diff --git a/.gitignore b/.gitignore
index 4dba1c9..0477015 100644
--- a/.gitignore
+++ b/.gitignore
@@ -128,7 +128,8 @@ dmypy.json
 # Pyre type checker
 .pyre/
 
-# vscode
+# IDE
 .vscode/
+.idea/
 
 **/*DS_Store*
\ No newline at end of file
diff --git a/notebooks/inference_custom_model_example.ipynb b/notebooks/inference_custom_model_example.ipynb
new file mode 100644
index 0000000..a2ff806
--- /dev/null
+++ b/notebooks/inference_custom_model_example.ipynb
@@ -0,0 +1,932 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "a6727d05-5009-49f5-8918-d5fdd22cb187",
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2025-05-15T11:50:39.093309Z",
+     "start_time": "2025-05-15T11:50:39.060430Z"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "%load_ext autoreload\n",
+    "%autoreload 2"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "de41f9a83014711d",
+   "metadata": {
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
+   "source": [
+    "#### Run the cell bellow, uncommented if using autogluon"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "b3766057c77adc9",
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2025-05-15T11:50:39.113487Z",
+     "start_time": "2025-05-15T11:50:39.096588Z"
+    },
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# pip install autogluon"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "c64a6fab-1676-4539-b460-5b2fdb456b04",
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2025-05-15T11:50:44.845976Z",
+     "start_time": "2025-05-15T11:50:39.117232Z"
+    },
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "from pathlib import Path\n",
+    "\n",
+    "import matplotlib.pyplot as plt\n",
+    "import pandas as pd\n",
+    "\n",
+    "from anonymeter.evaluators import InferenceEvaluator, MLModelInference\n",
+    "from autogluon.tabular import TabularPredictor\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "3429c0fa-8915-4320-93f2-016fedf2b570",
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2025-05-15T11:50:44.889226Z",
+     "start_time": "2025-05-15T11:50:44.848139Z"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "import warnings\n",
+    "warnings.filterwarnings('ignore')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ada19947-b895-4279-aac3-9b87fac2fa6b",
+   "metadata": {},
+   "source": [
+    "## Downloading the data\n",
+    "\n",
+    "For this example, we will use the famous `Adults` (more details [here](https://archive.ics.uci.edu/ml/datasets/adult)) dataset. This dataset contains aggregated census data, where every row represent a population segment. For the purpose of demonstrating `Anonymeter`, we will use this data as if each row would in fact refer to a real individual. \n",
+    "\n",
+    "The synthetic version has been generated by `CTGAN` from [SDV](https://sdv.dev/SDV/user_guides/single_table/ctgan.html), as explained in the paper accompanying this code release. For details on the generation process, e.g. regarding hyperparameters, see Section 6.2.1 of [the accompanying paper](https://petsymposium.org/popets/2023/popets-2023-0055.php)).\n",
+    "\n",
+    "We pull these datasets from the [Statice](https://www.statice.ai/) public GC bucket:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "fc128115-2f0c-43b1-9198-5c5594eae7f3",
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2025-05-15T11:50:45.192418Z",
+     "start_time": "2025-05-15T11:50:44.893155Z"
+    },
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "ori = pd.read_csv(\"datasets/adults_train.csv\")\n",
+    "syn = pd.read_csv(\"datasets/adults_syn_ctgan.csv\")\n",
+    "control = pd.read_csv(\"datasets/adults_control.csv\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "f6abeed8-23ae-4d4a-9cdb-006c0bba109c",
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2025-05-15T11:50:45.259297Z",
+     "start_time": "2025-05-15T11:50:45.194514Z"
+    },
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>age</th>\n",
+       "      <th>type_employer</th>\n",
+       "      <th>fnlwgt</th>\n",
+       "      <th>education</th>\n",
+       "      <th>education_num</th>\n",
+       "      <th>marital</th>\n",
+       "      <th>occupation</th>\n",
+       "      <th>relationship</th>\n",
+       "      <th>race</th>\n",
+       "      <th>sex</th>\n",
+       "      <th>capital_gain</th>\n",
+       "      <th>capital_loss</th>\n",
+       "      <th>hr_per_week</th>\n",
+       "      <th>country</th>\n",
+       "      <th>income</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>53</td>\n",
+       "      <td>Self-emp-not-inc</td>\n",
+       "      <td>138022</td>\n",
+       "      <td>11th</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Divorced</td>\n",
+       "      <td>Craft-repair</td>\n",
+       "      <td>Not-in-family</td>\n",
+       "      <td>White</td>\n",
+       "      <td>Male</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>60</td>\n",
+       "      <td>United-States</td>\n",
+       "      <td>&lt;=50K</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>31</td>\n",
+       "      <td>Private</td>\n",
+       "      <td>344200</td>\n",
+       "      <td>HS-grad</td>\n",
+       "      <td>9</td>\n",
+       "      <td>Married-civ-spouse</td>\n",
+       "      <td>Exec-managerial</td>\n",
+       "      <td>Husband</td>\n",
+       "      <td>White</td>\n",
+       "      <td>Male</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>40</td>\n",
+       "      <td>United-States</td>\n",
+       "      <td>&gt;50K</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>28</td>\n",
+       "      <td>Private</td>\n",
+       "      <td>242482</td>\n",
+       "      <td>HS-grad</td>\n",
+       "      <td>9</td>\n",
+       "      <td>Never-married</td>\n",
+       "      <td>Handlers-cleaners</td>\n",
+       "      <td>Own-child</td>\n",
+       "      <td>White</td>\n",
+       "      <td>Male</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>40</td>\n",
+       "      <td>United-States</td>\n",
+       "      <td>&lt;=50K</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>26</td>\n",
+       "      <td>Private</td>\n",
+       "      <td>193165</td>\n",
+       "      <td>Some-college</td>\n",
+       "      <td>10</td>\n",
+       "      <td>Married-civ-spouse</td>\n",
+       "      <td>Transport-moving</td>\n",
+       "      <td>Husband</td>\n",
+       "      <td>White</td>\n",
+       "      <td>Male</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>52</td>\n",
+       "      <td>United-States</td>\n",
+       "      <td>&gt;50K</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>27</td>\n",
+       "      <td>Private</td>\n",
+       "      <td>267989</td>\n",
+       "      <td>Some-college</td>\n",
+       "      <td>10</td>\n",
+       "      <td>Married-civ-spouse</td>\n",
+       "      <td>Machine-op-inspct</td>\n",
+       "      <td>Husband</td>\n",
+       "      <td>White</td>\n",
+       "      <td>Male</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>40</td>\n",
+       "      <td>United-States</td>\n",
+       "      <td>&lt;=50K</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   age     type_employer  fnlwgt     education  education_num  \\\n",
+       "0   53  Self-emp-not-inc  138022          11th              7   \n",
+       "1   31           Private  344200       HS-grad              9   \n",
+       "2   28           Private  242482       HS-grad              9   \n",
+       "3   26           Private  193165  Some-college             10   \n",
