Added docs

Signed-off-by: Stefano Savare <[email protected]>

.github/workflows/documentation.yml

@@ -5,7 +5,7 @@ on:
       - main
     paths:
       - 'README.rst'
-      - 'docs/*'
+      - 'docs/source/*'
       - '.github/workflows/documentation.yml'
 
 

docs/source/aws/terraform.rst

@@ -2,14 +2,16 @@ Terraform
 ==========
 
 We currently use `Terraform <https://www.terraform.io>`_ to configure the infrastructure.
+`This <https://github.com/the-deep/dfs-deepl-terraform>`_ is the repository.
 
 What is on terraform?
 ---------------------
 
 - All S3 buckets, except the Terraform state
 - The Sagemaker IAM. We do not store on Terraform the rest of the IAM
 - The Sagemaker Notebook Instances and Github repository
-- The Streamlit ECR repository
+- The test environments ECR repository
+- The MLFlow server and ECR repository
 
 What is not on Terraform?
 -------------------------

docs/source/modeling/testing_environment.rst

@@ -1,37 +1,6 @@
 Testing Environment
 ===================
 
-We have built a simple online test environment to check the predictions of our models on the subpillars.
-You can find it `here <http://test-env.eba-crsiq2wb.us-east-1.elasticbeanstalk.com>`_.
+We have a dedicated repository dedicated to our offline and online test environment.
+You can find it `here <https://github.com/the-deep/deep-test-environment>`_
 
-It is useful to show to externals our results. We plan to add more features to it.
-
-We have used Streamlit as Python library, and Docker + ECR + Beanstalk as deployment option.
-
-Deployment
------------------------
-
-We use the Github CI to build the test environment image and push it to ECR.
-We then use Beanstalk to serve it.
-
-Building the image
-~~~~~~~~~~~~~~~~~~
-
-The image is automatically built via the CI.
-If you want to change the code, check ``docker/Dockerfile`` and change the variables
-``ENTRYPOINT`` and ``CMD``.
-Currently the CI downloads the data from
-``s3://<DEV_BUCKET>/test-environment/data`` to ``data/streamlit``. If you want to change
-the input data, look at that.
-
-Pushing to ECR
-~~~~~~~~~~~~~~
-
-It is fully automated and a priori you should not do anything
-
-Update the image on Beanstalk
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-This step is not automated, and you should do it **every time you want to serve a new image**.
-Go in Beanstalk -> Environments -> Test env -> Upload and deploy.
-Upload the file ``docker/Dockerrun.aws.json``.

requirements/test-env-requirements.txt

@@ -1,4 +1,4 @@
-streamlit==0.82.0
+streamlit==0.84.1
 pdfminer==20191125
 nltk==3.6.2
 pandas==1.2.4