lecrapaud 0.11.2__py3-none-any.whl → 0.11.4__py3-none-any.whl

lecrapaud/api.py

@@ -27,11 +27,12 @@ Basic Usage:
 
 import joblib
 import pandas as pd
+import ast
+import os
 import logging
 import seaborn as sns
 import numpy as np
 import matplotlib.pyplot as plt
-from lecrapaud.utils import logger
 from lecrapaud.db.session import init_db
 from lecrapaud.feature_selection import FeatureSelectionEngine, PreprocessModel
 from lecrapaud.model_selection import (
@@ -46,6 +47,8 @@ from lecrapaud.feature_engineering import FeatureEngineeringEngine, PreprocessFe
 from lecrapaud.experiment import create_experiment
 from lecrapaud.db import Experiment
 from lecrapaud.search_space import normalize_models_idx
+from lecrapaud.utils import logger
+from lecrapaud.directories import tmp_dir
 
 
 class LeCrapaud:
@@ -108,6 +111,12 @@ class ExperimentEngine:
         if id:
             self.experiment = Experiment.get(id)
             kwargs.update(self.experiment.context)
+            experiment_dir = f"{tmp_dir}/{self.experiment.name}"
+            preprocessing_dir = f"{experiment_dir}/preprocessing"
+            data_dir = f"{experiment_dir}/data"
+            os.makedirs(experiment_dir, exist_ok=True)
+            os.makedirs(preprocessing_dir, exist_ok=True)
+            os.makedirs(data_dir, exist_ok=True)
         else:
             if data is None:
                 raise ValueError("Either id or data must be provided")
@@ -344,9 +353,13 @@ class ExperimentEngine:
         return pd.read_csv(f"{self.experiment.path}/feature_summary.csv")
 
     def get_threshold(self, target_number: int):
-        return joblib.load(
+        thresholds = joblib.load(
             f"{self.experiment.path}/TARGET_{target_number}/thresholds.pkl"
         )
+        if isinstance(thresholds, str):
+            thresholds = ast.literal_eval(thresholds)
+
+        return thresholds
 
     def load_model(self, target_number: int, model_name: str = None):
 

lecrapaud/db/alembic/env.py

@@ -15,7 +15,7 @@ config.set_main_option("sqlalchemy.url", DATABASE_URL)
 # Interpret the config file for Python logging.
 # This line sets up loggers basically.
 if config.config_file_name is not None:
-    fileConfig(config.config_file_name)
+    fileConfig(config.config_file_name, disable_existing_loggers=False)
 
 # add your model's MetaData object here
 # for 'autogenerate' support

lecrapaud/db/alembic.ini

@@ -84,11 +84,14 @@ sqlalchemy.url = %(DATABASE_URL)s
 [loggers]
 keys = root,sqlalchemy,alembic
 
+[loggers_root]
+disable_existing_loggers = False
+
 [handlers]
 keys = console
 
 [formatters]
-keys = generic
+keys = lecrapaud_format
 
 [logger_root]
 level = WARN
@@ -109,8 +112,8 @@ qualname = alembic
 class = StreamHandler
 args = (sys.stderr,)
 level = NOTSET
-formatter = generic
+formatter = lecrapaud_format
 
-[formatter_generic]
-format = %(levelname)-5.5s [%(name)s] %(message)s
-datefmt = %H:%M:%S
+[formatter_lecrapaud_format]
+format = %(asctime)s - %(name)s - %(levelname)s - %(funcName)s - %(message)s
+datefmt = %Y-%m-%d %H:%M:%S

lecrapaud/model_selection.py

@@ -1328,46 +1328,131 @@ def load_model(target_dir: str):
 # plots
 def plot_evaluation_for_classification(prediction: dict):
     """
-    Args
-        prediction (pd.DataFrame): Should be a df with TARGET, PRED, 0, 1 columns for y_true value (TARGET), y_pred (PRED), and probabilities (for classification only : 0 and 1)
+    Plot evaluation metrics for classification tasks (both binary and multiclass).
+
+    Args:
+        prediction (pd.DataFrame): Should be a df with:
+            - TARGET: true labels
+            - PRED: predicted labels
+            - For binary: column '1' or 1 for positive class probabilities
+            - For multiclass: columns 2 onwards for class probabilities
     """
     y_true = prediction["TARGET"]
     y_pred = prediction["PRED"]
-    y_pred_proba = prediction[1] if 1 in prediction.columns else prediction["1"]
 
