dir="outputs/RS_ATSI_demo"
+)
+print(results["chl_validation_overall"])
+print(results["atsi_validation_overall"])
+```
+
+## Validation design
+
+### 1. CHL validation
+- R²
+- RMSE
+- MAE
+- Bias
+- train / test / independent validation
+- site-wise validation
+
+### 2. ATSI validation
+- RS-ATSI vs in-situ ATSI
+- confusion matrix by trophic class
+- agreement by station and season
+
+### 3. Spatial validation
+- hotspot consistency
+- estuarine gradient realism
+- upper / middle / lower estuary comparison
+
+## Exported outputs
+
+The package exports:
+- Excel workbook
+- CSV files
+- TXT files
+- JSON files
+
+Typical output tables include:
+- `predictions_all`
+- `chl_validation_overall`
+- `chl_validation_by_site_test`
+- `atsi_validation_overall`
+- `atsi_validation_by_site_test`
+- `atsi_confusion_matrix_test`
+- `station_season_agreement_test`
+- `spatial_zone_summary_test`
+
+## Scientific notes
+
+- ATSI remains **CHL-driven**
+- SAL is used to determine **threshold context**
+- ATSI values are clipped to **0–100**
+- Current class translation is operational:
+  - Unpolluted
+  - Moderate
+  - Eutrophic
+  - Hypertrophic
+
+## Limitations
+
+This release does not yet include:
+- uncertainty quantification
+- prediction intervals
+- raster ingestion
+- geospatial map production
+- SHAP explainability
+
+## Contact
+
+**Dr Md Galal Uddin**  
+School of Engineering, University of Galway, Ireland  
+Email: **jalaluddinbd1987@gmail.com**

rsatsimodel-2.5.0/RSATSIModel.egg-info/SOURCES.txt

@@ -0,0 +1,31 @@
+LICENSE
+MANIFEST.in
+README.md
+pyproject.toml
+setup.py
+RSATSIModel.egg-info/PKG-INFO
+RSATSIModel.egg-info/SOURCES.txt
+RSATSIModel.egg-info/dependency_links.txt
+RSATSIModel.egg-info/requires.txt
+RSATSIModel.egg-info/top_level.txt
+atsimodel/__init__.py
+atsimodel/atsifunction.py
+atsimodel/chl_models.py
+atsimodel/classification.py
+atsimodel/core.py
+atsimodel/io.py
+atsimodel/metrics.py
+atsimodel/olci_features.py
+atsimodel/plotting.py
+atsimodel/preprocessing.py
+atsimodel/rs_core.py
+atsimodel/thresholds.py
+atsimodel/utils.py
+atsimodel/validation.py
+examples/sample_atsi_input.xlsx
+examples/sample_rs_atsi_input.csv
+examples/sample_rs_atsi_input.txt
+examples/sample_rs_atsi_input.xlsx
+examples/sample_rs_atsi_rhow_input.csv
+examples/sample_rs_atsi_rhow_input.txt
+examples/sample_rs_atsi_rhow_input.xlsx

rsatsimodel-2.5.0/RSATSIModel.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

rsatsimodel-2.5.0/RSATSIModel.egg-info/requires.txt

@@ -0,0 +1,5 @@
+pandas>=1.5.0
+numpy>=1.23.0
+matplotlib>=3.6.0
+openpyxl>=3.1.0
+scikit-learn>=1.2.0

rsatsimodel-2.5.0/RSATSIModel.egg-info/top_level.txt

@@ -0,0 +1 @@
+atsimodel

rsatsimodel-2.5.0/atsimodel/__init__.py

@@ -0,0 +1,5 @@
+from .core import run_atsi
+from .rs_core import run_rs_atsi_pipeline
+
+__version__ = "2.5.0"
+__author__ = "Dr Md Galal Uddin"

rsatsimodel-2.5.0/atsimodel/atsifunction.py

@@ -0,0 +1,20 @@
+import numpy as np
+import pandas as pd
+
+def clip_0_100(x):
+    if pd.isna(x):
+        return np.nan
+    return float(np.clip(x, 0.0, 100.0))
+
+def compute_atsi_score(chl_value, chl_threshold_low, chl_threshold_upper):
+    NFh = 100.0
+    NFl = 0.0
+    if pd.isna(chl_value) or pd.isna(chl_threshold_low) or pd.isna(chl_threshold_upper):
+        return np.nan
+    if chl_threshold_upper <= 0:
+        return np.nan
+    atsi = (NFh - NFl) - (((chl_value - chl_threshold_low) / chl_threshold_upper) * NFh)
+    return clip_0_100(atsi)
+
+def compute_atsi_series(df, chl_col="CHL", low_col="CHL_TL", upper_col="CHL_TU"):
+    return df.apply(lambda row: compute_atsi_score(row[chl_col], row[low_col], row[upper_col]), axis=1)

rsatsimodel-2.5.0/atsimodel/chl_models.py

@@ -0,0 +1,37 @@
+from dataclasses import dataclass
+import numpy as np
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.impute import SimpleImputer
+from sklearn.pipeline import Pipeline
+from .metrics import regression_metrics
+
+@dataclass
+class CHLModelBundle:
+    model: object
+    feature_cols: list
+    train_metrics: dict
+    test_metrics: dict
+
+def fit_rf_chl_model(train_df, test_df, feature_cols, target_col="CHL", random_state=42):
+    X_train = train_df[feature_cols].copy()
+    y_train = train_df[target_col].copy()
+    X_test = test_df[feature_cols].copy()
+    y_test = test_df[target_col].copy()
+    model = Pipeline(steps=[
+        ("imputer", SimpleImputer(strategy="median")),
+        ("rf", RandomForestRegressor(n_estimators=500, random_state=random_state, n_jobs=-1)),
+    ])
+    model.fit(X_train, y_train)
+    pred_train = np.clip(model.predict(X_train), 0, None)
+    pred_test = np.clip(model.predict(X_test), 0, None)
+    train_metrics = regression_metrics(y_train, pred_train, label="Train")
+    test_metrics = regression_metrics(y_test, pred_test, label="Test")
+    bundle = CHLModelBundle(model=model, feature_cols=feature_cols, train_metrics=train_metrics, test_metrics=test_metrics)
+    return bundle, pred_train, pred_test
+
+def predict_chl(df, bundle, output_col="CHL_RS"):
+    X = df[bundle.feature_cols].copy()
+    pred = np.clip(bundle.model.predict(X), 0, None)
+    out = df.copy()
+    out[output_col] = pred
+    return out

rsatsimodel-2.5.0/atsimodel/classification.py

@@ -0,0 +1,17 @@
+import pandas as pd
+
+def classify_atsi(score):
+    if pd.isna(score):
+        return None
+    score = max(0.0, min(100.0, float(score)))
+    if score >= 75:
+        return "Unpolluted"
+    elif score >= 50:
+        return "Moderate"
+    elif score >= 25:
+        return "Eutrophic"
+    else:
+        return "Hypertrophic"
+
+def classify_series(series):
+    return series.apply(classify_atsi)