+       "4   27           Private  267989  Some-college             10   \n",
+       "\n",
+       "              marital         occupation   relationship   race   sex  \\\n",
+       "0            Divorced       Craft-repair  Not-in-family  White  Male   \n",
+       "1  Married-civ-spouse    Exec-managerial        Husband  White  Male   \n",
+       "2       Never-married  Handlers-cleaners      Own-child  White  Male   \n",
+       "3  Married-civ-spouse   Transport-moving        Husband  White  Male   \n",
+       "4  Married-civ-spouse  Machine-op-inspct        Husband  White  Male   \n",
+       "\n",
+       "   capital_gain  capital_loss  hr_per_week        country income  \n",
+       "0             0             0           60  United-States  <=50K  \n",
+       "1             0             0           40  United-States   >50K  \n",
+       "2             0             0           40  United-States  <=50K  \n",
+       "3             0             0           52  United-States   >50K  \n",
+       "4             0             0           40  United-States  <=50K  "
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "ori.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f1d19013-b7cf-48e3-a068-6c1e5449884e",
+   "metadata": {},
+   "source": [
+    "As visible the dataset contains several demographic information, as well as information regarding the education, financial situation, and personal life of some tens of thousands of \"individuals\"."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "0429baae-424d-4ebe-b8ec-9205397515ba",
+   "metadata": {},
+   "source": [
+    "# Measuring the Inference Risk\n",
+    "\n",
+    "With this new implementation we allow for the user to pass their own model in order to do the attribute inference.\n",
+    "The requirements for the model are:\n",
+    "1. It is trained on the synthetic data with the `secret` as a target.\n",
+    "2. It supports the ::predict(x) method\n",
+    "\n",
+    "Finally, it should be passed as `ml_model=predictor` to the `InferenceEvaluator` like in the example below:\n",
+    "(in case we pass our own ml model and want to have the predictions for all original/control data, set sample_attacks=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "6c07054c-7ced-46c3-8a12-14123f6cc965",
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2025-05-15T11:53:00.988964Z",
+     "start_time": "2025-05-15T11:50:45.261347Z"
+    },
+    "scrolled": true,
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "No path specified. Models will be saved in: \"AutogluonModels/ag-20251121_092829\"\n",
+      "Verbosity: 2 (Standard Logging)\n",
+      "=================== System Info ===================\n",
+      "AutoGluon Version:  1.4.0\n",
+      "Python Version:     3.9.22\n",
+      "Operating System:   Darwin\n",
+      "Platform Machine:   x86_64\n",
+      "Platform Version:   Darwin Kernel Version 24.6.0: Mon Jul 14 11:28:17 PDT 2025; root:xnu-11417.140.69~1/RELEASE_X86_64\n",
+      "CPU Count:          12\n",
+      "Memory Avail:       4.33 GB / 16.00 GB (27.0%)\n",
+      "Disk Space Avail:   11.13 GB / 465.63 GB (2.4%)\n",
+      "===================================================\n",
+      "No presets specified! To achieve strong results with AutoGluon, it is recommended to use the available presets. Defaulting to `'medium'`...\n",
+      "\tRecommended Presets (For more details refer to https://auto.gluon.ai/stable/tutorials/tabular/tabular-essentials.html#presets):\n",
+      "\tpresets='extreme' : New in v1.4: Massively better than 'best' on datasets <30000 samples by using new models meta-learned on https://tabarena.ai: TabPFNv2, TabICL, Mitra, and TabM. Absolute best accuracy. Requires a GPU. Recommended 64 GB CPU memory and 32+ GB GPU memory.\n",
+      "\tpresets='best'    : Maximize accuracy. Recommended for most users. Use in competitions and benchmarks.\n",
+      "\tpresets='high'    : Strong accuracy with fast inference speed.\n",
+      "\tpresets='good'    : Good accuracy with very fast inference speed.\n",
+      "\tpresets='medium'  : Fast training time, ideal for initial prototyping.\n",
+      "Beginning AutoGluon training ...\n",
+      "AutoGluon will save models to \"/Users/ivanananevski/projects/anonymeter-copy/anonymeter/notebooks/AutogluonModels/ag-20251121_092829\"\n",
+      "Train Data Rows:    39040\n",
+      "Train Data Columns: 14\n",
+      "Label Column:       age\n",
+      "AutoGluon infers your prediction problem is: 'multiclass' (because dtype of label-column == int, but few unique label-values observed).\n",
+      "\tFirst 10 (of 73) unique label values:  [37, 53, 17, 30, 40, 34, 36, 25, 56, 46]\n",
+      "\tIf 'multiclass' is not the correct problem_type, please manually specify the problem_type parameter during Predictor init (You may specify problem_type as one of: ['binary', 'multiclass', 'regression', 'quantile'])\n",
+      "Problem Type:       multiclass\n",
+      "Preprocessing data ...\n",
+      "Warning: Some classes in the training set have fewer than 10 examples. AutoGluon will only keep 66 out of 73 classes for training and will not try to predict the rare classes. To keep more classes, increase the number of datapoints from these rare classes in the training data or reduce label_count_threshold.\n",
+      "Fraction of data from classes with at least 10 examples that will be kept for training models: 0.999359631147541\n",
+      "Train Data Class Count: 66\n",
+      "Using Feature Generators to preprocess the data ...\n",
+      "Fitting AutoMLPipelineFeatureGenerator...\n",
+      "\tAvailable Memory:                    4553.59 MB\n",
+      "\tTrain Data (Original)  Memory Usage: 23.48 MB (0.5% of available memory)\n",
+      "\tInferring data type of each feature based on column values. Set feature_metadata_in to manually specify special dtypes of the features.\n",
+      "\tStage 1 Generators:\n",
+      "\t\tFitting AsTypeFeatureGenerator...\n",
+      "\t\t\tNote: Converting 2 features to boolean dtype as they only contain 2 unique values.\n",
+      "\tStage 2 Generators:\n",
+      "\t\tFitting FillNaFeatureGenerator...\n",
+      "\tStage 3 Generators:\n",
+      "\t\tFitting IdentityFeatureGenerator...\n",
+      "\t\tFitting CategoryFeatureGenerator...\n",
+      "\t\t\tFitting CategoryMemoryMinimizeFeatureGenerator...\n",
+      "\tStage 4 Generators:\n",
+      "\t\tFitting DropUniqueFeatureGenerator...\n",
+      "\tStage 5 Generators:\n",
+      "\t\tFitting DropDuplicatesFeatureGenerator...\n",
+      "\tTypes of features in original data (raw dtype, special dtypes):\n",
+      "\t\t('int', [])    : 5 | ['fnlwgt', 'education_num', 'capital_gain', 'capital_loss', 'hr_per_week']\n",
+      "\t\t('object', []) : 9 | ['type_employer', 'education', 'marital', 'occupation', 'relationship', ...]\n",
+      "\tTypes of features in processed data (raw dtype, special dtypes):\n",
+      "\t\t('category', [])  : 7 | ['type_employer', 'education', 'marital', 'occupation', 'relationship', ...]\n",
+      "\t\t('int', [])       : 5 | ['fnlwgt', 'education_num', 'capital_gain', 'capital_loss', 'hr_per_week']\n",
+      "\t\t('int', ['bool']) : 2 | ['sex', 'income']\n",
+      "\t0.5s = Fit runtime\n",
+      "\t14 features in original data used to generate 14 features in processed data.\n",
+      "\tTrain Data (Processed) Memory Usage: 1.83 MB (0.0% of available memory)\n",
+      "Data preprocessing and feature engineering runtime = 0.62s ...\n",
