Regression for high-level timbral features - Keras NN smallest model.ipynb

{"cells":[{"cell_type":"code","source":["pip install mirdata"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"CEBMcOkW5dyQ","executionInfo":{"status":"ok","timestamp":1679784243968,"user_tz":-60,"elapsed":24584,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"182a6a6d-9f61-4908-ccc0-f35f9aacbf4e"},"execution_count":1,"outputs":[{"output_type":"stream","name":"stdout","text":["Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n","Collecting mirdata\n","  Downloading mirdata-0.3.7-py3-none-any.whl (14.9 MB)\n","\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m14.9/14.9 MB\u001b[0m \u001b[31m17.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[?25hRequirement already satisfied: pyyaml in /usr/local/lib/python3.9/dist-packages (from mirdata) (6.0)\n","Collecting pretty-midi>=0.2.8\n","  Downloading pretty_midi-0.2.10.tar.gz (5.6 MB)\n","\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m5.6/5.6 MB\u001b[0m \u001b[31m63.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n","Requirement already satisfied: pandas>=1.0.0 in /usr/local/lib/python3.9/dist-packages (from mirdata) (1.4.4)\n","Requirement already satisfied: chardet in /usr/lib/python3/dist-packages (from mirdata) (3.0.4)\n","Requirement already satisfied: smart-open>=5.0.0 in /usr/local/lib/python3.9/dist-packages (from mirdata) (6.3.0)\n","Requirement already satisfied: numpy>=1.16 in /usr/local/lib/python3.9/dist-packages (from mirdata) (1.22.4)\n","Requirement already satisfied: requests in /usr/local/lib/python3.9/dist-packages (from mirdata) (2.27.1)\n","Collecting jams\n","  Downloading jams-0.3.4.tar.gz (51 kB)\n","\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m51.3/51.3 KB\u001b[0m \u001b[31m3.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n","Requirement already satisfied: scipy in /usr/local/lib/python3.9/dist-packages (from mirdata) (1.10.1)\n","Requirement already satisfied: librosa>=0.8.0 in /usr/local/lib/python3.9/dist-packages (from mirdata) (0.10.0.post2)\n","Requirement already satisfied: tqdm in /usr/local/lib/python3.9/dist-packages (from mirdata) (4.65.0)\n","Requirement already satisfied: h5py>=2.10.0 in /usr/local/lib/python3.9/dist-packages (from mirdata) (3.8.0)\n","Collecting Deprecated>=1.2.13\n","  Downloading Deprecated-1.2.13-py2.py3-none-any.whl (9.6 kB)\n","Requirement already satisfied: wrapt<2,>=1.10 in /usr/local/lib/python3.9/dist-packages (from Deprecated>=1.2.13->mirdata) (1.15.0)\n","Requirement already satisfied: numba>=0.51.0 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (0.56.4)\n","Requirement already satisfied: audioread>=2.1.9 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (3.0.0)\n","Requirement already satisfied: scikit-learn>=0.20.0 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (1.2.2)\n","Requirement already satisfied: soundfile>=0.12.1 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (0.12.1)\n","Requirement already satisfied: pooch<1.7,>=1.0 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (1.6.0)\n","Requirement already satisfied: decorator>=4.3.0 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (4.4.2)\n","Requirement already satisfied: msgpack>=1.0 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (1.0.5)\n","Requirement already satisfied: typing-extensions>=4.1.1 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (4.5.0)\n","Requirement already satisfied: lazy-loader>=0.1 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (0.2)\n","Requirement already satisfied: soxr>=0.3.2 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (0.3.4)\n","Requirement already satisfied: joblib>=0.14 in /usr/local/lib/python3.9/dist-packages (from librosa>=0.8.0->mirdata) (1.1.1)\n","Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.9/dist-packages (from pandas>=1.0.0->mirdata) (2022.7.1)\n","Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.9/dist-packages (from pandas>=1.0.0->mirdata) (2.8.2)\n","Collecting mido>=1.1.16\n","  Downloading mido-1.2.10-py2.py3-none-any.whl (51 kB)\n","\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m51.1/51.1 KB\u001b[0m \u001b[31m3.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[?25hRequirement already satisfied: six in /usr/local/lib/python3.9/dist-packages (from pretty-midi>=0.2.8->mirdata) (1.16.0)\n","Requirement already satisfied: sortedcontainers>=2.0.0 in /usr/local/lib/python3.9/dist-packages (from jams->mirdata) (2.4.0)\n","Requirement already satisfied: jsonschema>=3.0.0 in /usr/local/lib/python3.9/dist-packages (from jams->mirdata) (4.3.3)\n","Collecting mir_eval>=0.5\n","  Downloading mir_eval-0.7.tar.gz (90 kB)\n","\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m90.7/90.7 KB\u001b[0m \u001b[31m7.