Is it top 10 as in Info tab page or top 20 as specify in the email

Hey Zindians! 👋 I’ve been thinking about how we can make our competitions even more impactful and aligned with real-world needs. One idea that came to mind is... See more

I found the data set is also too low to make a good clinical SLM, the Zindi team should redefine the challenge to higher the probability of getting a good clini... See more

The current reliance on the ROUGE score is fundamentally misaligned with the competition’s goals. ROUGE rewards surface-level word overlap, not clinical correct... See more

The ROUGE metric is poorly suited for this challenge because it focuses on surface-level n-gram overlap rather than capturing the correctness, clinical safety,... See more

The primary flaw was misalignment between the evaluation metric (ROUGE) and the task objective (clinical reasoning). ROUGE rewarded superficial overlaps, not se... See more

test = pd.read_csv('Data/test_raw.csv')test['Clinician'] = test['Prompt'].apply(lambda x: "summary "+ " ".join(x.split('.')[1:]).replace('\n', ' '))test[['Maste... See more

my model can generate summary but not diagnose and plan properly, anybody have tip about how to do that? please help me

I've recently returned to working on the Kenya Clinicians Response Generation competition, and after taking a closer look at how submissions are evaluated, I ge... See more

Is anyone also experiencing the submission issue - not able to make a submission. Is zindi down ?

any suggestion of model??

I need to adjust the clinician field from train.csv, am I allowed if given data in not change just rephrased

If I'm using multiple prompts per data point during inference, should I aim for a lower latency per prompt to keep overall processing time in check?

?

to generate 100-300 tokens takes few seconds for every model i tried

I was thinking to use open source API, can we do that?

Hi, Can someone please let me know what the judging criteria are?Thank you.

I am using a model from hf and I need my token import, can I do that?

some model need your token to install in pc, can we use that?

Hello Zindians,I hope all is well with everyone.I have manually reviewed all 400 “Prompt–Clinician” examples. I noticed some inconsistencies in the data that co... See more

Maxium time, I got the score of 0.25, and this is my first time participating in such hackathons, can any one give me some tips?? I would be glad if you do that... See more

How do we reduce inference time ?

It's on the skills you will learn part.

Hi! Getting 'Unknown Error:error running python script' Any ideas?

Are we authorized to use external data for training ?Can we use non commercial data ? Still open sourced but non commercial