rsatsimodel-2.5.0/atsimodel/core.py

@@ -0,0 +1,14 @@
+from .io import load_data
+from .preprocessing import validate_atsi_input
+from .thresholds import assign_thresholds
+from .atsifunction import compute_atsi_series
+from .classification import classify_series
+
+def run_atsi(file_path):
+    df = load_data(file_path)
+    df = validate_atsi_input(df)
+    df = assign_thresholds(df, sal_col="SAL")
+    df["ATSI"] = compute_atsi_series(df)
+    df["ATSI"] = df["ATSI"].clip(lower=0, upper=100)
+    df["ATSI_Class"] = classify_series(df["ATSI"])
+    return df

rsatsimodel-2.5.0/atsimodel/io.py

@@ -0,0 +1,30 @@
+from pathlib import Path
+import pandas as pd
+
+def load_data(file_path):
+    file_path = Path(file_path)
+    if not file_path.exists():
+        raise FileNotFoundError(f"File not found: {file_path}")
+    suffix = file_path.suffix.lower()
+    if suffix in [".xlsx", ".xls"]:
+        return pd.read_excel(file_path)
+    if suffix == ".csv":
+        return pd.read_csv(file_path)
+    if suffix in [".txt", ".tsv"]:
+        return pd.read_csv(file_path, sep=None, engine="python")
+    raise ValueError(f"Unsupported file format: {suffix}. Use xlsx/xls/csv/txt/tsv.")
+
+def save_table(df, file_path):
+    file_path = Path(file_path)
+    file_path.parent.mkdir(parents=True, exist_ok=True)
+    suffix = file_path.suffix.lower()
+    if suffix in [".xlsx", ".xls"]:
+        df.to_excel(file_path, index=False)
+    elif suffix == ".csv":
+        df.to_csv(file_path, index=False)
+    elif suffix in [".txt", ".tsv"]:
+        df.to_csv(file_path, index=False, sep="\\t")
+    elif suffix == ".json":
+        df.to_json(file_path, orient="records", indent=4)
+    else:
+        raise ValueError(f"Unsupported output format: {suffix}")

rsatsimodel-2.5.0/atsimodel/metrics.py

@@ -0,0 +1,41 @@
+import numpy as np
+import pandas as pd
+from sklearn.metrics import r2_score, mean_squared_error, mean_absolute_error, confusion_matrix
+
+def rmse(y_true, y_pred):
+    return float(np.sqrt(mean_squared_error(y_true, y_pred)))
+
+def mbe(y_true, y_pred):
+    y_true = np.asarray(y_true)
+    y_pred = np.asarray(y_pred)
+    return float(np.mean(y_pred - y_true))
+
+def agreement_rate(y_true_labels, y_pred_labels):
+    y_true_labels = pd.Series(y_true_labels).astype(str)
+    y_pred_labels = pd.Series(y_pred_labels).astype(str)
+    return float((y_true_labels == y_pred_labels).mean())
+
+def regression_metrics(y_true, y_pred, label="All"):
+    y_true = pd.Series(y_true).astype(float)
+    y_pred = pd.Series(y_pred).astype(float)
+    mask = y_true.notna() & y_pred.notna()
+    y_true = y_true[mask]
+    y_pred = y_pred[mask]
+    if len(y_true) == 0:
+        return {"Group": label, "N": 0, "R2": np.nan, "RMSE": np.nan, "MAE": np.nan, "Bias": np.nan}
+    return {
+        "Group": label,
+        "N": int(len(y_true)),
+        "R2": float(r2_score(y_true, y_pred)) if len(y_true) >= 2 else np.nan,
+        "RMSE": rmse(y_true, y_pred),
+        "MAE": float(mean_absolute_error(y_true, y_pred)),
+        "Bias": mbe(y_true, y_pred),
+    }
+
+def confusion_matrix_table(y_true_labels, y_pred_labels, labels=None):
+    y_true_labels = pd.Series(y_true_labels).astype(str)
+    y_pred_labels = pd.Series(y_pred_labels).astype(str)
+    if labels is None:
+        labels = sorted(set(y_true_labels.unique()).union(set(y_pred_labels.unique())))
+    cm = confusion_matrix(y_true_labels, y_pred_labels, labels=labels)
+    return pd.DataFrame(cm, index=[f"True_{x}" for x in labels], columns=[f"Pred_{x}" for x in labels])

rsatsimodel-2.5.0/atsimodel/olci_features.py

@@ -0,0 +1,55 @@
+import numpy as np
+import pandas as pd
+
+# Sentinel-3 OLCI reflectance style inputs expected from tabular exports
+DEFAULT_OLCI_FEATURES = [f"RHOW_{i}" for i in range(1, 12)]
+
+
+def add_olci_features(df):
+    """
+    Add engineered features from Sentinel-3 OLCI-style reflectance inputs.
+
+    Required raw inputs:
+    RHOW_1 ... RHOW_11
+
+    These are intended to match user-exported tabular reflectance columns such as:
+    Rhow_1, Rhow_2, ..., Rhow_11
+    """
+    df = df.copy()
+    eps = 1e-9
+
+    for c in DEFAULT_OLCI_FEATURES:
+        df[c] = pd.to_numeric(df[c], errors="coerce")
+
+    # Core ratios
+    df["R_6_1"] = df["RHOW_6"] / (df["RHOW_1"] + eps)
+    df["R_6_2"] = df["RHOW_6"] / (df["RHOW_2"] + eps)
+    df["R_7_3"] = df["RHOW_7"] / (df["RHOW_3"] + eps)
+    df["R_8_4"] = df["RHOW_8"] / (df["RHOW_4"] + eps)
+    df["R_9_5"] = df["RHOW_9"] / (df["RHOW_5"] + eps)
+    df["R_10_6"] = df["RHOW_10"] / (df["RHOW_6"] + eps)
+    df["R_11_7"] = df["RHOW_11"] / (df["RHOW_7"] + eps)
+
+    # Normalized differences
+    pairs = [(6,1),(6,2),(7,3),(8,4),(9,5),(10,6),(11,7)]
+    for a,b in pairs:
+        df[f"ND_{a}_{b}"] = (df[f"RHOW_{a}"] - df[f"RHOW_{b}"]) / (df[f"RHOW_{a}"] + df[f"RHOW_{b}"] + eps)
+
+    # Adjacent band slopes / differences
+    for i in range(1, 11):
+        df[f"DIFF_{i}_{i+1}"] = df[f"RHOW_{i+1}"] - df[f"RHOW_{i}"]
+
+    # Logs
+    for c in DEFAULT_OLCI_FEATURES:
+        df[f"LOG_{c}"] = np.log1p(np.clip(df[c], a_min=0, a_max=None))
+
+    return df
+
+
+def get_default_model_features():
+    feats = DEFAULT_OLCI_FEATURES.copy()
+    feats += ["R_6_1", "R_6_2", "R_7_3", "R_8_4", "R_9_5", "R_10_6", "R_11_7"]
+    feats += [f"ND_{a}_{b}" for a,b in [(6,1),(6,2),(7,3),(8,4),(9,5),(10,6),(11,7)]]
+    feats += [f"DIFF_{i}_{i+1}" for i in range(1, 11)]
+    feats += [f"LOG_RHOW_{i}" for i in range(1, 12)]
+    return feats