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n","Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.9/dist-packages (from requests->mirdata) (1.26.15)\n","Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.9/dist-packages (from requests->mirdata) (2022.12.7)\n","Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.9/dist-packages (from requests->mirdata) (3.4)\n","Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.9/dist-packages (from requests->mirdata) (2.0.12)\n","Requirement already satisfied: attrs>=17.4.0 in /usr/local/lib/python3.9/dist-packages (from jsonschema>=3.0.0->jams->mirdata) (22.2.0)\n","Requirement already satisfied: pyrsistent!=0.17.0,!=0.17.1,!=0.17.2,>=0.14.0 in /usr/local/lib/python3.9/dist-packages (from jsonschema>=3.0.0->jams->mirdata) (0.19.3)\n","Requirement already satisfied: future in /usr/local/lib/python3.9/dist-packages (from mir_eval>=0.5->jams->mirdata) (0.18.3)\n","Requirement already satisfied: setuptools in /usr/local/lib/python3.9/dist-packages (from numba>=0.51.0->librosa>=0.8.0->mirdata) (67.6.0)\n","Requirement already satisfied: llvmlite<0.40,>=0.39.0dev0 in /usr/local/lib/python3.9/dist-packages (from numba>=0.51.0->librosa>=0.8.0->mirdata) (0.39.1)\n","Requirement already satisfied: appdirs>=1.3.0 in /usr/local/lib/python3.9/dist-packages (from pooch<1.7,>=1.0->librosa>=0.8.0->mirdata) (1.4.4)\n","Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.9/dist-packages (from pooch<1.7,>=1.0->librosa>=0.8.0->mirdata) (23.0)\n","Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.9/dist-packages (from scikit-learn>=0.20.0->librosa>=0.8.0->mirdata) (3.1.0)\n","Requirement already satisfied: cffi>=1.0 in /usr/local/lib/python3.9/dist-packages (from soundfile>=0.12.1->librosa>=0.8.0->mirdata) (1.15.1)\n","Requirement already satisfied: pycparser in /usr/local/lib/python3.9/dist-packages (from cffi>=1.0->soundfile>=0.12.1->librosa>=0.8.0->mirdata) (2.21)\n","Building wheels for collected packages: pretty-midi, jams, mir_eval\n","  Building wheel for pretty-midi (setup.py) ... \u001b[?25l\u001b[?25hdone\n","  Created wheel for pretty-midi: filename=pretty_midi-0.2.10-py3-none-any.whl size=5592303 sha256=403f10b5ad5e53c3b087b6a4b5937f0a34519f2a3c51751540dc13d7b3f23594\n","  Stored in directory: /root/.cache/pip/wheels/75/ec/20/b8e937a5bcf1de547ea5ce465db7de7f6761e15e6f0a01e25f\n","  Building wheel for jams (setup.py) ... \u001b[?25l\u001b[?25hdone\n","  Created wheel for jams: filename=jams-0.3.4-py3-none-any.whl size=64923 sha256=2e10c97827467e108049bbd6084de8adaea11c9c130f46ecd0dafea1ca41c53e\n","  Stored in directory: /root/.cache/pip/wheels/3a/30/bd/89f3791651385a3a5f02865508db932f506b26304f0e428593\n","  Building wheel for mir_eval (setup.py) ... \u001b[?25l\u001b[?25hdone\n","  Created wheel for mir_eval: filename=mir_eval-0.7-py3-none-any.whl size=100718 sha256=4f2a50572c8b9de75bf9d65b336f6f0c25dcae1f7bbe8244996487ffc583cfab\n","  Stored in directory: /root/.cache/pip/wheels/e9/f5/d5/eb3db1d056253da195208853842bce745a84b29f44cab59b6c\n","Successfully built pretty-midi jams mir_eval\n","Installing collected packages: mido, pretty-midi, Deprecated, mir_eval, jams, mirdata\n","Successfully installed Deprecated-1.2.13 jams-0.3.4 mido-1.2.10 mir_eval-0.7 mirdata-0.3.7 pretty-midi-0.2.10\n"]}]},{"cell_type":"code","execution_count":2,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":47248,"status":"ok","timestamp":1679784291209,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"},"user_tz":-60},"id":"RbpAjqCULUiQ","outputId":"fe4053a6-43f7-4c1b-eb14-1af589bec9ea"},"outputs":[{"output_type":"stream","name":"stdout","text":["Mounted at /content/gdrive\n"]}],"source":["import pandas\n","import os \n","import pickle\n","\n","import time\n","\n","import mirdata\n","import numpy as np\n","import tensorflow as tf\n","\n","from sklearn.model_selection import train_test_split\n","from sklearn.preprocessing import StandardScaler\n","\n","from keras.models import Sequential\n","from keras.layers import Input\n","from keras.layers import Dense\n","from keras.layers import Conv1D\n","from keras.layers import Flatten\n","from keras.layers import Dropout\n","from keras.layers import MaxPooling1D\n","\n","from google.colab import drive\n","drive.mount('/content/gdrive')"]},{"cell_type":"code","source":["# GLOBAL PATHS and variables\n","pickleFiles_FolderName = '/content/gdrive/MyDrive/Colab Notebooks/Music Information Retrieval/Assignment/Final'"],"metadata":{"id":"FWF8RwZCL64w","executionInfo":{"status":"ok","timestamp":1679784291210,"user_tz":-60,"elapsed":10,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}}},"execution_count":3,"outputs":[]},{"cell_type":"code","source":["tensF_ShuffleSeed = 0\n","train_split = 0.8 # validation split is the same as train_split"],"metadata":{"id":"RuqWXqC_jlcc","executionInfo":{"status":"ok","timestamp":1679784291210,"user_tz":-60,"elapsed":8,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}}},"execution_count":4,"outputs":[]},{"cell_type":"code","source":["# https://mirdata.readthedocs.io/en/stable/source/mirdata.html#module-mirdata.datasets.freesound_one_shot_percussive_sounds\n","\n","percDataset = mirdata.initialize(\"freesound_one_shot_percussive_sounds\")\n","percDataset.download()  \n","print('Finished downloading freesound_one_shot_percussive_sounds dataset.')