Zero Shot Text Classification

zero-shot-text-classification

# pip install transformers
from transformers import pipeline

tags = ["travel", "surfing", "Indonesia", "Bali", "cryptocurrencies", "bitcoin", "mining", "gardening", "plants"]
sequences = ["Best surf camp in Bali", "How do I build a crypto mining rig?", "my plant has yellow leaves"]

classifer = pipeline("zero-shot-classification")


for text in sequences:
  result = classifer(text, tags)
  print(text + ":")
  for label, score in zip(result["labels"], result["scores"]):
      print(label, score)
  print("\n\n")
  
  #-----------------------------------------------------
  # Output 
 """
 
Best surf camp in Bali:
Bali 0.42244213819503784
surfing 0.40309280157089233
travel 0.10794368386268616
Indonesia 0.05909815430641174
bitcoin 0.003099546767771244
cryptocurrencies 0.0015756567008793354
plants 0.0012342673726379871
gardening 0.0008835656335577369
mining 0.0006302755209617317



How do I build a crypto mining rig?:
cryptocurrencies 0.6826867461204529
mining 0.2748754620552063
Bali 0.012076622806489468
bitcoin 0.010957781225442886
travel 0.005191634874790907
surfing 0.004757559858262539
Indonesia 0.004334508907049894
plants 0.002861511427909136
gardening 0.0022582190576940775



my plant has yellow leaves:
plants 0.9064661264419556
gardening 0.07110192626714706
Bali 0.006484213285148144
travel 0.004733263049274683
surfing 0.0045728315599262714
Indonesia 0.003091926220804453
mining 0.0013643732527270913
bitcoin 0.0011404716642573476
cryptocurrencies 0.0010448909597471356
  
"""