rsatsimodel-2.5.0/atsimodel/plotting.py

@@ -0,0 +1,19 @@
+import matplotlib.pyplot as plt
+
+def plot_atsi_distribution(df, score_col="ATSI"):
+    plt.figure(figsize=(8, 5))
+    plt.hist(df[score_col].dropna(), bins=20, edgecolor="black")
+    plt.xlabel("ATSI Score")
+    plt.ylabel("Frequency")
+    plt.title("Distribution of ATSI Scores")
+    plt.tight_layout()
+    plt.show()
+
+def plot_chl_vs_atsi(df, chl_col="CHL", score_col="ATSI"):
+    plt.figure(figsize=(8, 5))
+    plt.scatter(df[chl_col], df[score_col])
+    plt.xlabel("CHL (µg/L)")
+    plt.ylabel("ATSI Score")
+    plt.title("CHL vs ATSI")
+    plt.tight_layout()
+    plt.show()

rsatsimodel-2.5.0/atsimodel/preprocessing.py

@@ -0,0 +1,29 @@
+import pandas as pd
+
+def standardize_columns(df):
+    df = df.copy()
+    df.columns = [str(c).strip().upper() for c in df.columns]
+    return df
+
+def validate_atsi_input(df):
+    df = standardize_columns(df)
+    required = ["CHL", "SAL"]
+    missing = [c for c in required if c not in df.columns]
+    if missing:
+        raise ValueError(f"Missing required columns: {missing}. ATSI requires CHL and SAL.")
+    df["CHL"] = pd.to_numeric(df["CHL"], errors="coerce")
+    df["SAL"] = pd.to_numeric(df["SAL"], errors="coerce")
+    return df
+
+def validate_rs_input(df, feature_cols, sal_col="SAL", target_col=None):
+    df = standardize_columns(df)
+    feature_cols = [c.upper() for c in feature_cols]
+    sal_col = sal_col.upper()
+    target_col = target_col.upper() if target_col else None
+    required = feature_cols + [sal_col] + ([target_col] if target_col else [])
+    missing = [c for c in required if c not in df.columns]
+    if missing:
+        raise ValueError(f"Missing RS input columns: {missing}")
+    for c in feature_cols + [sal_col] + ([target_col] if target_col else []):
+        df[c] = pd.to_numeric(df[c], errors="coerce")
+    return df