\n","percDataset.validate() \n","print('Finished validating freesound_one_shot_percussive_sounds dataset.')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"JBvB8aJMA4-L","executionInfo":{"status":"ok","timestamp":1679784325719,"user_tz":-60,"elapsed":34516,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"75b6773b-87c9-4c1c-8d6e-f50a69c84426"},"execution_count":5,"outputs":[{"output_type":"stream","name":"stderr","text":["107MB [00:06, 16.8MB/s]                           \n","5.35MB [00:02, 2.73MB/s]                            \n","57.3MB [00:04, 13.6MB/s]                            \n","1.29MB [00:01, 752kB/s]                            \n","8.00kB [00:01, 5.76kB/s]\n"]},{"output_type":"stream","name":"stdout","text":["Finished downloading freesound_one_shot_percussive_sounds dataset.\n"]},{"output_type":"stream","name":"stderr","text":["100%|██████████| 1/1 [00:00<00:00, 283.59it/s]\n","100%|██████████| 10254/10254 [00:02<00:00, 3636.96it/s]"]},{"output_type":"stream","name":"stdout","text":["Finished validating freesound_one_shot_percussive_sounds dataset.\n"]},{"output_type":"stream","name":"stderr","text":["\n"]}]},{"cell_type":"code","source":["# Find min and max values of timbral synthesis features annotations, \n","# so that we can normalize later\n","with open(os.path.join(pickleFiles_FolderName, 'audioFileWavef_TimbralFeatures_DF.p'), 'rb') as audioFileWavef_TimbralFeatures_DF_Pickle:\n","  audioFileWavef_TimbralFeatures_DF = pickle.load(audioFileWavef_TimbralFeatures_DF_Pickle)\n","\n","audioFileWavef_TimbralFeatures_DF.head() # successfully retrieved pandas dataframe from pickle file"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":299},"id":"qaiaLNfrJ8uy","executionInfo":{"status":"ok","timestamp":1679784339933,"user_tz":-60,"elapsed":14233,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"b8c7c235-54c8-42c9-e392-ad84d688de2d"},"execution_count":6,"outputs":[{"output_type":"execute_result","data":{"text/plain":["   freesound_sound_id  audio_waveform_sampleN_0  audio_waveform_sampleN_1  \\\n","0              203887                 -0.000031                 -0.000031   \n","1              203893                 -0.000031                 -0.000031   \n","2              319613                  0.023254                 -0.362274   \n","3              361958                 -0.003174                 -0.002441   \n","4              277684                  0.001007                  0.001984   \n","\n","   audio_waveform_sampleN_2  audio_waveform_sampleN_3  \\\n","0                  0.000000                 -0.000031   \n","1                 -0.000031                 -0.000031   \n","2                 -0.115540                  0.606995   \n","3                 -0.009338                 -0.005646   \n","4                  0.002197                  0.002533   \n","\n","   audio_waveform_sampleN_4  audio_waveform_sampleN_5  \\\n","0                 -0.000031                 -0.000031   \n","1                  0.000000                  0.000000   \n","2                  0.025940                 -0.159943   \n","3                 -0.014252                 -0.008392   \n","4                  0.002716                  0.002594   \n","\n","   audio_waveform_sampleN_6  audio_waveform_sampleN_7  \\\n","0                  0.000000                  0.000031   \n","1                  0.000000                  0.000031   \n","2                  0.004211                 -0.385834   \n","3                 -0.015869                  0.002869   \n","4                  0.002838                  0.003174   \n","\n","   audio_waveform_sampleN_8  ...  audio_waveform_sampleN_15997  \\\n","0                  0.000031  ...                           0.0   \n","1                  0.000061  ...                           0.0   \n","2                  0.056091  ...                           0.0   \n","3                  0.007721  ...                           0.0   \n","4                  0.002899  ...                           