Skip to content

Instantly share code, notes, and snippets.

@fmnobar
Created June 5, 2023 01:39
Show Gist options
  • Save fmnobar/4f8165624c20bb050410001028bbffec to your computer and use it in GitHub Desktop.
Save fmnobar/4f8165624c20bb050410001028bbffec to your computer and use it in GitHub Desktop.
HPO - Hyperopt
Display the source blob
Display the rendered blob
Raw
{
"cells": [
{
"cell_type": "markdown",
"id": "d26e8eea-90a2-4100-b140-dcd765ef274c",
"metadata": {},
"source": [
"# Hyperparameter Optimization With Hyperopt - Intro & Implementation\n",
"\n",
"[Hyperopt](https://github.com/hyperopt/hyperopt) is an open-source hyperparameter optimization tool that I personally use to improve my machine learning projects and have found it to be quite easy to implement.\n",
"\n",
"Hyperparameter optimization, is the process of identifying the best combination of hyperparameters for a machine learning model to satisfy an objective function (usually this is defined as minimizing the objective function for consistency). To use a different analogy, each machine learning model comes with various knobs and levers that we can tune, until we get the outcome from the model that we have been looking for. The act of finding the right combination of hyperparameters that result in the outcome that we have been looking for is called hyperparameter optimization. Some examples of such parameters are: learning rate, architecture of a neural network (e.g. number of hidden layers), choice of the optimizer, etc.\n",
"\n",
"In this post we will implement Hyperopt to [ADD DETAILS]. If you are interested in exploring other hyperparameter optimization strategies, such as grid search, random search and bayesian optimization, check out the post below:\n",
"\n",
"[Intro & Implementation of Grid Search, Random Search and Bayesian Optimization](https://towardsdatascience.com/hyperparameter-optimization-intro-and-implementation-of-grid-search-random-search-and-bayesian-b2f16c00578a)\n",
"\n",
"Let's get started!\n",
"\n",
"## 1. Basics \n",
"\n",
"### 1.1. Concepts and Installation\n",
"\n",
"Let's first define some relevant concepts for using Hyperopt.\n",
"\n",
"- **Objective Function:** This is the function that hyperparameter optimization tries to minimize. More specifically, objective function accepts a combination of hyperparameters as input and returns the error level of the model, given those accepted hyperparameters. The goal of hyperparameter optimization is to find the combination of hyperparameters that minimize this error.\n",
"- **Search Space:** The range of input values (i.e. parameters) that the Objective Function accepts as arguments.\n",
"- **Optimization Algorithm:** As the name suggests, this is the algorithm used to minimize the Objective Function. Hyperopt utilizes different search algorithms, such as random search and Tree of Parzen Estimators (TPE) ([documentation](http://hyperopt.github.io/hyperopt/#algorithms)). We will discuss this further with examples.\n",
"\n",
"Now that we are familiar with the concepts, let's start by installing [Hyperopt](https://github.com/hyperopt/hyperopt) by running the command below:"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "6acc755e-a142-42a6-b313-e0df8c8404bd",
"metadata": {},
"outputs": [],
"source": [
"!pip install hyperopt"
]
},
{
"cell_type": "markdown",
"id": "14aafbd5-e4f3-46e6-aafd-f186641e4d72",
"metadata": {},
"source": [
"Now that we have the library installed, we will first walk through a very simple example to get a handle on how Hyperopt works. After that, we will move on to a more interesting and complicated example. \n",
"\n",
"### 1.2. Simple Example\n",
"\n",
"Let's start with a very simple one to help us understand how the overall process of hyperparameter optimization using Hyperopt works. We will start with a quadratic function of `f(x) = (x - 1)^2`. This function's optimized point is at `x = 1` and therefore we know what to expect. Let's implement this using Hyperopt and see how it works. \n",
"\n",
"In order to do so, we will take the following steps:\n",
"1. Import necessary libraries and packages\n",
"2. Define the objective function and the search space\n",
"3. Run the optimization process\n",