+      "AutoGluon will gauge predictive performance using evaluation metric: 'accuracy'\n",
+      "\tTo change this, specify the eval_metric parameter of Predictor()\n",
+      "Automatically generating train/validation split with holdout_frac=0.06403688524590163, Train Rows: 36516, Val Rows: 2499\n",
+      "User-specified model hyperparameters to be fit:\n",
+      "{\n",
+      "\t'RF': [{}],\n",
+      "}\n",
+      "Fitting 1 L1 models, fit_strategy=\"sequential\" ...\n",
+      "Fitting model: RandomForest ...\n",
+      "\tFitting with cpus=12, gpus=0, mem=1.5/4.6 GB\n",
+      "\tWarning: Reducing model 'n_estimators' from 300 -> 42 due to low memory. Expected memory usage reduced from 106.04% -> 15.0% of available memory...\n",
+      "\t0.0576\t = Validation score   (accuracy)\n",
+      "\t1.51s\t = Training   runtime\n",
+      "\t0.06s\t = Validation runtime\n",
+      "Fitting model: WeightedEnsemble_L2 ...\n",
+      "\tEnsemble Weights: {'RandomForest': 1.0}\n",
+      "\t0.0576\t = Validation score   (accuracy)\n",
+      "\t0.02s\t = Training   runtime\n",
+      "\t0.0s\t = Validation runtime\n",
+      "AutoGluon training complete, total runtime = 3.95s ... Best model: WeightedEnsemble_L2 | Estimated inference throughput: 42137.1 rows/s (2499 batch size)\n",
+      "TabularPredictor saved. To load, use: predictor = TabularPredictor.load(\"/Users/ivanananevski/projects/anonymeter-copy/anonymeter/notebooks/AutogluonModels/ag-20251121_092829\")\n",
+      "No path specified. Models will be saved in: \"AutogluonModels/ag-20251121_092838\"\n",
+      "Verbosity: 2 (Standard Logging)\n",
+      "=================== System Info ===================\n",
+      "AutoGluon Version:  1.4.0\n",
+      "Python Version:     3.9.22\n",
+      "Operating System:   Darwin\n",
+      "Platform Machine:   x86_64\n",
+      "Platform Version:   Darwin Kernel Version 24.6.0: Mon Jul 14 11:28:17 PDT 2025; root:xnu-11417.140.69~1/RELEASE_X86_64\n",
+      "CPU Count:          12\n",
+      "Memory Avail:       5.53 GB / 16.00 GB (34.5%)\n",
+      "Disk Space Avail:   24.42 GB / 465.63 GB (5.2%)\n",
+      "===================================================\n",
+      "No presets specified! To achieve strong results with AutoGluon, it is recommended to use the available presets. Defaulting to `'medium'`...\n",
+      "\tRecommended Presets (For more details refer to https://auto.gluon.ai/stable/tutorials/tabular/tabular-essentials.html#presets):\n",
+      "\tpresets='extreme' : New in v1.4: Massively better than 'best' on datasets <30000 samples by using new models meta-learned on https://tabarena.ai: TabPFNv2, TabICL, Mitra, and TabM. Absolute best accuracy. Requires a GPU. Recommended 64 GB CPU memory and 32+ GB GPU memory.\n",
+      "\tpresets='best'    : Maximize accuracy. Recommended for most users. Use in competitions and benchmarks.\n",
+      "\tpresets='high'    : Strong accuracy with fast inference speed.\n",
+      "\tpresets='good'    : Good accuracy with very fast inference speed.\n",
+      "\tpresets='medium'  : Fast training time, ideal for initial prototyping.\n",
+      "Beginning AutoGluon training ...\n",
+      "AutoGluon will save models to \"/Users/ivanananevski/projects/anonymeter-copy/anonymeter/notebooks/AutogluonModels/ag-20251121_092838\"\n",
+      "Train Data Rows:    39040\n",
+      "Train Data Columns: 14\n",
+      "Label Column:       type_employer\n",
+      "AutoGluon infers your prediction problem is: 'multiclass' (because dtype of label-column == object).\n",
+      "\t9 unique label values:  ['Private', '?', 'State-gov', 'Federal-gov', 'Local-gov', 'Self-emp-not-inc', 'Self-emp-inc', 'Without-pay', 'Never-worked']\n",
+      "\tIf 'multiclass' is not the correct problem_type, please manually specify the problem_type parameter during Predictor init (You may specify problem_type as one of: ['binary', 'multiclass', 'regression', 'quantile'])\n",
+      "Problem Type:       multiclass\n",
+      "Preprocessing data ...\n",
+      "Warning: Some classes in the training set have fewer than 10 examples. AutoGluon will only keep 8 out of 9 classes for training and will not try to predict the rare classes. To keep more classes, increase the number of datapoints from these rare classes in the training data or reduce label_count_threshold.\n",
+      "Fraction of data from classes with at least 10 examples that will be kept for training models: 0.9998206967213115\n",
+      "Train Data Class Count: 8\n",
+      "Using Feature Generators to preprocess the data ...\n",
+      "Fitting AutoMLPipelineFeatureGenerator...\n",
+      "\tAvailable Memory:                    5521.62 MB\n",
+      "\tTrain Data (Original)  Memory Usage: 21.38 MB (0.4% of available memory)\n",
+      "\tInferring data type of each feature based on column values. Set feature_metadata_in to manually specify special dtypes of the features.\n",
+      "\tStage 1 Generators:\n",
+      "\t\tFitting AsTypeFeatureGenerator...\n",
+      "\t\t\tNote: Converting 2 features to boolean dtype as they only contain 2 unique values.\n",
+      "\tStage 2 Generators:\n",
+      "\t\tFitting FillNaFeatureGenerator...\n",
+      "\tStage 3 Generators:\n",
+      "\t\tFitting IdentityFeatureGenerator...\n",
+      "\t\tFitting CategoryFeatureGenerator...\n",
+      "\t\t\tFitting CategoryMemoryMinimizeFeatureGenerator...\n",
+      "\tStage 4 Generators:\n",
+      "\t\tFitting DropUniqueFeatureGenerator...\n",
+      "\tStage 5 Generators:\n",
+      "\t\tFitting DropDuplicatesFeatureGenerator...\n",
+      "\tTypes of features in original data (raw dtype, special dtypes):\n",
+      "\t\t('int', [])    : 6 | ['age', 'fnlwgt', 'education_num', 'capital_gain', 'capital_loss', ...]\n",
+      "\t\t('object', []) : 8 | ['education', 'marital', 'occupation', 'relationship', 'race', ...]\n",
+      "\tTypes of features in processed data (raw dtype, special dtypes):\n",
+      "\t\t('category', [])  : 6 | ['education', 'marital', 'occupation', 'relationship', 'race', ...]\n",
+      "\t\t('int', [])       : 6 | ['age', 'fnlwgt', 'education_num', 'capital_gain', 'capital_loss', ...]\n",
+      "\t\t('int', ['bool']) : 2 | ['sex', 'income']\n",
+      "\t0.4s = Fit runtime\n",
+      "\t14 features in original data used to generate 14 features in processed data.\n",
+      "\tTrain Data (Processed) Memory Usage: 2.09 MB (0.0% of available memory)\n",
+      "Data preprocessing and feature engineering runtime = 0.55s ...\n",
+      "AutoGluon will gauge predictive performance using evaluation metric: 'accuracy'\n",
+      "\tTo change this, specify the eval_metric parameter of Predictor()\n",
+      "Automatically generating train/validation split with holdout_frac=0.06403688524590163, Train Rows: 36533, Val Rows: 2500\n",
+      "User-specified model hyperparameters to be fit:\n",
+      "{\n",
+      "\t'RF': [{}],\n",
+      "}\n",
+      "Fitting 1 L1 models, fit_strategy=\"sequential\" ...\n",
+      "Fitting model: RandomForest ...\n",
+      "\tFitting with cpus=12, gpus=0, mem=0.2/5.3 GB\n",
+      "\t0.7044\t = Validation score   (accuracy)\n",
+      "\t4.2s\t = Training   runtime\n",
+      "\t0.11s\t = Validation runtime\n",
+      "Fitting model: WeightedEnsemble_L2 ...\n",
+      "\tEnsemble Weights: {'RandomForest': 1.0}\n",
+      "\t0.7044\t = Validation score   (accuracy)\n",
+      "\t0.01s\t = Training   runtime\n",
+      "\t0.0s\t = Validation runtime\n",
+      "AutoGluon training complete, total runtime = 6.66s ... Best model: WeightedEnsemble_L2 | Estimated inference throughput: 23297.2 rows/s (2500 batch size)\n",