rsatsimodel-2.5.0/atsimodel/rs_core.py

@@ -0,0 +1,92 @@
+import pandas as pd
+from sklearn.model_selection import train_test_split
+from .io import load_data
+from .preprocessing import validate_rs_input, standardize_columns
+from .olci_features import add_olci_features, get_default_model_features
+from .chl_models import fit_rf_chl_model, predict_chl
+from .thresholds import assign_thresholds
+from .atsifunction import compute_atsi_series
+from .classification import classify_series
+from .validation import validate_chl_overall, validate_chl_by_group, validate_atsi_regression, validate_atsi_by_group, validate_atsi_classification, validate_station_season_agreement, spatial_zone_summary
+from .utils import export_results_dict
+
+def _split_train_test_independent(df, split_col="SPLIT"):
+    df = df.copy()
+    if split_col in df.columns:
+        train_df = df[df[split_col].astype(str).str.upper() == "TRAIN"].copy()
+        test_df = df[df[split_col].astype(str).str.upper() == "TEST"].copy()
+        ind_df = df[df[split_col].astype(str).str.upper().isin(["INDEPENDENT", "VALIDATION"])].copy()
+        if len(train_df) == 0 or len(test_df) == 0:
+            raise ValueError("When SPLIT exists, it must contain at least TRAIN and TEST rows.")
+        return train_df, test_df, ind_df
+    train_df, test_df = train_test_split(df, test_size=0.25, random_state=42)
+    return train_df.copy(), test_df.copy(), pd.DataFrame(columns=df.columns)
+
+def _compute_atsi_from_chl_sal(df, chl_col, sal_col="SAL", out_score_col="ATSI_RS", out_class_col="ATSI_RS_CLASS"):
+    out = df.copy()
+    out = out.rename(columns={sal_col: "SAL"})
+    out = assign_thresholds(out, sal_col="SAL")
+    out[out_score_col] = compute_atsi_series(out, chl_col=chl_col, low_col="CHL_TL", upper_col="CHL_TU").clip(lower=0, upper=100)
+    out[out_class_col] = classify_series(out[out_score_col])
+    return out
+
+def run_rs_atsi_pipeline(input_file, output_dir="outputs_rs_atsi", target_col="CHL", sal_col="SAL", feature_cols=None, site_col="SITE", season_col="SEASON", zone_col="ZONE"):
+    df = load_data(input_file)
+    df = standardize_columns(df)
+    df = add_olci_features(df)
+    if feature_cols is None:
+        feature_cols = get_default_model_features()
+    feature_cols = [c.upper() for c in feature_cols]
+    df = validate_rs_input(df, feature_cols=feature_cols, sal_col=sal_col, target_col=target_col)
+    train_df, test_df, ind_df = _split_train_test_independent(df, split_col="SPLIT")
+    bundle, pred_train, pred_test = fit_rf_chl_model(train_df, test_df, feature_cols=feature_cols, target_col=target_col.upper())
+    train_df["CHL_RS"] = pred_train
+    test_df["CHL_RS"] = pred_test
+    if len(ind_df) > 0:
+        ind_df = predict_chl(ind_df, bundle, output_col="CHL_RS")
+    train_df = _compute_atsi_from_chl_sal(train_df, chl_col=target_col.upper(), sal_col=sal_col.upper(), out_score_col="ATSI_INSITU", out_class_col="ATSI_INSITU_CLASS")
+    train_df = _compute_atsi_from_chl_sal(train_df, chl_col="CHL_RS", sal_col=sal_col.upper(), out_score_col="ATSI_RS", out_class_col="ATSI_RS_CLASS")
+    test_df = _compute_atsi_from_chl_sal(test_df, chl_col=target_col.upper(), sal_col=sal_col.upper(), out_score_col="ATSI_INSITU", out_class_col="ATSI_INSITU_CLASS")
+    test_df = _compute_atsi_from_chl_sal(test_df, chl_col="CHL_RS", sal_col=sal_col.upper(), out_score_col="ATSI_RS", out_class_col="ATSI_RS_CLASS")
+    if len(ind_df) > 0:
+        ind_df = _compute_atsi_from_chl_sal(ind_df, chl_col=target_col.upper(), sal_col=sal_col.upper(), out_score_col="ATSI_INSITU", out_class_col="ATSI_INSITU_CLASS")
+        ind_df = _compute_atsi_from_chl_sal(ind_df, chl_col="CHL_RS", sal_col=sal_col.upper(), out_score_col="ATSI_RS", out_class_col="ATSI_RS_CLASS")
+    validation_tables = {}
+    validation_tables["chl_validation_overall"] = pd.concat([
+        validate_chl_overall(train_df, actual_col=target_col.upper(), pred_col="CHL_RS", group_label="Train"),
+        validate_chl_overall(test_df, actual_col=target_col.upper(), pred_col="CHL_RS", group_label="Test"),
+        validate_chl_overall(ind_df, actual_col=target_col.upper(), pred_col="CHL_RS", group_label="Independent") if len(ind_df) > 0 else pd.DataFrame(),
+    ], ignore_index=True)
+    validation_tables["chl_validation_by_site_test"] = validate_chl_by_group(test_df, actual_col=target_col.upper(), pred_col="CHL_RS", group_col=site_col.upper())
+    if len(ind_df) > 0:
+        validation_tables["chl_validation_by_site_independent"] = validate_chl_by_group(ind_df, actual_col=target_col.upper(), pred_col="CHL_RS", group_col=site_col.upper())
+    validation_tables["atsi_validation_overall"] = pd.concat([
+        validate_atsi_regression(train_df, actual_col="ATSI_INSITU", pred_col="ATSI_RS", group_label="Train"),
+        validate_atsi_regression(test_df, actual_col="ATSI_INSITU", pred_col="ATSI_RS", group_label="Test"),
+        validate_atsi_regression(ind_df, actual_col="ATSI_INSITU", pred_col="ATSI_RS", group_label="Independent") if len(ind_df) > 0 else pd.DataFrame(),
+    ], ignore_index=True)
+    validation_tables["atsi_validation_by_site_test"] = validate_atsi_by_group(test_df, actual_col="ATSI_INSITU", pred_col="ATSI_RS", group_col=site_col.upper())
+    if len(ind_df) > 0:
+        validation_tables["atsi_validation_by_site_independent"] = validate_atsi_by_group(ind_df, actual_col="ATSI_INSITU", pred_col="ATSI_RS", group_col=site_col.upper())
+    class_test = validate_atsi_classification(test_df, actual_score_col="ATSI_INSITU", pred_score_col="ATSI_RS")
+    validation_tables["atsi_class_summary_test"] = class_test["summary"]
+    validation_tables["atsi_confusion_matrix_test"] = class_test["confusion_matrix"]
+    if len(ind_df) > 0:
+        class_ind = validate_atsi_classification(ind_df, actual_score_col="ATSI_INSITU", pred_score_col="ATSI_RS")
+        validation_tables["atsi_class_summary_independent"] = class_ind["summary"]
+        validation_tables["atsi_confusion_matrix_independent"] = class_ind["confusion_matrix"]
+    validation_tables["station_season_agreement_test"] = validate_station_season_agreement(test_df, site_col=site_col.upper(), season_col=season_col.upper(), actual_score_col="ATSI_INSITU", pred_score_col="ATSI_RS")
+    if len(ind_df) > 0:
+        validation_tables["station_season_agreement_independent"] = validate_station_season_agreement(ind_df, site_col=site_col.upper(), season_col=season_col.upper(), actual_score_col="ATSI_INSITU", pred_score_col="ATSI_RS")
+    validation_tables["spatial_zone_summary_test"] = spatial_zone_summary(test_df, zone_col=zone_col.upper(), actual_score_col="ATSI_INSITU", pred_score_col="ATSI_RS")
+    if len(ind_df) > 0:
+        validation_tables["spatial_zone_summary_independent"] = spatial_zone_summary(ind_df, zone_col=zone_col.upper(), actual_score_col="ATSI_INSITU", pred_score_col="ATSI_RS")
+    validation_tables["chl_model_metrics_summary"] = pd.DataFrame([bundle.train_metrics, bundle.test_metrics])
+    predictions_all = pd.concat([
+        train_df.assign(DATASET="Train"),
+        test_df.assign(DATASET="Test"),
+        ind_df.assign(DATASET="Independent") if len(ind_df) > 0 else pd.DataFrame()
+    ], ignore_index=True)
+    results = {"predictions_all": predictions_all, **validation_tables}
+    export_results_dict(results, output_dir=output_dir)
+    return results

rsatsimodel-2.5.0/atsimodel/thresholds.py

@@ -0,0 +1,62 @@
+import pandas as pd
+
+SAL_THRESHOLD_TABLE = {
+    0: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    1: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    2: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    3: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    4: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    5: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    6: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    7: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    8: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    9: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    10: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    11: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    12: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    13: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    14: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    15: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    16: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    17: {"CHL_TL": 15.0, "CHL_TU": 30.0},
+    18: {"CHL_TL": 14.7, "CHL_TU": 29.4},
+    19: {"CHL_TL": 14.4, "CHL_TU": 28.9},
+    20: {"CHL_TL": 14.2, "CHL_TU": 28.3},
+    21: {"CHL_TL": 13.9, "CHL_TU": 27.8},
+    22: {"CHL_TL": 13.6, "CHL_TU": 27.2},
+    23: {"CHL_TL": 13.3, "CHL_TU": 26.7},
+    24: {"CHL_TL": 13.1, "CHL_TU": 26.1},
+    25: {"CHL_TL": 12.8, "CHL_TU": 25.6},
+    26: {"CHL_TL": 12.5, "CHL_TU": 25.0},
+    27: {"CHL_TL": 12.2, "CHL_TU": 24.4},
+    28: {"CHL_TL": 11.9, "CHL_TU": 23.9},
+    29: {"CHL_TL": 11.7, "CHL_TU": 23.3},
+    30: {"CHL_TL": 11.4, "CHL_TU": 22.8},
+    31: {"CHL_TL": 11.1, "CHL_TU": 22.2},
+    32: {"CHL_TL": 10.8, "CHL_TU": 21.7},
+    33: {"CHL_TL": 10.6, "CHL_TU": 21.1},
+    34: {"CHL_TL": 10.3, "CHL_TU": 20.6},
+    35: {"CHL_TL": 10.0, "CHL_TU": 20.0}
+}
+
+def round_salinity_to_median_bin(sal):
+    if pd.isna(sal):
+        return None
+    sal_bin = int(round(float(sal)))
+    return max(0, min(35, sal_bin))
+
+def get_chl_thresholds_from_salinity(sal):
+    sal_bin = round_salinity_to_median_bin(sal)
+    if sal_bin is None:
+        return {"SAL_BIN": None, "CHL_TL": None, "CHL_TU": None}
+    vals = SAL_THRESHOLD_TABLE[sal_bin]
+    return {"SAL_BIN": sal_bin, "CHL_TL": vals["CHL_TL"], "CHL_TU": vals["CHL_TU"]}
+
+def assign_thresholds(df, sal_col="SAL"):
+    df = df.copy()
+    sal_col = sal_col.upper()
+    out = df[sal_col].apply(get_chl_thresholds_from_salinity).apply(pd.Series)
+    df["SAL_BIN"] = out["SAL_BIN"]
+    df["CHL_TL"] = out["CHL_TL"]
+    df["CHL_TU"] = out["CHL_TU"]
+    return df