"4. Print the results (i.e. the optimized point that we expect to be `x = 1`)\n",
"\n",
"Code block below, follows the above steps:"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "ec0687af-393a-4c41-b98a-86d3d448fe9a",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"100%|██████████████████████████████| 100/100 [00:00<00:00, 546.43trial/s, best loss: 8.974426119521573e-06]\n",
"{'x': 0.9970042653456086}\n"
]
}
],
"source": [
"# 1. Import necessary libraries and packages\n",
"from hyperopt import hp, fmin, tpe, Trials\n",
"\n",
"# 2. Define the objective function and the search space\n",
"def objective_function(x):\n",
" return (x - 1)**2\n",
"\n",
"search_space = hp.uniform('x', -2, 2)\n",
"\n",
"# 3. Run the optimization process\n",
"\n",
"# Trials object to store the results\n",
"trials = Trials()\n",
"\n",
"# Run the optimization\n",
"best = fmin(fn=objective_function, space=search_space, algo=tpe.suggest, trials=trials, max_evals=100)\n",
"\n",
"# 4. Print the results\n",
"print(best)"
]
},
{
"cell_type": "markdown",
"id": "a081ec2b-0869-4051-b5bc-75378b10f879",
"metadata": {},
"source": [
"\"best\" returns the best combination of hyperparameters that the model was able to find and in this case it is almost equal to `x = 1`, as we expected! The process to implement Hyperopt is generally the same and now that we have walked through a simple one, let's move to a more advanced example. "
]
},
{
"cell_type": "markdown",
"id": "b72e5795-5bfc-415c-91fd-1753d3e188fa",
"metadata": {},
"source": [
"## 2. Hyperopt Implementation\n",
"\n",
"### 2.1. Support Vector Machines and Iris Data Set\n",
"\n",
"In a previous [post](https://towardsdatascience.com/hyperparameter-optimization-intro-and-implementation-of-grid-search-random-search-and-bayesian-b2f16c00578a) I used grid search, random search and Bayesian optimization for hyperparameter optimization using the [Iris data set provided by scikit-learn](https://scikit-learn.org/stable/auto_examples/datasets/plot_iris_dataset.html). Iris data set includes 3 different irises petal and sepal length and is a commonly-used data set for classification exercise. In this post, we will use the same data set but we will use a support vector machine (SVM) as a model with two parameters that we can optimize as follows:\n",
"\n",
"- `C`: Regularization parameter, which trades off misclassification of training examples against simplicity of the decision surface\n",
"- `gamma`: Kernel coefficient, which defines how much influence a single training example has. The larger gamma is, the closer other examples must be to be affected\n",
"\n",
"Since the goal of this exercise is to go through the hyperparameter optimization, I will not go deeper into what SVMs do but if you are interested, I find [this](https://scikit-learn.org/stable/modules/svm.html) scikit-learn post helpful.\n",
"\n",
"We will generally follow the same steps that we used in the simple example earlier but will also visualize the process at the end:\n",
"\n",
"1. Import necessary libraries and packages\n",
"2. Define the objective function and the search space\n",
"3. Run the optimization process\n",
"4. Visualize the optimization\n",
"\n",
"#### 2.1.1. Step 1 - Import Libraries and Packages\n",
"Let's import the libraries and packages and then load the data set. "
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "d3aee0e8-a725-4e0a-a286-54178cc6e2c4",
"metadata": {},
"outputs": [],
"source": [
"# Import libraries and packages\n",
"from sklearn import datasets\n",
"from sklearn.svm import SVC\n",
"from sklearn.model_selection import cross_val_score\n",
"\n",
"# Load Iris dataset\n",
"iris = datasets.load_iris()\n",
"X = iris.data\n",
"y = iris.target"
]
},
{
"cell_type": "markdown",
"id": "b1c43a93-f22f-48ac-a81a-cb68f0d41303",
"metadata": {},
"source": [
"#### 2.1.2. Step 2 - Define Objective Function and Search Space\n",
"\n",
"Let's first start with defining the objective function, whihc will train an SVM and returns the negative of the cross-validation score, which is what we want to minimize. Note that we are minimizing the negative of cross-validation score to be consistent with the general goal of \"minimizing\" the objective function (instead of \"maximizing\" the cross-validation score)."
