{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "### 1. Umieszczenie danych w Digital Ocean Spaces" ] }, { "cell_type": "code", "execution_count": 45, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "
\n", "\n", "\n", "
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# lepsza wizualizacja DataFrame\n", "\n", "from itables import init_notebook_mode\n", "\n", "init_notebook_mode(all_interactive=True)" ] }, { "cell_type": "code", "execution_count": 46, "metadata": {}, "outputs": [], "source": [ "# praca z Cloud Storage\n", "\n", "from dotenv import load_dotenv\n", "import boto3\n", "import os\n", "\n", "load_dotenv()\n", "\n", "s3 = boto3.client(\n", " \"s3\",\n", " #aws_access_key_id=os.getenv(\"AWS_ACCESS_KEY_ID\"),\n", " #aws_secret_access_key=os.getenv(\"AWS_SECRET_ACCESS_KEY\"),\n", " #endpoint_url=os.getenv(\"AWS_ENDPOINT_URL_S3\")\n", ")" ] }, { "cell_type": "code", "execution_count": 47, "metadata": {}, "outputs": [], "source": [ "BUCKET_NAME = \"halfmarathon\"" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "# wysyłanie plików do Cloud Storage\n", "\n", "s3.upload_file(\n", " Filename='halfmarathon_wroclaw_2023__final.csv',\n", " Bucket=BUCKET_NAME,\n", " Key='stocks/year=2023/halfmarathon_wroclaw_2023__final.csv'\n", ")\n", "\n", "s3.upload_file(\n", " Filename='halfmarathon_wroclaw_2024__final.csv',\n", " Bucket=BUCKET_NAME,\n", " Key='stocks/year=2024/halfmarathon_wroclaw_2024__final.csv'\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 2.1. Wczytanie danych z Digital Ocean Spaces do DataFrame" ] }, { "cell_type": "code", "execution_count": 48, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", "
MiejsceNumer startowyImięNazwiskoMiastoKrajDrużynaPłećPłeć MiejsceKategoria wiekowaKategoria wiekowa MiejsceRocznik5 km Czas5 km Miejsce Open5 km Tempo10 km Czas10 km Miejsce Open10 km Tempo15 km Czas15 km Miejsce Open15 km Tempo20 km Czas20 km Miejsce Open20 km TempoTempo StabilnośćCzasTempo
\n", "\n", "
\n", "Loading ITables v2.2.2 from the init_notebook_mode cell...\n", "(need help?)
\n", "\n" ], "text/plain": [ " Miejsce Numer startowy Imię Nazwisko Miasto Kraj \\\n", "0 1.0 596 NIKODEM DWORCZAK KOŚCIAN POL \n", "1 2.0 616 MATEUSZ KACZOR RADOM POL \n", "2 3.0 154 PATRYK KOZŁOWSKI RADOM POL \n", "3 4.0 591 DARIUSZ BORATYŃSKI WROCŁAW POL \n", "4 5.0 521 SZYMON DOROŻYŃSKI LUBON POL \n", "\n", " Drużyna Płeć Płeć Miejsce Kategoria wiekowa \\\n", "0 NaN M 1.0 M20 \n", "1 RLTL OPTIMA RADOM M 2.0 M20 \n", "2 RLTL-ZTE-RADOM M 3.0 M20 \n", "3 WOSIEK TEAM AZS AWF WROCŁAW M 4.0 M20 \n", "4 SZYMI TEAM AZS POLITECHNIKA OPOLSKA M 5.0 M30 \n", "\n", " ... 10 km Tempo 15 km Czas 15 km Miejsce Open 15 km Tempo 20 km Czas \\\n", "0 ... 2.920000 00:45:07 2.0 3.083333 01:00:33 \n", "1 ... 2.920000 00:45:07 3.0 3.083333 01:00:38 \n", "2 ... 2.920000 00:45:07 1.0 3.083333 01:00:59 \n", "3 ... 3.110000 00:47:48 4.0 3.293333 01:05:40 \n", "4 ... 3.153333 00:48:09 5.0 3.453333 01:06:05 \n", "\n", " 20 km Miejsce Open 20 km Tempo Tempo Stabilność Czas Tempo \n", "0 1.0 3.086667 0.007267 01:04:03 3.036265 \n", "1 2.0 3.103333 0.008267 01:04:24 3.052856 \n", "2 3.0 3.173333 0.012467 01:04:40 3.065497 \n", "3 4.0 3.573333 0.028667 01:09:44 3.305681 \n", "4 5.0 3.586667 0.039800 01:10:05 3.322272 \n", "\n", "[5 rows x 27 columns]" ] }, "execution_count": 48, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import pandas as pd\n", "\n", "df = pd.read_csv(f's3://{BUCKET_NAME}/stocks/year=2024/halfmarathon_wroclaw_2024__final.csv', sep=';')\n", "df.