WIP: Fixes (#40)

* what type of pipeline * openAI clarifications * let's not force it. users should see choice between kserve and modelmesh * refreshed screenshots
rh-aiservices-bu · Apr 17, 2024 · b51f51c · b51f51c
1 parent 60d41a3
commit b51f51c
Show file tree

Hide file tree

Showing 12 changed files with 15 additions and 15 deletions.
diff --git a/bootstrap/ic-user-projects/create-projects-and-resources-job.yaml b/bootstrap/ic-user-projects/create-projects-and-resources-job.yaml
@@ -54,7 +54,7 @@ spec:
               openshift.io/display-name: $USER_PROJECT
             labels:
               kubernetes.io/metadata.name: $USER_PROJECT
-              modelmesh-enabled: 'true'
+              # modelmesh-enabled: 'true'
               opendatahub.io/dashboard: 'true'
             name: $USER_PROJECT
           spec:

diff --git a/bootstrap/ic-user-projects/create-projects-and-resources.bash b/bootstrap/ic-user-projects/create-projects-and-resources.bash
@@ -32,7 +32,7 @@ metadata:
     openshift.io/display-name: $USER_PROJECT
   labels:
     kubernetes.io/metadata.name: $USER_PROJECT
-    modelmesh-enabled: 'true'
+    # modelmesh-enabled: 'true'
     opendatahub.io/dashboard: 'true'
   name: $USER_PROJECT
 spec:

diff --git a/content/modules/ROOT/assets/images/04/deploy-a-model.png b/content/modules/ROOT/assets/images/04/deploy-a-model.png
diff --git a/content/modules/ROOT/assets/images/04/model-deployed-success.png b/content/modules/ROOT/assets/images/04/model-deployed-success.png
diff --git a/content/modules/ROOT/pages/02-02-auto-created-project.adoc b/content/modules/ROOT/pages/02-02-auto-created-project.adoc
@@ -23,4 +23,4 @@ image::02/02-02-pre-created-components.png[]
 ** A Workbench - this is your environment that you can experiment and train in.
 ** A Cluster storage - this is a persistent storage for your workbench.
 ** A Data Connection - it contains all the information needed to talk with an existing S3 storage, we use that to store models and pipeline artifacts.
-** A Pipeline Server - the pipeline server has already been set up here so you can import or run pipelines right away.
+** A Pipeline Server - the pipeline server has already been set up here so you can import or run Data Science pipelines right away.
diff --git a/content/modules/ROOT/pages/02-02-diy-creating-project.adoc b/content/modules/ROOT/pages/02-02-diy-creating-project.adoc
@@ -118,4 +118,4 @@ image::02/02-02-pipelineserver03.png[]
 
 At this point, your pipeline server is ready and deployed.
 
-IMPORTANT: You need to **wait** until that screen is ready. If it's still spinning, wait for it to complete. If you continue and create your workbench **before** the pipeline server is ready, your workbench will not be able to submit pipelines to it.
+IMPORTANT: You need to **wait** until that screen is ready. If it's still spinning, wait for it to complete. If you continue and create your workbench **before** the pipeline server is ready, your workbench will not be able to submit Data Science pipelines to it.
diff --git a/content/modules/ROOT/pages/05-05-process-claims.adoc b/content/modules/ROOT/pages/05-05-process-claims.adoc
@@ -115,7 +115,7 @@ image::05/05-import-pipeline.jpg[import pipeline]
 [.bordershadow]
 image::05/05-import-pipeline-highlighted.jpg[imported pipeline]
 
-* Click  **Import Pipeline** and you should see it pop up under your pipelines
+* Click  **Import Pipeline** and you should see it appear under the pipelines section of your Project
 
 * Now go into the settings at the right side
 * Click **Create Run** to create a new run of the pipeline you just added

diff --git a/content/modules/ROOT/pages/06-01-potential-imp-ref.adoc b/content/modules/ROOT/pages/06-01-potential-imp-ref.adoc
@@ -37,7 +37,7 @@ If you want to read what **we** thought could be improved, read below! (response
 * We could have something that analyzes the images and checks for discrepancies with the customer data, such as:
 ** Not the same make or color car as what is on file.
 ** Mismatch in license plate, if visible in the picture.
-* We've only scratched the surface with gitops and pipelines here
+* We've only scratched the surface with gitops and Data Science pipelines here
 ** There was no performance testing done. If too many users connect at the same time, it might overwhelm either the app, the database, the LLM, etc...
 * Currently, most simple changes would probably end up breaking the application. And the person who, for example decides to change Mistral7B for Flan-T5-Small would not necessarily realize that.
 ** It would be critical to have multiple instances (Dev/Test/UAT/Prod) of the application.