]
},
{
"cell_type": "code",
"execution_count": 28,
"id": "c9c89182-8e54-4a69-921d-318fc75b0473",
"metadata": {},
"outputs": [],
"source": [
"def objective_function(parameters):\n",
" clf = SVC(**parameters)\n",
" score = cross_val_score(clf, X, y, cv=5).mean()\n",
" return -score"
]
},
{
"cell_type": "markdown",
"id": "290a6280-6ac1-44e2-b49b-5d21e24e4b01",
"metadata": {},
"source": [
"Next we will define the search space, which consists of the values that our parameters of C and gamma can take. Note that we will use Hyperopt's `hp.uniform(label, low, high)`, which returns a value uniformly between \"low\" and \"high\" ([source](http://hyperopt.github.io/hyperopt/getting-started/search_spaces/))."
]
},
{
"cell_type": "code",
"execution_count": 29,
"id": "91734ab6-3935-414f-a0a1-cb37c2e15d02",
"metadata": {},
"outputs": [],
"source": [
"# Search Space\n",
"search_space = {\n",
" 'C': hp.uniform('C', 0.1, 10),\n",
" 'gamma': hp.uniform('gamma', 0.01, 1)\n",
"}"
]
},
{
"cell_type": "markdown",
"id": "93d0ab40-d40d-4f6e-a357-a35d894c0e7d",
"metadata": {},
"source": [
"#### 2.1.3. Run Optimization\n",
"\n",
"Same as the simple example earlier, we will use a TPE algorithm and store the results in a `Trials` object. "
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "3ab00845-773c-4423-bf86-c66154b1751a",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"100%|████████████████████████████████| 100/100 [00:00<00:00, 100.07trial/s, best loss: -0.9866666666666667]\n"
]
}
],
"source": [
"# Trials object to store the results\n",
"trials = Trials()\n",
"\n",
"# Run optimization\n",
"best = fmin(fn=objective_function, space=search_space, algo=tpe.suggest, trials=trials, max_evals=100)"
]
},
{
"cell_type": "markdown",
"id": "80b3e247-8214-4417-b5c4-6f39f3e62e72",
"metadata": {},
"source": [
"#### 2.1.4. Visualize Optimization\n",
"\n",
"As we remember from the simple example, \"best\" includes the selected set of hyperparameters that Hyperopt found based on the implemented optimization strategy. Let's look at the results!"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "c045544a-07c5-42e8-bfca-fe51fe43dea1",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"{'C': 5.699828691001509, 'gamma': 0.045618829804396965}\n"
]
}
],
"source": [
"print(best)"
]
},
{
"cell_type": "markdown",
"id": "0afc4666-a9b6-4794-9f3a-805f27887349",
"metadata": {},
"source": [
"As expected, now we have a combination of hyperparameters that minimize the optimization function, using Hyperopt. \n",
"\n",
"Let's visually look at how the objective function values change as the hyperparameters change. We will start with defining a function named `plot_obj_vs_hp()` that accomplishes this visualization. And then use that function to visualize the results. Make sure to look for the red dot - that one indicates the best combination of hyperparameters, according to our hyperparameter optimization!"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "adad4f67-b72f-40ef-8a7c-2973ba6c24e0",
"metadata": {},
"outputs": [],
"source": [
"# Import libraries\n",
"import matplotlib.pyplot as plt\n",
"\n",
"def plot_obj_vs_hp(trials, search_space, best):\n",
" # Extract the results\n",
" results = trials.trials\n",
" \n",
" # Create a list of hyperparameters\n",
" hyperparameters = list(search_space.keys())\n",
" \n",
" # Create a new figure with 2 subplots side by side\n",
" fig, axes = plt.subplots(1, 2, figsize=(12, 6))\n",
" \n",
" # Loop through hyperparameters and generate plots\n",
" for idx, hp in enumerate(hyperparameters):\n",
" # Extract the values of a given hyperparameter\n",
" hp_values = [res['misc']['vals'][f'{hp}'] for res in results]\n",
