mdbloice / pyrea Goto Github PK

Hi, when I use the package, I have the following issues:
code is:
d1 = np.random.rand(100,10)
d2 = np.random.rand(100,10)
d3 = np.random.rand(100,10)
data = [d1,d2,d3]
params = pyrea.parea_2_genetic(data, k_min=2, k_max=5)

it returns the error:
TypeError Traceback (most recent call last)
Cell In[14], line 1
----> 1 params_hierarchical = pyrea.parea_1_genetic(data, k_min=2, k_max=5, k_final=3, n_generations=10, n_population=10)

File ~/anaconda3/envs/eval/lib/python3.10/site-packages/pyrea/core.py:833, in parea_1_genetic(data, k_min, k_max, k_final, n_population, n_generations)
830 stats.register("max", np.max)
832 # Run the genetic algorithm
--> 833 pop, log = algorithms.eaSimple(population, toolbox, cxpb=0.7, mutpb=0.2, ngen=n_generations, stats=stats, halloffame=hall_of_fame, verbose=True)
835 print(f"\nSummary:\n{log}")
837 return hall_of_fame[0]

File ~/anaconda3/envs/eval/lib/python3.10/site-packages/deap/algorithms.py:151, in eaSimple(population, toolbox, cxpb, mutpb, ngen, stats, halloffame, verbose)
149 invalid_ind = [ind for ind in population if not ind.fitness.valid]
150 fitnesses = toolbox.map(toolbox.evaluate, invalid_ind)
--> 151 for ind, fit in zip(invalid_ind, fitnesses):
152 ind.fitness.values = fit
154 if halloffame is not None:

File ~/anaconda3/envs/eval/lib/python3.10/site-packages/pyrea/core.py:795, in parea_1_genetic..evaluate(individual)
792 with warnings.catch_warnings():
793 warnings.simplefilter("ignore")
--> 795 sil = parea_1(data,
796 individual[0],
797 individual[1],
...
839 xp, is_array_api_compliant = get_namespace(array)
841 # store reference to original array to check if copy is needed when
842 # function returns
TypeError: np.matrix is not supported. Please convert to a numpy array with np.asarray.

Do you know how to fix it?
Thanks.

Hi author, I read the pseudocode of the paper “A hierarchical clustering and data fusion approach for disease subtype discovery” you mentioned, but didn't quite understand how clusters c1 and c2 are obtained, and how in each view, each data is reused until there is only one cluster per view?