     # Plot confusion matrix
     plot_confusion_matrix(y_true, y_pred)
 
-    # Compute ROC curve and ROC area
-    fpr, tpr, thresholds = roc_curve(y_true, y_pred_proba)
-    roc_auc = auc(fpr, tpr)
+    # Determine if binary or multiclass
+    unique_labels = np.unique(y_true)
+    unique_labels = np.sort(unique_labels)
+    n_classes = len(unique_labels)
+
+    if n_classes <= 2:
+        # Binary classification
+        y_pred_proba = prediction[1] if 1 in prediction.columns else prediction["1"]
+
+        # Compute and plot ROC curve
+        fpr, tpr, _ = roc_curve(y_true, y_pred_proba)
+        roc_auc = auc(fpr, tpr)
+
+        plt.figure(figsize=(8, 8))
+        plt.plot(
+            fpr,
+            tpr,
+            color="darkorange",
+            lw=2,
+            label=f"ROC curve (area = {roc_auc:0.2f})",
+        )
+        plt.plot([0, 1], [0, 1], color="navy", lw=2, linestyle="--")
+        plt.xlim([0.0, 1.0])
+        plt.ylim([0.0, 1.05])
+        plt.xlabel("False Positive Rate")
+        plt.ylabel("True Positive Rate")
+        plt.title("ROC Curve")
+        plt.legend(loc="lower right")
+        plt.show()
+
+        # Compute and plot precision-recall curve
+        precision, recall, _ = precision_recall_curve(y_true, y_pred_proba)
+        average_precision = average_precision_score(y_true, y_pred_proba)
+
+        plt.figure(figsize=(8, 8))
+        plt.step(recall, precision, color="b", alpha=0.2, where="post")
+        plt.fill_between(recall, precision, step="post", alpha=0.2, color="b")
+        plt.xlabel("Recall")
+        plt.ylabel("Precision")
+        plt.ylim([0.0, 1.05])
+        plt.xlim([0.0, 1.0])
+        plt.title(f"Precision-Recall Curve: AP={average_precision:0.2f}")
+        plt.show()
 
-    plt.figure(figsize=(8, 8))
-    plt.plot(
-        fpr, tpr, color="darkorange", lw=2, label="ROC curve (area = %0.2f)" % roc_auc
-    )
-    plt.plot([0, 1], [0, 1], color="navy", lw=2, linestyle="--")
-    plt.xlim([0.0, 1.0])
-    plt.ylim([0.0, 1.05])
-    plt.xlabel("False Positive Rate")
-    plt.ylabel("True Positive Rate")
-    plt.title("ROC Curve")
-    plt.legend(loc="lower right")
-    plt.show()
+    else:
+        # Multiclass classification
+        # Get class probabilities
+        pred_cols = [
+            col for col in prediction.columns if col not in ["ID", "TARGET", "PRED"]
+        ]
+        y_pred_proba = prediction[pred_cols].values
 