"\n",
" # Flatten the list of values\n",
" hp_values = [item for sublist in hp_values for item in sublist]\n",
"\n",
" # Extract the corresponding objective function values\n",
" objective_values = [res['result']['loss'] for res in results]\n",
"\n",
" # Create the scatter plot\n",
" axes[idx].scatter(hp_values, objective_values, label='Trial Hyperparameter Combinations')\n",
" \n",
" # Highlight the best hyperparameters\n",
" axes[idx].scatter(best[hp], min(objective_values), color='red', label='Best Hyperparameter Combinations')\n",
" axes[idx].set_xlabel(f'{hp}')\n",
" axes[idx].set_ylabel('Loss')\n",
" axes[idx].set_title(f'Loss vs. {hp}')\n",
" axes[idx].legend(loc='upper right')\n",
" \n",
" plt.tight_layout()\n",
" plt.show()\n"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "e35faa3f-9c44-480b-a1e7-1d942310ba7e",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 864x432 with 2 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"# Plot optimization vs. hyperparameters in 2D\n",
"plot_obj_vs_hp(trials, search_space, best)"
]
},
{
"cell_type": "markdown",
"id": "dc470da0-d883-443f-a128-6b602d704938",
"metadata": {},
"source": [
"Note that since 'C' and 'gamma' are not really related to each other, we are showing them separately versus changes of the objective function. Since we want the objective function to be minimized, then we're looking for the furthest bottom side of the plots above and based on the results of the hyperparameter optimization, we know that what we are looking for is where `{'C': 5.164418859504847, 'gamma': 0.07084064498886927}`, which results in an objective function loss of around -0.986.\n",
"\n",
"I was also curious to look at these plots in a three-dimensional manner so I created the function below to accomplish that. Let's look at the plot. "
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "893ce2a6-ba9c-4774-9011-8826723caa02",
"metadata": {},
"outputs": [],
"source": [
"from mpl_toolkits.mplot3d import Axes3D\n",
"import matplotlib.pyplot as plt\n",
"\n",
"def plot_obj_vs_hp_3d(trials, search_space, best):\n",
" # Extract the results\n",
" results = trials.trials\n",
" \n",
" # Create a list of hyperparameters\n",
" hyperparameters = list(search_space.keys())\n",
"\n",
" # Extract the values of hyperparameters\n",
" hp_values_0 = [res['misc']['vals'][f'{hyperparameters[0]}'] for res in results]\n",
" hp_values_1 = [res['misc']['vals'][f'{hyperparameters[1]}'] for res in results]\n",
"\n",
" # Flatten the lists of values\n",
" hp_values_0 = [item for sublist in hp_values_0 for item in sublist]\n",
" hp_values_1 = [item for sublist in hp_values_1 for item in sublist]\n",
"\n",
" # Extract the corresponding objective function values\n",
" objective_values = [res['result']['loss'] for res in results]\n",
"\n",
" # Create a new figure\n",
" fig = plt.figure(figsize=(10, 7))\n",
"\n",
" # Add a 3D subplot\n",
" ax = fig.add_subplot(111, projection='3d')\n",
"\n",
" # Create the scatter plot\n",
" scatter = ax.scatter(hp_values_0, hp_values_1, objective_values, c=objective_values, cmap='viridis', label='Trial hyperparameters')\n",
"\n",
" # Highlight the best hyperparameters\n",
" ax.scatter(best[hyperparameters[0]], best[hyperparameters[1]], min(objective_values), color='red', label='Best hyperparameters')\n",
"\n",
" # Add labels using hyperparameters from search_space\n",
" ax.set_xlabel(hyperparameters[0])\n",
" ax.set_ylabel(hyperparameters[1])\n",
" ax.set_zlabel('Loss')\n",
" ax.set_title('Loss Across Hyperparameters')\n",
" fig.colorbar(scatter)\n",
" ax.legend(loc='upper right')\n",
"\n",
