audio_SVM.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[],"authorship_tag":"ABX9TyMuY6U8e1rq2f/ds9UtgmTy"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"markdown","source":["# Audio SVM\n","\n","### Importing all libraries"],"metadata":{"id":"1LwcwdupTwxK"}},{"cell_type":"code","execution_count":null,"metadata":{"id":"el8dRBphSEpv"},"outputs":[],"source":["from google.colab import drive\n","from sklearn import metrics\n","from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n","from sklearn.multiclass import OneVsRestClassifier\n","from sklearn.pipeline import make_pipeline\n","from sklearn.preprocessing import StandardScaler\n","from sklearn.svm import SVC\n","import numpy as np \n","import matplotlib.pyplot as plt\n","from sklearn.model_selection import cross_val_score\n","import pickle"]},{"cell_type":"markdown","source":["### Connecting to Google Drive"],"metadata":{"id":"5CFey2IqT3Bh"}},{"cell_type":"code","source":["drive.mount('/content/drive')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"lDwR2oqYS4h4","executionInfo":{"status":"ok","timestamp":1672072662579,"user_tz":-60,"elapsed":21505,"user":{"displayName":"Antonio Cozzolino","userId":"17154683919260126550"}},"outputId":"38ba7728-53cd-47fe-f410-340a55f52d45"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Mounted at /content/drive\n"]}]},{"cell_type":"markdown","source":["### Loading datasets (Numpy arrays)\n","Datasets are saved in numpy arrays for load its quickly "],"metadata":{"id":"W7OWCEdPT6oB"}},{"cell_type":"code","source":["X_train = np.load('/content/drive/MyDrive/ml_dl/audio-2/training_set/X_train.npy')\n","y_train = np.load('/content/drive/MyDrive/ml_dl/audio-2/training_set/y_train.npy')\n","\n","X_validation = np.load('/content/drive/MyDrive/ml_dl/audio-2/validation_set/X_validation.npy')\n","y_validation = np.load('/content/drive/MyDrive/ml_dl/audio-2/validation_set/y_validation.npy')\n","\n","X_test = np.load('/content/drive/MyDrive/ml_dl/audio-2/test_set/X_test.npy')\n","y_test = np.load('/content/drive/MyDrive/ml_dl/audio-2/test_set/y_test.npy')\n","\n","X = np.load('/content/drive/MyDrive/ml_dl/audio-2/dataset/X.npy')\n","y = np.load('/content/drive/MyDrive/ml_dl/audio-2/dataset/y.npy')"],"metadata":{"id":"qY3Nq3FeS7LN"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["### SVC kernel selection"],"metadata":{"id":"ycAveJuyUB2p"}},{"cell_type":"code","source":["kernel = ['poly', 'rbf', 'sigmoid']\n","acc = []\n","\n","for i in kernel:\n","  svm = make_pipeline(StandardScaler(), SVC(kernel = i))\n","  svm.fit(X_train, y_train)\n","  y_pred = svm.predict(X_validation)\n","  a = metrics.accuracy_score(y_validation, y_pred)\n","  acc.append(a)\n","  print(i + ': ' + str(a))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"mDF7nLnZTGFU","executionInfo":{"status":"ok","timestamp":1672072709344,"user_tz":-60,"elapsed":10823,"user":{"displayName":"Antonio Cozzolino","userId":"17154683919260126550"}},"outputId":"3a05850e-bd44-469f-ba4e-09c89e3cb7b4"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["poly: 0.6319444444444444\n","rbf: 0.9097222222222222\n","sigmoid: 0.7708333333333334\n"]}]},{"cell_type":"code","source":["plt.figure(figsize=(15, 6))\n","plt.plot(kernel, acc)\n","plt.title(\"Kernel performance\")\n","plt.xlabel(\"Kernel\")\n","plt.ylabel(\"Accuracy\")\n","plt.grid(True)\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":404},"id":"fwZLSBUYTJfr","executionInfo":{"status":"ok","timestamp":1672072722886,"user_tz":-60,"elapsed":462,"user":{"displayName":"Antonio Cozzolino","userId":"17154683919260126550"}},"outputId":"4b50a6a3-6256-4e5e-a60a-c6ecfb1b74ea"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1080x432 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["### Training the best model"],"metadata":{"id":"XeBfOaklUIHV"}},{"cell_type":"code","source":["svm = make_pipeline(StandardScaler(), SVC())\n","svm.fit(X_train, y_train)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"jtHX2PAITPSa","executionInfo":{"status":"ok","timestamp":1672072740265,"user_tz":-60,"elapsed":3841,"user":{"displayName":"Antonio Cozzolino","userId":"17154683919260126550"}},"outputId":"79910f28-dd85-4846-b417-d7cf8614d2e4"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["Pipeline(steps=[('standardscaler', StandardScaler()), ('svc', SVC())])"]},"metadata":{},"execution_count":6}]},{"cell_type":"markdown","source":["### Testing on test set"],"metadata":{"id":"E3xLgHajUOqG"}},{"cell_type":"code","source":["y_pred = svm.predict(X_test)\n","print(\"Accuracy:\", metrics.accuracy_score(y_test, y_pred))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"8eK79ufNTSyG","executionInfo":{"status":"ok","timestamp":1672072754223,"user_tz":-60,"elapsed":1965,"user":{"displayName":"Antonio Cozzolino","userId":"17154683919260126550"}},"outputId":"f4a95707-d320-4d8f-cd3f-988d13aab9bd"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Accuracy: 0.8969072164948454\n"]}]},{"cell_type":"markdown","source":["### 10-fold Cross Validation"],"metadata":{"id":"YrEkaiJlUR63"}},{"cell_type":"code","source":["scores = cross_val_score(svm, X = X, y = y, cv=10, n_jobs=-1)\n","accuracy_CV10 = np.mean(scores)\n","print('Cross Validation accuracy: ' + str(accuracy_CV10))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"FUFEP6yBTWir","executionInfo":{"status":"ok","timestamp":1672072819668,"user_tz":-60,"elapsed":45178,"user":{"displayName":"Antonio Cozzolino","userId":"17154683919260126550"}},"outputId":"3b59efc5-c400-4dfb-bc3d-607db3b5ea01"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Cross Validation accuracy: 0.941967213114754\n"]}]},{"cell_type":"markdown","source":["### Saving the best model"],"metadata":{"id":"5Frf2Z5WUVfw"}},{"cell_type":"code","source":["model_path = \"/content/drive/MyDrive/ml_dl/audio-2/models/audio_SVM.pkl\"\n","\n","# save\n","with open(model_path,'wb') as f:\n","    pickle.dump(svm,f)\n","\n","# load\n","with open(model_path, 'rb') as f:\n","    svm = pickle.load(f)"],"metadata":{"id":"969VjnbXTcDb"},"execution_count":null,"outputs":[]}]}