[FIXED] TypeError:unsupported format string passed to function .__format__

Issue

Working on jupyter notebooks, I came across this problem:

TypeError:unsupported format string passed to function .__format__

This is where the code starts:

def evaluate_preds(y_true, y_preds):
    """
    Performs Evaluation comp on y_true labels and y_preds labels on a classification
    """
    accuracy = accuracy_score(y_true,y_preds)
    precision = precision_score(y_true,y_preds)
    recall = recall_score(y_true,y_preds)
    f1 = f1_score(y_true,y_preds)
    metric_dict = {"accuracy": round(accuracy,2),
                  "precision": round(precision,2),
                  "recall": round(recall,2),
                  "f1": round(f1,2)}

    print(f"Accuracy :{ accuracy_score*100:.2f}%")
    print(f"Precision :{ precision_score:.2f}")
    print(f"Recall :{ recall_score:.2f}")
    print(f"F1 :{ f1_score:.2f}")

    return metric_dict

This below code is being run on a different cell in jupyter notebook

from sklearn.ensemble import RandomForestClassifier

np.random.seed(42)

heart_disease_shuffled= heart_disease.sample(frac=1)

X= heart_disease_shuffled.drop("target",axis =1)
y = heart_disease_shuffled["target"]

train_split = round(0.7 * len(heart_disease_shuffled))
valid_split = round(train_split + 0.15 * len(heart_disease_shuffled))

X_train,y_train = X[:train_split], y[:train_split]
X_valid,y_valid = X[train_split:valid_split], y[train_split:valid_split]

X_test,y_test = X[valid_split:], y[:valid_split]

clf = RandomForestClassifier()
clf.fit(X_train,y_train)


# Make Baseline preds

y_preds = clf.predict(X_valid)

# Evaluate classifier on validation set

baseline_metrics = evaluate_preds(y_valid, y_preds) 
baseline_metrics

How can I resolve it?

Tried changing the parameters and a bunch of other things but all of them popped up some errors like the one listed above

Solution

You are trying to print functions instead of the values they return. For instance, you assign

accuracy = accuracy_score(y_true,y_preds)

but then try to format the function, not the result

f"Accuracy :{ accuracy_score*100:.2f}%"

Instead, you should use the calculated value

f"Accuracy :{ accuracy*100:.2f}%"

But you've also created a dictionary with rounded values. You could use that dictionary with an f-string or use the .format method instead

f"Accuracy :{ metric_dict['accuracy']:.2f}"
# -- or --
"Accuracy : {accuracy:.2f}".format(**metric_dict)

Answered By - tdelaney