E te mana'o e amata i masini suʻesuʻe?
Ua ou fatuina se aʻoaʻoga faigofie ma faigofie mo tagata amata atoatoa. Faʻatasi, o le a tatou faʻasolosolo laasaga autu o le aʻoaʻoina o se faʻataʻitaʻiga aʻoaʻoga masini.
Aʻo faʻamatalaina laasaga o le aʻoaʻoina o se faʻataʻitaʻiga taʻitasi, o le a ou tuʻuina atu foʻi se faʻataʻitaʻiga sili ona taua o se faʻafitauli o le aʻoaʻoina o masini. O lea, afai e te manaʻo e mulimuli i ai, e mafai ona e sii maia lenei seti faʻamatalaga faʻataʻitaʻiga mai lenei fesoʻotaʻiga.
Ua na'o se fa'ata'ita'iga fa'amaumauga e fesoasoani ia te oe e amata ai le a'oa'oina o masini.
E 18 a matou tulaga taua o tagata o vaitausaga eseese ma itupa e faʻamatalaina a latou musika e sili ona fiafia i ai. I le faʻaaogaina, foliga o le "tausaga" ma le "itupa" o le a matou taumafai e mate po o le a le ituaiga musika e sili ona latou fiafia i ai.
Fa'aaliga: 1 ma le 0 o lo'o tu'uina atu i itupa o tama'ita'i ma tama i lenei fa'amaumauga.
Ae peitaʻi, afai e te le manaʻo e mulimuli i le faʻataʻitaʻiga, e lelei atoatoa. O le a ou faʻamatalaina auiliili uma nei laasaga. O lea, tatou maulu i totonu!
Mea Muamua e Iloa
Aʻo leʻi alu i laasaga o le aʻoaʻoina o se faʻataʻitaʻiga, seʻi o tatou faʻamalamalamaina nisi o manatu. O le aʻoaʻoina o masini o se Atamai fa'apitoa amio pulea e taulaʻi i le atinaʻeina o algorithms e mafai ona aʻoaʻoina mai faʻamaumauga.
Ina ia faia lenei mea, o loʻo aʻoaʻoina faʻataʻitaʻiga aʻoaʻoga masini i luga o se faʻamaumauga o loʻo aʻoaʻoina ai le faʻataʻitaʻiga pe faʻapefea ona fai faʻamatalaga saʻo poʻo faavasegaina i luga o faʻamatalaga fou, e leʻi iloa muamua.
O lea la, o a nei faʻataʻitaʻiga? A faataitaiga masini aʻoaʻo e talitutusa ma se fua e faʻaogaina e le komepiuta e faʻatupu ai faʻamatalaga faʻamatalaga poʻo filifiliga.
O se faʻataʻitaʻiga, e pei o se fua, e mulimuli i se seti o faʻatonuga e iloilo ai faʻamaumauga ma faʻatupuina faʻamatalaga poʻo faʻamasinoga e faʻatatau i mamanu o loʻo maua i faʻamaumauga. O le tele o faʻamatalaga o loʻo aʻoaʻoina ai le faʻataʻitaʻiga, o le sili atu ona saʻo o ana valoʻaga.
O a Ituaiga Fa'ata'ita'iga e Mafai Ona Tatou Toleni?
Se'i o tatou va'ai po'o a fa'ata'ita'iga fa'aa'oa'oga a masini.
- Linear Regression: o se faʻataʻitaʻiga e vaʻai ai se suiga faʻaauau mai le tasi poʻo le sili atu o mea e tuʻuina atu.
- Neural Networks: se feso'ota'iga o nodes so'otaga e mafai ona a'oa'o e iloa fa'ata'ita'iga lavelave i fa'amaumauga.
- La'au Fa'ai'uga: o se faiga fa'ai'uga e fausia i luga o se filifili o fa'amatalaga fa'alafua fa'apea-ese.
- Clustering: o se seti o faʻataʻitaʻiga e faʻapipiʻi faʻamaumauga faʻatusatusa e faʻavae i luga o mea tutusa.
- Logistic Regression: o se faʻataʻitaʻiga mo faʻafitauli faʻavasegaina o loʻo i ai i le fesuiaiga o loʻo faʻatatauina e lua tau aoga.