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 2.2. Czyszczenie danych" ] }, { "cell_type": "code", "execution_count": 49, "metadata": {}, "outputs": [], "source": [ "# Snippet pomocniczy - zmiana czasu na sekundy\n", "\n", "def convert_time_to_seconds(time):\n", " if pd.isnull(time) or time in ['DNS', 'DNF']: #DID NOT START / DID NOT FINISH\n", " return None\n", " time = time.split(':')\n", " return int(time[0]) * 3600 + int(time[1]) * 60 + int(time[2])" ] }, { "cell_type": "code", "execution_count": 50, "metadata": {}, "outputs": [], "source": [ "df['Czas'] = df['Czas'].apply(convert_time_to_seconds)\n", "df['5 km Czas'] = df['5 km Czas'].apply(convert_time_to_seconds)\n", "df['10 km Czas'] = df['10 km Czas'].apply(convert_time_to_seconds)\n", "df['15 km Czas'] = df['15 km Czas'].apply(convert_time_to_seconds)\n", "df['20 km Czas'] = df['20 km Czas'].apply(convert_time_to_seconds)" ] }, { "cell_type": "code", "execution_count": 51, "metadata": {}, "outputs": [], "source": [ "# zostawiam tylko potrzebne kolumny\n", "\n", "df = df[['Płeć', 'Rocznik', '5 km Czas', '5 km Tempo', '10 km Czas', '10 km Tempo',\n", " '15 km Czas', '15 km Tempo', '20 km Czas', '20 km Tempo', 'Czas', 'Tempo']].copy()" ] }, { "cell_type": "code", "execution_count": 52, "metadata": {}, "outputs": [], "source": [ "# zmieniam nazewnictwo\n", "\n", "df = df.rename(columns = {\n", " '5 km Czas' : '5km Czas [sek]', \n", " '10 km Czas' : '10km Czas [sek]', \n", " '15 km Czas' : '15km Czas [sek]', \n", " '20 km Czas' : '20km Czas [sek]',\n", " '5 km Tempo' : '5km Tempo [min/km]',\n", " '10 km Tempo' : '10km Tempo [min/km]',\n", " '15 km Tempo' : '15km Tempo [min/km]',\n", " '20 km Tempo' : '20km Tempo [min/km]',\n", " 'Tempo' : 'Tempo [min/km]',\n", " 'Czas' : 'Czas [sek]'})" ] }, { "cell_type": "code", "execution_count": 53, "metadata": {}, "outputs": [], "source": [ "# zmieniam kolumne rocznik na kolumne wiek\n", "\n", "df['Wiek'] = 2024 - df['Rocznik']\n", "df = df.drop(columns=['Rocznik'])" ] }, { "cell_type": "code", "execution_count": 55, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "RangeIndex: 13007 entries, 0 to 13006\n", "Data columns (total 12 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 Płeć 12998 non-null object \n", " 1 5km Czas [sek] 10288 non-null float64\n", " 2 5km Tempo [min/km] 10288 non-null float64\n", " 3 10km Czas [sek] 10288 non-null float64\n", " 4 10km Tempo [min/km] 10279 non-null float64\n", " 5 15km Czas [sek] 10287 non-null float64\n", " 6 15km Tempo [min/km] 10277 non-null float64\n", " 7 20km Czas [sek] 10295 non-null float64\n", " 8 20km Tempo [min/km] 10285 non-null float64\n", " 9 Czas [sek] 10300 non-null float64\n", " 10 Tempo [min/km] 10300 non-null float64\n", " 11 Wiek 12723 non-null float64\n", "dtypes: float64(11), object(1)\n", "memory usage: 1.2+ MB\n" ] } ], "source": [ "df.info()" ] }, { "cell_type": "code", "execution_count": 56, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", "
0
\n", "\n", "
\n", "Loading ITables v2.2.2 from the init_notebook_mode cell...\n", "(need help?)
