Final fixes for Nemotron-CC tutorial (#829)

* Final fixes for Nemotron-CC tutorial

Signed-off-by: Sarah Yurick <sarahyurick@gmail.com>

* add vineeth's fixes

Signed-off-by: Sarah Yurick <sarahyurick@gmail.com>

* add missing imports

Signed-off-by: Sarah Yurick <sarahyurick@gmail.com>

---------

Signed-off-by: Sarah Yurick <sarahyurick@gmail.com>

Author: sarahyurick

tutorials/nemotron-cc/nemotron_cc.ipynb

@@ -495,8 +495,8 @@
    },
    "outputs": [],
    "source": [
-    "# client = get_client(cluster_type=\"cpu\", n_workers=10, processes=True, memory_limit=\"16GiB\") #noqa: ERA001\n",
-    "# client"
+    "client = get_client(cluster_type=\"cpu\", n_workers=10, processes=True, memory_limit=\"16GiB\")\n",
+    "client"
    ]
   },
   {
@@ -742,8 +742,8 @@
    },
    "outputs": [],
    "source": [
-    "# client.cluster.close() #noqa: ERA001\n",
-    "# client.shutdown() #noqa: ERA001"
+    "client.cluster.close()\n",
+    "client.shutdown()"
    ]
   },
   {
@@ -784,13 +784,15 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": null,
    "id": "5f788b91",
    "metadata": {
     "tags": []
    },
    "outputs": [],
    "source": [
+    "from dask.distributed import Client, LocalCluster\n",
+    "\n",
     "from nemo_curator import AddId"
    ]
   },
@@ -811,8 +813,8 @@
    },
    "outputs": [],
    "source": [
-    "# cluster = LocalCluster(n_workers=10, processes=True, memory_limit=\"16GB\") #noqa: ERA001\n",
-    "# client = Client(cluster) #noqa: ERA001"
+    "cluster = LocalCluster(n_workers=10, processes=True, memory_limit=\"16GB\")\n",
+    "client = Client(cluster)"
    ]
   },
   {
@@ -893,8 +895,8 @@
    },
    "outputs": [],
    "source": [
-    "# client.cluster.close() #noqa: ERA001\n",
-    "# client.shutdown() #noqa: ERA001"
+    "client.cluster.close()\n",
+    "client.shutdown()"
    ]
   },
   {
@@ -1189,15 +1191,15 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": null,
    "id": "5ef2f05e",
    "metadata": {
     "tags": []
    },
    "outputs": [],
    "source": [
-    "# client.cluster.close() #noqa: ERA001\n",
-    "# client.shutdown() #noqa: ERA001"
+    "client.cluster.close()\n",
+    "client.shutdown()"
    ]
   },
   {
@@ -2583,7 +2585,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": null,
    "id": "6418114a",
    "metadata": {},
    "outputs": [
@@ -2615,7 +2617,7 @@
     }
    ],
    "source": [
-    "output_dataset = input_dataset\n",
+    "output_dataset = DocumentDataset(input_dataset.df.reset_index())\n",
     "for classifier in classifiers:\n",
     "    output_dataset = classifier(dataset=output_dataset)\n",
     "\n",
@@ -3185,25 +3187,33 @@
     "!mkdir -p {high_quality_output_data_dir}"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "095c99e5",
+   "metadata": {},
+   "source": [
+    "An NVIDIA API token can be obtained from: http://build.nvidia.com/"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "id": "aa31956f",
    "metadata": {},
    "outputs": [],
    "source": [
     "from openai import OpenAI\n",
     "from transformers import AutoTokenizer\n",
     "\n",
-    "openai_api_token = \"<your-openai-api-token>\"  # noqa: S105\n",
+    "nvidia_api_token = \"<your-nv-api-token>\"  # noqa: S105\n",
     "hf_token = \"<your-huggingface-token>\"  # noqa: S105\n",
     "hf_model_name = \"google/gemma-3-1b-it\"\n",
     "api_model_name = \"google/gemma-3-1b-it\"\n",
     "base_url = \"https://integrate.api.nvidia.com/v1\"\n",
     "\n",
     "\n",
     "tokenizer = AutoTokenizer.from_pretrained(hf_model_name, token=hf_token)\n",
-    "openai_client = OpenAI(base_url=base_url, api_key=openai_api_token)"
+    "openai_client = OpenAI(base_url=base_url, api_key=nvidia_api_token)"
    ]
   },
   {
@@ -3726,7 +3736,7 @@
    "id": "01b076ee",
    "metadata": {},
    "source": [
-    "## 5. Visualization"
+    "## 5. [Optional] Visualization"
    ]
   },
   {
@@ -3765,7 +3775,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": null,
    "id": "b23b66e6",
    "metadata": {},
    "outputs": [
@@ -4021,11 +4031,11 @@
    "source": [
     "from viz.text_comparison_widget import compare_row_by_id\n",
     "\n",
-    "low_quality_exmaple = \"EN_CC-0000000181\"\n",
+    "low_quality_example = \"<pick-one-low-quality-datapoint>\" # ex: EN_CC-0000000181 (based on our runs)\n",
     "\n",
     "compare_row_by_id(\n",
     "    final,\n",
-    "    row_id=low_quality_exmaple,  # Change the row id to see different examples\n",
+    "    row_id=low_quality_example,  # Change the row id to see different examples\n",
     "    id_column=\"id\",\n",
     "    col1=\"text\",\n",
     "    col2=\"rephrased\",\n",
@@ -4038,7 +4048,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": null,
    "id": "200b87a8",
    "metadata": {},
    "outputs": [
@@ -4360,12 +4370,12 @@
    "source": [
     "from viz.text_comparison_widget import compare_row_by_id\n",
     "\n",
-    "high_quality_exmaple = \"EN_CC-0000031171\"\n",
+    "high_quality_example = \"<pick-one-high-quality-datapoint>\" # ex: EN_CC-EN_CC-0000031171 (based on our runs)\n",
     "\n",
     "method = \"distill\"  # Choose between diverse_qa, distill, extract_knowledge, knowledge_list\n",
     "compare_row_by_id(\n",
     "    final,\n",
-    "    row_id=high_quality_exmaple,  # Change the row id to see different examples\n",
+    "    row_id=high_quality_example,  # Change the row id to see different examples\n",
     "    id_column=\"id\",\n",
     "    col1=\"text\",\n",
     "    col2=method,\n",