rsatsimodel-2.5.0/atsimodel/utils.py

@@ -0,0 +1,28 @@
+from pathlib import Path
+import pandas as pd
+
+def export_results_dict(results_dict, output_dir):
+    output_dir = Path(output_dir)
+    output_dir.mkdir(parents=True, exist_ok=True)
+    excel_path = output_dir / "RS_ATSI_results_bundle.xlsx"
+    with pd.ExcelWriter(excel_path, engine="openpyxl") as writer:
+        for key, value in results_dict.items():
+            if isinstance(value, pd.DataFrame):
+                value.to_excel(writer, sheet_name=key[:31], index=False)
+    for key, value in results_dict.items():
+        if isinstance(value, pd.DataFrame):
+            value.to_csv(output_dir / f"{key}.csv", index=False)
+            value.to_csv(output_dir / f"{key}.txt", index=False, sep="\t")
+            value.to_json(output_dir / f"{key}.json", orient="records", indent=4)
+
+def export_single(df, out_base="ATSI_results", formats=("xlsx", "csv", "txt", "json")):
+    out_base = Path(out_base)
+    out_base.parent.mkdir(parents=True, exist_ok=True)
+    if "xlsx" in formats:
+        df.to_excel(f"{out_base}.xlsx", index=False)
+    if "csv" in formats:
+        df.to_csv(f"{out_base}.csv", index=False)
+    if "txt" in formats:
+        df.to_csv(f"{out_base}.txt", index=False, sep="\t")
+    if "json" in formats:
+        df.to_json(f"{out_base}.json", orient="records", indent=4)

rsatsimodel-2.5.0/atsimodel/validation.py

@@ -0,0 +1,70 @@
+import pandas as pd
+from .metrics import regression_metrics, agreement_rate, confusion_matrix_table
+from .classification import classify_series
+
+def validate_chl_overall(df, actual_col="CHL", pred_col="CHL_RS", group_label="Overall"):
+    return pd.DataFrame([regression_metrics(df[actual_col], df[pred_col], label=group_label)])
+
+def validate_chl_by_group(df, actual_col="CHL", pred_col="CHL_RS", group_col="SITE"):
+    rows = []
+    if group_col not in df.columns:
+        return pd.DataFrame()
+    for g, sub in df.groupby(group_col):
+        rows.append(regression_metrics(sub[actual_col], sub[pred_col], label=str(g)))
+    return pd.DataFrame(rows)
+
+def validate_atsi_regression(df, actual_col="ATSI_INSITU", pred_col="ATSI_RS", group_label="Overall"):
+    return pd.DataFrame([regression_metrics(df[actual_col], df[pred_col], label=group_label)])
+
+def validate_atsi_by_group(df, actual_col="ATSI_INSITU", pred_col="ATSI_RS", group_col="SITE"):
+    rows = []
+    if group_col not in df.columns:
+        return pd.DataFrame()
+    for g, sub in df.groupby(group_col):
+        rows.append(regression_metrics(sub[actual_col], sub[pred_col], label=str(g)))
+    return pd.DataFrame(rows)
+
+def validate_atsi_classification(df, actual_score_col="ATSI_INSITU", pred_score_col="ATSI_RS",
+                                 actual_class_col=None, pred_class_col=None):
+    out = {}
+    df = df.copy()
+    if actual_class_col is None:
+        actual_class_col = "__ATSI_CLASS_TRUE__"
+        df[actual_class_col] = classify_series(df[actual_score_col])
+    if pred_class_col is None:
+        pred_class_col = "__ATSI_CLASS_PRED__"
+        df[pred_class_col] = classify_series(df[pred_score_col])
+    out["summary"] = pd.DataFrame([{
+        "Agreement_Rate": agreement_rate(df[actual_class_col], df[pred_class_col]),
+        "N": int(len(df))
+    }])
+    out["confusion_matrix"] = confusion_matrix_table(df[actual_class_col], df[pred_class_col])
+    return out
+
+def validate_station_season_agreement(df, site_col="SITE", season_col="SEASON",
+                                      actual_score_col="ATSI_INSITU", pred_score_col="ATSI_RS"):
+    rows = []
+    df = df.copy()
+    df["ATSI_CLASS_TRUE"] = classify_series(df[actual_score_col])
+    df["ATSI_CLASS_PRED"] = classify_series(df[pred_score_col])
+    group_cols = []
+    if site_col in df.columns: group_cols.append(site_col)
+    if season_col in df.columns: group_cols.append(season_col)
+    if not group_cols: return pd.DataFrame()
+    for g, sub in df.groupby(group_cols):
+        if not isinstance(g, tuple): g = (g,)
+        row = {col: val for col, val in zip(group_cols, g)}
+        row["Agreement_Rate"] = agreement_rate(sub["ATSI_CLASS_TRUE"], sub["ATSI_CLASS_PRED"])
+        row["N"] = int(len(sub))
+        rows.append(row)
+    return pd.DataFrame(rows)
+
+def spatial_zone_summary(df, zone_col="ZONE", actual_score_col="ATSI_INSITU", pred_score_col="ATSI_RS"):
+    if zone_col not in df.columns:
+        return pd.DataFrame()
+    rows = []
+    for g, sub in df.groupby(zone_col):
+        row = regression_metrics(sub[actual_score_col], sub[pred_score_col], label=str(g))
+        row[zone_col] = g
+        rows.append(row)
+    return pd.DataFrame(rows)