\n", "\n" ], "text/plain": [ "Płeć 0.000692\n", "5km Czas [sek] 0.209041\n", "5km Tempo [min/km] 0.209041\n", "10km Czas [sek] 0.209041\n", "10km Tempo [min/km] 0.209733\n", "15km Czas [sek] 0.209118\n", "15km Tempo [min/km] 0.209887\n", "20km Czas [sek] 0.208503\n", "20km Tempo [min/km] 0.209272\n", "Czas [sek] 0.208119\n", "Tempo [min/km] 0.208119\n", "Wiek 0.021834\n", "dtype: float64" ] }, "execution_count": 56, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# sprawdzam wartości brakujace w ujęciu procentowym\n", "\n", "df.isnull().sum() / len(df)" ] }, { "cell_type": "code", "execution_count": 57, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "77.00468978242485" ] }, "execution_count": 57, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(df.dropna(subset=['Wiek', 'Płeć', 'Czas [sek]'])) / len(df) * 100" ] }, { "cell_type": "code", "execution_count": 58, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", "
Płeć5km Czas [sek]5km Tempo [min/km]10km Czas [sek]10km Tempo [min/km]15km Czas [sek]15km Tempo [min/km]20km Czas [sek]20km Tempo [min/km]Czas [sek]Tempo [min/km]Wiek
\n", "\n", "
\n", "Loading ITables v2.2.2 from the init_notebook_mode cell...\n", "(need help?)
\n", "\n" ], "text/plain": [ " Płeć 5km Czas [sek] 5km Tempo [min/km] 10km Czas [sek] \\\n", "11206 M NaN NaN NaN \n", "11407 M NaN NaN NaN \n", "12285 M NaN NaN NaN \n", "11488 K NaN NaN NaN \n", "12871 M NaN NaN NaN \n", "\n", " 10km Tempo [min/km] 15km Czas [sek] 15km Tempo [min/km] \\\n", "11206 NaN NaN NaN \n", "11407 NaN NaN NaN \n", "12285 NaN NaN NaN \n", "11488 NaN NaN NaN \n", "12871 NaN NaN NaN \n", "\n", " 20km Czas [sek] 20km Tempo [min/km] Czas [sek] Tempo [min/km] Wiek \n", "11206 NaN NaN NaN NaN 35.0 \n", "11407 NaN NaN NaN NaN 33.0 \n", "12285 NaN NaN NaN NaN 33.0 \n", "11488 NaN NaN NaN NaN 32.0 \n", "12871 NaN NaN NaN NaN 63.0 " ] }, "execution_count": 58, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# widzę, że brakuje wszystkich pomiarów\n", "\n", "df[df['Czas [sek]'].isna()].sample(5)" ] }, { "cell_type": "code", "execution_count": 59, "metadata": {}, "outputs": [], "source": [ "# usuwam NaN z kolumn wiek, płeć bowiem to tylko nieco ponad 2% danych, ktore nie maja takiego wplywu\n", "# dodatkowo usuwam ponad 20% danych z kolumny Czas, bowiem Ci zawodnicy nie zostali zmierzeni\n", "\n", "df.dropna(subset=['Wiek', 'Płeć', 'Czas [sek]'], inplace=True)" ] }, { "cell_type": "code", "execution_count": 60, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", "
0
\n", "\n", "
\n", "Loading ITables v2.2.2 from the init_notebook_mode cell...\n", "(need help?)
\n", "\n" ], "text/plain": [ "Płeć 0\n", "5km Czas [sek] 12\n", "5km Tempo [min/km] 12\n", "10km Czas [sek] 11\n", "10km Tempo [min/km] 20\n", "15km Czas [sek] 12\n", "15km Tempo [min/km] 21\n", "20km Czas [sek] 5\n", "20km Tempo [min/km] 14\n", "Czas [sek] 0\n", "Tempo [min/km] 0\n", "Wiek 0\n", "dtype: int64" ] }, "execution_count": 60, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.isna().sum()\n", "\n", "# najwazniejsze, ze nie ma NaN w kolumnie target= 'Czas [sek]'\n", "# z reszta poradzi sobie PyCaret" ] }, { "cell_type": "code", "execution_count": 61, "metadata": {}, "outputs": [], "source": [ "#sprawdzam przykładowe wartości odstajace\n", "\n", "df[['Wiek', 'Czas [sek]', 'Tempo [min/km]']].plot(kind='box');" ] }, { "cell_type": "code", "execution_count": 62, "metadata": {}, "outputs": [], "source": [ "#usuwam wszystkie wartości odstajace z kolumn 'Czas', 'Tempo' oraz 'Wiek'\n", "\n", "columns_with_outliers = [\n", " 'Wiek', 'Tempo [min/km]', '5km Czas [sek]', '10km Czas [sek]', '15km Czas [sek]', '20km Czas [sek]', 'Czas [sek]',\n", " '5km Tempo [min/km]', '10km Tempo [min/km]', '15km Tempo [min/km]', '20km Tempo [min/km]'\n", "]\n", "\n", "clear_df = df.copy()\n", "\n", "for column in columns_with_outliers:\n", " Q1 = clear_df[column].quantile(0.25)\n", " Q3 = clear_df[column].quantile(0.75)\n", " IQR = Q3 - Q1\n", "\n", " lower_bound = Q1 - 1.5 * IQR\n", " upper_bound = Q3 + 1.5 * IQR\n", "\n", " clear_df = clear_df[~((clear_df[column] < lower_bound) | (clear_df[column] > upper_bound))]\n", "\n", "clear_df[['Wiek', 'Czas [sek]', 'Tempo [min/km]']].plot(kind='box');" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 2.3. Trenowanie modelu" ] }, { "cell_type": "code", "execution_count": 63, "metadata": {}, "outputs": [], "source": [ "# importuję biblioteki dla modelu regresji (model -> value)\n", "\n", "from pycaret.regression import setup, compare_models, finalize_model, plot_model, save_model, predict_model" ] }, { "cell_type": "code", "execution_count": 71, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DescriptionValue
Session id123
TargetCzas [sek]
Target typeRegression
Original data shape(9554, 12)
Transformed data shape(9554, 6)
Transformed train set shape(6687, 6)
Transformed test set shape(2867, 6)
Ignore features6
Numeric features4
Categorical features1
Rows with missing values0.3%
PreprocessTrue
Imputation typesimple
Numeric imputationmean
Categorical imputationmode
Maximum one-hot encoding25
Encoding methodNone
Fold GeneratorKFold
Fold Number10
CPU Jobs-1
Use GPUFalse
Log ExperimentFalse
Experiment Namereg-default-name
USI41cb
\n", "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# tworzę eksperyment\n", "# po kilku próbach odtworzeń eksperymentu zdecydowałem zignorować kolumny z tempem, gdyż jest to wynikowa czasu\n", "# dodatkowo pozostawiłem jedynie czas na 5 i 10km, gdyż sa to typowe dystansy startowe (pomimo duzo wiekszego MAE!)\n", "# + dodałem w v2: tempo na 15km żeby zmniejszyć MAE\n", "\n", "exp = setup(\n", " data=clear_df, \n", " target='Czas [sek]',\n", " ignore_features=['Tempo [min/km]', '5km Tempo [min/km]', '10km Tempo [min/km]', \n", " '20km Tempo [min/km]','15km Czas [sek]', '20km Czas [sek]'],\n", " session_id=123)" ] }, { "cell_type": "code", "execution_count": 72, "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)
lrLinear Regression78.768914552.6933119.97120.98730.01560.01050.1840
brBayesian Ridge78.771614552.6874119.97100.98730.01560.01050.0780
ridgeRidge Regression78.794914552.9929119.97040.98730.01560.01050.0800
lassoLasso Regression78.951014575.6010120.05530.98730.01570.01050.1560
llarLasso Least Angle Regression78.951114575.7035120.05560.98730.01570.01050.0860
