corection bug figsize

Reviewed-on: #15

.drone.yml

@@ -0,0 +1,44 @@
+kind: pipeline
+name: default
+type: docker
+
+trigger:
+  event:
+    - push
+
+steps:
+  - name: lint
+    image: python:3.12
+    commands:
+      - pip install --root-user-action=ignore -r requirements.txt
+      - ruff check .
+
+  - name: docker-image
+    image: plugins/docker
+    settings:
+      dockerfile: Dockerfile
+      registry: hub.codefirst.iut.uca.fr
+      repo: hub.codefirst.iut.uca.fr/bastien.ollier/miner
+      username:
+        from_secret: REGISTRY_USER
+      password:
+        from_secret: REGISTRY_PASSWORD
+      cache_from:
+        - hub.codefirst.iut.uca.fr/bastien.ollier/miner:latest
+    depends_on: [ lint ]
+
+  - name: deploy-miner
+    image: hub.codefirst.iut.uca.fr/clement.freville2/codefirst-dockerproxy-clientdrone:latest
+    settings:
+      image: hub.codefirst.iut.uca.fr/bastien.ollier/miner:latest
+      container: miner
+      command: create
+      overwrite: true
+      admins: bastienollier,clementfreville2,hugopradier2
+    environment:
+      DRONE_REPO_OWNER: bastien.ollier
+    depends_on: [ docker-image ]
+    when:
+      branch:
+        - main
+        - ci/*

.gitignore

@@ -1 +1,2 @@
 __pycache__
+.venv

Dockerfile

@@ -0,0 +1,9 @@
+FROM python:3.12-slim
+
+WORKDIR /app
+
+COPY . . 
+RUN pip3 install -r requirements.txt
+
+EXPOSE 80
+ENTRYPOINT ["streamlit", "run", "frontend/exploration.py", "--server.port=80", "--server.address=0.0.0.0", "--server.baseUrlPath=/containers/bastienollier-miner"]

frontend/clusters.py

@@ -0,0 +1,83 @@
+from sklearn.cluster import DBSCAN, KMeans
+import numpy as np
+from dataclasses import dataclass
+from abc import ABC, abstractmethod
+from typing import Any, Optional
+
+@dataclass
+class ClusterResult:
+    labels: np.array
+    centers: Optional[np.array]
+    statistics: list[dict[str, Any]]
+
+
+class Cluster(ABC):
+    @abstractmethod
+    def run(self, data: np.array) -> ClusterResult:
+        pass
+
+
+class DBSCANCluster(Cluster):
+    def __init__(self, eps: float = 0.5, min_samples: int = 5):
+        self.eps = eps
+        self.min_samples = min_samples
+
+    #@typing.override
+    def run(self, data: np.array) -> ClusterResult:
+        dbscan = DBSCAN(eps=self.eps, min_samples=self.min_samples)
+        labels = dbscan.fit_predict(data)
+        return ClusterResult(labels, None, self.get_statistics(data, labels))
+
+    def get_statistics(self, data: np.array, labels: np.array) -> list[dict[str, Any]]:
+        unique_labels = np.unique(labels)
+        stats = []
+        for label in unique_labels:
+            if label == -1:
+                continue
+            cluster_points = data[labels == label]
+            num_points = len(cluster_points)
+            density = num_points / (np.max(cluster_points, axis=0) - np.min(cluster_points, axis=0)).prod()
+            stats.append({
+                "cluster": label,
+                "num_points": num_points,
+                "density": density
+            })
+        return stats
+
+    def __str__(self) -> str:
+        return "DBScan"
+
+
+class KMeansCluster(Cluster):
+    def __init__(self, n_clusters: int = 8, n_init: int = 1, max_iter: int = 300):
+        self.n_clusters = n_clusters
+        self.n_init = n_init
+        self.max_iter = max_iter
+
+    #@typing.override
+    def run(self, data: np.array) -> ClusterResult:
+        kmeans = KMeans(n_clusters=self.n_clusters, init="random", n_init=self.n_init, max_iter=self.max_iter, random_state=111)
+        labels = kmeans.fit_predict(data)
+        centers = kmeans.cluster_centers_
+        return ClusterResult(labels, centers, self.get_statistics(data, labels, centers))
+
+    def get_statistics(self, data: np.array, labels: np.array, centers: np.array) -> list[dict[str, Any]]:
+        unique_labels = np.unique(labels)
+        stats = []
+
+        for label in unique_labels:
+            cluster_points = data[labels == label]
+            num_points = len(cluster_points)
+            center = centers[label]
+            stats.append({
+                "cluster": label,
+                "num_points": num_points,
+                "center": center,
+            })
+        return stats
+
+    def __str__(self) -> str:
+        return "KMeans"
+
+
+CLUSTERING_STRATEGIES = [DBSCANCluster(), KMeansCluster()]

