Official announcements

All aboard ūüöÜ

Attention all contributors! As a reminder, the third and final milestone of this challenge ends today. Your contributions will continue to count for points up until midnight UTC. After that, no more points will be awarded for new contributions. So now is the time to get in any last commits! We'll follow up in the coming days with an announcement of final results and next steps for the project. Stay tuned!
5 months, 3 weeks ago

AWS Credits for Contributions to the Algorithms!

Thanks for all the awesome work so far! We're in a great place to use our remaining time improving the quality of our algorithms and how they integrate with the application. Towards that end, we want to make sure that contributors have the resources to train deep learning models and test their predictions. Our friends at AWS have generously provided us with credits to help with the cause. If you're planning to improve the algorithms, email info [at] drivendata [dot] org, and tell us what you're planning to work on, and why you need cloud compute resources to do that. We'll provide $1,000 in credits to the first 6 contributors that are planning on tackling big issues!
6 months, 3 weeks ago

Join us for a walkthrough of the current state of the application!

We'll be running a walkthrough of the application for all contributors on Monday, November 27 at 4pm UTC ([check your time zone](https://time.is/1600_27_Nov_2017_in_UTC)). This will be a great chance to spin up on the project and see the latest version of the end-to-end application. Thanks so much for all of your hardwork so far! Come see what we've accomplished and get motivated for what's next. Details for joining the webinar are below: #### Join from PC, Mac, Linux, iOS or Android: - https://zoom.us/j/378483110 #### Or iPhone one-tap: - US: +16465588656,,378483110# or +16699006833,,378483110# #### Or Telephone: - Dial(for higher quality, dial a number based on your current location)Ôľö - US: +1 646 558 8656 or +1 669 900 6833 (Meeting ID: 378 483 110) - [International numbers available here](https://zoom.us/zoomconference?m=iLlwfthYMV1QcYZrVsiHQY91HnZ5jVUj)
7 months, 3 weeks ago

MVP phase coming to a close!

Hi all, We're excited by the frenzy of progress being made on the project over the last few weeks. Some logistical notes for how this phase will get closed out: * PRs that are opened up until the MVP phase officially closes at midnight UTC tonight will be eligible for points on the MVP leaderboard. Naturally, that means the MVP leaderboard will be subject to change for a few days after the phase ends as the PRs are resolved. * We will treat these PRs normally and they will go through the customary review process ‚ÄĒ as usual, our policy within the open source side of the house is to pretend the points don't exist! ‚ÄĒ except that we expect contributors to be extremely responsive in addressing questions or requested changes so that we can finalize the MVP leaderboard in a timely fashion. * Once the phase ends, the milestone tag on Github will be switched over for issues that now properly belong in the feature building phase. This may not happen instantaneously, but no new points will be entered for the MVP phase after midnight UTC regardless of the Github milestone. For a reminder of how prizes will be allocated at the end of a milestone, see the ['Prizes' page](https://concepttoclinic.drivendata.org/prizes). Happy development! - The C2C team
8 months, 3 weeks ago

Private Docker Cloud repos for early contributors

Just to sweeten the deal: the first 100 submitters of accepted PRs will earn the submitter 5 private Docker Cloud repos thanks to the generous sponsorship of our friends at Docker! Check out the Docker Cloud [site](https://cloud.docker.com/) to learn more about what you can do with private repositories.
11 months ago

Launch Announcement

The Concept to Clinic challenge is live! Thanks for joining in this community effort to bring advances in AI to the front lines of lung cancer detection. Read up on the challenge and Sign Up to start contributing today!
11 months, 2 weeks ago

Newsfeed

2 days, 13 hours ago
3 days, 6 hours ago
3 days, 12 hours ago
4 days ago
3 weeks, 2 days ago

@pjbull commented on issue #312: Build fails on step 5/25 with error code 1

@sunnnset The only supported configuration is to build within Docker, which creates the image on top of Ubuntu. We just merged the above PR, so building on Docker will work now.
4 weeks, 1 day ago

@sunnnset commented on issue #312: Build fails on step 5/25 with error code 1

I met exactly the same problem while running the bulid command on my macOS 10.13, what should I do to fix it? It seems that there's no python3-distutils on mac platform.
4 weeks, 1 day ago
4 weeks, 1 day ago
4 weeks, 1 day ago

@guyucowboy created a new issue: #314: How to add noise in generators?