larLeast Angle Regression79.116515025.0174122.00550.98690.01580.01050.1010
lightgbmLight Gradient Boosting Machine79.624714359.2411119.30010.98750.01560.01060.5580
huberHuber Regressor83.376615676.1121124.60780.98640.01630.01110.2810
enElastic Net149.160545253.9670212.20360.96070.02790.01990.0960
knnK Neighbors Regressor172.323356894.9785238.35810.95070.03130.02300.0560
parPassive Aggressive Regressor183.706165007.1198253.01770.94330.03330.02470.0840
ompOrthogonal Matching Pursuit199.248770389.5637265.20950.93900.03480.02660.0840
dummyDummy Regressor874.24931159780.61251076.5937-0.00150.14860.12280.0570
\n", "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# szukam najlepszego modelu pod katem metryki MAE - średni bład bezw.\n", "\n", "best_model = exp.compare_models(sort='MAE')" ] }, { "cell_type": "code", "execution_count": 75, "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
 MAEMSERMSER2RMSLEMAPE
Fold      
078.365612967.8820113.87660.98930.01470.0105
173.264410805.3990103.94900.99130.01350.0097
274.248212164.9272110.29470.98910.01440.0099
375.564713983.3737118.25130.98780.01530.0100
478.546514506.2915120.44210.98720.01550.0104
583.115618237.1329135.04490.98360.01700.0110
678.639712678.2960112.59790.98970.01470.0104
786.166522120.5840148.72990.98050.02020.0116
877.494912437.9261111.52550.98960.01450.0104
982.282815625.1201125.00050.98510.01680.0111
Mean78.768914552.6933119.97120.98730.01560.0105
Std3.85783202.582012.63300.00320.00180.0005
\n", "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Fitting 10 folds for each of 2 candidates, totalling 20 fits\n", "Original model was better than the tuned model, hence it will be returned. NOTE: The display metrics are for the tuned model (not the original one).\n" ] } ], "source": [ "# próba strojenia hiperparametrów\n", "\n", "best_tuned_model = exp.tune_model(best_model, n_iter=200, optimize='MAE')" ] }, { "cell_type": "code", "execution_count": 76, "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPETT (Sec)
0Linear Regression78.768914552.6933119.97120.98730.01560.01050.0700
1Linear Regression78.768914552.6933119.97120.98730.01560.01050.0510
\n", "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# strojenie hiperparametrow nie przynioslo rezultatow\n", "\n", "best_final_model = compare_models([best_model, best_tuned_model])" ] }, { "cell_type": "code", "execution_count": 77, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
 ModelMAEMSERMSER2RMSLEMAPE
0Linear Regression76.127612832.4781113.28050.98830.01490.0102
\n", "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n", "\n", "\n", "\n", "
Płeć5km Czas [sek]10km Czas [sek]15km Tempo [min/km]WiekCzas [sek]prediction_label
\n", "\n", "
\n", "Loading ITables v2.2.2 from the init_notebook_mode cell...\n", "(need help?)
\n", "\n" ], "text/plain": [ " Płeć 5km Czas [sek] 10km Czas [sek] 15km Tempo [min/km] Wiek \\\n", "3300 M 1610.0 3209.0 5.426667 23.0 \n", "5272 K 1808.0 3463.0 5.780000 33.0 \n", "9033 K 2068.0 4065.0 7.010000 25.0 \n", "7292 M 1814.0 3657.0 6.380000 43.0 \n", "3171 M 1568.0 3070.0 5.386667 47.0 \n", "... ... ... ... ... ... \n", "1387 K 1387.0 2771.0 4.966667 39.0 \n", "4225 M 1755.0 3404.0 5.530000 23.0 \n", "1550 M 1399.0 2773.0 4.990000 35.0 \n", "8912 M 2027.0 4010.0 7.123333 68.0 \n", "8057 M 1963.0 3943.0 6.553333 51.0 \n", "\n", " Czas [sek] prediction_label \n", "3300 6845.0 6890.403662 \n", "5272 7350.0 7384.118025 \n", "9033 8918.0 8828.856485 \n", "7292 8026.0 8006.799982 \n", "3171 6805.0 6772.471948 \n", "... ... ... \n", "1387 6171.0 6153.154133 \n", "4225 7081.0 7147.795519 \n", "1550 6248.0 6208.970183 \n", "8912 8831.0 8910.319505 \n", "8057 8349.0 8368.285442 \n", "\n", "[2867 rows x 7 columns]" ] }, "execution_count": 77, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# sprawdzam best model na Holdout Dataset - jeszcze lepsza skuteczność!\n", "\n", "exp.predict_model(best_final_model)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Wizualizacje" ] }, { "cell_type": "code", "execution_count": 78, "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# istotność cech\n", "\n", "plot_model(best_final_model, plot='feature')" ] }, { "cell_type": "code", "execution_count": 79, "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# widać, że model przewiduje niemalże doskonale\n", "\n", "plot_model(best_final_model, plot='error')" ] }, { "cell_type": "code", "execution_count": 80, "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# widać, że model nie jest przeuczony - bardzo dobrze przewiduje dla danych testowych jak i treningowych\n", "\n", "plot_model(best_final_model, plot='residuals')" ] }, { "cell_type": "code", "execution_count": 81, "metadata": {}, "outputs": [], "source": [ "# trenuję model na całym Dataset\n", "\n", "final_model = finalize_model(best_final_model)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Zapisywanie modelu (lokalnie + do DigitalOcean)" ] }, { "cell_type": "code", "execution_count": 98, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Transformation Pipeline and Model Successfully Saved\n" ] }, { "data": { "text/plain": [ "(Pipeline(memory=Memory(location=None),\n", " steps=[('numerical_imputer',\n", " TransformerWrapper(include=['5km Czas [sek]',\n", " '10km Czas [sek]',\n", " '15km Tempo [min/km]', 'Wiek'],\n", " transformer=SimpleImputer())),\n", " ('categorical_imputer',\n", " TransformerWrapper(include=['Płeć'],\n", " transformer=SimpleImputer(strategy='most_frequent'))),\n", " ('ordinal_encoding',\n", " TransformerWrapper(include=['Płeć'],\n", " transformer=OrdinalEncoder(cols=['Płeć'],\n", " handle_missing='return_nan',\n", " mapping=[{'col': 'Płeć',\n", " 'data_type': dtype('O'),\n", " 'mapping': K 0\n", " M 1\n", " NaN -1\n", " dtype: int64}]))),\n", " ('clean_column_names',\n", " TransformerWrapper(transformer=CleanColumnNames())),\n", " ('actual_estimator', LinearRegression(n_jobs=-1))]),\n", " 'linear_regression_pipeline.pkl')" ] }, "execution_count": 98, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# zapisuję pipeline lokalnie\n", "\n", "save_model(final_model, 'linear_regression_pipeline')" ] }, { "cell_type": "code", "execution_count": 99, "metadata": {}, "outputs": [], "source": [ "# zapisuję model do Digital Ocean Space\n", "\n", "s3.upload_file(\n", " Filename='linear_regression_pipeline.pkl',\n", " Bucket=BUCKET_NAME,\n", " Key='stocks/models/linear_regression_pipeline.pkl'\n", ")" ] }, { "cell_type": "code", "execution_count": 100, "metadata": {}, "outputs": [], "source": [ "user_data_df = pd.DataFrame(\n", " {'Wiek' : 28,\n", " 'Płeć' : 'M',\n", " '5km Czas [sek]' : 1500, \n", " '10km Czas [sek]' : 5000,\n", " '15km Tempo [min/km]' : 4.45,\n", " }, index=[0])" ] }, { "cell_type": "code", "execution_count": 101, "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Estymowany czas ukończenia półmaratonu w formacie: H:M:S wynosi 1:59:25\n" ] } ], "source": [ "import datetime\n", "\n", "prediction = predict_model(final_model, data=user_data_df)\n", "\n", "prediction_seconds = round(prediction[\"prediction_label\"][0], 2)\n", "\n", "prediction_time = str(datetime.timedelta(seconds=int(prediction_seconds)))\n", "\n", "print(f'Estymowany czas ukończenia półmaratonu w formacie: H:M:S wynosi {prediction_time}')" ] }, { "cell_type": "code", "execution_count": null, "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": "3.11.9" } }, "nbformat": 4, "nbformat_minor": 4 }