-    # Compute precision-recall curve
-    precision, recall, _ = precision_recall_curve(y_true, y_pred_proba)
-    average_precision = average_precision_score(y_true, y_pred_proba)
-
-    plt.figure(figsize=(8, 8))
-    plt.step(recall, precision, color="b", alpha=0.2, where="post")
-    plt.fill_between(recall, precision, step="post", alpha=0.2, color="b")
-    plt.xlabel("Recall")
-    plt.ylabel("Precision")
-    plt.ylim([0.0, 1.05])
-    plt.xlim([0.0, 1.0])
-    plt.title("Precision-Recall Curve: AP={0:0.2f}".format(average_precision))
-    plt.show()
+        # Compute ROC curve and ROC area for each class
+        fpr = dict()
+        tpr = dict()
+        roc_auc = dict()
+
+        plt.figure(figsize=(10, 8))
+        colors = plt.cm.get_cmap("tab10")(np.linspace(0, 1, n_classes))
+
+        for i, (label, color) in enumerate(zip(unique_labels, colors)):
+            y_true_binary = (y_true == label).astype(int)
+            y_score = y_pred_proba[:, i]
+
+            fpr[i], tpr[i], _ = roc_curve(y_true_binary, y_score)
+            roc_auc[i] = auc(fpr[i], tpr[i])
+
+            plt.plot(
+                fpr[i],
+                tpr[i],
+                color=color,
+                lw=2,
+                label=f"Class {label} (area = {roc_auc[i]:0.2f})",
+            )
+
+        plt.plot([0, 1], [0, 1], "k--", lw=2)
+        plt.xlim([0.0, 1.0])
+        plt.ylim([0.0, 1.05])
+        plt.xlabel("False Positive Rate")
+        plt.ylabel("True Positive Rate")
+        plt.title("Multiclass ROC Curves (One-vs-Rest)")
+        plt.legend(loc="lower right")
+        plt.show()
+
+        # Compute PR curve for each class
+        plt.figure(figsize=(10, 8))
+
+        for i, (label, color) in enumerate(zip(unique_labels, colors)):
+            y_true_binary = (y_true == label).astype(int)
+            y_score = y_pred_proba[:, i]
+
+            precision, recall, _ = precision_recall_curve(y_true_binary, y_score)
+            average_precision = average_precision_score(y_true_binary, y_score)
+
+            plt.step(
+                recall,
+                precision,
+                color=color,
+                alpha=0.8,
+                where="post",
+                label=f"Class {label} (AP = {average_precision:0.2f})",
+            )
+
+        plt.xlabel("Recall")
+        plt.ylabel("Precision")
+        plt.ylim([0.0, 1.05])
+        plt.xlim([0.0, 1.0])
+        plt.title("Multiclass Precision-Recall Curves")
+        plt.legend(loc="lower left")
+        plt.show()
 
 
 def plot_confusion_matrix(y_true, y_pred):

lecrapaud/utils.py

@@ -9,7 +9,7 @@ from ftfy import fix_text
 import unicodedata
 import re
 import string
-
+import sys
 from lecrapaud.directories import logger_dir
 from lecrapaud.config import LOGGING_LEVEL, PYTHON_ENV, LECRAPAUD_LOCAL
 

{lecrapaud-0.11.2.dist-info → lecrapaud-0.11.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: lecrapaud
-Version: 0.11.2
+Version: 0.11.4
 Summary: Framework for machine and deep learning, with regression, classification and time series analysis
 License: Apache License
 Author: Pierre H. Gallet

