CODEWITHSUNDEEP
pythongithubkerasstreamlit
adhok
adhok

Reputation: 431

Unable to deploy a streamlit app on streamlit cloud

I am new to stream-lit and I am trying to deploy an app using the stream-lit cloud facility(free version) that can be found here -> https://streamlit.io/cloud

The app that we are building, accepts image inputs through a drag and drop interface and displays outputs to the user. This app is working fine on my local machine. The GitHub repository for the same can be found here -> https://github.com/adhok/SeeFood

While deploying on stream-lit cloud using the procedure provided, it gives me the following error. enter image description here

I tried restarting the process multiple times but it is giving the same error.

I also tried changing the name of the python file that contains stream-lit front-end functions from app.py to streamlit_app.py to no avail.

My hypothesis is that the repository is too big (in memory) for the streamlit-cloud interface as there are Keras Model Weights present, without which the model cannot make a prediction.

Please do let me know if there is anything I should change in my repository to make this work. Thanks in advance!

Upvotes: 1

Views: 964

Answers (1)

Md Owais
Md Owais

Reputation: 1

You can try following steps:

  1. Delete you requirements.txt file.
  2. Open your CMD and run the pip command in your app directory for creating the requirements.txt file with all packages along with their version command: pip install pipreqs.
  3. Once pipreqs installed then type pipreqs in CMD and hit enter button
  4. Commit all changes in the Github repo.
  5. Try to deploy.

Upvotes: 0

Related Questions