diff --git a/lab-materials/03/03-01-nb-llm-example.ipynb b/lab-materials/03/03-01-nb-llm-example.ipynb
@@ -70,8 +70,8 @@
     "inference_server_url = \"http://llm.ic-shared-llm.svc.cluster.local:8000\"\n",
     "\n",
     "# LLM definition\n",
-    "llm = VLLMOpenAI(\n",
-    "    openai_api_key=\"EMPTY\",\n",
+    "llm = VLLMOpenAI(           # we are using the vLLM OpenAI-compatible API client. But the Model is running on OpenShift, not OpenAI.\n",
+    "    openai_api_key=\"EMPTY\",   # and that is why we don't need an OpenAI key for this.\n",
     "    openai_api_base= f\"{inference_server_url}/v1\",\n",
     "    model_name=\"mistralai/Mistral-7B-Instruct-v0.2\",\n",
     "    top_p=0.92,\n",

diff --git a/lab-materials/03/03-02-summarization.ipynb b/lab-materials/03/03-02-summarization.ipynb
@@ -66,8 +66,8 @@
     "inference_server_url = \"http://llm.ic-shared-llm.svc.cluster.local:8000\"\n",
     "\n",
     "# LLM definition\n",
-    "llm = VLLMOpenAI(\n",
-    "    openai_api_key=\"EMPTY\",\n",
+    "llm = VLLMOpenAI(           # we are using the vLLM OpenAI-compatible API client. But the Model is running on OpenShift, not OpenAI.\n",
+    "    openai_api_key=\"EMPTY\",   # and that is why we don't need an OpenAI key for this.\n",
     "    openai_api_base= f\"{inference_server_url}/v1\",\n",
     "    model_name=\"mistralai/Mistral-7B-Instruct-v0.2\",\n",
     "    top_p=0.92,\n",

diff --git a/lab-materials/03/03-03-information-extraction.ipynb b/lab-materials/03/03-03-information-extraction.ipynb
@@ -66,8 +66,8 @@
     "inference_server_url = \"http://llm.ic-shared-llm.svc.cluster.local:8000\"\n",
     "\n",
     "# LLM definition\n",
-    "llm = VLLMOpenAI(\n",
-    "    openai_api_key=\"EMPTY\",\n",
+    "llm = VLLMOpenAI(           # we are using the vLLM OpenAI-compatible API client. But the Model is running on OpenShift, not OpenAI.\n",
+    "    openai_api_key=\"EMPTY\",   # and that is why we don't need an OpenAI key for this.\n",
     "    openai_api_base= f\"{inference_server_url}/v1\",\n",
     "    model_name=\"mistralai/Mistral-7B-Instruct-v0.2\",\n",
     "    top_p=0.92,\n",

diff --git a/lab-materials/03/03-04-comparing-model-servers.ipynb b/lab-materials/03/03-04-comparing-model-servers.ipynb
@@ -50,7 +50,7 @@
    "source": [
     "### Langchain pipeline\n",
     "\n",
-    "We are now going to define two different LLM endpoints, and two different pipelines."
+    "We are now going to define two different LLM endpoints, and two different Langchain pipelines."
    ]
   },
   {
@@ -66,8 +66,8 @@
     "inference_server_url = \"http://llm.ic-shared-llm.svc.cluster.local:8000\"\n",
     "\n",
     "# LLM definition\n",
-    "llm = VLLMOpenAI(\n",
-    "    openai_api_key=\"EMPTY\",\n",
+    "llm = VLLMOpenAI(           # we are using the vLLM OpenAI-compatible API client. But the Model is running on OpenShift, not OpenAI.\n",
+    "    openai_api_key=\"EMPTY\",   # and that is why we don't need and OpenAI key for this.\n",
     "    openai_api_base= f\"{inference_server_url}/v1\",\n",
     "    model_name=\"mistralai/Mistral-7B-Instruct-v0.2\",\n",
     "    top_p=0.92,\n",