tkassis / orgaquant Goto Github PK
View Code? Open in Web Editor NEWOrganoid Localization and Quantification Using Deep Convolutional Neural Networks
Home Page: https://www.nature.com/articles/s41598-019-48874-y
License: MIT License
Organoid Localization and Quantification Using Deep Convolutional Neural Networks
Home Page: https://www.nature.com/articles/s41598-019-48874-y
License: MIT License
I got an error message when I attempted to open OrgaQuant using: streamlit run orgaquant.py
I solved the issue by installing keras==2.3.1
Installation Step 5
Code:
pip install keras-resnet==0.2.0 cython keras==2.3.1 matplotlib opencv-python progressbar2 streamlit
I got an error message when I attempted to open OrgaQuant using: streamlit run orgaquant.py I solved the issue by installing keras==2.3.1
Installation Step 5 Code: pip install keras-resnet==0.2.0 cython keras==2.3.1 matplotlib opencv-python progressbar2 streamlit
Dear, I also got error and solved what you suggested.
However, the results are not same as the user image shot, I attached the results with test images.
How can I solve it?
Originally posted by @180910ldh in #7 (comment)
Hi Dr. Kassis,
Thank you so much for developing this very useful program! I was trying to install it on a Mac computer and ran into the issue of Mac not supporting TensorFlow gpu. I was wondering if you had any advice about getting around this. Would it be possible to use TensorFlow CPU or TensorFlow instead?
Thanks!
I first encountered closed issue #7 and solved that by following the instructions; i.e. changing keras to version 2.2.5
pip install keras==2.2.5
Secondly, I was getting the error in the title "AttributeError: 'str' object has no attribute 'decode'" which I tracked down to the version of h5py used to load previously trained models. You can fix it by reverting to an older version of h5py. Further details on this error are given in a Tensorflow github issue here: tensorflow/tensorflow#44467
pip install h5py==2.10.0 --force-reinstall
Would you mind updating the docs in README.md to help out other users?
Is there anyway around the requirement of NVIDIA graphics card. From my understanding, the newer MacOS does not use this - does this mean we cannot run it on Macs?
Thank you for your share. i don't know how to get your datesets.I am looking forword to your reply.
Hi Dr. Kassis,
I have used OrgaQuant before and it worked great, really useful, and very easy to use.
However, we got computers updated and the previous installed OrgaQuant was lost, we changed from window 7 to windows 10 and I tried installing the OrgaQuant with the specified python version and tensorflow versions but for a reason I keep getting an error.
version: python 3.6.9 and Tensorflow-gpu 2.3.1
I attached the error window in case you are able to give me a hand.
Thank you so much,
Juan Martin
Dear Dr. Kassis,
I am a postdoctoral fellow at the University of British Columbia and wanted to thank you for this great resource! Our lab has used OrgaQuant with great success for profiling our growing organoids. We are currently tracking organoids over time to evaluate their proliferation and run into difficulties because our brightfield microscope does not have an automatic stage. Therefore we have to cycle back and forth from picture and csv file to try to match an organoid of interest over several pictures. By any chance could the program label each individual bounding box numerically which is also linked to the csv file?
Thank you in advance!
Hi,
when installing OrgaQuant I encountered a few minor issues; I'm currently trying to run OrgaQuant with streamlit using the command "streamlit run orgaquant.py --server.port XXX" (antivirus programs tend to block certain ports), but I'm encountering the following error:
`Traceback:
File "c:\anaconda3\envs\orgaquant\lib\site-packages\streamlit\script_runner.py", line 332, in run_script
exec(code, module.dict)
File "E:\Tools\OrgaQuant-master\orgaquant.py", line 24, in
model = load_orga_model()
File "E:\Tools\OrgaQuant-master\orgaquant.py", line 22, in load_orga_model
return models.load_model(os.path.join('trained_models', model_path + '.h5'), backbone_name='resnet50')
File "E:\Tools\OrgaQuant-master\keras_retinanet\models_init.py", line 82, in load_model
import keras.models
File "c:\anaconda3\envs\orgaquant\lib\site-packages\keras_init_.py", line 6, in
'Keras requires TensorFlow 2.2 or higher. '`
Now, updating tensorflow is obviously a problem as it conflicts with the settings of the environment that are given in the beginning: 'conda install tensorflow-gpu=1.14 git'
Can you suggest a way to resolve these issues? Thanks in advance! :)
Edit: Nevermind - I saw that the issue has already been solved in a previous issue.
Dear Dr Kassis,
I am a research student at Kings College London and I am part of the Diabetes Research Group. We use mouse and human intestinal organoids to carry out our research. Your approach to quantifying organoids is amazing and would save us a lot of time. I tried creating my own model to recognize images we took our-self(the OrgaQuant training model wouldn't recognize them) and to also quantify organoids when budding has occurred. May I share some images with you and we can possibly create a new trained model.
Kind regards,
Ryan Pitaluga
Hello Dr. Kassis,
Thank you so much for developing the algorithm. Counting organoids is a long process and I really appreciate your efforts to shorten that process significantly.
I noticed, however, that the trained model does not work for our samples. Hence, I am wondering if you are able to create a custom model for organoid count for our lab.
Thank you!
I am following the README.md file and was having issues under the "Usage" heading. When I run streamlit run orgaquant.py
, I get the following error in Chrome: "InternalError: cudaGetDevice() failed. Status: CUDA driver version is insufficient for CUDA runtime version". I'm not sure what this means or how to solve it. Can you offer any advice? Thanks.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.