- La'au Fa'ai'uga: o se faiga fa'ai'uga e fausia i luga o se filifili o fa'amatalaga fa'alafua fa'apea-ese.
- Random Forest: o se fa'ata'ita'iga tu'ufa'atasi e aofia ai le tele o la'au fa'ai'uga. E masani ona faʻaaogaina mo le faʻavasegaina ma le toe faʻaleleia o talosaga.
- K-Nearest Neighbors: o se faʻataʻitaʻiga e vaʻai ai le fesuiaiga o loʻo faʻaogaina i le faʻaogaina o faʻamatalaga k-latalata i le seti aʻoaʻoga.
Fa'alagolago i la matou fa'afitauli ma fa'amaumauga, matou te filifili po'o le fea masini a'oa'oga fa'ata'ita'iga e sili ona fetaui ma lo matou tulaga. Ae ui i lea, o le a tatou toe foi mai i lenei mea mulimuli ane. Ia, tatou amata a'oa'oina la tatou fa'ata'ita'iga. Ou te faamoemoe ua uma ona e siiina mai le talafaamaumau pe afai e te fia mulimuli i la matou faataitaiga.
E le gata i lea, ou te fautuaina le i ai Tusigāmanatu Jupyter fa'apipi'i i lau masini fa'apitonu'u ma fa'aaogaina mo au masini a'oa'oga poloketi.
1: Fa'amatala le fa'afitauli
O le laasaga muamua i totonu aʻoaʻoina se aʻoaʻoga masini fa'ata'ita'iga o lo'o fa'amalamalamaina le mataupu e fo'ia. O lenei mea e aofia ai le filifilia o fesuiaiga e te manaʻo e vaʻai (e taʻua o le fesuiaiga faʻatatau) ma fesuiaiga o le a faʻaaogaina e faʻatupu ai na valoʻaga (faʻaalia o foliga poʻo faʻamatalaga).
E tatau fo'i ona e filifili po'o le a le fa'afitauli o le a'oa'oina o masini o lo'o e taumafai e fa'atalanoaina (fa'avasegaina, toe fa'afo'i, fa'aputuina, ma isi) ma pe o le a le ituaiga fa'amaumauga e te mana'omia e fa'aputuina pe e a'oa'oina ai lau fa'ata'ita'iga.
O le ituaiga fa'ata'ita'iga e te fa'afaigaluegaina e fa'atatau i le ituaiga fa'afitauli a'oa'oga masini o lo'o e fa'amoemoe e fo'ia. Fa'avasegaga, toe fa'afo'i, ma le fa'aputuina o vaega muamua ia e tolu o lu'itau a'oa'oga masini. A e manaʻo e vaʻai se fesuiaiga faʻavasega, pei o se imeli o se spam pe leai, e te faʻaogaina le faʻavasegaina.
A e manaʻo e vaʻai se fesuiaiga faifaipea, pei o le tau o se fale, e te faʻaaogaina le faʻasologa. E fa'aogaina le fa'aputuga e tu'u fa'atasi ai mea fa'amaumauga e fa'atatau i mea e masani ai.
Afai tatou te vaavaai i la tatou faataitaiga; O la matou lu'i o le fuafuaina lea o le ituaiga musika e fiafia i ai le tagata mai lona itupa ma tausaga. O le a matou fa'aogaina se fa'amaumauga o tagata e 18 mo lenei fa'ata'ita'iga ma fa'amatalaga e uiga io latou tausaga, itupa, ma sitaili musika e fiafia i ai.
2. Saunia faamatalaga
A maeʻa ona e faʻamaonia le faʻafitauli, e tatau ona e saunia faʻamaumauga mo le aʻoaʻoina o le faʻataʻitaʻiga. E aofia ai le fa'amamāina ma le fa'agaioiina o fa'amaumauga. O lea la, e mafai ona tatou mautinoa o loʻo i totonu o se faʻatulagaga o le masini aʻoaʻoga algorithm mafai ona faʻaaoga.
