renaming todo to exercise

notebooks/vertex_genai/labs/gen_ai_evaluation_service.ipynb

@@ -173,7 +173,7 @@
     "\n",
     "Let's say that we are trying to evaluate how well different prompts works for summarization using Gemini. We'll start by defining a few articles in `context` and since we are using computation based metrics for evaluation, we will also need to define the ground truth for the summaries. This will be defined in `reference`. `eval_dataset` should be a dataframe that contains columns needed for evaluation.\n",
     "\n",
-    "**TODO** Create a `eval_dataset` DataFrame with the columns `context`, `reference` and `instruction`"
+    "**Exercise** Create a `eval_dataset` DataFrame with the columns `context`, `reference` and `instruction`"
    ]
   },
   {
@@ -309,7 +309,7 @@
     "\n",
     "To start with, we'll be using model inference with prompt templates.\n",
     "\n",
-    "**TODO** Define `EvalTask` with the parameters dataset, metrics and experiment. See the documentation for [EvalTask](https://cloud.google.com/vertex-ai/generative-ai/docs/reference/python/latest/vertexai.preview.evaluation.EvalTask) "
+    "**Exercise** Define `EvalTask` with the parameters dataset, metrics and experiment. See the documentation for [EvalTask](https://cloud.google.com/vertex-ai/generative-ai/docs/reference/python/latest/vertexai.preview.evaluation.EvalTask) "
    ]
   },
   {
@@ -512,7 +512,7 @@
    "id": "5cf4de49-8165-45a7-a743-bc3696c9d0d1",
    "metadata": {},
    "source": [
-    "**TODO:** For the above `context` list, generate the summary responses from Gemini using the `generate_content()` function. Refer to the [documentation](https://cloud.google.com/vertex-ai/generative-ai/docs/model-reference/inference#non-streaming) "
+    "**Exercise:** For the above `context` list, generate the summary responses from Gemini using the `generate_content()` function. Refer to the [documentation](https://cloud.google.com/vertex-ai/generative-ai/docs/model-reference/inference#non-streaming) "
    ]
   },
   {
@@ -538,7 +538,7 @@
    "id": "3610f476-b48a-487d-9c5f-fb6db55ef1a8",
    "metadata": {},
    "source": [
-    "**TODO**: Define a dataset for evaluation `eval_dataset` as a pandas Dataframe. This eval_dataset only has one column, the responses you generated above"
+    "**Exercise**: Define a dataset for evaluation `eval_dataset` as a pandas Dataframe. This eval_dataset only has one column, the responses you generated above"
    ]
   },
   {
@@ -727,7 +727,7 @@
    "source": [
     "Once we have defined the dataset and the evaluation crieteria. We are ready to kick off the evaluation job.\n",
     "\n",
-    "**TODO:** Define your `EvalTask` function here. The parameters that are needed for pairwise evaluateion are, dataset, metrics and experiment\n",
+    "**Exercise** Define your `EvalTask` function here. The parameters that are needed for pairwise evaluateion are, dataset, metrics and experiment\n",
     "\n",
     "Please note: the cell below is going to take 10-15 mins to finish execution."
    ]