Hi Thanks for sharing the codes. How to add noise ( Gaussian,Poisson, Salt and Pepper (S&P) and Speckle Noise) in generators in file prediction\src\preprocess\generators.py ? It does not provide the 3D data augmentation method with adding noise. Any help is appreciated. By the way, this is my first time opening an issue in this project. I do not know how to use the template. Best regards
4 weeks, 1 day ago
4 weeks, 1 day ago
1 month, 1 week ago

@daikankan commented on issue #312: Build fails on step 5/25 with error code 1

sudo apt-get install python3-distutils from https://github.com/pypa/pip/issues/5356 https://github.com/pypa/pip/issues/5367
1 month, 1 week ago
1 month, 2 weeks ago

@Askleapios opened a new pull request: #313: Build was failing at step 5/21 with code 1.

<!--- Provide a general summary of your changes in the Title above --> ## Description Just added `python3-distutils` in the installation packages because of a new requirement of Ubuntu 18.04 ## Reference to official issue https://github.com/concept-to-clinic/concept-to-clinic/issues/312: ## Motivation and Context <!--- Why is this change required? What problem does it solve? --> 'python3-pip' now requires 'python3-distutils' or else it exits with an error. **Without this change the build fails.** <!--- If adding a new feature or making improvements not already reflected in an official issue, please reference the relevant sections of the design doc --> ## How Has This Been Tested? <!--- Please describe in detail how you tested your changes. --> After the change the stack has been built successfully. <!--- Include details of your testing environment, and the tests you ran to --> My test environment is Win10 Pro 1803 64Bit with Docker v.18.03.1-ce-win65 (17513) stable. <!--- see how your change affects other areas of the code, etc. --> This should affect all platforms. ## CLA - [ ] I have signed the CLA; if other committers are in the commit history, they have signed the CLA as well
1 month, 2 weeks ago

@Askleapios created a new issue: #312: Build fails on step 5/25 with error code 1

## Expected Behavior It should complete building. ## Current Behavior After executing "docker-compose -f local.yml build" per manual, it fails with error code 1 on step 5/25. This is the error: Step 5/21 : RUN python3.6 get-pip.py ---> Running in 13b3cd635736 Traceback (most recent call last): File "get-pip.py", line 20649, in <module> main() File "get-pip.py", line 197, in main bootstrap(tmpdir=tmpdir) File "get-pip.py", line 82, in bootstrap import pip._internal File "/tmp/tmp_qu4qa61/pip.zip/pip/_internal/__init__.py", line 42, in <module> File "/tmp/tmp_qu4qa61/pip.zip/pip/_internal/cmdoptions.py", line 16, in <module> File "/tmp/tmp_qu4qa61/pip.zip/pip/_internal/index.py", line 25, in <module> File "/tmp/tmp_qu4qa61/pip.zip/pip/_internal/download.py", line 35, in <module> File "/tmp/tmp_qu4qa61/pip.zip/pip/_internal/locations.py", line 10, in <module> ImportError: cannot import name 'sysconfig' ERROR: Service 'base' failed to build: The command '/bin/sh -c python3.6 get-pip.py' returned a non-zero code: 1 ## Possible Solution According to this source (https://bugs.launchpad.net/ubuntu/+source/python-pip/+bug/1773794); there is a bug, or should I say a new requirement, in Ubuntu 18.04. So, we just have to add ‚Äúpython3-distutils‚ÄĚ in the installation packages of the file .\concept-to-clinic-master\compose\base\Dockerfile-dev and everything builds fine. The final file should look like this: RUN apt-get update && apt-get install -y tcl tk python3.6 python3.6-tk wget python-opencv python3-distutils ## Steps to Reproduce http://concept-to-clinic.readthedocs.io/en/latest/developing-locally-docker.html#run-tests 1.$ docker-compose -f local.yml build 2. 3. 4. ## Context (Environment) It should affect all platforms. ## Checklist before submitting - [ ] I have confirmed this using the officially supported Docker Compose setup using the `local.yml` configuration and ensured that I built the containers again and they reflect the most recent version of the project at the `HEAD` commit on the `master` branch - [ ] I have searched through the other currently open issues and am confident this is not a duplicate of an existing bug - [ ] I provided a **minimal code snippet** or list of steps that reproduces the bug. - [ ] I provided **screenshots** where appropriate - [ ] I filled out all the relevant sections of this template
1 month, 2 weeks ago
1 month, 2 weeks ago
1 month, 2 weeks ago