+      "TabularPredictor saved. To load, use: predictor = TabularPredictor.load(\"/Users/ivanananevski/projects/anonymeter-copy/anonymeter/notebooks/AutogluonModels/ag-20251121_092838\")\n",
+      "No path specified. Models will be saved in: \"AutogluonModels/ag-20251121_092850\"\n",
+      "Verbosity: 2 (Standard Logging)\n",
+      "=================== System Info ===================\n",
+      "AutoGluon Version:  1.4.0\n",
+      "Python Version:     3.9.22\n",
+      "Operating System:   Darwin\n",
+      "Platform Machine:   x86_64\n",
+      "Platform Version:   Darwin Kernel Version 24.6.0: Mon Jul 14 11:28:17 PDT 2025; root:xnu-11417.140.69~1/RELEASE_X86_64\n",
+      "CPU Count:          12\n",
+      "Memory Avail:       5.02 GB / 16.00 GB (31.4%)\n",
+      "Disk Space Avail:   23.66 GB / 465.63 GB (5.1%)\n",
+      "===================================================\n",
+      "No presets specified! To achieve strong results with AutoGluon, it is recommended to use the available presets. Defaulting to `'medium'`...\n",
+      "\tRecommended Presets (For more details refer to https://auto.gluon.ai/stable/tutorials/tabular/tabular-essentials.html#presets):\n",
+      "\tpresets='extreme' : New in v1.4: Massively better than 'best' on datasets <30000 samples by using new models meta-learned on https://tabarena.ai: TabPFNv2, TabICL, Mitra, and TabM. Absolute best accuracy. Requires a GPU. Recommended 64 GB CPU memory and 32+ GB GPU memory.\n",
+      "\tpresets='best'    : Maximize accuracy. Recommended for most users. Use in competitions and benchmarks.\n",
+      "\tpresets='high'    : Strong accuracy with fast inference speed.\n",
+      "\tpresets='good'    : Good accuracy with very fast inference speed.\n",
+      "\tpresets='medium'  : Fast training time, ideal for initial prototyping.\n",
+      "Beginning AutoGluon training ...\n",
+      "AutoGluon will save models to \"/Users/ivanananevski/projects/anonymeter-copy/anonymeter/notebooks/AutogluonModels/ag-20251121_092850\"\n",
+      "Train Data Rows:    39040\n",
+      "Train Data Columns: 14\n",
+      "Label Column:       fnlwgt\n",
+      "AutoGluon infers your prediction problem is: 'regression' (because dtype of label-column == int and many unique label-values observed).\n",
+      "\tLabel info (max, min, mean, stddev): (814369, 13492, 181263.22961, 100607.94976)\n",
+      "\tIf 'regression' is not the correct problem_type, please manually specify the problem_type parameter during Predictor init (You may specify problem_type as one of: ['binary', 'multiclass', 'regression', 'quantile'])\n",
+      "Problem Type:       regression\n",
+      "Preprocessing data ...\n",
+      "Using Feature Generators to preprocess the data ...\n",
+      "Fitting AutoMLPipelineFeatureGenerator...\n",
+      "\tAvailable Memory:                    5106.42 MB\n",
+      "\tTrain Data (Original)  Memory Usage: 23.49 MB (0.5% of available memory)\n",
+      "\tInferring data type of each feature based on column values. Set feature_metadata_in to manually specify special dtypes of the features.\n",
+      "\tStage 1 Generators:\n",
+      "\t\tFitting AsTypeFeatureGenerator...\n",
+      "\t\t\tNote: Converting 2 features to boolean dtype as they only contain 2 unique values.\n",
+      "\tStage 2 Generators:\n",
+      "\t\tFitting FillNaFeatureGenerator...\n",
+      "\tStage 3 Generators:\n",
+      "\t\tFitting IdentityFeatureGenerator...\n",
+      "\t\tFitting CategoryFeatureGenerator...\n",
+      "\t\t\tFitting CategoryMemoryMinimizeFeatureGenerator...\n",
+      "\tStage 4 Generators:\n",
+      "\t\tFitting DropUniqueFeatureGenerator...\n",
+      "\tStage 5 Generators:\n",
+      "\t\tFitting DropDuplicatesFeatureGenerator...\n",
+      "\tTypes of features in original data (raw dtype, special dtypes):\n",
+      "\t\t('int', [])    : 5 | ['age', 'education_num', 'capital_gain', 'capital_loss', 'hr_per_week']\n",
+      "\t\t('object', []) : 9 | ['type_employer', 'education', 'marital', 'occupation', 'relationship', ...]\n",
+      "\tTypes of features in processed data (raw dtype, special dtypes):\n",
+      "\t\t('category', [])  : 7 | ['type_employer', 'education', 'marital', 'occupation', 'relationship', ...]\n",
+      "\t\t('int', [])       : 5 | ['age', 'education_num', 'capital_gain', 'capital_loss', 'hr_per_week']\n",
+      "\t\t('int', ['bool']) : 2 | ['sex', 'income']\n",
+      "\t0.4s = Fit runtime\n",
+      "\t14 features in original data used to generate 14 features in processed data.\n",
+      "\tTrain Data (Processed) Memory Usage: 1.83 MB (0.0% of available memory)\n",
+      "Data preprocessing and feature engineering runtime = 0.46s ...\n",
+      "AutoGluon will gauge predictive performance using evaluation metric: 'root_mean_squared_error'\n",
+      "\tThis metric's sign has been flipped to adhere to being higher_is_better. The metric score can be multiplied by -1 to get the metric value.\n",
+      "\tTo change this, specify the eval_metric parameter of Predictor()\n",
+      "Automatically generating train/validation split with holdout_frac=0.06403688524590163, Train Rows: 36540, Val Rows: 2500\n",
+      "User-specified model hyperparameters to be fit:\n",
+      "{\n",
+      "\t'RF': [{}],\n",
+      "}\n",
+      "Fitting 1 L1 models, fit_strategy=\"sequential\" ...\n",
+      "Fitting model: RandomForest ...\n",
+      "\tFitting with cpus=12, gpus=0, mem=0.0/5.0 GB\n",
+      "\t-96603.1958\t = Validation score   (-root_mean_squared_error)\n",
+      "\t10.78s\t = Training   runtime\n",
+      "\t0.13s\t = Validation runtime\n",
+      "Fitting model: WeightedEnsemble_L2 ...\n",
+      "\tEnsemble Weights: {'RandomForest': 1.0}\n",
+      "\t-96603.1958\t = Validation score   (-root_mean_squared_error)\n",
+      "\t0.0s\t = Training   runtime\n",
+      "\t0.0s\t = Validation runtime\n",
+      "AutoGluon training complete, total runtime = 12.3s ... Best model: WeightedEnsemble_L2 | Estimated inference throughput: 19535.7 rows/s (2500 batch size)\n",
+      "TabularPredictor saved. To load, use: predictor = TabularPredictor.load(\"/Users/ivanananevski/projects/anonymeter-copy/anonymeter/notebooks/AutogluonModels/ag-20251121_092850\")\n",
+      "No path specified. Models will be saved in: \"AutogluonModels/ag-20251121_093000\"\n",
+      "Verbosity: 2 (Standard Logging)\n",
+      "=================== System Info ===================\n",
+      "AutoGluon Version:  1.4.0\n",
+      "Python Version:     3.9.22\n",
+      "Operating System:   Darwin\n",
+      "Platform Machine:   x86_64\n",
+      "Platform Version:   Darwin Kernel Version 24.6.0: Mon Jul 14 11:28:17 PDT 2025; root:xnu-11417.140.69~1/RELEASE_X86_64\n",
+      "CPU Count:          12\n",
+      "Memory Avail:       6.53 GB / 16.00 GB (40.8%)\n",
+      "Disk Space Avail:   35.48 GB / 465.63 GB (7.6%)\n",
+      "===================================================\n",
+      "No presets specified! To achieve strong results with AutoGluon, it is recommended to use the available presets. Defaulting to `'medium'`...\n",
+      "\tRecommended Presets (For more details refer to https://auto.gluon.ai/stable/tutorials/tabular/tabular-essentials.html#presets):\n",
+      "\tpresets='extreme' : New in v1.4: Massively better than 'best' on datasets <30000 samples by using new models meta-learned on https://tabarena.ai: TabPFNv2, TabICL, Mitra, and TabM. Absolute best accuracy. Requires a GPU. Recommended 64 GB CPU memory and 32+ GB GPU memory.\n",