frontend/exploration.py

@@ -1,5 +1,6 @@
 import pandas as pd
 import streamlit as st
+import codecs
 
 st.set_page_config(
     page_title="Project Miner",
@@ -9,10 +10,13 @@ st.set_page_config(
 st.title("Home")
 
 ### Exploration
-uploaded_file = st.file_uploader("Upload your CSV file", type=["csv"])
+uploaded_file = st.file_uploader("Upload your CSV file", type=["csv", "tsv"])
+separator = st.selectbox("Separator", [",", ";", "\\t"])
+separator = codecs.getdecoder("unicode_escape")(separator)[0]
+has_header = st.checkbox("Has header", value=True)
 
 if uploaded_file is not None:
-    st.session_state.data = pd.read_csv(uploaded_file)
+    st.session_state.data = pd.read_csv(uploaded_file, sep=separator, header=0 if has_header else 1)
     st.session_state.original_data = st.session_state.data
     st.success("File loaded successfully!")
 

frontend/normstrategy.py

@@ -1,6 +1,7 @@
 from abc import ABC, abstractmethod
 from pandas import DataFrame, Series
 from pandas.api.types import is_numeric_dtype
+from sklearn.neighbors import KNeighborsClassifier
 from typing import Any, Union
 
 class DataFrameFunction(ABC):
@@ -18,11 +19,14 @@ class MVStrategy(DataFrameFunction):
     """A way to handle missing values in a dataframe."""
 
     @staticmethod
-    def list_available(df: DataFrame, series: Series) -> list['MVStrategy']:
+    def list_available(df: DataFrame, label: str, series: Series) -> list['MVStrategy']:
         """Get all the strategies that can be used."""
         choices = [DropStrategy(), ModeStrategy()]
         if is_numeric_dtype(series):
             choices.extend((MeanStrategy(), MedianStrategy(), LinearRegressionStrategy()))
+            other_columns = df.select_dtypes(include="number").drop(label, axis=1).columns.to_list()
+            if len(other_columns):
+                choices.append(KNNStrategy(other_columns))
         return choices
 
 
@@ -97,6 +101,43 @@ class LinearRegressionStrategy(MVStrategy):
         return "Use linear regression"
 
 
+class KNNStrategy(MVStrategy):
+    def __init__(self, training_features: list[str]):
+        self.available_features = training_features
+        self.training_features = training_features
+        self.n_neighbors = 3
+
+    def apply(self, df: DataFrame, label: str, series: Series) -> DataFrame:
+        # Remove any training column that have any missing values
+        usable_data = df.dropna(subset=self.training_features)
+        # Select columns to impute from
+        train_data = usable_data.dropna(subset=label)
+        # Create train dataframe
+        x_train = train_data.drop(label, axis=1)
+        y_train = train_data[label]
+
+        reg = KNeighborsClassifier(self.n_neighbors).fit(x_train, y_train)
+
+        # Create test dataframe
+        test_data = usable_data[usable_data[label].isnull()]
+        if test_data.empty:
+            return df
+        x_test = test_data.drop(label, axis=1)
+        predicted = reg.predict(x_test)
+
+        # Fill with predicated values and patch the original data
+        usable_data[label].fillna(Series(predicted), inplace=True)
+        df.fillna(usable_data, inplace=True)
+        return df
+
+    def count_max(self, df: DataFrame, label: str) -> int:
+        usable_data = df.dropna(subset=self.training_features)
+        return usable_data[label].count()
+
+    def __str__(self) -> str:
+        return "kNN"
+
+
 class KeepStrategy(ScalingStrategy):
     #@typing.override
     def apply(self, df: DataFrame, label: str, series: Series) -> DataFrame:

frontend/pages/clustering.py

@@ -0,0 +1,86 @@
+import streamlit as st
+import matplotlib.pyplot as plt
+from clusters import DBSCANCluster, KMeansCluster, CLUSTERING_STRATEGIES
+from sklearn.decomposition import PCA
+from sklearn.metrics import silhouette_score
+import numpy as np
+
+st.header("Clustering")
+
+if "data" in st.session_state:
+    data = st.session_state.data
+
+    general_row = st.columns([1, 1, 1])
+    clustering = general_row[0].selectbox("Clustering method", CLUSTERING_STRATEGIES)
+    data_name = general_row[1].multiselect("Columns", data.select_dtypes(include="number").columns)
+    n_components = general_row[2].number_input("Reduce dimensions to (PCA)", min_value=1, max_value=3, value=2)
+
+    with st.form("cluster_form"):
+        if isinstance(clustering, KMeansCluster):
+            row1 = st.columns([1, 1, 1])
+            clustering.n_clusters = row1[0].number_input("Number of clusters", min_value=1, max_value=data.shape[0], value=clustering.n_clusters)
+            clustering.n_init = row1[1].number_input("n_init", min_value=1, value=clustering.n_init)
+            clustering.max_iter = row1[2].number_input("max_iter", min_value=1, value=clustering.max_iter)
+        elif isinstance(clustering, DBSCANCluster):
+            row1 = st.columns([1, 1])
+            clustering.eps = row1[0].slider("eps", min_value=0.0001, max_value=1.0, step=0.05, value=clustering.eps)
+            clustering.min_samples = row1[1].number_input("min_samples", min_value=1, value=clustering.min_samples)
+
+        st.form_submit_button("Launch")
+
+    if len(data_name) > 0:
+        x = data[data_name].to_numpy()
+        n_components = min(n_components, len(data_name))
+        if len(data_name) > n_components:
+            pca = PCA(n_components)
+            x = pca.fit_transform(x)
+            if n_components == 2:
+                (fig, ax) = plt.subplots(figsize=(8, 8))
+                for i in range(0, pca.components_.shape[1]):
+                    ax.arrow(
+                        0,
+                        0,
+                        pca.components_[0, i],
+                        pca.components_[1, i],
+                        head_width=0.1,
+                        head_length=0.1
+                    )
+
+                    plt.text(
+                        pca.components_[0, i] + 0.05,
+                        pca.components_[1, i] + 0.05,
+                        data_name[i]
+                    )
+                circle = plt.Circle((0, 0), radius=1, edgecolor='b', facecolor='None')
+                ax.add_patch(circle)
+                plt.axis("equal")
+                ax.set_title("PCA result - Correlation circle")
+                st.pyplot(fig)
+
+        result = clustering.run(x)
+        st.write("## Cluster stats")
+        st.table(result.statistics)
+
+        st.write("## Graphical representation")
+        fig = plt.figure()
+        if n_components == 1:
+            plt.scatter(x, np.zeros_like(x))
+        elif n_components == 2:
+            ax = fig.add_subplot(projection='rectilinear')
+            plt.scatter(x[:, 0], x[:, 1], c=result.labels, s=50, cmap="viridis")
+            if result.centers is not None:
+                plt.scatter(result.centers[:, 0], result.centers[:, 1], c="black", s=200, marker="X")
+        else:
+            ax = fig.add_subplot(projection='3d')
+
+            ax.scatter(x[:, 0], x[:, 1],x[:, 2], c=result.labels, s=50, cmap="viridis")
+            if result.centers is not None:
+                ax.scatter(result.centers[:, 0], result.centers[:, 1], result.centers[:, 2], c="black", s=200, marker="X")
+        st.pyplot(fig)
+        if not (result.labels == 0).all():
+            st.write("Silhouette score:", silhouette_score(x, result.labels))
+    else:
+        st.error("Select at least one column")
+
+else:
+    st.error("file not loaded")

