diff --git a/notebooks/01_simple-tests.ipynb b/notebooks/01_simple-tests.ipynb index d90baee..de74dcb 100644 --- a/notebooks/01_simple-tests.ipynb +++ b/notebooks/01_simple-tests.ipynb @@ -307,10 +307,120 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 23, "id": "23e6f86c", "metadata": {}, "outputs": [], + "source": [ + "# No Noise case\n", + "x = np.linspace(-1, 1, 1000)\n", + "y = datamanip.runge_function(x)\n", + "x_train, x_test, y_train, y_test = train_test_split(\n", + " x, y, test_size=0.2, random_state=datamanip.get_RNG().integers(0, 1e6)\n", + ")\n", + "\n", + "x_train = x_train.reshape(-1, 1)\n", + "x_test = x_test.reshape(-1, 1)\n", + "y_train = y_train.reshape(-1, 1)\n", + "y_test = y_test.reshape(-1, 1)\n", + "\n", + "X_train_scaled, X_test_scaled = datamanip.scale_data(x_train, x_test)\n", + "y_train_scaled, y_test_scaled = datamanip.scale_data(y_train, y_test)" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "id": "63a37bc9", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Hidden Layers: 1, Neurons: 50, Test MSE: 0.15527058910285813\n", + "Hidden Layers: 1, Neurons: 100, Test MSE: 0.10180048256207899\n", + "Hidden Layers: 2, Neurons: 50, Test MSE: 3.143921430819739e-05\n", + "Hidden Layers: 2, Neurons: 100, Test MSE: 1.0371793804978556e-05\n" + ] + } + ], + "source": [ + "for n_hidden in [1, 2]:\n", + " for n_neurons in [50, 100]:\n", + " layers = get_regression_model(n_hidden, n_neurons)\n", + " network = FFNN(\n", + " layers,\n", + " AdamScheduler(learning_rate=0.001, epochs=1000),\n", + " MSELoss(),\n", + " )\n", + " network.fit(X_train_scaled, y_train_scaled)\n", + " y_pred_nn = network.predict(X_test_scaled)\n", + " test_mse = mean_squared_error(y_test_scaled, y_pred_nn)\n", + " print(f\"Hidden Layers: {n_hidden}, Neurons: {n_neurons}, Test MSE: {test_mse}\")" + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "9b568bde", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "OLS Test MSE: 0.019835005963678293\n" + ] + } + ], + "source": [ + "x_train = datamanip.polynomial_features(x_train.flatten(), p=10)\n", + "x_test = datamanip.polynomial_features(x_test.flatten(), p=10)\n", + "X_train_scaled, X_test_scaled = datamanip.scale_data(x_train, x_test)\n", + "\n", + "\n", + "beta = optimizers.Ridge_parameters(X_train_scaled, y_train_scaled, lam=1e-10)\n", + "y_pred = X_test_scaled @ beta\n", + "test_mse = mean_squared_error(y_test_scaled, y_pred)\n", + "print(f\"OLS Test MSE: {test_mse}\")\n" + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "id": "9db82101", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import plotting\n", + "\n", + "plt.scatter(x_test[:,1], y_test_scaled, s=6, label=\"True Data\")\n", + "plt.scatter(x_test[:,1], y_pred, s=6, label=\"OLS Prediction\")\n", + "plt.scatter(x_test[:,1], y_pred_nn, s=6, label=\"NN Prediction\")\n", + "plt.legend()\n", + "plt.xlabel(\"$x$\")\n", + "plt.ylabel(r\"$y / \\sigma_y$\")\n", + "plt.savefig(\"runge_no_noise_comparison.pdf\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "79016960", + "metadata": {}, + "outputs": [], "source": [] } ], diff --git a/notebooks/runge_no_noise_comparison.pdf b/notebooks/runge_no_noise_comparison.pdf new file mode 100644 index 0000000..e10d489 Binary files /dev/null and b/notebooks/runge_no_noise_comparison.pdf differ diff --git a/report/main.pdf b/report/main.pdf index 6161ee8..def6310 100644 Binary files a/report/main.pdf and b/report/main.pdf differ diff --git a/report/main.tex b/report/main.tex index 44dc526..e719555 100644 --- a/report/main.tex +++ b/report/main.tex @@ -281,6 +281,13 @@ As we mentioned before, there is the possibility for our model to overadjust for A further hindrance in our approach with using the output of a regression model to make classifications is, that for an efficient workflow, we need a full but independent prediction of the dataset. To achieve this, we use a workflow inspired by cross-validation. In cross-validation you use multiple folds of the dataset, whereby for every fold, a different part of the dataset is used as testing data, and the rest of the dataset is used for training purposes. This provides independence of training and testing data while being able to test the models general performance for the entire dataset. For out-of-fold prediction, we use the same approach of changing the training data to leave out a specific part each time, but furthermore we use the model trained on each training set, to make predictions for the remaining set. Combining all the predicted sets into one set, allows for having independent predictions for the entire dataset. Having the predictions be independent is important, as otherwise no statement can be made on the generalization ability of the workflow. Using this approach we can expect the same performance on new and yet unseen data. \subsection{Implementation} +\begin{figure} + \centering + \includegraphics[width=\columnwidth]{../notebooks/runge_no_noise_comparison.pdf} + \caption{Comparison of predictions made by OLS regression of polynomial degree 10 and a FFNN with two hidden layers of 100 nodes each on the Runge function without noise.