+      "\tpresets='best'    : Maximize accuracy. Recommended for most users. Use in competitions and benchmarks.\n",
+      "\tpresets='high'    : Strong accuracy with fast inference speed.\n",
+      "\tpresets='good'    : Good accuracy with very fast inference speed.\n",
+      "\tpresets='medium'  : Fast training time, ideal for initial prototyping.\n",
+      "Beginning AutoGluon training ...\n",
+      "AutoGluon will save models to \"/Users/ivanananevski/projects/anonymeter-copy/anonymeter/notebooks/AutogluonModels/ag-20251121_093000\"\n",
+      "Train Data Rows:    39040\n",
+      "Train Data Columns: 14\n",
+      "Label Column:       education\n",
+      "AutoGluon infers your prediction problem is: 'multiclass' (because dtype of label-column == object).\n",
+      "\tFirst 10 (of 16) unique label values:  ['HS-grad', '11th', 'Masters', 'Some-college', '5th-6th', 'Bachelors', 'Assoc-acdm', '9th', 'Doctorate', 'Prof-school']\n",
+      "\tIf 'multiclass' is not the correct problem_type, please manually specify the problem_type parameter during Predictor init (You may specify problem_type as one of: ['binary', 'multiclass', 'regression', 'quantile'])\n",
+      "Problem Type:       multiclass\n",
+      "Preprocessing data ...\n",
+      "Train Data Class Count: 16\n",
+      "Using Feature Generators to preprocess the data ...\n",
+      "Fitting AutoMLPipelineFeatureGenerator...\n",
+      "\tAvailable Memory:                    6593.09 MB\n",
+      "\tTrain Data (Original)  Memory Usage: 21.36 MB (0.3% of available memory)\n",
+      "\tInferring data type of each feature based on column values. Set feature_metadata_in to manually specify special dtypes of the features.\n",
+      "\tStage 1 Generators:\n",
+      "\t\tFitting AsTypeFeatureGenerator...\n",
+      "\t\t\tNote: Converting 2 features to boolean dtype as they only contain 2 unique values.\n",
+      "\tStage 2 Generators:\n",
+      "\t\tFitting FillNaFeatureGenerator...\n",
+      "\tStage 3 Generators:\n",
+      "\t\tFitting IdentityFeatureGenerator...\n",
+      "\t\tFitting CategoryFeatureGenerator...\n",
+      "\t\t\tFitting CategoryMemoryMinimizeFeatureGenerator...\n",
+      "\tStage 4 Generators:\n",
+      "\t\tFitting DropUniqueFeatureGenerator...\n",
+      "\tStage 5 Generators:\n",
+      "\t\tFitting DropDuplicatesFeatureGenerator...\n",
+      "\tTypes of features in original data (raw dtype, special dtypes):\n",
+      "\t\t('int', [])    : 6 | ['age', 'fnlwgt', 'education_num', 'capital_gain', 'capital_loss', ...]\n",
+      "\t\t('object', []) : 8 | ['type_employer', 'marital', 'occupation', 'relationship', 'race', ...]\n",
+      "\tTypes of features in processed data (raw dtype, special dtypes):\n",
+      "\t\t('category', [])  : 6 | ['type_employer', 'marital', 'occupation', 'relationship', 'race', ...]\n",
+      "\t\t('int', [])       : 6 | ['age', 'fnlwgt', 'education_num', 'capital_gain', 'capital_loss', ...]\n",
+      "\t\t('int', ['bool']) : 2 | ['sex', 'income']\n",
+      "\t0.4s = Fit runtime\n",
+      "\t14 features in original data used to generate 14 features in processed data.\n",
+      "\tTrain Data (Processed) Memory Usage: 2.09 MB (0.0% of available memory)\n",
+      "Data preprocessing and feature engineering runtime = 0.56s ...\n",
+      "AutoGluon will gauge predictive performance using evaluation metric: 'accuracy'\n",
+      "\tTo change this, specify the eval_metric parameter of Predictor()\n",
+      "Automatically generating train/validation split with holdout_frac=0.06403688524590163, Train Rows: 36540, Val Rows: 2500\n",
+      "User-specified model hyperparameters to be fit:\n",
+      "{\n",
+      "\t'RF': [{}],\n",
+      "}\n",
+      "Fitting 1 L1 models, fit_strategy=\"sequential\" ...\n",
+      "Fitting model: RandomForest ...\n",
+      "\tFitting with cpus=12, gpus=0, mem=0.4/6.4 GB\n",
+      "\t0.8064\t = Validation score   (accuracy)\n",
+      "\t4.32s\t = Training   runtime\n",
+      "\t0.12s\t = Validation runtime\n",
+      "Fitting model: WeightedEnsemble_L2 ...\n",
+      "\tEnsemble Weights: {'RandomForest': 1.0}\n",
+      "\t0.8064\t = Validation score   (accuracy)\n",
+      "\t0.01s\t = Training   runtime\n",
+      "\t0.0s\t = Validation runtime\n",
+      "AutoGluon training complete, total runtime = 7.13s ... Best model: WeightedEnsemble_L2 | Estimated inference throughput: 20566.3 rows/s (2500 batch size)\n",
+      "TabularPredictor saved. To load, use: predictor = TabularPredictor.load(\"/Users/ivanananevski/projects/anonymeter-copy/anonymeter/notebooks/AutogluonModels/ag-20251121_093000\")\n",
+      "No path specified. Models will be saved in: \"AutogluonModels/ag-20251121_093013\"\n",
+      "Verbosity: 2 (Standard Logging)\n",
+      "=================== System Info ===================\n",
+      "AutoGluon Version:  1.4.0\n",
+      "Python Version:     3.9.22\n",
+      "Operating System:   Darwin\n",
+      "Platform Machine:   x86_64\n",
+      "Platform Version:   Darwin Kernel Version 24.6.0: Mon Jul 14 11:28:17 PDT 2025; root:xnu-11417.140.69~1/RELEASE_X86_64\n",
+      "CPU Count:          12\n",
+      "Memory Avail:       5.17 GB / 16.00 GB (32.3%)\n",
+      "Disk Space Avail:   35.78 GB / 465.63 GB (7.7%)\n",
+      "===================================================\n",
+      "No presets specified! To achieve strong results with AutoGluon, it is recommended to use the available presets. Defaulting to `'medium'`...\n",
+      "\tRecommended Presets (For more details refer to https://auto.gluon.ai/stable/tutorials/tabular/tabular-essentials.html#presets):\n",
+      "\tpresets='extreme' : New in v1.4: Massively better than 'best' on datasets <30000 samples by using new models meta-learned on https://tabarena.ai: TabPFNv2, TabICL, Mitra, and TabM. Absolute best accuracy. Requires a GPU. Recommended 64 GB CPU memory and 32+ GB GPU memory.\n",
+      "\tpresets='best'    : Maximize accuracy. Recommended for most users. Use in competitions and benchmarks.\n",
+      "\tpresets='high'    : Strong accuracy with fast inference speed.\n",
+      "\tpresets='good'    : Good accuracy with very fast inference speed.\n",
+      "\tpresets='medium'  : Fast training time, ideal for initial prototyping.\n",
+      "Beginning AutoGluon training ...\n",
+      "AutoGluon will save models to \"/Users/ivanananevski/projects/anonymeter-copy/anonymeter/notebooks/AutogluonModels/ag-20251121_093013\"\n",
+      "Train Data Rows:    39040\n",
+      "Train Data Columns: 14\n",
+      "Label Column:       education_num\n",
+      "AutoGluon infers your prediction problem is: 'multiclass' (because dtype of label-column == int, but few unique label-values observed).\n",
+      "\tFirst 10 (of 14) unique label values:  [9, 7, 14, 10, 13, 12, 4, 16, 15, 11]\n",
+      "\tIf 'multiclass' is not the correct problem_type, please manually specify the problem_type parameter during Predictor init (You may specify problem_type as one of: ['binary', 'multiclass', 'regression', 'quantile'])\n",
+      "Problem Type:       multiclass\n",
+      "Preprocessing data ...\n",
+      "Warning: Some classes in the training set have fewer than 10 examples. AutoGluon will only keep 13 out of 14 classes for training and will not try to predict the rare classes. To keep more classes, increase the number of datapoints from these rare classes in the training data or reduce label_count_threshold.\n",
