Hello everyone,you can find my solution on my GitHub: https://github.com/anashas/Instadeep-CompetitionThank you

Congratulations to the winners and many thanks to Zindi and InstaDeep for hosting this interesting competition. My solution is a transformer model with no ensem... See more

The approach was simple. Tokenized using keras tokenizer at a character level. Used an Embedding layer with each amino acid being represented in 100 dimensions.... See more

My model is a fine-tuned pretrained model (ProtBert) that was trained on billions of protein sequence data (BFD). It is based on the Transformers architecture -... See more

Hi all,this is a quick overview on the Zindi Package that we are trying to build and that is in testing now.==>CLICK HERE TO OPEN THE NOTEBOOK<==This pack... See more

Hi guys, I created a notebook on kaggle using some EDA and deep learning (2 bilstm) as architecture using TPU...https://www.kaggle.com/hamzaghanmi/enzyme-classi... See more

Hello. I have this error when submitting "Missing entries for IDs 08841052, 103933E1, 33476E28" but these IDs don't exist in the samplesubmission.csv and test.c... See more

My model is taking forever (about 45-60 min) to complete an epoch (which I think is due to the long sequences) for a daily simple (baseline) network. I haven't... See more

How many epochs do one have to train to see reasonable predictions. Training 10 epochs, my model predicts only 'class19' with 29% accuracy

I don't understand - train accuracy is higher then 90, validation accuracy close to 90, curve is coverging nice, BUT test accuracy - lower then 80! Why is that?... See more

Guys, when I've tried to submit, it gives me error - "Missing entries for IDs 08841052, 103933E1, 33476E28". But ... I don't see such IDs in test? Why this may... See more

There is a problem while downloading the datasets, the download either crushes or doesn't finish, help!

Hi everybody !Maybe it is not very important but I would like to know if sbd can explain the variable CREATURE in the training data ...

m

Goodday zindians, please do i necessarily need the unlabelled_data for training?Can i still get an accurate model without using the unlabelled_data?

Hi all,I've made a minimal starter notebook: https://colab.research.google.com/drive/1nkzr5B_miJpyFpTqX4UWmaSrcX57gFw9?usp=sharingIt's worth trimming down the d... See more

Dear Zindians, We have introduced a competition submission limit. You may make a maximum of 300 submissions for this competition. You may make 10 submissions pe... See more