0.0   \n","\n","   audio_waveform_sampleN_15998  audio_waveform_sampleN_15999   hardness  \\\n","0                           0.0                           0.0  58.964800   \n","1                           0.0                           0.0  68.950243   \n","2                           0.0                           0.0  70.543572   \n","3                           0.0                           0.0  54.433781   \n","4                           0.0                           0.0  71.738502   \n","\n","       depth  brightness  roughness     warmth  sharpness  boominess  \n","0  52.583343   51.286288  54.450442  53.450765  43.850985  36.458663  \n","1  35.996983   71.406639  67.818761  34.422775  58.811990  21.432983  \n","2  54.408973   62.367384  55.920217  39.524543  53.265036  37.984857  \n","3  70.611045   38.445561  43.515676  55.998536  40.757533  43.517493  \n","4  57.402842   73.531486  50.322603  36.656637  56.843763  28.371194  \n","\n","[5 rows x 16008 columns]"],"text/html":["\n","  <div id=\"df-813d6b75-fbae-4dee-bcc1-da66a7b7b05d\">\n","    <div class=\"colab-df-container\">\n","      <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>freesound_sound_id</th>\n","      <th>audio_waveform_sampleN_0</th>\n","      <th>audio_waveform_sampleN_1</th>\n","      <th>audio_waveform_sampleN_2</th>\n","      <th>audio_waveform_sampleN_3</th>\n","      <th>audio_waveform_sampleN_4</th>\n","      <th>audio_waveform_sampleN_5</th>\n","      <th>audio_waveform_sampleN_6</th>\n","      <th>audio_waveform_sampleN_7</th>\n","      <th>audio_waveform_sampleN_8</th>\n","      <th>...</th>\n","      <th>audio_waveform_sampleN_15997</th>\n","      <th>audio_waveform_sampleN_15998</th>\n","      <th>audio_waveform_sampleN_15999</th>\n","      <th>hardness</th>\n","      <th>depth</th>\n","      <th>brightness</th>\n","      <th>roughness</th>\n","      <th>warmth</th>\n","      <th>sharpness</th>\n","      <th>boominess</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>203887</td>\n","      <td>-0.000031</td>\n","      <td>-0.000031</td>\n","      <td>0.000000</td>\n","      <td>-0.000031</td>\n","      <td>-0.000031</td>\n","      <td>-0.000031</td>\n","      <td>0.000000</td>\n","      <td>0.000031</td>\n","      <td>0.000031</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>58.964800</td>\n","      <td>52.583343</td>\n","      <td>51.286288</td>\n","      <td>54.450442</td>\n","      <td>53.450765</td>\n","      <td>43.850985</td>\n","      <td>36.458663</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>203893</td>\n","      <td>-0.000031</td>\n","      <td>-0.000031</td>\n","      <td>-0.000031</td>\n","      <td>-0.000031</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","      <td>0.000000</td>\n","      <td>0.000031</td>\n","      <td>0.000061</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>68.950243</td>\n","      <td>35.996983</td>\n","      <td>71.406639</td>\n","      <td>67.818761</td>\n","      <td>34.422775</td>\n","      <td>58.811990</td>\n","      <td>21.432983</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>319613</td>\n","      <td>0.023254</td>\n","      <td>-0.362274</td>\n","      <td>-0.115540</td>\n","      <td>0.606995</td>\n","      <td>0.025940</td>\n","      <td>-0.159943</td>\n","      <td>0.004211</td>\n","      <td>-0.385834</td>\n","      <td>0.056091</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>70.543572</td>\n","      <td>54.408973</td>\n","      <td>62.367384</td>\n","      <td>55.920217</td>\n","      <td>39.524543</td>\n","      <td>53.265036</td>\n","      <td>37.984857</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>361958</td>\n","      <td>-0.003174</td>\n","      <td>-0.002441</td>\n","      <td>-0.009338</td>\n","      <td>-0.005646</td>\n","      <td>-0.014252</td>\n","      <td>-0.008392</td>\n","      <td>-0.015869</td>\n","      <td>0.002869</td>\n","      <td>0.007721</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>54.433781</td>\n","      <td>70.611045</td>\n","      <td>38.445561</td>\n","      <td>43.515676</td>\n","      <td>55.998536</td>\n","      <td>40.757533</td>\n","      <td>43.517493</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>277684</td>\n","      <td>0.001007</td>\n","      <td>0.001984</td>\n","      <td>0.002197</td>\n","      <td>0.002533</td>\n","      <td>0.002716</td>\n","      <td>0.002594</td>\n","      <td>0.002838</td>\n","      <td>0.003174</td>\n","      <td>0.002899</td>\n","      <td>...