+      "Fraction of data from classes with at least 10 examples that will be kept for training models: 0.9998463114754098\n",
+      "Train Data Class Count: 13\n",
+      "Using Feature Generators to preprocess the data ...\n",
+      "Fitting AutoMLPipelineFeatureGenerator...\n",
+      "\tAvailable Memory:                    5330.31 MB\n",
+      "\tTrain Data (Original)  Memory Usage: 23.49 MB (0.4% of available memory)\n",
+      "\tInferring data type of each feature based on column values. Set feature_metadata_in to manually specify special dtypes of the features.\n",
+      "\tStage 1 Generators:\n",
+      "\t\tFitting AsTypeFeatureGenerator...\n",
+      "\t\t\tNote: Converting 2 features to boolean dtype as they only contain 2 unique values.\n",
+      "\tStage 2 Generators:\n",
+      "\t\tFitting FillNaFeatureGenerator...\n",
+      "\tStage 3 Generators:\n",
+      "\t\tFitting IdentityFeatureGenerator...\n",
+      "\t\tFitting CategoryFeatureGenerator...\n",
+      "\t\t\tFitting CategoryMemoryMinimizeFeatureGenerator...\n",
+      "\tStage 4 Generators:\n",
+      "\t\tFitting DropUniqueFeatureGenerator...\n",
+      "\tStage 5 Generators:\n",
+      "\t\tFitting DropDuplicatesFeatureGenerator...\n",
+      "\tTypes of features in original data (raw dtype, special dtypes):\n",
+      "\t\t('int', [])    : 5 | ['age', 'fnlwgt', 'capital_gain', 'capital_loss', 'hr_per_week']\n",
+      "\t\t('object', []) : 9 | ['type_employer', 'education', 'marital', 'occupation', 'relationship', ...]\n",
+      "\tTypes of features in processed data (raw dtype, special dtypes):\n",
+      "\t\t('category', [])  : 7 | ['type_employer', 'education', 'marital', 'occupation', 'relationship', ...]\n",
+      "\t\t('int', [])       : 5 | ['age', 'fnlwgt', 'capital_gain', 'capital_loss', 'hr_per_week']\n",
+      "\t\t('int', ['bool']) : 2 | ['sex', 'income']\n",
+      "\t0.4s = Fit runtime\n",
+      "\t14 features in original data used to generate 14 features in processed data.\n",
+      "\tTrain Data (Processed) Memory Usage: 1.83 MB (0.0% of available memory)\n",
+      "Data preprocessing and feature engineering runtime = 0.44s ...\n",
+      "AutoGluon will gauge predictive performance using evaluation metric: 'accuracy'\n",
+      "\tTo change this, specify the eval_metric parameter of Predictor()\n",
+      "Automatically generating train/validation split with holdout_frac=0.06403688524590163, Train Rows: 36534, Val Rows: 2500\n",
+      "User-specified model hyperparameters to be fit:\n",
+      "{\n",
+      "\t'RF': [{}],\n",
+      "}\n",
+      "Fitting 1 L1 models, fit_strategy=\"sequential\" ...\n",
+      "Fitting model: RandomForest ...\n",
+      "\tFitting with cpus=12, gpus=0, mem=0.3/5.1 GB\n",
+      "\t0.8564\t = Validation score   (accuracy)\n",
+      "\t4.43s\t = Training   runtime\n",
+      "\t0.14s\t = Validation runtime\n",
+      "Fitting model: WeightedEnsemble_L2 ...\n",
+      "\tEnsemble Weights: {'RandomForest': 1.0}\n",
+      "\t0.8564\t = Validation score   (accuracy)\n",
+      "\t0.02s\t = Training   runtime\n",
+      "\t0.0s\t = Validation runtime\n",
+      "AutoGluon training complete, total runtime = 7.3s ... Best model: WeightedEnsemble_L2 | Estimated inference throughput: 17247.9 rows/s (2500 batch size)\n",
+      "TabularPredictor saved. To load, use: predictor = TabularPredictor.load(\"/Users/ivanananevski/projects/anonymeter-copy/anonymeter/notebooks/AutogluonModels/ag-20251121_093013\")\n"
+     ]
+    }
+   ],
+   "source": [
+    "columns = ori.columns[:5] # Using the first 5 columns only for this example\n",
+    "results = []\n",
+    "\n",
+    "results_for_groups = {}\n",
+    "\n",
+    "for secret in columns:\n",
+    "    predictor = TabularPredictor(label=secret)\n",
+    "    predictor.fit(\n",
+    "        train_data=syn,\n",
+    "        hyperparameters={\n",
+    "            'RF': {},  \n",
+    "        }\n",
+    "    )\n",
+    "    inference_model = MLModelInference(predictor)\n",
+    "    aux_cols = [col for col in columns if col != secret]\n",
+    "    \n",
+    "    # Needed for anonymeter (is the target categorical or numerical variable?)\n",
+    "    regression = True if syn[secret].dtype.kind in 'iuf' else False\n",
+    "    \n",
+    "    # Init the InferenceEvaluator\n",
+    "    evaluator = InferenceEvaluator(ori=ori, \n",
+    "                                   syn=syn, \n",
+    "                                   control=control,\n",
+    "                                   aux_cols=aux_cols,\n",
+    "                                   secret=secret,\n",
+    "                                   n_attacks=6000, #todo important! this parameter is obsolete if sample_attacks is False\n",
+    "                                   regression = regression,\n",
+    "                                   inference_model=inference_model) # Setting to true for this example\n",
+    "    # Evaluate\n",
+    "    evaluator.evaluate(n_jobs=-2)\n",
+    "    \n",
+    "    # Gather the results for the current group\n",
+    "    results_for_groups[secret] = evaluator.risk_for_groups()\n",
+    "    results.append((secret, evaluator.results()))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "59651586-891c-4ae6-8b28-fe7a7de66aeb",
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2025-05-15T11:53:46.772503Z",
+     "start_time": "2025-05-15T11:53:00.992149Z"
+    },
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "for secret in columns:\n",
+    "    curr_res = results_for_groups[secret]\n",
+    "\n",
+    "    fig, ax = plt.subplots()\n",
+    "\n",
+    "    risks = [res[\"risk\"].value for _, res in curr_res.items()]\n",
+    "    columns = [key for key,_ in curr_res.items()]\n",
+    "\n",
+    "    ax.bar(x=columns, height=risks, alpha=0.5, ecolor='black', capsize=10)\n",
+    "\n",
+    "    plt.xticks(rotation=45, ha='right')\n",
+    "    plt.title(secret)\n",
+    "    ax.set_ylabel(\"Measured inference risk\")\n",
+    "    _ = ax.set_xlabel(\"Secret column\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "c790d894-e3d9-4cee-bd08-f4ef5f364d4e",
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2025-05-15T11:53:47.147242Z",
+     "start_time": "2025-05-15T11:53:46.776658Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig, ax = plt.subplots()\n",
+    "\n",
+    "risks = [res[1].risk().value for res in results]\n",
+    "columns = [res[0] for res in results]\n",
+    "\n",
+    "ax.bar(x=columns, height=risks, alpha=0.5, ecolor='black', capsize=10)\n",
+    "\n",
+    "plt.xticks(rotation=45, ha='right')\n",
+    "plt.title(\"Risk scores per secret\")\n",
+    "ax.set_ylabel(\"Measured inference risk\")\n",
+    "_ = ax.set_xlabel(\"Secret column\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "79503a44-6fc3-4ba4-a56d-97cb8994d176",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.22"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "237cf5f6b3dcd73bf2688629baee50bd53e43ee0aa8f2bde7060bbd4d3c193da"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/src/anonymeter/evaluators/__init__.py b/src/anonymeter/evaluators/__init__.py
index 26b2848..d31cb0a 100644
--- a/src/anonymeter/evaluators/__init__.py
+++ b/src/anonymeter/evaluators/__init__.py
@@ -5,5 +5,6 @@
 from anonymeter.evaluators.inference_evaluator import InferenceEvaluator
 from anonymeter.evaluators.linkability_evaluator import LinkabilityEvaluator
 from anonymeter.evaluators.singling_out_evaluator import SinglingOutEvaluator
+from anonymeter.evaluators.inference_predictor import KNNInference, MLModelInference
 
