TypeError: _binary_search_perplexity() takes exactly 3 positional arguments (4 given)

[iasiavnvai]

TypeError Traceback (most recent call last) Cell In[58], line 1 ----> 1 model, loss = train( 2 model_, X, device=device, nb_iterations=500, batch_size=1000, lambda_kl=0.01) 3 torch.save(model.state_dict(), "../models/tsne_500_001.pth") 4 plot_latent(X, model, device)

File /workspace/Pythoncluster/artefact/training.py:87, in train(model, X, device, nb_iterations, batch_size, lambda_kl, distance_trajectory, lr, weight_decay) 83 v = batch[0].to(device) 85 if lambda_kl > 0: 86 P = torch.as_tensor( ---> 87 make_P(batch[0], metric=distance_trajectory), device=device 88 ) 90 for iteration in tqdm(range(nb_iterations), leave=False): 91 lat, output = model(v)

File /workspace/Pythoncluster/artefact/training.py:35, in make_P(X, perplexity, metric) 32 def make_P(X, perplexity=30, metric="euclidiean"): 33 distances = pairwise_distances(X, metric=metric) ---> 35 P = _joint_probabilities(distances, perplexity) 36 assert np.all(np.isfinite(P)), "All probabilities should be finite" 37 assert np.all(P >= 0), "All probabilities should be non-negative"

File /workspace/Pythoncluster/artefact/training.py:23, in _joint_probabilities(distances, desired_perplexity, verbose) 21 def _joint_probabilities(distances, desired_perplexity, verbose=0): 22 distances = distances.astype(np.float32, copy=True) ---> 23 conditional_P = _binary_search_perplexity( 24 distances, None, desired_perplexity, verbose 25 ) 26 P = conditional_P + conditional_P.T 27 sum_P = np.maximum(np.sum(P), MACHINE_EPSILON_NP)

File sklearn/manifold/_utils.pyx:15, in sklearn.manifold._utils._binary_search_perplexity() TypeError: _binary_search_perplexity() takes exactly 3 positional arguments (4 given) This error occurred when I used sklearn0.23.2. do you know what is going on?Thank you very much for your answer