@xuchao-25824839 created a new issue: #311: Algorithms/Segment/Training not work

<!--- Provide a general summary of the issue in the Title above --> ## Expected Behavior <!--- Tell us what should happen --> Traning done then output model ## Current Behavior <!--- Tell us what happens instead of the expected behavior --> 1. [\d+].xml is no valid DICOM 2. ValueError: could not broadcast input array from shape (332,360,360) into shape (512,512,512) ## Possible Solution <!--- Not obligatory, but suggest a fix/reason for the bug, --> ## Steps to Reproduce <!--- Provide a link to a live example, or an unambiguous set of steps to --> <!--- reproduce this bug. Include code to reproduce, if relevant --> 1. export COMPOSE_FILE=local.yml && docker-compose up 2. docker exec -it concept-to-clinic_prediction_1 bash 3. python > import matplotlib > matplotlib.use('Agg') > from src.algorithms.segment.src.data_generation import prepare_training_data > prepare_training_data() > from src.algorithms.segment.src.training import train > train() `output` ERROR:root:69.xml is no valid DICOM Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/app/src/algorithms/segment/src/training.py", line 78, in train :CUBOID_IMAGE_SHAPE[2]] ValueError: could not broadcast input array from shape (332,360,360) into shape (512,512,512) 4. ## Context (Environment) <!--- How has this issue affected you? What are you trying to accomplish? --> <!--- Providing context helps us come up with a solution that is most useful in the real world --> ## Detailed Description <!--- Provide a detailed description of the change or addition you are proposing --> `concept-to-clinic/docs/_build/html/design-doc.html` Training In order to train the segmentation model, the following steps are necessary: Download the LIDC dataset. Put it in tests/assets/test_image_data/full optionally using a symbolic link pylidc wraps the LIDC dataset and provides information such as annotated nodules, visualization methods etc. You must create a .pylidrc file that specifies the path to your local LIDC dataset like so Run prepare_training_data to generate the binary segmentation masks in prediction/src/algorithms/segment/assets. Finally, to train the model run: from src.algorithms.segment.src.training import train train() ## Possible Implementation <!--- Not obligatory, but suggest an idea for implementing addition or change --> ## Checklist before submitting - [‚ąö] I have confirmed this using the officially supported Docker Compose setup using the `local.yml` configuration and ensured that I built the containers again and they reflect the most recent version of the project at the `HEAD` commit on the `master` branch - [‚ąö ] I have searched through the other currently open issues and am confident this is not a duplicate of an existing bug - [‚ąö ] I provided a **minimal code snippet** or list of steps that reproduces the bug. - [ ‚ąö] I provided **screenshots** where appropriate - [‚ąö ] I filled out all the relevant sections of this template
1 month, 2 weeks ago
1 month, 2 weeks ago
1 month, 2 weeks ago
1 month, 2 weeks ago
1 month, 2 weeks ago

@ryant26 commented on issue #310: Duplicate Status Checks After Enabling Github Apps

Not sure how this issue got posted here I was trying to post it on travisci ... sorry
1 month, 3 weeks ago

@ryant26 created a new issue: #310: Duplicate Status Checks After Enabling Github Apps

<!--- Provide a general summary of the issue in the Title above --> ## Expected Behavior <!--- Tell us what should happen --> Pull requests should show a single "check" for TravisCI, "Travis CI - Pull Request" ## Current Behavior <!--- Tell us what happens instead of the expected behavior --> I am getting 2 checks that reference the same build: * Travis CI - Pull Request * continuous-integration/travis-ci/pr ![duplicate checks](https://user-images.githubusercontent.com/3953165/40578428-051d39ec-60d1-11e8-8ef0-389479b43fee.PNG) ## Context (Environment) This is happening on a travisci.com build. I did not use the "Activate and Migrate" button to activate GitHub Apps integration. 1. I deleted all the ssh keys for my github.com account to generate new ones. 2. I logged into travisci.com and noticed that my build was no longer listed. 3. I added a repository and was taken through the GitHub Apps workflow to authorize an application 4. I opened a PR and noticed this duplicate checks ## Checklist before submitting - [x] I have searched through the other currently open issues and am confident this is not a duplicate of an existing bug - [x] I provided **screenshots** where appropriate - [x] I filled out all the relevant sections of this template
1 month, 3 weeks ago
2 months, 1 week ago
2 months, 2 weeks ago
2 months, 3 weeks ago
2 months, 3 weeks ago
2 months, 3 weeks ago
2 months, 3 weeks ago
2 months, 3 weeks ago
2 months, 3 weeks ago
2 months, 4 weeks ago
3 months, 1 week ago
3 months, 1 week ago
3 months, 2 weeks ago
3 months, 2 weeks ago
3 months, 3 weeks ago
4 months, 1 week ago
4 months, 1 week ago
4 months, 2 weeks ago
4 months, 2 weeks ago
4 months, 2 weeks ago
4 months, 2 weeks ago

