CODEWITHSUNDEEP
pythonsortingmatplotlib
Perl Del Rey
Perl Del Rey

Reputation: 1059

list sorted in ascending order when passed to matplotlib

I defined a python function that gets, as input, a list. The list I am passing is sorted in descending order. I am making this list the x-axis of my line plot. However, matplotlib is sorting it back in descending order. I want to retain the descending order. Here is my code:

    def produce_curves_topK(self, topKs, metric):
        for i in range(len(clfs)):
            risk_df = self.risk_dfs[i]
            metrics = []
            for topk in topKs:
                risk_df_curr = risk_df.head(n=topk)
                # test_indices = list(risk_df_curr['test_indices'])
                y_pred_curr = list(risk_df_curr['y_pred'])
                # y_true_curr = list(self.y_test_df.loc[test_indices, self.target_variable])
                y_true_curr = list(risk_df_curr['y_test'])
                if metric == 'precision':
                    precision_curr = precision_score(y_true_curr, y_pred_curr)
                    metrics.append(precision_curr)
                else:
                    recall_curr = recall_score(y_true_curr, y_pred_curr)
                    metrics.append(recall_curr)

            # vals = {
            #     'topKs': topKs,
            #     'metrics': metrics
            # }
            # df = pd.DataFrame(vals, columns=['topKs', 'metrics'])

            # HERE IT IS BEING SORTED IN ASCENDING ORDER
            plt.plot(topKs, metrics, label=self.model_names[i], marker='o')

        plt.legend(loc='best')
        plt.xlabel('Top K')
        if metric == 'precision':
            plt.ylabel('Precision')
            plt.savefig('precisions_topK.png')
        else:
            plt.ylabel('Recall')
            plt.savefig('recalls_topK.png')
        plt.close()

# the list is originally in descending order
produce_curves_topK(topKs=[60, 50, 40, 30, 20, 10], metric='precision')

And here is the plot (x-axis sorted in ascending order - I want it in descending order) enter image description here

Upvotes: 0

Views: 422

Answers (1)

BenB
BenB

Reputation: 658

You can grab the axes object and then set the xlim to be in reverse order.

fig, ax = plt.subplots()
plt.plot(x,y)
ax.set_xlim(ax.get_xlim()[::-1])

Here is an example:

import matplotlib.pyplot as plt

x = [60,50,40,30,20,10]
y = [0,1,2,3,4,5]
fig, ax = plt.subplots()

plt.plot(x,y)
ax.set_xlim(ax.get_xlim()[::-1])
plt.show()

Upvotes: 1

Related Questions