} + \label{fig:runge_comparison} +\end{figure} + All the code necessary to use a FFNN is implemented from scratch in \texttt{Python3} as the \texttt{easynn} library. The source code therefore is located in the \texttt{/src/easynn/} folder of the accompanying git repository \footnote{\url{https://github.uio.no/larsbog/FYSSTK-Project2}}. It is separated into two main modules, called \texttt{schedulers} and \texttt{feedforward}. The scheduler modules use modified versions of our previous work on using gradient descent techniques for ordinary least squares techniques. It is also structured in a modular way, so that scheduling tasks like early stopping and dynamic learning rate adjustment are easy to implement in further versions. The base scheduler implements an update method of signature \texttt{def update(self, gradients: gradient\_\-type) -> gradient\_\-type:}, with \texttt{gradient\_\-type = List[\-Tuple[\-np.ndarray, np.ndarray]] }. This can be overwritten to implement any optimization algorithm. The method takes the gradients of the cost functions with respect to the weights and biases of each layer. Each layer is assigned a tuple of two arrays of the gradients. It should then return the updates for each of the weights and biases in the same format. The training process which fetches new updates after each evaluation of the loss and gradients will continue will the \texttt{cont} property of the base class is true. @@ -300,7 +307,8 @@ As a further validation of the functionality, we use the Runge function \begin{equation} f(x) = \frac{1}{1 + 25 x^2}, \end{equation} -with added Gaussian noise of standard deviation \num{1.0} to create a regression dataset. On this dataset of size $N = \num{100000}$ and \qty{80}{\percent}-\qty{20}{\percent} train-test-splitting we achieve a MSE of \numrange{0.9136}{0.9178}, using between one and two hidden layers with either 50 or 100 nodes per hidden layer. This exceeds the performance of our previous work on least squares regression using different regularization and optimization techniques \cite{bognerRegularizationOptimizationAll2025}, where a MSE of \num{0.92} was achieved using the same dataset \footnote{The $x$ values were expanded into multiple of features containing polynomial multiples $x^n$ of themselves. For optimal performance $n \in \{0, 1, \dots, \geq 10\}$ were required.} and train-test-splitting. This motivates further the use of neural networks for regression tasks, as possible non-linearities in the data can be modeled more easily. Least squares methods on the other hand provide only linear combinations of the input features. +with added Gaussian noise of standard deviation \num{1.0} to create a regression dataset. On this dataset of size $N = \num{100000}$ and \qty{80}{\percent}-\qty{20}{\percent} train-test-splitting we achieve a MSE of \numrange{0.9136}{0.9178}, using between one and two hidden layers with either 50 or 100 nodes per hidden layer. This exceeds the performance of our previous work on least squares regression using different regularization and optimization techniques \cite{bognerRegularizationOptimizationAll2025}, where a MSE of \num{0.92} was achieved using the same dataset \footnote{The $x$ values were expanded into multiple of features containing polynomial multiples $x^n$ of themselves. For optimal performance $n \in \{0, 1, \dots, \geq 10\}$ were required.} and train-test-splitting. +In the case of no added noise, the difference is even more stark. In the case of 2 hidden layers with Leaky ReLU activation and 100 nodes per hidden layer on $N= \num{1000}$ points, we achieve $\mathrm{MSE} = \num{1.037e-5}$. OLS methods achieve a performance of \num{0.01984} in this case. \Cref{fig:runge_comparison} shows the predictions by both methods in form of a plot. This motivates further the use of neural networks for regression tasks, as possible non-linearities in the data can be modeled more easily. Least squares methods on the other hand provide only linear combinations of the input features. \subsection{Use of AI tools}