Having trouble deploying to Streamlit cloud. Model not found

Community Cloud
,
1

I am having trouble deploying my model into streamlit cloud. It works just fine on my local machine when run through command prompt but I get errors that it can’t find my model when trying to deploy the new app.

File "/home/adminuser/venv/lib/python3.9/site-packages/streamlit/runtime/scriptrunner/script_runner.py", line 535, in _run_script
    exec(code, module.__dict__)File "/mount/src/laptop-price-predict/streamlit_app.py", line 11, in <module>
    model = load_model()File "/mount/src/laptop-price-predict/streamlit_app.py", line 8, in load_model
    model = joblib.load(model_path)File "/home/adminuser/venv/lib/python3.9/site-packages/joblib/numpy_pickle.py", line 579, in load
    with open(filename, 'rb') as f:

Here is my github that has the streamlit_app.py and my requirements.txt
Am I supposed to have something else in github to be able to run the app?
I’m sure its something that has to do with a path but I’m just not figuring it out.

Any help is appreciated
Thank you.

2

You are supposed to have 'laptop_price_prediction_model.joblib'. Otherwise how could joblib load it?

3

This is where I show how new I am to all this. Where am I supposed to put it so it can find it?

4

Same place as the other files.

5

This is my first time using GitHub so I’m sorry if I’m not following. Are you able to give me a step by step of what I’m supposed to do with the file? How and where it’s supposed to be so it can load properly?

6

Just put the model files in the same folder as the other source files.

7

If you are using just the github web interface, click the “Add file” button to add files to your repository.

8

I added the file but Im still getting the same error. ile “/home/adminuser/venv/lib/python3.9/site-packages/streamlit/runtime/scriptrunner/script_runner.py”, line 535, in _run_script
exec(code, module.dict)
File “/mount/src/laptop-price-predict/streamlit_app.py”, line 11, in
model = load_model()
File “/mount/src/laptop-price-predict/streamlit_app.py”, line 8, in load_model
model = joblib.load(model_path)
File “/home/adminuser/venv/lib/python3.9/site-packages/joblib/numpy_pickle.py”, line 587, in load
obj = _unpickle(fobj, filename, mmap_mode)
File “/home/adminuser/venv/lib/python3.9/site-packages/joblib/numpy_pickle.py”, line 506, in _unpickle
obj = unpickler.load()
File “/usr/local/lib/python3.9/pickle.py”, line 1212, in load
dispatchkey[0]

What else do I need to fix?

9

Which ML library was used to generate the model?

10

I just uploaded my notebook into github.

11

sorry wrong one

12

Now i loaded my notebook into github. laptop_interface.ipynb

13
  • use the file below :arrow_down:
  • if necessary, pin the versions of scikit-learn and joblib to the same versions you used on your local machine
  • do a reboot of the app

requirements.txt

pandas
streamlit
scikit-learn
joblib
14

Do I take everything out of my requirments.txt and then put in what you mentioned? I only ask because the ones you mentioned are already in my file.

15
  • Yes, you don’t need the rest
  • Maybe you have to pin the versions
  • Do a reboot after every change in the requirements file
16

Thank you so much. It is working. I really appreciate you both.

1 Like
17

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.