{lecrapaud-0.11.2.dist-info → lecrapaud-0.11.4.dist-info}/RECORD

@@ -1,15 +1,15 @@
 lecrapaud/__init__.py,sha256=oCxbtw_nk8rlOXbXbWo0RRMlsh6w-hTiZ6e5PRG_wp0,28
-lecrapaud/api.py,sha256=M3TlXXQ_2HSlXoIgz-wAXOyz5D2fvWtEWj1Yac1VcEU,16647
+lecrapaud/api.py,sha256=nh1dRcqDpEnyOMjvayUNg_DR1D26gXCQ7hZpsYENqk0,17178
 lecrapaud/config.py,sha256=eYnrktVq457xMIMGcUSilJdNxCsaGP_gRAlzCSwd6Vo,1047
 lecrapaud/db/__init__.py,sha256=82o9fMfaqKXPh2_rt44EzNRVZV1R4LScEnQYvj_TjK0,34
 lecrapaud/db/alembic/README,sha256=MVlc9TYmr57RbhXET6QxgyCcwWP7w-vLkEsirENqiIQ,38
-lecrapaud/db/alembic/env.py,sha256=rseEi8oR_eKXYYW3UwOKiCMuDEwT4lxsT7llySOUpgk,2305
+lecrapaud/db/alembic/env.py,sha256=0VdxHNIxhPCgUHnx6EwlVZLUMLlbqZ_eV7i0Ho2XqeI,2337
 lecrapaud/db/alembic/script.py.mako,sha256=MEqL-2qATlST9TAOeYgscMn1uy6HUS9NFvDgl93dMj8,635
 lecrapaud/db/alembic/versions/2025_06_23_1748-f089dfb7e3ba_.py,sha256=MNPyqWaQSHNV8zljD1G9f-LzrVz-nOKlgOhHEE0U8Oo,13060
 lecrapaud/db/alembic/versions/2025_06_24_1216-c62251b129ed_.py,sha256=g6aLRV6jAKXkPUEcs9FAeGfsYpe9rMTxfqbNib3U0-U,809
 lecrapaud/db/alembic/versions/2025_06_24_1711-86457e2f333f_.py,sha256=dl6tfvcqErgJ6NKvjve0euu7l0BWyEAKSS-ychsEAl8,1139
 lecrapaud/db/alembic/versions/2025_06_25_1759-72aa496ca65b_.py,sha256=sBgPLvvqI_HmPqQ0Kime1ZL1AHSeuYJHlmFJOnXWeuU,835
-lecrapaud/db/alembic.ini,sha256=zgvur-5jnLsT66_98FaTOTNgjwObGZCE0HqMwRAeJrs,3587
+lecrapaud/db/alembic.ini,sha256=TXrZB4pWVLn2EUg867yp6paA_19vGeirO95mTPA3nbs,3699
 lecrapaud/db/models/__init__.py,sha256=Lhyw9fVLdom0Fc6yIP-ip8FjkU1EwVwjae5q2VM815Q,740
 lecrapaud/db/models/base.py,sha256=CYtof_UjFwX3C7XUifequh_UtLHJ25bU7LCwT501uGE,7508
 lecrapaud/db/models/experiment.py,sha256=IeS-TWPT-4l9xCMIdR2S2O-foXNt3Ru6WmtPMWToK7c,4035
@@ -35,10 +35,10 @@ lecrapaud/misc/tabpfn_tests.ipynb,sha256=VkgsCUJ30d8jaL2VaWtQAgb8ngHPNtPgnXLs7QQ
 lecrapaud/misc/test-gpu-bilstm.ipynb,sha256=4nLuZRJVe2kn6kEmauhRiz5wkWT9AVrYhI9CEk_dYUY,9608
 lecrapaud/misc/test-gpu-resnet.ipynb,sha256=27Vu7nYwujYeh3fOxBNCnKJn3MXNPKZU-U8oDDUbymg,4944
 lecrapaud/misc/test-gpu-transformers.ipynb,sha256=k6MBSs_Um1h4PykvE-LTBcdpbWLbIFST_xl_AFW2jgI,8444
-lecrapaud/model_selection.py,sha256=PQGEWVWN-4ZeHCqrmXBpHgq1QZi_1nOOeu5gazXGDLQ,60487
+lecrapaud/model_selection.py,sha256=h4WPtGCUeuWIXDJ8L2-i1I7RwrZlnxAresGW5l8bGwE,63195
 lecrapaud/search_space.py,sha256=-JkzuMhaomdwiWi4HvVQY5hiw3-oREemJA16tbwEIp4,34854
-lecrapaud/utils.py,sha256=MUgDoJ31GOF8WRLn_WLzDbHw7OTKxq_ldnZT6dpxdQo,8295
-lecrapaud-0.11.2.dist-info/LICENSE,sha256=MImCryu0AnqhJE_uAZD-PIDKXDKb8sT7v0i1NOYeHTM,11350
-lecrapaud-0.11.2.dist-info/METADATA,sha256=Ft9avufHygUCmyQ4Fbkp2k_W5B4HkIY_v-zl3EDX4LI,11017
-lecrapaud-0.11.2.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-lecrapaud-0.11.2.dist-info/RECORD,,
+lecrapaud/utils.py,sha256=JdBB1NvbNIx4y0Una-kSZdo1_ZEocc5hwyYFIZKHmGg,8305
+lecrapaud-0.11.4.dist-info/LICENSE,sha256=MImCryu0AnqhJE_uAZD-PIDKXDKb8sT7v0i1NOYeHTM,11350
+lecrapaud-0.11.4.dist-info/METADATA,sha256=3O_6bcQaCfragLXVyAsZWVpP9lvxEziUo9DqWzRE1r4,11017
+lecrapaud-0.11.4.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+lecrapaud-0.11.4.dist-info/RECORD,,