E ono aofia ai gaioiga e pei o le tapeina o mea taua o lo'o misi, suia fa'amaumauga fa'avasega i fa'amaumauga fa'anumera, ma le fa'avasegaina po'o le fa'avasegaina o fa'amaumauga ina ia mautinoa o lo'o tutusa uma uiga.
Mo se faʻataʻitaʻiga, o le auala lenei e te tapeina ai mea taua o loʻo misi:
import pandas as pd
# Load the data into a pandas DataFrame
data = pd.read_csv('data.csv')
# Check for missing values
print(data.isnull().sum())
# Drop rows with missing values
data.dropna(inplace=True)
# Check that all missing values have been removed
print(data.isnull().sum())
Laititi fa'amatalaga: I le laina o "import pandas as pd",
matou te faʻaulufale mai le faletusi Pandas ma tuʻuina atu i ai le igoa "pd" e faʻafaigofie ai ona faʻasino ana galuega ma mea mulimuli ane i le code.
O Pandas o se taʻutaʻua taʻutaʻua mo le Python mo le faʻaogaina o faʻamatalaga ma auʻiliʻiliga, aemaise lava pe a galue i faʻamaumauga faʻatulagaina poʻo faʻamaumauga.
I la matou faʻataʻitaʻiga o le fuafuaina o ituaiga musika. Matou te fa'aulufale muamua atu le fa'amaumauga. Ua ou faaigoaina music.csv, peita'i, e mafai ona e fa'aigoa i le mea e te mana'o ai.
Ina ia saunia faʻamaumauga mo le aʻoaʻoina o se faʻataʻitaʻiga aʻoaʻoga masini, matou te vaevaeina i uiga (tausaga ma le itupa) ma sini (ituaiga musika).
O le a matou fa'aopoopoina le vaevaeina o fa'amaumauga i le 80:20 toleniga ma seti su'ega e iloilo ai le fa'atinoga o la matou fa'ata'ita'iga ma 'alofia ai le fa'aogaina.
# Import necessary libraries
import pandas as pd
from sklearn.model_selection import train_test_split
# Load data from CSV file/code>
music_data = pd.read_csv('music.csv')
# Split data into features and target
X = music_data.drop(columns=['genre'])
y = music_data['genre']
# Split data into training and testing sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
3. Filifili se fa'ata'ita'iga a'oa'oga masini.
A uma ona e saunia faʻamaumauga, e tatau ona e filifilia se faʻataʻitaʻiga aʻoaʻoga masini e fetaui ma lau galuega.
E tele algorithms e filifili mai ai, e pei o laʻau faʻaiʻuga, faʻasologa o mea faʻapitoa, lagolago masini vector, neural networks, ma isi. O le algorithm e te filifilia o le a fuafuaina i le ituaiga o mataupu o loʻo e taumafai e tali, le ituaiga o faʻamatalaga o loʻo ia te oe, ma au faʻatinoga e manaʻomia.
Matou te fa'aogaina se fa'avasegaga la'au fa'ai'uga mo lenei fa'ata'ita'iga ona o lo'o matou galulue fa'atasi ma se fa'afitauli fa'avasegaina (va'aiga fa'amaumauga fa'avasega).
# Import necessary libraries
from sklearn.tree import DecisionTreeClassifier
Ole fa'aaliga lea ole fa'aogaina ole Decision Tree Classifier:
4. Aoao le faataitaiga
E mafai ona e amata a'oa'oina le fa'ata'ita'iga pe a e filifilia se algorithm e a'oa'oina ai masini. O lenei mea e aofia ai le faʻaaogaina o faʻamatalaga na faia muamua e aʻoaʻo ai le algorithm i le auala e fai ai faʻamatalaga i luga o faʻamatalaga fou, e leʻi vaʻaia muamua.
O le algorithm o le a suia ona faʻamaufaʻailoga i totonu i le taimi o aʻoaʻoga e faʻaitiitia ai le eseesega i le va o ana tau faʻatatau ma tau moni i faʻamaumauga aʻoaʻoga. Ole aofa'i o fa'amaumauga e fa'aogaina mo a'oa'oga, fa'apea fo'i fa'asologa fa'apitoa a le algorithm, e mafai uma ona i ai se a'afiaga i le sa'o o le fa'ata'ita'iga e maua.
