Spaces:
Sleeping
Sleeping
Delete task_prioritizer.py
Browse files- task_prioritizer.py +0 -105
task_prioritizer.py
DELETED
|
@@ -1,105 +0,0 @@
|
|
| 1 |
-
import torch
|
| 2 |
-
from transformers import BertTokenizer, BertForSequenceClassification, DistilBertForSequenceClassification
|
| 3 |
-
from datetime import datetime
|
| 4 |
-
|
| 5 |
-
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
| 6 |
-
|
| 7 |
-
# Load the intent classifier model and tokenizer
|
| 8 |
-
num_intent_labels = 151 # Set the correct number of labels for the intent classifier
|
| 9 |
-
intent_model = BertForSequenceClassification.from_pretrained("bert-base-uncased", num_labels=num_intent_labels)
|
| 10 |
-
intent_model.load_state_dict(torch.load("intent_classifier.pth"))
|
| 11 |
-
intent_tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
|
| 12 |
-
intent_model.to(device)
|
| 13 |
-
intent_model.eval()
|
| 14 |
-
|
| 15 |
-
# Load the emotions model and tokenizer
|
| 16 |
-
emotions_model = DistilBertForSequenceClassification.from_pretrained("./saved_model")
|
| 17 |
-
emotions_tokenizer = BertTokenizer.from_pretrained("./saved_model")
|
| 18 |
-
emotions_model.to(device)
|
| 19 |
-
emotions_model.eval()
|
| 20 |
-
|
| 21 |
-
# Define the label names for emotions
|
| 22 |
-
emotion_label_names = ["admiration", "amusement", "anger", "annoyance", "approval", "caring", "confusion", "curiosity", "desire", "disappointment", "disapproval", "disgust", "embarrassment", "excitement", "fear", "gratitude", "grief", "joy", "love", "nervousness", "optimism", "pride", "realization", "relief", "remorse", "sadness", "surprise", "neutral"]
|
| 23 |
-
|
| 24 |
-
def predict_intent(sentence):
|
| 25 |
-
inputs = intent_tokenizer(sentence, return_tensors="pt", padding="max_length", truncation=True, max_length=128)
|
| 26 |
-
inputs = {key: val.to(device) for key, val in inputs.items()}
|
| 27 |
-
|
| 28 |
-
with torch.no_grad():
|
| 29 |
-
outputs = intent_model(**inputs)
|
| 30 |
-
predicted_class = torch.argmax(outputs.logits, dim=1).cpu().numpy()[0]
|
| 31 |
-
|
| 32 |
-
return predicted_class
|
| 33 |
-
|
| 34 |
-
def predict_emotion(sentence):
|
| 35 |
-
inputs = emotions_tokenizer(sentence, return_tensors="pt", padding="max_length", truncation=True, max_length=128)
|
| 36 |
-
inputs = {key: val.to(device) for key, val in inputs.items() if key != "token_type_ids"}
|
| 37 |
-
|
| 38 |
-
with torch.no_grad():
|
| 39 |
-
outputs = emotions_model(**inputs)
|
| 40 |
-
predicted_class = torch.argmax(outputs.logits, dim=1).cpu().numpy()[0]
|
| 41 |
-
|
| 42 |
-
return predicted_class, emotion_label_names[predicted_class]
|
| 43 |
-
|
| 44 |
-
def calculate_priority_score(intent, emotion, time_remaining):
|
| 45 |
-
# Example priority score calculation
|
| 46 |
-
intent_weight = 0.4
|
| 47 |
-
emotion_weight = 0.3
|
| 48 |
-
time_weight = 0.3
|
| 49 |
-
|
| 50 |
-
# Normalize time_remaining to a score between 0 and 1
|
| 51 |
-
time_score = max(0, min(1, 1 - (time_remaining.total_seconds() / (24 * 3600))))
|
| 52 |
-
|
| 53 |
-
# Calculate priority score
|
| 54 |
-
priority_score = (intent * intent_weight) + (emotion * emotion_weight) + (time_score * time_weight)
|
| 55 |
-
return priority_score
|
| 56 |
-
|
| 57 |
-
def prioritize_task(task_description, due_date_time, predicted_emotion, predicted_label_name):
|
| 58 |
-
predicted_intent = predict_intent(task_description)
|
| 59 |
-
|
| 60 |
-
# Calculate time remaining until the due date and time
|
| 61 |
-
due_date_time = datetime.strptime(due_date_time, "%Y-%m-%d %H:%M:%S")
|
| 62 |
-
time_remaining = due_date_time - datetime.now()
|
| 63 |
-
|
| 64 |
-
priority_score = calculate_priority_score(predicted_intent, predicted_emotion, time_remaining)
|
| 65 |
-
|
| 66 |
-
return {
|
| 67 |
-
"description": task_description,
|
| 68 |
-
"due_date_time": due_date_time,
|
| 69 |
-
"time_remaining": time_remaining,
|
| 70 |
-
"predicted_intent": predicted_intent,
|
| 71 |
-
"predicted_emotion": predicted_emotion,
|
| 72 |
-
"predicted_label_name": predicted_label_name,
|
| 73 |
-
"priority_score": priority_score
|
| 74 |
-
}
|
| 75 |
-
|
| 76 |
-
# Example tasks
|
| 77 |
-
tasks = [
|
| 78 |
-
{"description": "Finish the report by tomorrow.", "due_date_time": "2025-03-02 09:00:00"},
|
| 79 |
-
{"description": "meeting", "due_date_time": "2025-03-02 12:00:00"},
|
| 80 |
-
{"description": "listen to music.", "due_date_time": "2025-03-02 15:00:00"},
|
| 81 |
-
{"description": "daily linkedin queens game.", "due_date_time": "2025-03-02 18:00:00"},
|
| 82 |
-
{"description": "prepare ppt", "due_date_time": "2025-03-02 21:00:00"}
|
| 83 |
-
]
|
| 84 |
-
|
| 85 |
-
# Overall emotion sentence
|
| 86 |
-
emotion_sentence = "I am feeling very tired and stressed now"
|
| 87 |
-
predicted_emotion, predicted_label_name = predict_emotion(emotion_sentence)
|
| 88 |
-
|
| 89 |
-
# Prioritize tasks
|
| 90 |
-
prioritized_tasks = []
|
| 91 |
-
for task in tasks:
|
| 92 |
-
prioritized_tasks.append(prioritize_task(task["description"], task["due_date_time"], predicted_emotion, predicted_label_name))
|
| 93 |
-
|
| 94 |
-
# Reorder tasks based on priority score (descending order)
|
| 95 |
-
prioritized_tasks.sort(key=lambda x: x["priority_score"], reverse=True)
|
| 96 |
-
|
| 97 |
-
# Print prioritized tasks
|
| 98 |
-
for task in prioritized_tasks:
|
| 99 |
-
print(f"Task Description: '{task['description']}'")
|
| 100 |
-
print(f"Due Date and Time: {task['due_date_time']}")
|
| 101 |
-
print(f"Time Remaining: {task['time_remaining']}")
|
| 102 |
-
print(f"Predicted Intent: {task['predicted_intent']}")
|
| 103 |
-
print(f"Predicted Emotion: {task['predicted_emotion']} ({task['predicted_label_name']})")
|
| 104 |
-
print(f"Priority Score: {task['priority_score']:.4f}")
|
| 105 |
-
print()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|