diff --git a/README.md b/README.md index 50dfdfe..6ef1190 100644 --- a/README.md +++ b/README.md @@ -75,6 +75,10 @@ docker rm -v $id ### Tests Instructions for automated and manual tests [here](https://github.com/code4sac/trash-ai/tree/production/frontend/__tests__). +## Functionality + +Documentation on typical workflow and functionality of the tool can be found [here](https://github.com/code4sac/trash-ai/blob/review2/docs/functionality) + ## Contribute We welcome contributions of all kinds. diff --git a/docs/README.md b/docs/README.md new file mode 100644 index 0000000..4979937 --- /dev/null +++ b/docs/README.md @@ -0,0 +1,17 @@ +# Top Level Overview of Docs + +## functionality.md +Provides information about the common workflows and functions of the app. + +## git-aws-account-setup.md +Describes how you can set up git based deployment of the app to AWS. + +## github-actions-deployment-role.md +How to set up the github deployment role for deploying with github actions. + +## github-actions.md +Information about the basic github action files and deployment strategy. + +## localdev.md +Instructions on how to deploy the app locally. + diff --git a/docs/functionality.md b/docs/functionality.md new file mode 100644 index 0000000..b1b0bf9 --- /dev/null +++ b/docs/functionality.md @@ -0,0 +1,36 @@ +# Batch Upload +The best way to use the tool is by batch uploading image files using the upload button. Alternatively if you just want to test the tool you can use the show samples button and it will give you some sample data to work with. +![image](https://github.com/code4sac/trash-ai/assets/26821843/c47b35f0-003d-4b32-b0d4-f3067884bd68) + +# Batch Download +After uploading your data, the AI routine will automatically label trash found within your images and format your data for download. You can click download all to download the data at this point. +![image](https://github.com/code4sac/trash-ai/assets/26821843/4cf3f4c6-01e6-457e-8322-179cd4e3e611) + +From the downloaded dataset you will get several file types. +1) the image you uploaded in its raw from labled with a unique hash +2) a json file with the same hash that contains the image level data from the AI +3) a second image with the same hash ending in -detect with overlayed bounding boxed and labels showing where the AI found what trash in your image. +4) a summary json file with summary data from all the images (useful when uploading trash from a whole survey). +5) a summary_detected and summary_totals csv file with data from the summary table formatted as a csv for folks who are more familar with that format. +6) a schema folder with an image_schema and summary_schema json files that can be used to validate the data formats and partnered markdown files describing the schema. +![image](https://github.com/code4sac/trash-ai/assets/26821843/941f7803-98fa-4d94-a320-3d7ca8280943) + +# Data Inspection +The tool can also be used to inspect your data. + +Clicking on uploads will allow you to inspect the AI classifications. +![image](https://github.com/code4sac/trash-ai/assets/26821843/74c71947-64d8-439e-b1ab-cc27e40db866) + +Selecting an image here will allow you to inspect detailed class information and metadata about the classifications. +![image](https://github.com/code4sac/trash-ai/assets/26821843/ee812b67-c5d0-42f1-b86b-f63c4ebe3634) + +Clicking summary will show information across the uploaded images which can also be selected. +![image](https://github.com/code4sac/trash-ai/assets/26821843/3084e433-fe01-418e-ad44-9aba86d7046b) + +If your images have spatial information, a selectable map can be viewed of where they trash was found. +![image](https://github.com/code4sac/trash-ai/assets/26821843/da9e6baa-c22d-468b-9915-63a312ea372b) + + + + + diff --git a/frontend/__tests__/README.md b/frontend/__tests__/README.md index 38d9652..54c3f1c 100644 --- a/frontend/__tests__/README.md +++ b/frontend/__tests__/README.md @@ -59,4 +59,11 @@ When the METADATA tab is opened, the relevant metadata for the image is displaye 4. After recieving results, does the summary page contain the correct data? -*The summary page shows the results of processing a random sample of images, as long as it's not empty it can be considered as correct.* \ No newline at end of file +*The summary page shows the results of processing a random sample of images, as long as it's not empty it can be considered as correct.* + +## Checklist for data schema tests +- [ ] Download the dataset ![image](https://github.com/code4sac/trash-ai/assets/26821843/b0bbd7e7-9554-477d-b2e4-ceb3e1b2a9b6) +- [ ] Inside of the schema folder you should have these 4 schema files. The JSON files are the JSON Schema and the Markdown files describe them. ![image](https://github.com/code4sac/trash-ai/assets/26821843/657b18ee-975f-4c83-bf84-20a31d1cb30b) +- [ ] Go to jsonschemavalidator.net ![image](https://github.com/code4sac/trash-ai/assets/26821843/aa91e1cd-4ce7-4600-9082-e6f815f07ede) +- [ ] Paste your schema in the left hand side image schema corresponds to the image data files, summary schema corresponds to the summary files. Using the image schema for this example. ![image](https://github.com/code4sac/trash-ai/assets/26821843/5833680c-6963-4952-948d-441b14090bb4) +- [ ] Paste the json file you want to validate in the right hand side, if the test worked, you will get a success message. ![image](https://github.com/code4sac/trash-ai/assets/26821843/88508de7-a895-4aa7-997e-89b95de5e95d)