-__all__ = ["InferenceEvaluator", "LinkabilityEvaluator", "SinglingOutEvaluator"]
+__all__ = ["InferenceEvaluator", "LinkabilityEvaluator", "SinglingOutEvaluator", "KNNInference", "MLModelInference"]
diff --git a/src/anonymeter/evaluators/inference_evaluator.py b/src/anonymeter/evaluators/inference_evaluator.py
index 94c87dd..341a40c 100644
--- a/src/anonymeter/evaluators/inference_evaluator.py
+++ b/src/anonymeter/evaluators/inference_evaluator.py
@@ -2,49 +2,58 @@
 # Copyright (c) 2022 Anonos IP LLC.
 # See https://github.com/statice/anonymeter/blob/main/LICENSE.md for details.
 """Privacy evaluator that measures the inference risk."""
-
-from typing import Optional
+from typing import Optional, Union
 
 import numpy as np
 import numpy.typing as npt
 import pandas as pd
+from pandas import DataFrame
 
-from anonymeter.neighbors.mixed_types_kneighbors import MixedTypeKNeighbors
 from anonymeter.stats.confidence import EvaluationResults, PrivacyRisk
 
+from anonymeter.evaluators.inference_predictor import InferencePredictor, KNNInference
+
 
 def _run_attack(
-    target: pd.DataFrame,
-    syn: pd.DataFrame,
-    n_attacks: int,
-    aux_cols: list[str],
-    secret: str,
-    n_jobs: int,
-    naive: bool,
-    regression: Optional[bool],
-) -> int:
+        target: pd.DataFrame,
+        syn: pd.DataFrame,
+        n_attacks: int,
+        aux_cols: list[str],
+        secret: str,
+        n_jobs: int,
+        naive: bool,
+        regression: Optional[bool],
+        inference_model: Optional[InferencePredictor],
+) -> tuple[int, Union[tuple[npt.NDArray, npt.NDArray], npt.NDArray, None], DataFrame]:
     if regression is None:
         regression = pd.api.types.is_numeric_dtype(target[secret])
 
-    targets = target.sample(n_attacks, replace=False)
-
     if naive:
         guesses = syn.sample(n_attacks)[secret]
-
+        targets = target.sample(n_attacks, replace=False)
     else:
-        nn = MixedTypeKNeighbors(n_jobs=n_jobs, n_neighbors=1).fit(candidates=syn[aux_cols])
+        # Instantiate the default KNN model if no other model is passed through `inference_model`.
+        if inference_model is None:
+            inference_model = KNNInference(data=syn, columns=aux_cols, target_col=secret, n_jobs=n_jobs)
 
-        guesses_idx = nn.kneighbors(queries=targets[aux_cols])
-        if isinstance(guesses_idx, tuple):
-            raise RuntimeError("guesses_idx cannot be a tuple")
+        # Depending on the `inference_model.sample_targets` we might not need to sample,
+        # but rather predict for all targets as we assume it can scale to all target samples.
+        if inference_model.sample_targets:
+            targets = target.sample(n_attacks, replace=False)
+        else:
+            targets = target
 
-        guesses = syn.iloc[guesses_idx.flatten()][secret]
+        guesses = inference_model.predict(targets)
 
-    return evaluate_inference_guesses(guesses=guesses, secrets=targets[secret], regression=regression).sum()
+    assert len(guesses) == len(targets), "Predictions and targets have different lengths"
+    guesses.index = targets.index
+
+    return evaluate_inference_guesses(guesses=guesses, secrets=targets[secret],
+                                      regression=regression).sum(), guesses, targets
 
 
 def evaluate_inference_guesses(
-    guesses: pd.Series, secrets: pd.Series, regression: bool, tolerance: float = 0.05
+        guesses: pd.Series, secrets: pd.Series, regression: bool, tolerance: float = 0.05
 ) -> npt.NDArray:
     """Evaluate the success of an inference attack.
 
@@ -142,23 +151,36 @@ class InferenceEvaluator:
         the variable.
     n_attacks : int, default is 500
         Number of attack attempts.
+    inference_model: InferencePredictor
+        An ml model fitted on `syn` as training data, and `secret` as target, that supports ::predict(x).
+        If not None, it will be used over the MixedTypeKNeighbors in the attack.
 
     """
 
     def __init__(
-        self,
-        ori: pd.DataFrame,
-        syn: pd.DataFrame,
-        aux_cols: list[str],
-        secret: str,
-        regression: Optional[bool] = None,
-        n_attacks: int = 500,
-        control: Optional[pd.DataFrame] = None,
+            self,
+            ori: pd.DataFrame,
+            syn: pd.DataFrame,
+            aux_cols: list[str],
+            secret: str,
+            regression: Optional[bool] = None,
+            n_attacks: int = 500,
+            control: Optional[pd.DataFrame] = None,
+            inference_model: Optional[InferencePredictor] = None
     ):
         self._ori = ori
         self._syn = syn
         self._control = control
         self._n_attacks = n_attacks
+        self._inference_model = inference_model
+        if self._inference_model is not None and not self._inference_model.sample_targets:
+            self._n_attacks_ori = self._ori.shape[0]
+            self._n_attacks_baseline = min(self._syn.shape[0], self._n_attacks_ori)
+            self._n_attacks_control = self._control.shape[0]
+        else:
+            self._n_attacks_ori = self._n_attacks
+            self._n_attacks_baseline = self._n_attacks
+            self._n_attacks_control = self._n_attacks
 
         # check if secret is a string column
         if not isinstance(secret, str):
@@ -172,17 +194,21 @@ def __init__(
         self._regression = regression
         self._aux_cols = aux_cols
         self._evaluated = False
+        self._data_groups = self._ori[self._secret].unique().tolist()
 
-    def _attack(self, target: pd.DataFrame, naive: bool, n_jobs: int) -> int:
+    def _attack(self, target: pd.DataFrame, naive: bool, n_jobs: int, n_attacks: int) -> tuple[
+        int, Union[tuple[npt.NDArray, npt.NDArray],
+        pd.Series, None], DataFrame]:
         return _run_attack(
             target=target,
             syn=self._syn,
-            n_attacks=self._n_attacks,
+            n_attacks=n_attacks,
             aux_cols=self._aux_cols,
             secret=self._secret,
             n_jobs=n_jobs,
             naive=naive,
             regression=self._regression,
+            inference_model=self._inference_model,
         )
 
     def evaluate(self, n_jobs: int = -2) -> "InferenceEvaluator":
@@ -199,10 +225,17 @@ def evaluate(self, n_jobs: int = -2) -> "InferenceEvaluator":
             The evaluated ``InferenceEvaluator`` object.
 
