# define path to dataset
pathToFile = "/home/shared/astro/PLAsTiCC/" if trainingOnGuanaco else "/home/leo/Downloads/plasticData/"
# ## Loading dataset with pytorch tool
# In[9]:
# torch_dataset_lazy = get_plasticc_datasets(pathToFile)
# Light curves are tensors are now [bands, [mjd, flux, err, mask],
# lc_data, lc_label, lc_plasticc_id
torch_dataset_lazy = get_plasticc_datasets(pathToFile, only_these_labels=only_these_labels, max_elements_per_class = max_elements_per_class)
# In[10]:
assert torch_dataset_lazy.__len__() != 494096, "dataset should be smaller"
print("dataset test ok")
# # Spliting data (train/test)
# In[11]:
# splitting the data
# # Load data
# In[5]:
# define path to dataset
pathToFile = "/home/shared/astro/PLAsTiCC/" if trainingOnGuanaco else "/home/leo/Downloads/plasticc_torch-master/"
# ## Loading dataset with pytorch tool
# In[6]:
# torch_dataset_lazy = get_plasticc_datasets(pathToFile)
# Light curves are tensors are now [bands, [mjd, flux, err, mask],
# lc_data, lc_label, lc_plasticc_id
torch_dataset_lazy = get_plasticc_datasets(pathToFile,
only_these_labels=only_these_labels)
# # Ploting one light curve
# In[7]:
# lc_data, lc_label, lc_plasticc_id = torch_dataset_lazy.__getitem__(123)
# display(lc_plasticc_id, lc_label)
# 6 bands: u g r i z Y
# 4 sequences: mjd, flux, error, mask
# 72 samples
# display(lc_data.shape, lc_data.dtype)
# print(lc_data.detach().numpy()[0, 0, :])
# In[8]: