Free Sematic from its shackles

sematic-ai · Jan 8, 2025 · 99c4de0 · 99c4de0
1 parent 5311664
commit 99c4de0
Show file tree

Hide file tree

Showing 15 changed files with 23 additions and 123 deletions.
diff --git a/LICENSE b/LICENSE
@@ -1,11 +1,3 @@
-Copyright (c) 2023-present Sematic Inc.
-
-Portions of this software are licensed as follows:
-
-* All content that resides under the "sematic/ee/" directory of this repository, if that directory exists, is licensed under the license defined in "sematic/ee/LICENSE".
-* All third party components incorporated into the Sematic Software are licensed under the original license provided by the owner of the applicable component.
-* Content outside of the above mentioned directories or restrictions above is available under the "Apache 2" license as defined below.
-
 Apache License
 ==============
 

diff --git a/Makefile b/Makefile
@@ -86,9 +86,11 @@ release:
 release-server:
 	rm -f docker/*.whl
 	cp ./dist/*sematic*.whl docker/
-	cd docker; docker build --build-arg EXTRA=default -t sematic/sematic-server:${TAG} -f Dockerfile.server .
+	cd docker; docker build --build-arg EXTRA=all -t sematic/sematic-server:${TAG} -f Dockerfile.server .
 	docker push sematic/sematic-server:${TAG}
-	cd docker; docker build --build-arg EXTRA=all -t sematic/sematic-server-ee:${TAG} -f Dockerfile.server .
+	# alias the ee server to be the same as the non-ee server for anybody
+	# who continues to rely on the ee tag.
+	docker image tag sematic/sematic-server:${TAG} sematic/sematic-server-ee:${TAG}
 	docker push sematic/sematic-server-ee:${TAG}
 
 test:

diff --git a/docs/changelog.md b/docs/changelog.md
@@ -96,7 +96,7 @@ with nothing else on the line.
     * [improvement] Add constraints to the DB schema in order to improve validations. In case you
       get any errors during the upgrade, please contact us on
       [Discord](https://discord.gg/4KZJ6kYVax) so that we can assist you.
-    * [improvement] Added documentation for custom user metrics[^1]
+    * [improvement] Added documentation for custom user metrics
     * [improvement] Improve error messaging for unschedulable pipeline runs
     * [improvement] When running the CLI via Bazel, use the current directory as the working
       directory
@@ -119,7 +119,7 @@ with nothing else on the line.
     * [improvement] Add index on edges for source/dest run ids
 * [0.31.0](https://pypi.org/project/sematic/0.31.0/)
     * [feature] Enable remote execution using pure-Docker, without bazel
-    * [feature] Support live-metrics during Sematic Function execution[^1]
+    * [feature] Support live-metrics during Sematic Function execution
     * [feature] Add visualization for Prompt/Response pairs
     * [example] Add Hacker News summarization example
     * [improvement] Expose all Kubernetes classes in the base `sematic` module
@@ -232,7 +232,7 @@ with nothing else on the line.
     * [bugfix] Fixed a bug where failure to notify the Dashboard or Resolver with Run updates via
       socketio messages would cause the Resolution to fail
 * [0.26.0](https://pypi.org/project/sematic/0.26.0/)
-    * [feature] Added optional anonymized user analytics to track Sematic usage[^2]
+    * [feature] Added optional anonymized user analytics to track Sematic usage[^1]
     * [improvement] Atomic database migrations with Helm hooks
     * [improvement] Move future pickle storage to new server-generated locations
     * [improvement] Refactor in-app links to use react router
@@ -255,7 +255,7 @@ with nothing else on the line.
     * [improvement] Show more detailed errors on failures of DB migrations
     * [bugfix] Reduce wheel size
 * [0.24.0](https://pypi.org/project/sematic/0.24.0/)
-    * [feature] Introduction of RayCluster[^1]
+    * [feature] Introduction of RayCluster
     * [feature] "Tee" cloud function logs so they appear in Sematic dashboard and the pod logs
     * [improvement] Visualize better which run was the true "root failure" when a resolution fails
     * [improvement] Improve an error message when using untyped dicts in type annotations
@@ -486,7 +486,5 @@ with nothing else on the line.
 * [0.0.2.alpha.1654828599](https://pypi.org/project/sematic/0.0.2a1654828599/)
     * Initial release
 
-[^1]: This feature is for Sematic's "Enterprise Edition" only. Please reach out
-to [email protected] if you are interested in using Sematic EE.
-[^2]: This release adds opt-out anonymized user analytics tracking to the Sematic
+[^1]: This release adds opt-out anonymized user analytics tracking to the Sematic
 code.  You can opt-out of this tracking on the Sematic Dashboard home page.
diff --git a/docs/container-images.md b/docs/container-images.md
@@ -19,15 +19,11 @@ Sematic currently supports two build systems:
 ### Docker
 
 {% hint style="warning" %}
-This feature is in "Beta."
+The Native Docker Build System only supports Debian GNU/Linux.
 {% endhint %}
 
 {% hint style="warning" %}
-The Native Docker Build System currently only supports Debian GNU/Linux.
-{% endhint %}
-
-{% hint style="warning" %}
-The Native Docker Build System currently does not support multiple base images.
+The Native Docker Build System does not support multiple base images.
 {% endhint %}
 
 If you don't want to use [Bazel](https://bazel.build) to manage your project,

diff --git a/docs/github.md b/docs/github.md
@@ -1,9 +1,4 @@
-# GitHub[^1]
-
-{% hint style="info" %}
-This feature is only available in Sematic
-Enterprise Edition.
-{% endhint %}
+# GitHub
 
 ## Summary
 
@@ -12,7 +7,7 @@ it be "traditional" code, [unit testing of ML pipelines](./testing.md), or runni
 [regression pipelines](https://www.sematic.dev/blog/continuous-learning-for-safer-and-better-ml-models)
 involving your data and models.
 
-If you have deployed Sematic EE[^1], you can configure it to integrate with GitHub such
+If you have deployed Sematic, you can configure it to integrate with GitHub such
 that you can validate your commits (for PRs or any other purpose) using Sematic runs.
 
 <img
@@ -287,7 +282,3 @@ github:
   enabled: true
   github_access_token: <TOKEN FROM GITHUB>
 ```
-
-[^1]: This feature of Sematic is only available with the "Enterprise Edition."
-Before using, please reach out to Sematic via [email protected] to obtain a
-license for "Sematic EE."
diff --git a/docs/metrics.md b/docs/metrics.md
@@ -1,7 +1,7 @@
 # Metrics
 
 Sematic surfaces a number of high-level metrics in the Dashboard, and lets
-Enterprise users log their own metrics at runtime.
+users log their own metrics at runtime.
 
 ## Pipeline metrics
 

diff --git a/docs/ray.md b/docs/ray.md
@@ -1,13 +1,4 @@
-# Ray[^1]
-
-{% hint style="warning" %}
-This feature is in "Beta."
-{% endhint %}
-
-{% hint style="info" %}
-This feature is only available in Sematic
-Enterprise Edition.
-{% endhint %}
+# Ray
 
 ## What is Ray?
 
@@ -217,18 +208,15 @@ of the Ray head AND of the autoscaler.
 
 ## Installation and configuration
 
-Before using this feature, make sure you are set up with a Sematic EE
-license. Once that's taken care of, you will need to make the following
-changes:
-
 ### Pip package
 
 Instead of depending on `sematic`, you will need to depend on `sematic[ray]`
 or `sematic[all]`.
 
 ### Helm chart
 
-Ensure that you are using an [EE server](./upgrades.md#foss-to-enterprise-edition1).
+Ensure that you are using an [EE server](https://hub.docker.com/r/sematic/sematic-server-ee)
+OR v0.41.0 or higher of the standard server.
 Then you will need to install Kuberay into your Kubernetes environment.
 
 #### Installing Kuberay
@@ -268,7 +256,3 @@ ray:
   non_gpu_tolerations: []
   gpu_resource_request_key: null
 ```
-
-[^1]: This feature of Sematic is only available with the "Enterprise Edition."
-Before using, please reach out to Sematic via [email protected] to obtain a
-license for "Sematic EE."
diff --git a/docs/slack.md b/docs/slack.md
@@ -1,7 +1,3 @@
-{% hint style="warning" %}
-This feature is in "Beta."
-{% endhint %}
-
 ## Using Slack with Sematic
 
 Sematic currently automatically sends [Pipeline Run](glossary.md#pipeline-run)

diff --git a/docs/testing.md b/docs/testing.md
@@ -10,7 +10,7 @@ have your pipeline execute for hours or days, and then have it fail near
 the end for a simple error!
 