@lamby commented on PR #309: Improve various code styles

Any luck all? Would love to get this merged! :)
4 months, 2 weeks ago

@lamby commented on PR #309: Improve various code styles [WIP]

@isms I'm not blocked, more that I wouldn't want to be *introducing* any new issue. Indeed, this PR can be merged (renaming to match)
4 months, 2 weeks ago

@reubano commented on PR #309: Improve various code styles [WIP]

> @reubano Can you dig in and find out what the source is? @isms @lamby can either of you verify that bug?
4 months, 2 weeks ago

@lamby commented on PR #309: Improve various code styles [WIP]

As in, I can't *quite* understand the error message!
4 months, 2 weeks ago
4 months, 2 weeks ago

@isms commented on PR #309: Improve various code styles [WIP]

@reubano Can you dig in and find out what the source is? @lamby I say we let that tests fail for a while while Reuben looks into it so you're not blocked :)
4 months, 2 weeks ago

@reubano commented on PR #309: Improve various code styles [WIP]

Possibly related to https://github.com/concept-to-clinic/concept-to-clinic/pull/305#issuecomment-363400089
4 months, 2 weeks ago

@lamby commented on PR #309: Improve various code styles [WIP]

@isms @reubano Any ideas on this testsuite failure? Can't seem to work out what is actually really broken, let alone how I could have done it...
4 months, 2 weeks ago

@lamby opened a new pull request: #309: Improve various code styles [WIP]

This PR includes a large number of code refactoring within the backend (Django) part of the application.
4 months, 2 weeks ago
4 months, 3 weeks ago
4 months, 3 weeks ago
4 months, 3 weeks ago
4 months, 3 weeks ago
4 months, 4 weeks ago

@reubano commented on PR #305: Refactor imagery loading EventEmitter into store value

When trying to run this on my own, I keep getting a blank screen at port `8080` and the following error: `GET http://34.227.31.217:8080/app.js net::ERR_CONTENT_LENGTH_MISMATCH` Anyone else? FYI, it's not specific to this PR either. I see even at [666fb1](https://github.com/concept-to-clinic/concept-to-clinic/commit/666fb1bf410cbd80eb7928d0e3c5f2433ad668a5).
5 months, 1 week ago
5 months, 2 weeks ago
5 months, 2 weeks ago

@vessemer commented on PR #306: Reactivate tests and clear models

All rebased after #307, tests have been passed.
5 months, 2 weeks ago
5 months, 2 weeks ago
5 months, 2 weeks ago
5 months, 2 weeks ago
5 months, 2 weeks ago

@louisgv commented on PR #305: Refactor imagery loading EventEmitter into store value

@lamby should I rebase/repush in order to make it pass Travis?
5 months, 2 weeks ago

@reubano commented on PR #307: Fix memory and styling errors

Yes, sorry for all the back and forth. The main idea was to just include the relevant (and minimum set of) code changes needed to get the travis test passing.
5 months, 2 weeks ago

@vessemer commented on PR #307: Fix memory and styling errors

@reubano, I'm a bit confused with all those permutations, but it seems to be done :)
5 months, 2 weeks ago

@reubano commented on PR #307: Chose an appropriate DATA_SHAPE

#306 should just be refactoring code. So it seems that the changes in `evaluate_detection.py` are actually fixes which would go here.
5 months, 2 weeks ago
5 months, 2 weeks ago

@reubano commented on PR #307: Chose an appropriate DATA_SHAPE

Tests look good now. Just the styling that is making travis unhappy ```bash $ flake8 prediction prediction/src/algorithms/evaluation/evaluate_detection.py:233:80: W291 trailing whitespace prediction/src/algorithms/evaluation/evaluate_detection.py:246:20: E999 SyntaxError: invalid syntax The command "flake8 prediction" exited with 1. $ pycodestyle prediction prediction/src/algorithms/evaluation/evaluate_detection.py:233:80: W291 trailing whitespace ```
5 months, 2 weeks ago