frontend/pages/normalization.py

@@ -1,5 +1,5 @@
 import streamlit as st
-from normstrategy import MVStrategy, ScalingStrategy
+from normstrategy import MVStrategy, ScalingStrategy, KNNStrategy
 
 if "data" in st.session_state:
     data = st.session_state.original_data
@@ -8,13 +8,16 @@ if "data" in st.session_state:
     for column, series in data.items():
         col1, col2 = st.columns(2)
         missing_count = series.isna().sum()
-        choices = MVStrategy.list_available(data, series)
+        choices = MVStrategy.list_available(data, column, series)
         option = col1.selectbox(
             f"Missing values of {column} ({missing_count})",
             choices,
             index=1,
             key=f"mv-{column}",
         )
+        if isinstance(option, KNNStrategy):
+            option.training_features = st.multiselect("Training columns", option.training_features, default=option.available_features, key=f"cols-{column}")
+            option.n_neighbors = st.number_input("Number of neighbors", min_value=1, max_value=option.count_max(data, column), value=option.n_neighbors, key=f"neighbors-{column}")
         # Always re-get the series to avoid reusing an invalidated series pointer
         data = option.apply(data, column, data[column])
 

frontend/pages/prediction_classification.py

@@ -0,0 +1,79 @@
+import streamlit as st
+from sklearn.linear_model import LogisticRegression
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import accuracy_score,confusion_matrix
+from sklearn.preprocessing import LabelEncoder
+import pandas as pd
+import matplotlib.pyplot as plt
+import seaborn as sns
+
+st.header("Prediction: Classification")
+
+if "data" in st.session_state:
+    data = st.session_state.data
+
+    with st.form("classification_form"):
+        st.subheader("Classification Parameters")
+        data_name = st.multiselect("Features", data.columns)
+        target_name = st.selectbox("Target", data.columns)
+        test_size = st.slider("Test Size", min_value=0.1, max_value=0.5, value=0.2, step=0.1)
+        st.form_submit_button('Train and Predict')
+
+    if data_name and target_name:
+        X = data[data_name]
+        y = data[target_name]
+
+        label_encoders = {}
+        for column in X.select_dtypes(include=['object']).columns:
+            le = LabelEncoder()
+            X[column] = le.fit_transform(X[column])
+            label_encoders[column] = le
+        
+        if y.dtype == 'object':
+            le = LabelEncoder()
+            y = le.fit_transform(y)
+            label_encoders[target_name] = le
+
+        X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=test_size, random_state=42)
+
+        model = LogisticRegression()
+        model.fit(X_train, y_train)
+        y_pred = model.predict(X_test)
+        accuracy = accuracy_score(y_test, y_pred)
+
+        st.subheader("Model Accuracy")
+        st.write(f"Accuracy on test data: {accuracy:.2f}")
+
+        st.subheader("Enter values for prediction")
+        pred_values = []
+        for feature in data_name:
+            if feature in label_encoders:
+                values = list(label_encoders[feature].classes_)
+                value = st.selectbox(f"Value for {feature}", values)
+                value_encoded = label_encoders[feature].transform([value])[0]
+                pred_values.append(value_encoded)
+            else:
+                value = st.number_input(f"Value for {feature}", value=0.0)
+                pred_values.append(value)
+        
+        prediction = model.predict(pd.DataFrame([pred_values], columns=data_name))
+
+        if target_name in label_encoders:
+            prediction = label_encoders[target_name].inverse_transform(prediction)
+        
+        st.write("Prediction:", prediction[0])
+
+        if len(data_name) == 1:
+            fig = plt.figure()
+
+            y_pred = [model.predict(pd.DataFrame([pred_value[0]], columns=data_name)) for pred_value in X.values.tolist()]
+            cm = confusion_matrix(y, y_pred)
+
+            sns.heatmap(cm, annot=True, fmt="d")
+
+            plt.xlabel('Predicted')
+            plt.ylabel('True')
+
+            st.pyplot(fig)
+else:
+    st.error("File not loaded")

