Kijan pou m fè yon premye modèl machine learning senp ak Python?

Question

Mwen konn Python debaz. Mwen vle eseye fè yon ti modèl ki predi yon bagay. Ki kote pou m kòmanse?

Accepted Answer

Kòmanse ak **scikit-learn** ak yon dataset ki deja enstale. Egzanp: predi kalite yon flè (dataset Iris):

```python
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score

X, y = load_iris(return_X_y=True)

# 1. Separe done yo: youn pou aprann, youn pou teste
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)

# 2. Antrene modèl la
model = RandomForestClassifier(random_state=0)
model.fit(X_train, y_train)

# 3. Evalye l sou done li pa janm wè
pred = model.predict(X_test)
print('Presizyon:', accuracy_score(y_test, pred))
```

Se twa etap sa yo ki baz tout machine learning: **separe → antrene → evalye**.

Apre sa, eseye ak done ki enterese w (pri machin, rezilta egzamen, lapli): netwaye yo ak pandas, epi eseye plizyè modèl. Sit **Kaggle** gen anpil dataset ak kou gratis.

1 réponse