" plt.show()\n"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "25b24a02-d546-4b27-9b9e-cdcdf4de1f36",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 720x504 with 2 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"# Plot optimization vs. hyperparameters in 3D\n",
"plot_obj_vs_hp_3d(trials, search_space, best)"
]
},
{
"cell_type": "markdown",
"id": "135436f6-a756-4f7e-83b4-aff446b38887",
"metadata": {},
"source": [
"Admittedly, this is not very easy to read but let's give it a shot. We are looking for the lowest loss, which is the darkest dots on the plot. Visually it aligns with the two-dimensional plots that we had generated before. \n",
"\n",
"Next, let's focus on a different example. \n",
"\n",
"### 2.2. Random Forest and Diabetes Data Set\n",
"\n",
"This example focuses on a regression model that attempts at measuring the progression of the disease, one year after baseline. This data set is also taken from [scikit-learn](https://scikit-learn.org/stable/datasets/toy_dataset.html#diabetes-dataset) but the difference is that the data set is mainly used for regression (instead of classification that we looked at with the Iris example). If you are interested in learning more about the differences between regression and classification, check out the post below. \n",
"\n",
"[Classification vs. Regression in Machine Learning — Which One Should I Use?](https://medium.com/@fmnobar/classification-vs-regression-in-machine-learning-which-one-should-i-use-f6b24d251c46)\n",
"\n",
"We will use a [Random Forest Regressor](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html) model for this example and will optimize the objective function for two hyperparameters as follows:\n",
"- `n_estimators`: Number of trees in the random forest\n",
"- `max_depth`: Maximum depth of trees in the random forest\n",
"\n",
"The overall process of optimization is the same as what we have done so far. So, let's break it down into our four usual steps!\n",
"\n",
"#### 2.2.1. Step 1 - Import Libraries and Packages\n",
"\n",
"We will start with importing the libraries and packages and then loading the data set. "
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "c79b6701-53aa-4ae0-96c4-a739fe2c3733",
"metadata": {},
"outputs": [],
"source": [
"# Import libraries and packages\n",
"from sklearn import datasets\n",
"from sklearn.ensemble import RandomForestRegressor\n",
"from sklearn.model_selection import cross_val_score\n",
"from hyperopt import fmin, tpe, hp, Trials\n",
"import matplotlib.pyplot as plt\n",
"from mpl_toolkits.mplot3d import Axes3D\n",
"\n",
"# Load Diabetes dataset\n",
"diabetes = datasets.load_diabetes()\n",
"X = diabetes.data\n",
"y = diabetes.target"
]
},
{
"cell_type": "markdown",
"id": "256da06e-54bd-437c-a54d-e2f746067a46",
"metadata": {},
"source": [
"#### 2.2.2. Step 2 - Define Objective Function and Search Space\n",
"\n",
"Let's first start with defining the objective function, whihc will train our Random Forest Regressor and returns the negative of the cross-validation score, which is what we want to minimize. Note that we are minimizing the negative of cross-validation score to be consistent with the general goal of \"minimizing\" the objective function (instead of \"maximizing\" the cross-validation score).\n",
"\n",
"Next we will define the search space, which consists of the values that our parameters of `n_estimators` and `max_depth` can take. Note that we will use Hyperopt's `hp.choice(label, options)`, which takes a value for the hyperparameter and the possible values for that hyperparameter ([source](http://hyperopt.github.io/hyperopt/getting-started/search_spaces/))."