</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>0.0</td>\n","      <td>71.738502</td>\n","      <td>57.402842</td>\n","      <td>73.531486</td>\n","      <td>50.322603</td>\n","      <td>36.656637</td>\n","      <td>56.843763</td>\n","      <td>28.371194</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>5 rows × 16008 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-813d6b75-fbae-4dee-bcc1-da66a7b7b05d')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-813d6b75-fbae-4dee-bcc1-da66a7b7b05d button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-813d6b75-fbae-4dee-bcc1-da66a7b7b05d');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":6}]},{"cell_type":"code","source":["timbFeat_List = [\"hardness\", \"depth\", \"brightness\", \"roughness\", \"warmth\", \"sharpness\", \"boominess\"]"],"metadata":{"id":"onbfuE4vOBra","executionInfo":{"status":"ok","timestamp":1679784339934,"user_tz":-60,"elapsed":10,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}}},"execution_count":7,"outputs":[]},{"cell_type":"code","source":["timbFeat_MinList = audioFileWavef_TimbralFeatures_DF[timbFeat_List].min(axis = 0, numeric_only = True)\n","timbFeat_MaxList = audioFileWavef_TimbralFeatures_DF[timbFeat_List].max(axis = 0, numeric_only = True)"],"metadata":{"id":"PZdazT3aMhBc","executionInfo":{"status":"ok","timestamp":1679784339935,"user_tz":-60,"elapsed":10,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}}},"execution_count":8,"outputs":[]},{"cell_type":"code","source":["print(timbFeat_MinList)\n","print(timbFeat_MaxList)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"2rIA8PPsNgPf","executionInfo":{"status":"ok","timestamp":1679784339935,"user_tz":-60,"elapsed":10,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"81bbc013-4e40-43c5-86d7-48ecf7fd71bb"},"execution_count":9,"outputs":[{"output_type":"stream","name":"stdout","text":["hardness      11.586026\n","depth          0.140980\n","brightness    10.874446\n","roughness      0.000000\n","warmth         7.696248\n","sharpness      0.000000\n","boominess      0.000000\n","dtype: float64\n","hardness       82.596457\n","depth         100.000000\n","brightness     84.309024\n","roughness      77.090665\n","warmth         78.870422\n","sharpness      92.585410\n","boominess      71.482249\n","dtype: float64\n"]}]},{"cell_type":"code","source":["timbFeat_List = [\"ac_hardness\", \"ac_depth\", \"ac_brightness\", \"ac_roughness\", \"ac_warmth\", \"ac_sharpness\", \"ac_boominess\"]"],"metadata":{"id":"G5AcQybyEV13","executionInfo":{"status":"ok","timestamp":1679784339936,"user_tz":-60,"elapsed":9,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}}},"execution_count":10,"outputs":[]},{"cell_type":"code","source":["# https://mirdata.readthedocs.io/en/stable/source/tutorial.html\n","\n","def freeSound_PercDataset_generator():\n","    track_ids = percDataset.track_ids\n","    for track_id in track_ids:\n","      track = percDataset.track(track_id)\n","      audio_signal, _ = track.audio\n","      audioSignal = np.array(audio_signal)\n","      audioSignal.resize(16000, refcheck=False)\n","      # print(len(audioSignal))\n","      # consider training example as valid only if all annotations are present\n","      # (there are some missing keys in the mir_data track.audiocommons_analysis dict)\n","      timbFeaturesGroundTruth = []\n","      for index, timbFeatName in enumerate(timbFeat_List):\n","        # apply normalization between 0 and 1 (y = x-min(x) / max(x) - min(x))\n","        if timbFeatName in track.audiocommons_analysis.keys():\n","          normValue = (track.audiocommons_analysis[timbFeatName] - timbFeat_MinList[index]) / (timbFeat_MaxList[index] - timbFeat_MinList[index])\n","          timbFeaturesGroundTruth.append(np.float32(normValue))\n","        else:\n","          timbFeaturesGroundTruth.append(0.)\n","      # Keras' Input layer accepts either 2 different 'datasets' for x (examples) and y (ground truth),\n","      # or a single tuple object as (x, y)\n","      # https://keras.io/api/models/model_training_apis/ \n","      yield (audioSignal,\n","              timbFeaturesGroundTruth)\n","\n","dataset = tf.data.Dataset.from_generator(\n","    freeSound_PercDataset_generator,\n","    output_signature=(\n","         tf.TensorSpec(shape = (16000), dtype = tf.float32, name = 'audio_waveform'),\n","         tf.TensorSpec(shape = (7), dtype = tf.float32, name = 'annotations')))"],"metadata":{"id":"-lko-5hw6N78","executionInfo":{"status":"ok","timestamp":1679784343367,"user_tz":-60,"elapsed":2885,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}}},"execution_count":11,"outputs":[]},{"cell_type":"code","source":["dataset"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"x9I9OmbvOrkV","executionInfo":{"status":"ok","timestamp":1679784343368,"user_tz":-60,"elapsed":8,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"2443f6cc-a514-4877-9a92-de0bb0fc8a93"},"execution_count":12,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<FlatMapDataset element_spec=(TensorSpec(shape=(16000,), dtype=tf.float32, name='audio_waveform'), TensorSpec(shape=(7,), dtype=tf.float32, name='annotations'))>"]},"metadata":{},"execution_count":12}]},{"cell_type":"code","source":["# check dataset \n","# for element in dataset:\n","#  print(element)"],"metadata":{"id":"poLnP0se4haF","executionInfo":{"status":"ok","timestamp":1679784343369,"user_tz":-60,"elapsed":7,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}}},"execution_count":13,"outputs":[]},{"cell_type":"code","source":["# Shuffle