rsatsimodel-2.5.0/examples/sample_rs_atsi_input.csv

@@ -0,0 +1,25 @@
+SITE,SEASON,ZONE,SPLIT,CHL,SAL,Rhow_1,Rhow_2,Rhow_3,Rhow_4,Rhow_5,Rhow_6,Rhow_7,Rhow_8,Rhow_9,Rhow_10,Rhow_11
+S1,Winter,Upper,TRAIN,17.694,20.95,0.0154,0.01285,0.00539,0.02886,0.0062,0.00795,0.02391,0.00521,0.02326,0.00531,0.02649
+S2,Spring,Middle,TRAIN,3.238,18.24,0.01458,0.01779,0.02632,0.01676,0.02376,0.01314,0.02592,0.02152,0.02945,0.00215,0.02053
+S3,Summer,Lower,TRAIN,7.068,25.54,0.0047,0.00304,0.00619,0.0207,0.02697,0.00515,0.01458,0.01374,0.00182,0.01121,0.01388
+S4,Autumn,Upper,TRAIN,6.241,29.65,0.00847,0.01516,0.00188,0.02096,0.01436,0.01993,0.00716,0.02757,0.00408,0.02278,0.02706
+S5,Winter,Middle,TRAIN,6.046,32.7,0.00109,0.01869,0.01722,0.00542,0.00212,0.01428,0.0029,0.01287,0.00545,0.02678,0.01874
+S6,Spring,Lower,TRAIN,20.678,28.01,0.01205,0.02257,0.01344,0.00521,0.01014,0.01242,0.00331,0.0018,0.01785,0.00777,0.02077
+S1,Summer,Upper,TRAIN,23.237,32.67,0.0177,0.01347,0.0259,0.0144,0.00144,0.02434,0.00497,0.02206,0.01349,0.02838,0.02167
+S2,Autumn,Middle,TRAIN,8.361,31.84,0.01339,0.00978,0.00566,0.01614,0.01113,0.01431,0.00881,0.02959,0.00286,0.0033,0.01374
+S3,Winter,Lower,TRAIN,20.854,21.49,0.02522,0.00117,0.01145,0.00889,0.02012,0.02106,0.02354,0.01452,0.0091,0.00327,0.0095
+S4,Spring,Upper,TRAIN,22.468,31.86,0.01888,0.02294,0.01957,0.00622,0.00869,0.02368,0.00766,0.02863,0.01233,0.02146,0.02672
+S5,Summer,Middle,TRAIN,13.798,29.69,0.00872,0.00325,0.028,0.01561,0.01991,0.01883,0.01323,0.00572,0.01567,0.00297,0.0117
+S6,Autumn,Lower,TRAIN,7.634,22.45,0.02452,0.01521,0.00859,0.02013,0.0279,0.02888,0.01429,0.00717,0.02292,0.02915,0.01367
+S1,Winter,Upper,TRAIN,20.958,30.75,0.01549,0.00983,0.02095,0.00763,0.01452,0.01698,0.00998,0.0109,0.01518,0.00895,0.00901
+S2,Spring,Middle,TRAIN,6.917,31.84,0.02301,0.02538,0.00532,0.00733,0.00701,0.00735,0.00613,0.01277,0.02495,0.00473,0.01706
+S3,Summer,Lower,TEST,19.054,22.79,0.01742,0.02856,0.02246,0.02869,0.02776,0.01671,0.00877,0.02708,0.01667,0.02329,0.01661
+S4,Autumn,Upper,TEST,16.489,26.43,0.01369,0.01025,0.0095,0.02202,0.02843,0.00809,0.02572,0.0292,0.01136,0.01325,0.01352
+S5,Winter,Middle,TEST,23.33,19.14,0.01249,0.02704,0.00807,0.002,0.00491,0.00342,0.02983,0.0158,0.02437,0.0271,0.01125
+S6,Spring,Lower,TEST,7.334,27.33,0.00164,0.01079,0.00766,0.02947,0.02709,0.02293,0.01175,0.02121,0.0064,0.01528,0.01671
+S1,Summer,Upper,TEST,20.38,21.81,0.01461,0.02455,0.01175,0.00125,0.00947,0.00867,0.01439,0.0192,0.02679,0.00123,0.01208
+S2,Autumn,Middle,TEST,13.918,30.24,0.01908,0.02417,0.01053,0.00868,0.00472,0.0101,0.01071,0.02596,0.02053,0.00574,0.02359
+S3,Winter,Lower,INDEPENDENT,7.326,20.78,0.02844,0.02,0.01877,0.0276,0.00442,0.02966,0.0116,0.00413,0.02865,0.00388,0.00545
+S4,Spring,Upper,INDEPENDENT,5.816,23.0,0.01362,0.00763,0.0168,0.00333,0.02302,0.01262,0.02093,0.01729,0.00709,0.00626,0.02189
+S5,Summer,Middle,INDEPENDENT,13.449,18.23,0.01508,0.00499,0.02274,0.02579,0.00356,0.0138,0.02232,0.02444,0.00775,0.00885,0.00705
+S6,Autumn,Lower,INDEPENDENT,15.403,18.52,0.01605,0.01331,0.00955,0.0052,0.02313,0.0277,0.01019,0.02345,0.01424,0.00319,0.02803

rsatsimodel-2.5.0/examples/sample_rs_atsi_input.txt

@@ -0,0 +1,25 @@
+SITE	SEASON	ZONE	SPLIT	CHL	SAL	Rhow_1	Rhow_2	Rhow_3	Rhow_4	Rhow_5	Rhow_6	Rhow_7	Rhow_8	Rhow_9	Rhow_10	Rhow_11
+S1	Winter	Upper	TRAIN	17.694	20.95	0.0154	0.01285	0.00539	0.02886	0.0062	0.00795	0.02391	0.00521	0.02326	0.00531	0.02649
+S2	Spring	Middle	TRAIN	3.238	18.24	0.01458	0.01779	0.02632	0.01676	0.02376	0.01314	0.02592	0.02152	0.02945	0.00215	0.02053
+S3	Summer	Lower	TRAIN	7.068	25.54	0.0047	0.00304	0.00619	0.0207	0.02697	0.00515	0.01458	0.01374	0.00182	0.01121	0.01388
+S4	Autumn	Upper	TRAIN	6.241	29.65	0.00847	0.01516	0.00188	0.02096	0.01436	0.01993	0.00716	0.02757	0.00408	0.02278	0.02706
+S5	Winter	Middle	TRAIN	6.046	32.7	0.00109	0.01869	0.01722	0.00542	0.00212	0.01428	0.0029	0.01287	0.00545	0.02678	0.01874
+S6	Spring	Lower	TRAIN	20.678	28.01	0.01205	0.02257	0.01344	0.00521	0.01014	0.01242	0.00331	0.0018	0.01785	0.00777	0.02077
+S1	Summer	Upper	TRAIN	23.237	32.67	0.0177	0.01347	0.0259	0.0144	0.00144	0.02434	0.00497	0.02206	0.01349	0.02838	0.02167
+S2	Autumn	Middle	TRAIN	8.361	31.84	0.01339	0.00978	0.00566	0.01614	0.01113	0.01431	0.00881	0.02959	0.00286	0.0033	0.01374
+S3	Winter	Lower	TRAIN	20.854	21.49	0.02522	0.00117	0.01145	0.00889	0.02012	0.02106	0.02354	0.01452	0.0091	0.00327	0.0095
+S4	Spring	Upper	TRAIN	22.468	31.86	0.01888	0.02294	0.01957	0.00622	0.00869	0.02368	0.00766	0.02863	0.01233	0.02146	0.02672
+S5	Summer	Middle	TRAIN	13.798	29.69	0.00872	0.00325	0.028	0.01561	0.01991	0.01883	0.01323	0.00572	0.01567	0.00297	0.0117
+S6	Autumn	Lower	TRAIN	7.634	22.45	0.02452	0.01521	0.00859	0.02013	0.0279	0.02888	0.01429	0.00717	0.02292	0.02915	0.01367
+S1	Winter	Upper	TRAIN	20.958	30.75	0.01549	0.00983	0.02095	0.00763	0.01452	0.01698	0.00998	0.0109	0.01518	0.00895	0.00901
+S2	Spring	Middle	TRAIN	6.917	31.84	0.02301	0.02538	0.00532	0.00733	0.00701	0.00735	0.00613	0.01277	0.02495	0.00473	0.01706
+S3	Summer	Lower	TEST	19.054	22.79	0.01742	0.02856	0.02246	0.02869	0.02776	0.01671	0.00877	0.02708	0.01667	0.02329	0.01661
+S4	Autumn	Upper	TEST	16.489	26.43	0.01369	0.01025	0.0095	0.02202	0.02843	0.00809	0.02572	0.0292	0.01136	0.01325	0.01352
+S5	Winter	Middle	TEST	23.33	19.14	0.01249	0.02704	0.00807	0.002	0.00491	0.00342	0.02983	0.0158	0.02437	0.0271	0.01125
+S6	Spring	Lower	TEST	7.334	27.33	0.00164	0.01079	0.00766	0.02947	0.02709	0.02293	0.01175	0.02121	0.0064	0.01528	0.01671
+S1	Summer	Upper	TEST	20.38	21.81	0.01461	0.02455	0.01175	0.00125	0.00947	0.00867	0.01439	0.0192	0.02679	0.00123	0.01208
+S2	Autumn	Middle	TEST	13.918	30.24	0.01908	0.02417	0.01053	0.00868	0.00472	0.0101	0.01071	0.02596	0.02053	0.00574	0.02359
+S3	Winter	Lower	INDEPENDENT	7.326	20.78	0.02844	0.02	0.01877	0.0276	0.00442	0.02966	0.0116	0.00413	0.02865	0.00388	0.00545
+S4	Spring	Upper	INDEPENDENT	5.816	23.0	0.01362	0.00763	0.0168	0.00333	0.02302	0.01262	0.02093	0.01729	0.00709	0.00626	0.02189
+S5	Summer	Middle	INDEPENDENT	13.449	18.23	0.01508	0.00499	0.02274	0.02579	0.00356	0.0138	0.02232	0.02444	0.00775	0.00885	0.00705
+S6	Autumn	Lower	INDEPENDENT	15.403	18.52	0.01605	0.01331	0.00955	0.0052	0.02313	0.0277	0.01019	0.02345	0.01424	0.00319	0.02803

