Przygotowania pod convlaba

requirements.txt

@@ -2,5 +2,5 @@ flair==0.13.1
 conllu==4.5.3
 pandas==1.5.3
 numpy==1.26.4
-torch==2.3.0
+torch==1.13
 convlab==3.0.2a0

src/main.py

@@ -3,6 +3,7 @@ from service.dialog_policy import DialogPolicy
 from service.natural_languag_understanding import NaturalLanguageUnderstanding
 from service.natural_language_generation import NaturalLanguageGeneration, parse_frame
 from service.templates import templates
+from convlab.dialog_agent import PipelineAgent
 
 # initialize classes
 nlu = NaturalLanguageUnderstanding() # NLU
@@ -10,11 +11,21 @@ monitor = DialogStateMonitor() # DSM
 dialog_policy = DialogPolicy() # DP
 language_generation = NaturalLanguageGeneration(templates) # NLG
 
+agent = PipelineAgent(nlu=nlu, dst=monitor, policy=None, nlg=language_generation, name='sys')
+resp = agent.response("Dzień dobry")
+print(resp)
 # Main loop
+dial_num = 0
+print("CTRL+C aby zakończyć program.")
+while True:
+    monitor.reset()
+
+    print(f"\n==== Rozpoczynasz rozmowę nr {dial_num} ====\n")
     user_input = input("Możesz zacząć pisać.\n")
+
     while True:
         # NLU
-    frame = nlu.process_input(user_input)
+        frame = nlu.predict(user_input)
         # print(frame)
 
         # DSM

src/service/dialog_state_monitor.py

@@ -1,4 +1,4 @@
-from src.model.frame import Frame
+from model.frame import Frame
 from convlab.dst.dst import DST
 import copy
 

src/service/natural_languag_understanding.py

@@ -1,3 +1,4 @@
+from convlab.nlu.nlu import NLU
 from flair.models import SequenceTagger
 from utils.nlu_utils import predict_single, predict_and_annotate
 from model.frame import Frame, Slot
@@ -41,7 +42,7 @@ SLOTS:
     sauce
 """
 
-class NaturalLanguageUnderstanding:
+class NaturalLanguageUnderstanding(NLU):
     def __init__(self):
         print("\n========================================================")
         print("Models are loading, it may take a moment, please wait...")
@@ -85,7 +86,7 @@ class NaturalLanguageUnderstanding:
 
         return slots
 
-    def process_input(self, text: str):
+    def predict(self, text: str, context: list):
         act = self.__predict_intention(text)
         slots = self.__predict_slot(text)
         frame = Frame(source = 'user', act = act, slots = slots)

src/service/natural_language_generation.py

@@ -1,6 +1,8 @@
 import re
 from service.template_selector import select_template
 import random
+from convlab.nlg.nlg import NLG
+
 # from service.templates import templates
 
 def parse_frame(frame):
@@ -12,7 +14,7 @@ def parse_frame(frame):
     
     return act, slots
 
-class NaturalLanguageGeneration:
+class NaturalLanguageGeneration(NLG):
     def __init__(self, templates):
         self.templates = templates