Hi everyone! 👋
We’re halfway through this challenge, and I wanted to check in to see how things are going for you all. Are you finding the challenge exciting, or has it been tougher than expected?
Here are a few questions to spark some conversation:
1️⃣ How’s your progress so far?
2️⃣ Any interesting discoveries or strategies you’d like to share with the community?
3️⃣ Have you hit any roadblocks or struggles? Let’s brainstorm together—someone might have just the tip you need to get unstuck!
Remember, this is a great opportunity to connect with others, share ideas, and grow as a data scientist. If you’re feeling stuck or overwhelmed, don’t hesitate to reach out here in the discussion.
Can’t wait to hear how you’re all doing—let’s keep the momentum going! 💪
Good luck with the second half of the challenge. You've got this! 🌟
@Amy_Bray
Can we train three models for this task/ we have to single model for three tasks?
Hey @engmatix, one model can handle all the three tasks
Is the evaluation code in the starter code, the same as the one used in the leaderboard?
For now yes. But in stater notebook the model is used it is not making any prediction on first two tasks but it is doing the third tasks of translation okay so i was wondering if we can use three model.
How is it not doing inference for the first two tasks?
I am checking it . why it doing it ? as soon i will find i will post it
The scoring script is not right. It only scores the translation. I will write a new topic on it
I also check it it is accurate. Start a new topic and i will also post message in it.
There is a topic : https://zindi.africa/competitions/lelapa-ai-buzuzu-mavi-challenge/discussions/25343
Thanks . i will see it.