I la matou faʻataʻitaʻiga faʻapitoa, o lea ua matou filifili i se metotia, e mafai ona matou aʻoaʻoina la matou faʻataʻitaʻiga ma faʻamatalaga aʻoaʻoga.
# Train the decision tree classifier
model = DecisionTreeClassifier()
model.fit(X_train, y_train)
5. Iloilo le faʻataʻitaʻiga
A maeʻa ona aʻoaʻoina le faʻataʻitaʻiga, e tatau ona iloiloina i luga o faʻamaumauga fou ina ia mautinoa e saʻo ma faʻalagolago. O lenei mea e aofia ai le suʻeina o le faʻataʻitaʻiga ma faʻamaumauga e leʻi faʻaaogaina i le taimi o aʻoaʻoga ma faʻatusatusa ana tau faʻatatau i tau moni i faʻamaumauga o suʻega.
O lenei iloiloga e mafai ona fesoasoani i le fa'ailoaina o so'o se fa'ata'ita'iga fa'aletonu, e pei o le fa'apipi'i po'o le fa'aletonu, ma e mafai ona o'o atu ai i so'o se fa'alelei e ono mana'omia.
I le fa'aaogaina o fa'amaumauga o su'ega, o le a matou iloiloina le sa'o o la matou fa'ata'ita'iga.
# Import necessary libraries
from sklearn.metrics import accuracy_score
# Predict the music genre for the test data
predictions = model.predict(X_test)
# Evaluate the model's accuracy
accuracy = accuracy_score(y_test, predictions)
print("Accuracy: ", accuracy)
E le leaga tele le togi sa'o mo le taimi nei. 🙂 Ina ia faʻaleleia lau sikoa saʻo, e mafai ona e faʻamama atili faʻamaumauga pe faʻataʻitaʻi faʻataʻitaʻiga eseese masini-aʻoaʻoga e iloa ai po o fea e maua ai le togi maualuga.
6. Fa'alelei le fa'ata'ita'iga
Afai e le lava le lelei o le faʻataʻitaʻiga, e mafai ona e faʻaleleia lelei e ala i le suia o faʻasologa algorithm eseese poʻo le faʻataʻitaʻiina o algorithms fou atoa.
O lenei faiga e mafai ona aofia ai le fa'ata'ita'iga i isi fua faatatau o a'oa'oga, fa'aleleia o fa'atonuga masani, po'o le suia o le numera po'o le lapopo'a o fa'afanua natia i totonu o se neural network.
7. Fa'aaoga le fa'ata'ita'iga
O le taimi lava e te fiafia ai i le faʻatinoga o le faʻataʻitaʻiga, e mafai ona e amata faʻaaogaina e faʻatupu ai valoʻaga i faʻamatalaga fou.
Atonu e a'afia ai le fafagaina o fa'amatalaga fou i totonu o le fa'ata'ita'iga ma le fa'aogaina o fa'ata'ita'iga a'oa'oina o le fa'ata'ita'iga e fa'atupu ai valo'aga i lena fa'amaumauga, po'o le tu'ufa'atasia o le fa'ata'ita'iga i se fa'aoga lautele po'o se faiga.
E mafai ona matou fa'aogaina la matou fa'ata'ita'iga e fa'atupu ai fa'amatalaga i fa'amatalaga fou pe a matou fiafia i lona sa'o. E mafai ona e fa'ata'ita'iina tulaga taua o le itupa ma le matua.
# Test the model with new data
new_data = [[25, 1], [30, 0]]
predictions = model.predict(new_data)
print("Predictions: ", predictions)
Faamae'a mai
Ua mae'a a matou a'oa'oga muamua i masini a'oa'oga.
Ou te faʻamoemoe ua e maua le aoga. E mafai nei ona e taumafai e fa'aoga masini fa'ata'ita'iga eseese e pei ole Linear Regression po'o Random Forest.
E tele faʻamaumauga ma luʻitau i totonu Komepiuta pe afai e te manaʻo e faʻaleleia lau faʻailoga ma le malamalama i le aʻoaʻoina o masini.
Tuua se tali