fraenkel-lab / mogp Goto Github PK

Thanks for providing this amazing work. My use case is mapping from a 2d sequence to another 2d sequence. How can I modify the code to achieve this goal?

Sorry to bother you, thanks for the amazing model! I'm just having trouble training the model with some of my own ALSFRS data. despite not having any missing or infinite values, I keep getting this message at the training stage.

ValueError: Input contains NaN, infinity or a value too large for dtype('float64').

Any ideas what I'm doing wrong?

Thanks!

The full error message is as follows:

ValueError Traceback (most recent call last)
/var/folders/s0/t1rd1wdx11s0sngn007g49n40000gn/T/ipykernel_76381/238700026.py in
1 mix = mogp.MoGP_constrained(X= dfTime, Y= dfALS, alpha=1., num_iter=10, savepath=save_dir, rand_seed=0, normalize=True)
----> 2 mix.sample()

~/opt/anaconda3/lib/python3.9/site-packages/mogp/mogp_constrained.py in sample(self)
217 """Sampler for Mixture of Gaussian Process model"""
218
--> 219 self.initialize_sampler(init_K=self.num_init_clusters, onset_anchor=self.onset_anchor)
220 self.allocmodel.calc_suffstats(self.z)
221 idx = np.arange(self.allocmodel.N)

~/opt/anaconda3/lib/python3.9/site-packages/mogp/mogp_constrained.py in initialize_sampler(self, init_K, onset_anchor)
203 # if fewer than specified init K clusters are found, then use largest number of clusters returned by kmeans
204 # kmeans = KMeans(n_clusters=init_K, random_state=self.rand_seed).fit(x_data)
--> 205 kmeans = self.k_means_init(init_K, x_data)
206
207 self.z = kmeans.labels

~/opt/anaconda3/lib/python3.9/site-packages/mogp/mogp_constrained.py in _k_means_init(self, init_K, x_data)
182 def _k_means_init(self, init_K, x_data):
183 # if fewer than specified init K clusters are found, then use largest number of clusters returned by kmeans
--> 184 kmeans = KMeans(n_clusters=init_K, random_state=self.rand_seed, n_init=10).fit(x_data)
185 return kmeans
186

~/opt/anaconda3/lib/python3.9/site-packages/sklearn/cluster/_kmeans.py in fit(self, X, y, sample_weight)
1135 Fitted estimator.
1136 """
-> 1137 X = self._validate_data(
1138 X,
1139 accept_sparse="csr",

~/opt/anaconda3/lib/python3.9/site-packages/sklearn/base.py in _validate_data(self, X, y, reset, validate_separately, **check_params)
564 raise ValueError("Validation should be done on X, y or both.")
565 elif not no_val_X and no_val_y:
--> 566 X = check_array(X, **check_params)
567 out = X
568 elif no_val_X and not no_val_y:

~/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/validation.py in check_array(array, accept_sparse, accept_large_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, ensure_min_samples, ensure_min_features, estimator)
798
799 if force_all_finite:
--> 800 _assert_all_finite(array, allow_nan=force_all_finite == "allow-nan")
801
802 if ensure_min_samples > 0:

~/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/validation.py in _assert_all_finite(X, allow_nan, msg_dtype)
112 ):
113 type_err = "infinity" if allow_nan else "NaN, infinity"
--> 114 raise ValueError(
115 msg_err.format(
116 type_err, msg_dtype if msg_dtype is not None else X.dtype

ValueError: Input contains NaN, infinity or a value too large for dtype('float64').