frontend/pages/prediction_regression.py

@@ -0,0 +1,63 @@
+import streamlit as st
+from sklearn.linear_model import LinearRegression
+from sklearn.metrics import r2_score
+import pandas as pd
+import matplotlib.pyplot as plt
+
+st.header("Prediction: Regression")
+
+if "data" in st.session_state:
+    data = st.session_state.data
+
+    with st.form("regression_form"):
+        st.subheader("Linear Regression Parameters")
+        data_name = st.multiselect("Features", data.select_dtypes(include="number").columns)
+        target_name = st.selectbox("Target", data.select_dtypes(include="number").columns)
+        st.form_submit_button('Train and Predict')
+
+    if data_name and target_name:
+        X = data[data_name]
+        y = data[target_name]
+
+        model = LinearRegression()
+        model.fit(X, y)
+        
+        st.subheader("Enter values for prediction")
+        pred_values = [st.number_input(f"Value for {feature}", value=0.0) for feature in data_name]
+        prediction = model.predict(pd.DataFrame([pred_values], columns=data_name))
+
+        st.write("Prediction:", prediction[0])
+
+        fig = plt.figure()
+        dataframe_sorted = pd.concat([X, y], axis=1).sort_values(by=data_name)
+
+        if len(data_name) == 1:
+            y_pred = [model.predict(pd.DataFrame([pred_value[0]], columns=data_name)) for pred_value in X.values.tolist()]
+            r2 = r2_score(y, y_pred)
+            st.write('R-squared score:', r2)
+
+            X = dataframe_sorted[data_name[0]]
+            y = dataframe_sorted[target_name]
+
+            prediction_array_y = [
+                model.predict(pd.DataFrame([[dataframe_sorted[data_name[0]].iloc[i]]], columns=data_name))[0]
+                for i in range(dataframe_sorted.shape[0])
+            ]
+
+            plt.scatter(dataframe_sorted[data_name[0]], dataframe_sorted[target_name], color='b')
+            plt.plot(dataframe_sorted[data_name[0]], prediction_array_y, color='r')
+        elif len(data_name) == 2:
+            ax = fig.add_subplot(111, projection='3d')
+
+            prediction_array_y = [
+                model.predict(pd.DataFrame([[dataframe_sorted[data_name[0]].iloc[i], dataframe_sorted[data_name[1]].iloc[i]]], columns=data_name))[0]
+                for i in range(dataframe_sorted.shape[0])
+            ]
+
+            ax.scatter(dataframe_sorted[data_name[0]], dataframe_sorted[data_name[1]], dataframe_sorted[target_name], color='b')
+            ax.plot(dataframe_sorted[data_name[0]], dataframe_sorted[data_name[1]], prediction_array_y, color='r')
+
+        st.pyplot(fig)
+
+else:
+    st.error("File not loaded")

requirements.txt

@@ -0,0 +1,6 @@
+matplotlib>=3.5.0
+pandas>=1.5.0
+seaborn>=0.12.0
+scikit-learn>=0.23.0
+streamlit>=1.35.0
+ruff>=0.4.8