 Luckily, Sematic provides out-of-the-box support for unit testing your
-Sematic pipelines. With Sematic Enterprise Edition, you can also
+Sematic pipelines. You can also
 integrate Sematic with your version control system to block PRs
 and perform other commit checking. See our [GitHub](./github.md)
 integration for more information.
@@ -104,4 +104,4 @@ def test_mock_sematic_funcs():
 
 Note that this is able to test the connections between your pipeline
 using familiar python mechanisms. This is one more advantage of using Sematic
-over traditional CI tools or container-oriented tooling for ML pipelines.
+over traditional CI tools or container-oriented tooling for ML pipelines.
diff --git a/docs/upgrades.md b/docs/upgrades.md
@@ -272,8 +272,7 @@ need to duplicate them in the `deps` field.
 You will need to make the following changes to the `values.yaml`
 for your Helm deployment:
 
-- If you are NOT using Sematic "Enterprise Edition",
-change `image.repository` from `sematicai/sematic-server`
+- Change `image.repository` from `sematicai/sematic-server`
 to `sematic/sematic-server`.
 - You may have hard-coded a Sematic version in `image.tag`
 to work around the bug referenced in
@@ -286,17 +285,6 @@ for the chart you install.
 parameter to
 [`ingress.sematic_dashboard_url`](https://github.com/sematic-ai/sematic/pull/547)
 
-## FOSS to "Enterprise Edition"
-
-0. Reach out to Sematic via [[email protected]](mailto:[email protected]) to obtain
-an Enterprise License.
-1. In your helm deployment, change `image.repository` to `sematic/sematic-server-ee`.
-2. In your `pip` installation (client installation), change from depending on `sematic`
-to `sematic[<extra>]`, where `<extra>` is the appropriate variant of Sematic containing
-the EE features you are interested in. See [integrations](#integrations) below for
-specific extras. Note that `sematic[all]` will equip clients with ALL Enterprise Edition
-features.
-
 ## Integrations
 
 Some integrations require changes to how pip packages and Helm deployments

diff --git a/helm/sematic-server/values.yaml b/helm/sematic-server/values.yaml
@@ -32,7 +32,7 @@ slack:
   slack_webhook_token: XXX/YYY/ZZZ # REPLACE ME
 
 github:
-  enabled: false  # This feature can only be enabled with Sematic EE.
+  enabled: false
   github_access_token: github_pat_1234 # REPLACE ME
 
 secret:

diff --git a/sematic/ee/LICENSE b/sematic/ee/LICENSE
diff --git a/sematic/ee/README.md b/sematic/ee/README.md
diff --git a/sematic/examples/lightning_resnet/README.md b/sematic/examples/lightning_resnet/README.md
@@ -7,8 +7,7 @@ to do distributed training of a model classifying images in the
 using ResNet.
 
 This example can only be run on Kubernetes clusters set up with sufficiently powerful
-GPU nodes enabled. It makes use of Sematic's Ray integration, which is available in
-Sematic EE.
+GPU nodes enabled. It makes use of Sematic's Ray integration.
 
 The modeling code in `train_eval` is based on an example from
 [the pytorch lightning documentation](https://pytorch-lightning.readthedocs.io/en/stable/notebooks/lightning_examples/cifar10-baseline.html)

diff --git a/sematic/examples/summarization_finetune/train_eval.py b/sematic/examples/summarization_finetune/train_eval.py
@@ -77,10 +77,7 @@ def trim_context(context):
 
 
 def log_metric(name: str, value: float) -> None:
-    """Stand-in for Sematic's `log_metric` that prints to stdout.
-
-    Sematic's log_metric is only available for EE users.
-    """
+    """Stand-in for Sematic's `log_metric` that prints to stdout."""
     print(f"Metric '{name}': {value}")