rsatsimodel-2.5.0/examples/sample_rs_atsi_rhow_input.csv

@@ -0,0 +1,25 @@
+SITE,SEASON,ZONE,SPLIT,CHL,SAL,Rhow_1,Rhow_2,Rhow_3,Rhow_4,Rhow_5,Rhow_6,Rhow_7,Rhow_8,Rhow_9,Rhow_10,Rhow_11
+S1,Winter,Upper,TRAIN,17.694,20.95,0.0154,0.01285,0.00539,0.02886,0.0062,0.00795,0.02391,0.00521,0.02326,0.00531,0.02649
+S2,Spring,Middle,TRAIN,3.238,18.24,0.01458,0.01779,0.02632,0.01676,0.02376,0.01314,0.02592,0.02152,0.02945,0.00215,0.02053
+S3,Summer,Lower,TRAIN,7.068,25.54,0.0047,0.00304,0.00619,0.0207,0.02697,0.00515,0.01458,0.01374,0.00182,0.01121,0.01388
+S4,Autumn,Upper,TRAIN,6.241,29.65,0.00847,0.01516,0.00188,0.02096,0.01436,0.01993,0.00716,0.02757,0.00408,0.02278,0.02706
+S5,Winter,Middle,TRAIN,6.046,32.7,0.00109,0.01869,0.01722,0.00542,0.00212,0.01428,0.0029,0.01287,0.00545,0.02678,0.01874
+S6,Spring,Lower,TRAIN,20.678,28.01,0.01205,0.02257,0.01344,0.00521,0.01014,0.01242,0.00331,0.0018,0.01785,0.00777,0.02077
+S1,Summer,Upper,TRAIN,23.237,32.67,0.0177,0.01347,0.0259,0.0144,0.00144,0.02434,0.00497,0.02206,0.01349,0.02838,0.02167
+S2,Autumn,Middle,TRAIN,8.361,31.84,0.01339,0.00978,0.00566,0.01614,0.01113,0.01431,0.00881,0.02959,0.00286,0.0033,0.01374
+S3,Winter,Lower,TRAIN,20.854,21.49,0.02522,0.00117,0.01145,0.00889,0.02012,0.02106,0.02354,0.01452,0.0091,0.00327,0.0095
+S4,Spring,Upper,TRAIN,22.468,31.86,0.01888,0.02294,0.01957,0.00622,0.00869,0.02368,0.00766,0.02863,0.01233,0.02146,0.02672
+S5,Summer,Middle,TRAIN,13.798,29.69,0.00872,0.00325,0.028,0.01561,0.01991,0.01883,0.01323,0.00572,0.01567,0.00297,0.0117
+S6,Autumn,Lower,TRAIN,7.634,22.45,0.02452,0.01521,0.00859,0.02013,0.0279,0.02888,0.01429,0.00717,0.02292,0.02915,0.01367
+S1,Winter,Upper,TRAIN,20.958,30.75,0.01549,0.00983,0.02095,0.00763,0.01452,0.01698,0.00998,0.0109,0.01518,0.00895,0.00901
+S2,Spring,Middle,TRAIN,6.917,31.84,0.02301,0.02538,0.00532,0.00733,0.00701,0.00735,0.00613,0.01277,0.02495,0.00473,0.01706
+S3,Summer,Lower,TEST,19.054,22.79,0.01742,0.02856,0.02246,0.02869,0.02776,0.01671,0.00877,0.02708,0.01667,0.02329,0.01661
+S4,Autumn,Upper,TEST,16.489,26.43,0.01369,0.01025,0.0095,0.02202,0.02843,0.00809,0.02572,0.0292,0.01136,0.01325,0.01352
+S5,Winter,Middle,TEST,23.33,19.14,0.01249,0.02704,0.00807,0.002,0.00491,0.00342,0.02983,0.0158,0.02437,0.0271,0.01125
+S6,Spring,Lower,TEST,7.334,27.33,0.00164,0.01079,0.00766,0.02947,0.02709,0.02293,0.01175,0.02121,0.0064,0.01528,0.01671
+S1,Summer,Upper,TEST,20.38,21.81,0.01461,0.02455,0.01175,0.00125,0.00947,0.00867,0.01439,0.0192,0.02679,0.00123,0.01208
+S2,Autumn,Middle,TEST,13.918,30.24,0.01908,0.02417,0.01053,0.00868,0.00472,0.0101,0.01071,0.02596,0.02053,0.00574,0.02359
+S3,Winter,Lower,INDEPENDENT,7.326,20.78,0.02844,0.02,0.01877,0.0276,0.00442,0.02966,0.0116,0.00413,0.02865,0.00388,0.00545
+S4,Spring,Upper,INDEPENDENT,5.816,23.0,0.01362,0.00763,0.0168,0.00333,0.02302,0.01262,0.02093,0.01729,0.00709,0.00626,0.02189
+S5,Summer,Middle,INDEPENDENT,13.449,18.23,0.01508,0.00499,0.02274,0.02579,0.00356,0.0138,0.02232,0.02444,0.00775,0.00885,0.00705
+S6,Autumn,Lower,INDEPENDENT,15.403,18.52,0.01605,0.01331,0.00955,0.0052,0.02313,0.0277,0.01019,0.02345,0.01424,0.00319,0.02803

