Skip to content
Discussion options

You must be logged in to vote

match_threshold=80
fallback_steps=15
max_length=35

import re

def parse_obs(observation: str, env, max_length, match_threshold, step, fallback_steps):
    observation = observation.split(" [SEP] ")
    observation = observation[observation.index("Instruction:"):]
    if step <= max_length - fallback_steps:
        note = """
- Try to make a purchase in {} steps.
- Puchase a product with no less than {}% match with the task. No need to look for a perfect match. If multiple products match the task, purchase the one with the highest match.
- The search engine uses bag-of-words. Optimize the search terms accordingly and don't put price in search. If you cannot find relevant items, consider go…

Replies: 1 comment 3 replies

Comment options

You must be logged in to vote
3 replies
@Holmeswww
Comment options

@Holmeswww
Comment options

Answer selected by Ashwini4869
@Ashwini4869
Comment options

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
enhancement New feature or request
2 participants