Pneumonia Classification in CXRs | DPhi

This is a companion discussion topic for the original entry at

Hello guys! Thanks for your interest in our challenge! We are really excited about your participation. Please make sure you sign up at for -

  1. Starter notebook
  2. Free CPU/GPU cloud notebooks
  3. Prizes!!! (Only registered users on Segmind will be eligible to win the prizes)

The dataset’s sharing policy seems to be changed, not able to download the dataset using the google drive downloader API, can anyone please check what’s going on ?

I can’t open the PyTorch environment of the notebook , whenever I try to open, it always open with Tensorflow environment. Help please.

Is there a submission limit in this challenge??

You have to select from the drop down menu in the Environment field.

@chanukya Could you please look into this?

I did that , but still it opens Tensorflow 1.15.5 environment for every option.

Give us till tomorrow noon, Bhavya. We will fix the issue.

1 Like

Hi @ppvalluri09
Could you please try again? It’s working fine for me. Looks like others are also not facing this issue.
Please let me know if you still face the issue.

There is no limit for submissions per day.

Can we use pretrained models for this competition?

1 Like

Yup it’s working fine, yesterday it was done for a while… Thanks

Anyone using Segmind’s cloud notebooks please take note that our servers are down due to increased demand and we’ll have it fixed by the evening. Sorry for the trouble!

It is up and running now.

I am not able to Import Libraries (like pandas, keras) . Whenever I try, It always show ModuleNotFound error. Help me please.

Hi @rutuja_padgil
In segmind Notebook, you need to install the libraries first and then only you will be able to use it. Use the below command to install libraries:
!pip install numpy
!pip install pandas

Hi, can we use transfer learning techniques like pre-trained models

Is it compulsory to run the models on segmind?

The environment selected is PyTorch but actually it is Tensorflow !