entire dataset and split it into train/validation/test datasets\n","\n","num_elements = len(list(dataset))\n","print(f'There are {num_elements} elements in the Dataset')\n","\n","dataset = dataset.shuffle(num_elements, seed=tensF_ShuffleSeed)\n","    \n","train_size = int(train_split * num_elements)\n","test_size = int((1. - train_split) * num_elements)\n","    \n","test_ds = dataset.skip(train_size).take(test_size)\n","\n","train_size_withValSplit = int(train_split * train_size)\n","validation_size = int((1. - train_split) * train_size)\n","\n","train_ds = dataset.take(train_size_withValSplit)    \n","validation_ds = dataset.skip(train_size_withValSplit).take(validation_size)\n","\n","print(f'There are {len(list(train_ds))} elements in the training Dataset')\n","print(f'There are {len(list(validation_ds))} elements in the validation Dataset')\n","print(f'There are {len(list(test_ds))} elements in the test Dataset')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"9v2Ho2pqjTtL","executionInfo":{"status":"ok","timestamp":1679784411546,"user_tz":-60,"elapsed":68184,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"8613f673-5915-4672-b077-a787eac7fc92"},"execution_count":14,"outputs":[{"output_type":"stream","name":"stdout","text":["There are 10254 elements in the Dataset\n","There are 6562 elements in the training Dataset\n","There are 1640 elements in the validation Dataset\n","There are 2050 elements in the test Dataset\n"]}]},{"cell_type":"code","source":["batch_size = 1"],"metadata":{"id":"F0syowcfzl_v","executionInfo":{"status":"ok","timestamp":1679784411546,"user_tz":-60,"elapsed":8,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}}},"execution_count":15,"outputs":[]},{"cell_type":"code","source":["train_ds = train_ds.batch(batch_size)\n","validation_ds = validation_ds.batch(batch_size)\n","test_ds = test_ds.batch(batch_size)"],"metadata":{"id":"cG2JGPybRUh0","executionInfo":{"status":"ok","timestamp":1679784411547,"user_tz":-60,"elapsed":7,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}}},"execution_count":16,"outputs":[]},{"cell_type":"code","execution_count":17,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":483,"status":"ok","timestamp":1679784467921,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"},"user_tz":-60},"id":"Olbeoyxm23wI","outputId":"8c8276a3-fc62-4f81-c42c-1b2fd51a5785"},"outputs":[{"output_type":"stream","name":"stdout","text":["Model: \"sequential\"\n","_________________________________________________________________\n"," Layer (type)                Output Shape              Param #   \n","=================================================================\n"," Conv1D_1 (Conv1D)           (1, 5332, 7)              56        \n","                                                                 \n"," MaxPooling_1D_1 (MaxPooling  (1, 2666, 7)             0         \n"," 1D)                                                             \n","                                                                 \n"," flatten (Flatten)           (1, 18662)                0         \n","                                                                 \n"," Dense_1 (Dense)             (1, 7)                    130641    \n","                                                                 \n","=================================================================\n","Total params: 130,697\n","Trainable params: 130,697\n","Non-trainable params: 0\n","_________________________________________________________________\n"]}],"source":["model = Sequential()\n","model.add(Input(shape = (16000, 1), batch_size = batch_size))\n","\n","# with stride 1, the output size == input size - filter size + 1 (16000 - 7 + 1)\n","model.add(Conv1D(7, 7, strides = 3, activation=\"relu\", padding = 'valid', name = \"Conv1D_1\"))\n","model.add(MaxPooling1D(pool_size = 2, name = \"MaxPooling_1D_1\"))\n","\n","model.add(Flatten())\n","\n","model.add(Dense(7, activation = 'linear', name = \"Dense_1\"))\n","\n","# In a regression task, since the outputs and ground truth values are continuous values,\n","# rather than classes, and hence ezact prediction are not possible, accuracy is not used \n","# https://www.freecodecamp.org/news/evaluation-metrics-for-regression-problems-machine-learning/\n","# https://keras.io/api/metrics/\n","_metrics = ['mean_absolute_error', 'mean_squared_error']\n","model.compile(loss = 'mean_absolute_error', loss_weights = [1/7,1/7,1/7,1/7,1/7,1/7,1/7], optimizer = 'adam', metrics = _metrics)\n","model.summary()"]},{"cell_type":"code","source":["pngFilePath = '/content/gdrive/MyDrive/Colab Notebooks/Music Information Retrieval/Assignment/Final/smallest_model_plot.png'\n","tf.keras.utils.plot_model(model, to_file = pngFilePath, show_shapes = True, show_dtype = True, show_layer_names = True, rankdir = 'LR', dpi = 96, show_layer_activations = True)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":109},"id":"48cOGwtP8V5v","executionInfo":{"status":"ok","timestamp":1679784494847,"user_tz":-60,"elapsed":757,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"457a1f63-8ac9-4266-8d30-69b13a7a4f54"},"execution_count":19,"outputs":[{"output_type":"execute_result","data":{"image/png":"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\n","text/plain":["<IPython.core.display.Image object>"]},"metadata":{},"execution_count":19}]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"n6qV_e4vp8fn","executionInfo":{"status":"ok","timestamp":1679412740874,"user_tz":-60,"elapsed":1001582,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"14707d29-6f15-4757-a2ff-63e2ecec6162"},"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/20\n","6562/6562 [==============================] - 56s 5ms/step - loss: 0.0159 - mean_absolute_error: 0.1113 - mean_squared_error: 0.0245 - val_loss: 0.0135 - val_mean_absolute_error: 0.0948 - val_mean_squared_error: 0.0177\n","Epoch 2/20\n","6562/6562 [==============================] - 62s 8ms/step - loss: 0.0139 - mean_absolute_error: 0.0972 - mean_squared_error: 0.0196 - val_loss: 0.0122 - val_mean_absolute_error: 0.0854 - val_mean_squared_error: 0.0152\n","Epoch 3/20\n","6562/6562 [==============================] - 54s 7ms/step - loss: 0.0131 - mean_absolute_error: 0.0914 - mean_squared_error: 0.0165 - val_loss: 0.0124 - val_mean_absolute_error: 0.0871 - val_mean_squared_error: 0.0148\n","Epoch 4/20\n","6562/6562 [==============================] - 42s 5ms/step - loss: 0.0123 - mean_absolute_error: 0.0864 - mean_squared_error: 0.0161 - val_loss: 0.0120 - val_mean_absolute_error: 0.0839 - val_mean_squared_error: 0.0151\n","Epoch 5/20\n","6562/6562 [==============================] - 43s 5ms/step - loss: 0.0121 - mean_absolute_error: 0.0849 - mean_squared_error: 0.0146 - val_loss: 0.0134 - val_mean_absolute_error: 0.0941 - val_mean_squared_error: 0.0169\n","Epoch 6/20\n","6562/6562 [==============================] - 44s 5ms/step - loss: 0.0120 - mean_absolute_error: 0.0841 - mean_squared_error: 0.0147 - val_loss: 0.0111 - val_mean_absolute_error: 0.0779 - val_mean_squared_error: 0.0126\n","Epoch 7/20\n","6562/6562 [==============================] - 42s 5ms/step - loss: 0.0118 - mean_absolute_error: 0.0827 - mean_squared_error: 0.0140 - val_loss: 0.0126 - val_mean_absolute_error: 0.0884 - val_mean_squared_error: 0.0157\n","Epoch 8/20\n","6562/6562 [==============================] - 42s 5ms/step - loss: 0.0118 - mean_absolute_error: 0.0823 - mean_squared_error: 0.0140 - val_loss: 0.0111 - val_mean_absolute_error: 0.0777 - val_mean_squared_error: 0.0126\n","Epoch 9/20\n","6562/6562 [==============================] - 41s 5ms/step - loss: 0.0118 - mean_absolute_error: 0.0823 - mean_squared_error: 0.0140 - val_loss: 0.0105 - val_mean_absolute_error: 0.0734 - val_mean_squared_error: 0.0110\n","Epoch 10/20\n","6562/6562 [==============================] - 42s 5ms/step - loss: 0.0116 - mean_absolute_error: 0.0811 - mean_squared_error: 0.0135 - val_loss: 0.0134 - val_mean_absolute_error: 0.0936 - val_mean_squared_error: 0.0174\n","Epoch 11/20\n","6562/6562 [==============================] - 52s 7ms/step - loss: 0.0118 - mean_absolute_error: 0.0828 - mean_squared_error: 0.0142 - val_loss: 0.0106 - val_mean_absolute_error: 0.0742 - val_mean_squared_error: 0.0121\n","Epoch 12/20\n","6562/6562 [==============================] - 42s 5ms/step - loss: 0.0115 - mean_absolute_error: 0.0807 - mean_squared_error: 0.0133 - val_loss: 0.0110 - val_mean_absolute_error: 0.0772 - val_mean_squared_error: 0.0124\n","Epoch 13/20\n","6562/6562 [==============================] - 42s 5ms/step - loss: 0.0115 - mean_absolute_error: 0.0804 - mean_squared_error: 0.0133 - val_loss: 0.0108 - val_mean_absolute_error: 0.0753 - val_mean_squared_error: 0.0128\n","Epoch 14/20\n","6562/6562 [==============================] - 42s 5ms/step - loss: 0.0115 - mean_absolute_error: 0.0804 - mean_squared_error: 0.0136 - val_loss: 0.0106 - val_mean_absolute_error: 0.0742 - val_mean_squared_error: 0.0112\n","Epoch 15/20\n","6562/6562 [==============================] - 41s 5ms/step - loss: 0.0114 - mean_absolute_error: 0.0800 - mean_squared_error: 0.0131 - val_loss: 0.0109 - val_mean_absolute_error: 0.0763 - val_mean_squared_error: 0.0130\n","Epoch 16/20\n","6562/6562 [==============================] - 41s 5ms/step - loss: 0.0116 - mean_absolute_error: 0.0813 - mean_squared_error: 0.0136 - val_loss: 0.0114 - val_mean_absolute_error: 0.0800 - val_mean_squared_error: 0.0124\n","Epoch 17/20\n","6562/6562 [==============================] - 41s 5ms/step - loss: 0.0116 - mean_absolute_error: 0.0809 - mean_squared_error: 0.0135 - val_loss: 0.0114 - val_mean_absolute_error: 0.0799 - val_mean_squared_error: 0.0138\n","Epoch 18/20\n","6562/6562 [==============================] - 42s 5ms/step - loss: 0.0115 - mean_absolute_error: 0.0808 - mean_squared_error: 0.0136 - val_loss: 0.0111 - val_mean_absolute_error: 0.0779 - val_mean_squared_error: 0.0124\n","Epoch 19/20\n","6562/6562 [==============================] - 42s 5ms/step - loss: 0.0114 - mean_absolute_error: 0.0796 - mean_squared_error: 0.0135 - val_loss: 0.0114 - val_mean_absolute_error: 0.0795 - val_mean_squared_error: 0.0133\n","Epoch 20/20\n","6562/6562 [==============================] - 41s 5ms/step - loss: 0.0114 - mean_absolute_error: 0.0801 - mean_squared_error: 0.0135 - val_loss: 0.0116 - val_mean_absolute_error: 0.0813 - val_mean_squared_error: 0.0130\n","Finished training. It took 1001.6484275540001 seconds\n"]}],"source":["start = time.monotonic()\n","history = model.fit(train_ds, epochs = 20, batch_size = batch_size, validation_data = validation_ds)\n","stop = time.monotonic()\n","print(f'Finished training. It took {stop - start} seconds')"]},{"cell_type":"code","source":["model.metrics_names"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Z-nAJCnuTO4V","executionInfo":{"status":"ok","timestamp":1679412740874,"user_tz":-60,"elapsed":12,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"ef7a48a1-40eb-4e66-e105-7ca606092588"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["['loss', 'mean_absolute_error', 'mean_squared_error']"]},"metadata":{},"execution_count":33}]},{"cell_type":"code","source":["testLoss, mae, mean_squared_error = model.evaluate(test_ds, batch_size = batch_size)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"l3F-jEia2oA8","executionInfo":{"status":"ok","timestamp":1679412761550,"user_tz":-60,"elapsed":20682,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"71c172c9-ed85-46ce-f1ed-8c6e40d6c4e1"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["2050/2050 [==============================] - 13s 3ms/step - loss: 0.0116 - mean_absolute_error: 0.0811 - mean_squared_error: 0.0135\n"]}]},{"cell_type":"code","source":["print(type(testLoss))\n","\n","print(model.metrics_names)\n","print(testLoss)\n","print(mae)\n","print(mean_squared_error)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"VNN-wa0v3klB","executionInfo":{"status":"ok","timestamp":1679412761554,"user_tz":-60,"elapsed":34,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"e30ce575-ac61-4a48-9170-3b27cb77cce9"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["<class 'float'>\n","['loss', 'mean_absolute_error', 'mean_squared_error']\n","0.011580346152186394\n","0.08106247335672379\n","0.013520305044949055\n"]}]},{"cell_type":"code","source":["# https://keras.io/guides/training_with_built_in_methods/\n","\n","import matplotlib.pyplot as plt\n","\n","def plot_loss_and_acc(history):\n","    train_loss = history.history[\"loss\"]\n","    val_loss = history.history[\"val_loss\"]\n","\n","    fig = plt.figure(figsize=(12,5))\n","\n","    ax = fig.add_subplot(121)\n","    ax.plot(range(len(train_loss)), train_loss, label=\"Train Loss\");\n","    plt.xlabel(\"Epochs\"); plt.ylabel(\"Loss (M. A. E.)\");\n","    plt.title(\"Train Loss (M. A. E.)\");\n","    plt.legend(loc=\"best\");\n","\n","    ax = fig.add_subplot(122)\n","    ax.plot(range(len(val_loss)), val_loss, label=\"Validation loss\");\n","    plt.xlabel(\"Epochs\"); plt.ylabel(\"Validation loss (M. A. E.)\");\n","    plt.title(\"Validation loss (M. A. E.)\");\n","    plt.legend(loc=\"best\");\n","\n","plot_loss_and_acc(history)\n","print(f'Test loss; {testLoss}')\n","\n","'''\n","Overfitting if: training loss << validation loss\n","Underfitting if: training loss >> validation loss\n","Just right if training loss ~ validation loss\n","'''"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":403},"id":"s211IRtjISZy","executionInfo":{"status":"ok","timestamp":1679412851810,"user_tz":-60,"elapsed":1078,"user":{"displayName":"MATTEO FABBRI","userId":"11369581659480367990"}},"outputId":"1d58aa3b-83a1-4e87-bc9a-2399c5a119f6"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Test loss; 0.011580346152186394\n"]},{"output_type":"execute_result","data":{"text/plain":["'\\nOverfitting if: training loss << validation loss\\nUnderfitting if: training loss >> validation loss\\nJust right if training loss ~ validation loss\\n'"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":37},{"output_type":"display_data","data":{"text/plain":["<Figure size 864x360 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]}],"metadata":{"colab":{"provenance":[],"authorship_tag":"ABX9TyOMZqecr51NJTK7xOol8xX+"},"gpuClass":"standard","kernelspec":{"display_name":"Python 3","name":"python3"},"language_info":{"name":"python"},"accelerator":"GPU"},"nbformat":4,"nbformat_minor":0}