         """
-        self._n_baseline = self._attack(target=self._ori, naive=True, n_jobs=n_jobs)
-        self._n_success = self._attack(target=self._ori, naive=False, n_jobs=n_jobs)
-        self._n_control = (
-            None if self._control is None else self._attack(target=self._control, naive=False, n_jobs=n_jobs)
+        # n_attacks is effective here
+        self._n_baseline, _, _ = self._attack(target=self._ori, naive=True, n_jobs=n_jobs,
+                                              n_attacks=self._n_attacks_baseline)
+
+        # n_attacks is not effective here, just needed for the baseline
+        self._n_success, self.guesses_success, self.target = self._attack(target=self._ori, naive=False, n_jobs=n_jobs,
+                                                                          n_attacks=self._n_attacks_ori)
+        # n_attacks is not effective here, just needed for the baseline
+        self._n_control, self.guesses_control, self.target_control = (
+            None if self._control is None else self._attack(target=self._control, naive=False, n_jobs=n_jobs,
+                                                            n_attacks=self._n_attacks_control)
         )
 
         self._evaluated = True
@@ -227,6 +260,9 @@ def results(self, confidence_level: float = 0.95) -> EvaluationResults:
 
         return EvaluationResults(
             n_attacks=self._n_attacks,
+            n_attacks_ori=self._n_attacks_ori,
+            n_attacks_baseline=self._n_attacks_baseline,
+            n_attacks_control=self._n_attacks_control,
             n_success=self._n_success,
             n_baseline=self._n_baseline,
             n_control=self._n_control,
@@ -258,3 +294,73 @@ def risk(self, confidence_level: float = 0.95, baseline: bool = False) -> Privac
         """
         results = self.results(confidence_level=confidence_level)
         return results.risk(baseline=baseline)
+
+    def risk_for_groups(self, confidence_level: float = 0.95) -> dict[
+        str, dict[str, Union[EvaluationResults, PrivacyRisk]]]:
+        """Compute the attack risks on a group level, for every unique value of `self._data_groups`.
+
+            Parameters
+            ----------
+            confidence_level : float, default is 0.95
+                Confidence level for the error bound calculation.
+
+            Returns
+            -------
+            dict[str, dict[str, EvaluationResults | PrivacyRisk]]
+                The group as a key, and then for every group the results (EvaluationResults),
+                and the risks (PrivacyRisk).
+
+            """
+        if not self._evaluated:
+            self.evaluate(n_jobs=-2)
+
+        all_results = {}
+
+        # For every unique group in `self._data_groups`
+        for group in self._data_groups:
+            # Get the targets for the current group
+            target = self.target[self.target[self._secret] == group]
+
+            # Get the guesses for the current group
+            guess = self.guesses_success.loc[target.index]
+
+            # Count the number of success attacks
+            n_success = evaluate_inference_guesses(guesses=guess,
+                                                   secrets=target[self._secret],
+                                                   regression=self._regression).sum()
+
+            if self._control is not None:
+                # Get the targets for the current control group
+                target_control = self.target_control[self.target_control[self._secret] == group]
+
+                # Get the guesses for the current control group
+                guesses_control = self.guesses_control.loc[target_control.index]
+
+                # Count the number of success control attacks
+                n_control = evaluate_inference_guesses(guesses=guesses_control,
+                                                       secrets=target_control[self._secret],
+                                                       regression=self._regression).sum()
+            else:
+                n_control = None
+
+            # Recreate the EvaluationResults for the current group
+            results = EvaluationResults(
+                n_attacks=self._n_attacks,  # passing for
+                n_attacks_ori=self._n_attacks_ori,
+                n_attacks_baseline=self._n_attacks_baseline,
+                # We leave the overall n_attacks_baseline here, it doesn't change the risk
+                n_attacks_control=self._n_attacks_control,
+                n_success=n_success,
+                n_baseline=self._n_baseline,  # We leave the overall _n_baseline here, it doesn't change the risk
+                n_control=n_control,
+                confidence_level=confidence_level,
+            )
+            # Compute the risk
+            risk = results.risk()
+
+            all_results[group] = {
+                "results": results,
+                "risk": risk
+            }
+
+        return all_results
diff --git a/src/anonymeter/evaluators/inference_predictor.py b/src/anonymeter/evaluators/inference_predictor.py
new file mode 100644
index 0000000..8b5e62c
--- /dev/null
+++ b/src/anonymeter/evaluators/inference_predictor.py
@@ -0,0 +1,45 @@
+from typing import Protocol
+
+import pandas as pd
+
+from anonymeter.neighbors.mixed_types_kneighbors import MixedTypeKNeighbors
+
+
+class InferencePredictor(Protocol):
+    def predict(self, X: pd.DataFrame) -> pd.Series:
+        ...
+
+    @property
+    def sample_targets(self) -> bool:
+        ...
+
+
+class MLModelInference:
+    def __init__(self, model, sample_targets: bool = False):
+        self._model = model
+        self._sample_targets = sample_targets
+
+    @property
+    def sample_targets(self) -> bool:
+        return self._sample_targets
+
+    def predict(self, x: pd.DataFrame) -> pd.Series:
+        return self._model.predict(x)
+
+
+class KNNInference:
+    def __init__(self, data: pd.DataFrame, columns: list[str], target_col: str, n_jobs: int):
+        self._nn = MixedTypeKNeighbors(n_jobs=n_jobs, n_neighbors=1).fit(candidates=data[columns])
+        self._data = data
+        self._target_col = target_col
+        self._columns = columns
+
+    @property
+    def sample_targets(self) -> bool:
+        return True
+
+    def predict(self, x: pd.DataFrame) -> pd.Series:
+        guesses_idx = self._nn.kneighbors(queries=x[self._columns])
+        if isinstance(guesses_idx, tuple):
+            raise RuntimeError("guesses_idx cannot be a tuple")
+        return self._data.iloc[guesses_idx.flatten()][self._target_col]
\ No newline at end of file
diff --git a/src/anonymeter/stats/confidence.py b/src/anonymeter/stats/confidence.py
index b95049c..7f3469c 100644
--- a/src/anonymeter/stats/confidence.py
+++ b/src/anonymeter/stats/confidence.py
@@ -199,18 +199,30 @@ def __init__(
         n_baseline: int,
         n_control: Optional[int] = None,
         confidence_level: float = 0.95,
+        n_attacks_ori: Optional[int] = None,
+        n_attacks_baseline: Optional[int] = None,
+        n_attacks_control: Optional[int] = None,
     ):
-        self.attack_rate = success_rate(n_total=n_attacks, n_success=n_success, confidence_level=confidence_level)
+        self.n_attacks = n_attacks
+        if None in [n_attacks_ori, n_attacks_baseline, n_attacks_control]:
+            # Setting all to self.n_attacks because of at least one missing value..
+            n_attacks_ori = self.n_attacks
+            n_attacks_baseline = self.n_attacks
+            n_attacks_control = self.n_attacks
+
+        self.attack_rate = success_rate(n_total=n_attacks_ori, n_success=n_success, confidence_level=confidence_level)
 
-        self.baseline_rate = success_rate(n_total=n_attacks, n_success=n_baseline, confidence_level=confidence_level)
+        self.baseline_rate = success_rate(n_total=n_attacks_baseline, n_success=n_baseline, confidence_level=confidence_level)
 
         self.control_rate = (
             None
             if n_control is None
-            else success_rate(n_total=n_attacks, n_success=n_control, confidence_level=confidence_level)
+            else success_rate(n_total=n_attacks_control, n_success=n_control, confidence_level=confidence_level)
         )
 
-        self.n_attacks = n_attacks
+        self.n_attacks_ori = n_attacks_ori
+        self.n_attacks_baseline = n_attacks_baseline
+        self.n_attacks_control = n_attacks_control
         self.n_success = n_success
         self.n_baseline = n_baseline
         self.n_control = n_control