rsatsimodel-2.5.0/examples/sample_rs_atsi_rhow_input.txt

@@ -0,0 +1,25 @@
+SITE	SEASON	ZONE	SPLIT	CHL	SAL	Rhow_1	Rhow_2	Rhow_3	Rhow_4	Rhow_5	Rhow_6	Rhow_7	Rhow_8	Rhow_9	Rhow_10	Rhow_11
+S1	Winter	Upper	TRAIN	17.694	20.95	0.0154	0.01285	0.00539	0.02886	0.0062	0.00795	0.02391	0.00521	0.02326	0.00531	0.02649
+S2	Spring	Middle	TRAIN	3.238	18.24	0.01458	0.01779	0.02632	0.01676	0.02376	0.01314	0.02592	0.02152	0.02945	0.00215	0.02053
+S3	Summer	Lower	TRAIN	7.068	25.54	0.0047	0.00304	0.00619	0.0207	0.02697	0.00515	0.01458	0.01374	0.00182	0.01121	0.01388
+S4	Autumn	Upper	TRAIN	6.241	29.65	0.00847	0.01516	0.00188	0.02096	0.01436	0.01993	0.00716	0.02757	0.00408	0.02278	0.02706
+S5	Winter	Middle	TRAIN	6.046	32.7	0.00109	0.01869	0.01722	0.00542	0.00212	0.01428	0.0029	0.01287	0.00545	0.02678	0.01874
+S6	Spring	Lower	TRAIN	20.678	28.01	0.01205	0.02257	0.01344	0.00521	0.01014	0.01242	0.00331	0.0018	0.01785	0.00777	0.02077
+S1	Summer	Upper	TRAIN	23.237	32.67	0.0177	0.01347	0.0259	0.0144	0.00144	0.02434	0.00497	0.02206	0.01349	0.02838	0.02167
+S2	Autumn	Middle	TRAIN	8.361	31.84	0.01339	0.00978	0.00566	0.01614	0.01113	0.01431	0.00881	0.02959	0.00286	0.0033	0.01374
+S3	Winter	Lower	TRAIN	20.854	21.49	0.02522	0.00117	0.01145	0.00889	0.02012	0.02106	0.02354	0.01452	0.0091	0.00327	0.0095
+S4	Spring	Upper	TRAIN	22.468	31.86	0.01888	0.02294	0.01957	0.00622	0.00869	0.02368	0.00766	0.02863	0.01233	0.02146	0.02672
+S5	Summer	Middle	TRAIN	13.798	29.69	0.00872	0.00325	0.028	0.01561	0.01991	0.01883	0.01323	0.00572	0.01567	0.00297	0.0117
+S6	Autumn	Lower	TRAIN	7.634	22.45	0.02452	0.01521	0.00859	0.02013	0.0279	0.02888	0.01429	0.00717	0.02292	0.02915	0.01367
+S1	Winter	Upper	TRAIN	20.958	30.75	0.01549	0.00983	0.02095	0.00763	0.01452	0.01698	0.00998	0.0109	0.01518	0.00895	0.00901
+S2	Spring	Middle	TRAIN	6.917	31.84	0.02301	0.02538	0.00532	0.00733	0.00701	0.00735	0.00613	0.01277	0.02495	0.00473	0.01706
+S3	Summer	Lower	TEST	19.054	22.79	0.01742	0.02856	0.02246	0.02869	0.02776	0.01671	0.00877	0.02708	0.01667	0.02329	0.01661
+S4	Autumn	Upper	TEST	16.489	26.43	0.01369	0.01025	0.0095	0.02202	0.02843	0.00809	0.02572	0.0292	0.01136	0.01325	0.01352
+S5	Winter	Middle	TEST	23.33	19.14	0.01249	0.02704	0.00807	0.002	0.00491	0.00342	0.02983	0.0158	0.02437	0.0271	0.01125
+S6	Spring	Lower	TEST	7.334	27.33	0.00164	0.01079	0.00766	0.02947	0.02709	0.02293	0.01175	0.02121	0.0064	0.01528	0.01671
+S1	Summer	Upper	TEST	20.38	21.81	0.01461	0.02455	0.01175	0.00125	0.00947	0.00867	0.01439	0.0192	0.02679	0.00123	0.01208
+S2	Autumn	Middle	TEST	13.918	30.24	0.01908	0.02417	0.01053	0.00868	0.00472	0.0101	0.01071	0.02596	0.02053	0.00574	0.02359
+S3	Winter	Lower	INDEPENDENT	7.326	20.78	0.02844	0.02	0.01877	0.0276	0.00442	0.02966	0.0116	0.00413	0.02865	0.00388	0.00545
+S4	Spring	Upper	INDEPENDENT	5.816	23.0	0.01362	0.00763	0.0168	0.00333	0.02302	0.01262	0.02093	0.01729	0.00709	0.00626	0.02189
+S5	Summer	Middle	INDEPENDENT	13.449	18.23	0.01508	0.00499	0.02274	0.02579	0.00356	0.0138	0.02232	0.02444	0.00775	0.00885	0.00705
+S6	Autumn	Lower	INDEPENDENT	15.403	18.52	0.01605	0.01331	0.00955	0.0052	0.02313	0.0277	0.01019	0.02345	0.01424	0.00319	0.02803

rsatsimodel-2.5.0/pyproject.toml

@@ -0,0 +1,21 @@
+[build-system]
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "RSATSIModel"
+version = "2.5.0"
+description = "Python package for computing ATSI and RS-ATSI using CHL, SAL, and Sentinel-3 OLCI-style data."
+readme = "README.md"
+requires-python = ">=3.9"
+license = {text = "MIT"}
+authors = [
+    {name = "Dr Md Galal Uddin", email = "jalaluddinbd1987@gmail.com"}
+]
+dependencies = [
+    "pandas>=1.5.0",
+    "numpy>=1.23.0",
+    "matplotlib>=3.6.0",
+    "openpyxl>=3.1.0",
+    "scikit-learn>=1.2.0"
+]

rsatsimodel-2.5.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

rsatsimodel-2.5.0/setup.py

@@ -0,0 +1,2 @@
+from setuptools import setup, find_packages
+setup(packages=find_packages())