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "d323cc70-aa21-46c1-8bc8-a5c33cd3aafc",
"metadata": {},
"outputs": [],
"source": [
"# Define objective function\n",
"def objective_function(parameters):\n",
" # Initiate RandomForestRegressor\n",
" regressor = RandomForestRegressor(**parameters)\n",
" \n",
" # Calculate the mean cross-validation score using 5 folds\n",
" score = cross_val_score(regressor, X, y, cv=5).mean()\n",
" \n",
" return -score\n",
"\n",
"# Define search Space\n",
"search_space = {\n",
" 'n_estimators': hp.choice('n_estimators', range(10, 300)),\n",
" 'max_depth': hp.choice('max_depth', range(1, 30)),\n",
"}"
]
},
{
"cell_type": "markdown",
"id": "f841ee99-f7fd-4198-92e9-cbb873e5904d",
"metadata": {},
"source": [
"#### 2.2.3. Run Optimization\n",
"\n",
"Same as the examples earlier, we will use a TPE algorithm and store the results in a `Trials` object. "
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "dc70a57d-900e-4cbf-a451-32fae7450f28",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████| 100/100 [01:53<00:00, 1.14s/trial, best loss: -0.4403690631798076]\n"
]
}
],
"source": [
"# Trials object to store the results\n",
"trials = Trials()\n",
"\n",
"# Run optimization\n",
"best = fmin(fn=objective_function, space=search_space, algo=tpe.suggest, trials=trials, max_evals=100)"
]
},
{
"cell_type": "markdown",
"id": "db93c59e-79f8-4160-8a94-26bae5edbb90",
"metadata": {},
"source": [
"#### 2.2.4. Visualize Optimization\n",
"\n",
"As we remember from the simple example, \"best\" includes the selected set of hyperparameters that Hyperopt found based on the implemented optimization strategy. Let's look at the results! "
]
},
{
"cell_type": "code",
"execution_count": 15,
"id": "fbcbb1f2-e0f7-4d4f-8781-7e5963a97b43",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"{'max_depth': 3, 'n_estimators': 92}\n"
]
}
],
"source": [
"print(best)"
]
},
{
"cell_type": "markdown",
"id": "f631803f-2786-44ff-b7e8-8a097a74c9c2",
"metadata": {},
"source": [
"As expected, now we have a combination of hyperparameters that minimize the optimization function, using Hyperopt. \n",
"\n",
"Let's visually look at how the objective function values change as the hyperparameters change, using the functions that we had defined before to create the plots in two and three dimensions. "
]
},
{
"cell_type": "code",
"execution_count": 16,
"id": "65167751-0d5c-40d7-871a-572132b63267",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 864x432 with 2 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"# Plot optimization vs. hyperparameters in 2D\n",
"plot_obj_vs_hp(trials, search_space, best)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"id": "93394aad-2114-4c3f-8bf7-a3ebd9203eb1",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 720x504 with 2 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"# Plot optimization vs. hyperparameters in 3D\n",
"plot_obj_vs_hp_3d(trials, search_space, best)"
]
},
{
"cell_type": "markdown",
"id": "10534613-7fd8-4aec-8491-6172ecd8cb78",
"metadata": {},
"source": [
"## Conclusion\n",
"\n",
"In this post, we introduced [Hyperopt](https://github.com/hyperopt/hyperopt), a powerful and open-source hyperparameter optimization tool and then walked through examples of implementation in the context of classification via a Support Vector Machine and regression via a Random Forest Regressor. Then we looked the best combination of hyperparameters found through these processes and visualized the results."
]
},
{
"cell_type": "markdown",
"id": "d0011a65-2fc3-4534-ac69-e07006722aac",
"metadata": {},
"source": [
"## X. Extras"
]
},
{
"cell_type": "code",
"execution_count": 27,
"id": "b2de5a5b-7aff-429e-9be9-8839ee36cc97",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"# Import libraries\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"\n",
"# Define the function\n",
"def f(x):\n",
" return (x - 1) ** 2\n",
"\n",
"# Generate x values from -5 to 5\n",
"x = np.linspace(-4, 6, 100)\n",
"\n",
"# Calculate corresponding y values\n",
"y = f(x)\n",
"\n",
"# Find the minimum point\n",
"min_point = np.min(y)\n",
"\n",
"# Create the plot\n",
"plt.plot(x, y, label='f(x) = (x-1)^2')\n",
"plt.xlabel('x')\n",
"plt.ylabel('f(x)')\n",
"plt.title('f(x) = (x-1)^2')\n",
"\n",
"# Set the x-axis limits\n",
"plt.xlim(-4, 6)\n",
"\n",
"# Add a horizontal dashed line at the minimum point\n",
"plt.axhline(y=min_point, color='red', linestyle='dashed', label='Minimum Point')\n",
"\n",
"# Add a legend\n",
"plt.legend()\n",
"\n",
"# Display the plot\n",
"plt.show()\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "187f4eaf-3d99-4a4f-b6cf-5f347775cb39",
"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": "3.9.7"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment