![\"Open](\"https://colab.research.google.com/assets/colab-badge.svg\"){kind=icon}
\n",
+ "\n",
+ ""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "collapsed": true,
+ "executionInfo": {
+ "elapsed": 19538,
+ "status": "ok",
+ "timestamp": 1756743355864,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "MlVgdTfI4mVF",
+ "outputId": "48eefd32-5470-44fe-db52-f3a799248c94"
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\u001b[?25l \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0.0/491.5 kB\u001b[0m \u001b[31m?\u001b[0m eta \u001b[36m-:--:--\u001b[0m\r\u001b[2K \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[32m471.0/491.5 kB\u001b[0m \u001b[31m15.8 MB/s\u001b[0m eta \u001b[36m0:00:01\u001b[0m\r\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m491.5/491.5 kB\u001b[0m \u001b[31m8.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[?25h"
+ ]
+ }
+ ],
+ "source": [
+ "pip install --quiet datasets==3.6.0 gcsfs==2025.3.0 fsspec==2025.3.0"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "collapsed": true,
+ "executionInfo": {
+ "elapsed": 16093,
+ "status": "ok",
+ "timestamp": 1756743375636,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "a_rQlPqIwJoe",
+ "outputId": "c8baf8bc-c1ac-40bb-99c3-7cd82dc477e0"
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.7/1.7 MB\u001b[0m \u001b[31m12.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m313.6/313.6 kB\u001b[0m \u001b[31m16.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[?25h"
+ ]
+ }
+ ],
+ "source": [
+ "pip install --quiet pymongo voyageai pandas==2.2.2 matplotlib"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "collapsed": true,
+ "executionInfo": {
+ "elapsed": 19250,
+ "status": "ok",
+ "timestamp": 1756744205537,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "v9WNOdlUESGQ",
+ "outputId": "096a4f12-42b1-4af7-b0f9-c0eaf0cbeaaf"
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Enter your Voyage API Key: ··········\n",
+ "Enter your MongoDB URI: ··········\n"
+ ]
+ }
+ ],
+ "source": [
+ "import getpass\n",
+ "import os\n",
+ "import voyageai\n",
+ "\n",
+ "# Function to securely get and set environment variables\n",
+ "def set_env_securely(var_name, prompt):\n",
+ " value = getpass.getpass(prompt)\n",
+ " os.environ[var_name] = value\n",
+ "\n",
+ "# Environment Variables\n",
+ "set_env_securely(\"VOYAGE_API_KEY\", \"Enter your Voyage API Key: \")\n",
+ "set_env_securely(\"MONGO_URI\", \"Enter your MongoDB URI: \")\n",
+ "MONGO_URI = os.environ.get(\"MONGO_URI\")\n",
+ "if not MONGO_URI:\n",
+ " raise ValueError(\"MONGO_URI not set in environment variables.\")\n",
+ "\n",
+ "# Voyage Client\n",
+ "voyage_client = voyageai.Client()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {
+ "executionInfo": {
+ "elapsed": 48,
+ "status": "ok",
+ "timestamp": 1756744211284,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "VdidySCdPv1b"
+ },
+ "outputs": [],
+ "source": [
+ "import pandas as pd\n",
+ "from datasets import load_dataset\n",
+ "from bson.binary import Binary, BinaryVectorDtype\n",
+ "import pymongo\n",
+ "\n",
+ "# Connect to Cluster\n",
+ "def get_mongo_client(uri):\n",
+ " \"\"\"Connect to MongoDB and confirm the connection.\"\"\"\n",
+ " client = pymongo.MongoClient(uri)\n",
+ " if client.admin.command(\"ping\").get(\"ok\") == 1.0:\n",
+ " print(\"Connected to MongoDB successfully.\")\n",
+ " return client\n",
+ " print(\"Failed to connect to MongoDB.\")\n",
+ " return None\n",
+ "\n",
+ "# Generate BSON Vector\n",
+ "def generate_bson_vector(array, data_type):\n",
+ " \"\"\"Convert an array to BSON vector format.\"\"\"\n",
+ " array = [float(val) for val in eval(array)]\n",
+ " return Binary.from_vector(array, BinaryVectorDtype(data_type))\n",
+ "\n",
+ "# Load Datasets\n",
+ "def load_and_prepare_data(dataset_name, amount):\n",
+ " \"\"\"Load and prepare streaming datasets for DataFrame.\"\"\"\n",
+ " data = load_dataset(dataset_name, streaming=True, split=\"train\").take(amount)\n",
+ " return pd.DataFrame(data)\n",
+ "\n",
+ "# Insert datasets into MongoDB Collection\n",
+ "def insert_dataframe_into_collection(df, collection):\n",
+ " \"\"\"Insert Dataset records into MongoDB collection.\"\"\"\n",
+ " collection.insert_many(df.to_dict(\"records\"))\n",
+ " print(f\"Inserted {len(df)} records into '{collection.name}' collection.\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "collapsed": true,
+ "executionInfo": {
+ "elapsed": 34182,
+ "status": "ok",
+ "timestamp": 1756744424113,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "WG--GrYFPzV9",
+ "outputId": "766e33fe-48a9-41c8-a823-9f45b5453a70"
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Connected to MongoDB successfully.\n",
+ "Collection 'wikipedia-22-12-en' created successfully.\n",
+ "Inserted 2500 records into 'wikipedia-22-12-en' collection.\n",
+ "Collection 'wikipedia-22-12-en-annotation' created successfully.\n",
+ "Inserted 2500 records into 'wikipedia-22-12-en-annotation' collection.\n"
+ ]
+ }
+ ],
+ "source": [
+ "import pandas as pd\n",
+ "from bson.binary import Binary, BinaryVectorDtype\n",
+ "from pymongo.errors import CollectionInvalid\n",
+ "\n",
+ "wikipedia_data_df = load_and_prepare_data(\"MongoDB/wikipedia-22-12-en-voyage-embed\", amount=2500)\n",
+ "wikipedia_annotation_data_df = load_and_prepare_data(\"MongoDB/wikipedia-22-12-en-annotation\", amount=2500)\n",
+ "wikipedia_annotation_data_df.drop(columns=[\"_id\"], inplace=True)\n",
+ "\n",
+ "# Convert embeddings to BSON format\n",
+ "wikipedia_data_df[\"embedding\"] = wikipedia_data_df[\"embedding\"].apply(\n",
+ " lambda x: generate_bson_vector(x, BinaryVectorDtype.FLOAT32)\n",
+ ")\n",
+ "\n",
+ "# MongoDB Setup\n",
+ "mongo_client = get_mongo_client(MONGO_URI)\n",
+ "DB_NAME = \"testing_datasets\"\n",
+ "db = mongo_client[DB_NAME]\n",
+ "\n",
+ "collections = {\n",
+ " \"wikipedia-22-12-en\": wikipedia_data_df,\n",
+ " \"wikipedia-22-12-en-annotation\": wikipedia_annotation_data_df,\n",
+ "}\n",
+ "\n",
+ "# Create Collections and Insert Data\n",
+ "for collection_name, df in collections.items():\n",
+ " if collection_name not in db.list_collection_names():\n",
+ " try:\n",
+ " db.create_collection(collection_name)\n",
+ " print(f\"Collection '{collection_name}' created successfully.\")\n",
+ " except CollectionInvalid:\n",
+ " print(f\"Error creating collection '{collection_name}'.\")\n",
+ " else:\n",
+ " print(f\"Collection '{collection_name}' already exists.\")\n",
+ "\n",
+ " # Clear collection and insert fresh data\n",
+ " collection = db[collection_name]\n",
+ " collection.delete_many({})\n",
+ " insert_dataframe_into_collection(df, collection)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {
+ "executionInfo": {
+ "elapsed": 49,
+ "status": "ok",
+ "timestamp": 1756744459328,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "SXeaGqmI1TsC"
+ },
+ "outputs": [],
+ "source": [
+ "import time\n",
+ "from pymongo.operations import SearchIndexModel\n",
+ "\n",
+ "def setup_vector_search_index(collection, index_definition, index_name=\"vector_index\"):\n",
+ " new_vector_search_index_model = SearchIndexModel(\n",
+ " definition=index_definition, name=index_name, type=\"vectorSearch\"\n",
+ " )\n",
+ "\n",
+ " # Create the new index\n",
+ " try:\n",
+ " result = collection.create_search_index(model=new_vector_search_index_model)\n",
+ " print(f\"Creating index '{index_name}'...\")\n",
+ "\n",
+ " # Wait for initial sync to complete\n",
+ " print(\"Polling to check if the index is ready. This may take around a minute.\")\n",
+ " predicate=None\n",
+ " if predicate is None:\n",
+ " predicate = lambda index: index.get(\"queryable\") is True\n",
+ " while True:\n",
+ " indices = list(collection.list_search_indexes(result))\n",
+ " if len(indices) and predicate(indices[0]):\n",
+ " break\n",
+ " time.sleep(5)\n",
+ " print(f\"Index '{index_name}' is ready for querying.\")\n",
+ " return result\n",
+ "\n",
+ " except Exception as e:\n",
+ " print(f\"Error creating new vector search index '{index_name}': {e!s}\")\n",
+ " return None"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {
+ "executionInfo": {
+ "elapsed": 50,
+ "status": "ok",
+ "timestamp": 1756744468384,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "y0vqX0Ed1dgc"
+ },
+ "outputs": [],
+ "source": [
+ "vector_index_definition_scalar_quantized = {\n",
+ " \"fields\": [\n",
+ " {\n",
+ " \"type\": \"vector\",\n",
+ " \"path\": \"embedding\",\n",
+ " \"quantization\": \"scalar\", # Added quantization for scalar vector quantization\n",
+ " \"numDimensions\": 1024,\n",
+ " \"similarity\": \"cosine\",\n",
+ " }\n",
+ " ]\n",
+ "}\n",
+ "\n",
+ "vector_index_definition_binary_quantized = {\n",
+ " \"fields\": [\n",
+ " {\n",
+ " \"type\": \"vector\",\n",
+ " \"path\": \"embedding\",\n",
+ " \"quantization\": \"binary\", # Changed quantization to binary for binary vector quantization\n",
+ " \"numDimensions\": 1024,\n",
+ " \"similarity\": \"cosine\",\n",
+ " }\n",
+ " ]\n",
+ "}\n",
+ "\n",
+ "vector_index_definition_float32_ann = {\n",
+ " \"fields\": [\n",
+ " {\n",
+ " \"type\": \"vector\",\n",
+ " \"path\": \"embedding\",\n",
+ " \"numDimensions\": 1024,\n",
+ " \"similarity\": \"cosine\",\n",
+ " }\n",
+ " ]\n",
+ "}"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {
+ "executionInfo": {
+ "elapsed": 47,
+ "status": "ok",
+ "timestamp": 1756744474089,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "jlQHLYxA1mNW"
+ },
+ "outputs": [],
+ "source": [
+ "import pymongo\n",
+ "vector_search_scalar_quantized_index_name = \"vector_index_scalar_quantized\"\n",
+ "vector_search_binary_quantized_index_name = \"vector_index_binary_quantized\"\n",
+ "vector_search_float32_ann_index_name = \"vector_index_float32_ann\"\n",
+ "db = mongo_client[DB_NAME]\n",
+ "wiki_data_collection = db[\"wikipedia-22-12-en\"]\n",
+ "wiki_annotation_data_collection = db[\"wikipedia-22-12-en-annotation\"]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 196
+ },
+ "collapsed": true,
+ "executionInfo": {
+ "elapsed": 87075,
+ "status": "ok",
+ "timestamp": 1756744563624,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "oKw8HI3H1r7w",
+ "outputId": "8e447700-a341-411b-815e-f97e6f61a920"
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Creating index 'vector_index_scalar_quantized'...\n",
+ "Polling to check if the index is ready. This may take around a minute.\n",
+ "Index 'vector_index_scalar_quantized' is ready for querying.\n",
+ "Creating index 'vector_index_binary_quantized'...\n",
+ "Polling to check if the index is ready. This may take around a minute.\n",
+ "Index 'vector_index_binary_quantized' is ready for querying.\n",
+ "Creating index 'vector_index_float32_ann'...\n",
+ "Polling to check if the index is ready. This may take around a minute.\n",
+ "Index 'vector_index_float32_ann' is ready for querying.\n"
+ ]
+ },
+ {
+ "data": {
+ "application/vnd.google.colaboratory.intrinsic+json": {
+ "type": "string"
+ },
+ "text/plain": [
+ "'vector_index_float32_ann'"
+ ]
+ },
+ "execution_count": 14,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "from pymongo.operations import SearchIndexModel\n",
+ "\n",
+ "setup_vector_search_index(\n",
+ " wiki_data_collection,\n",
+ " vector_index_definition_scalar_quantized,\n",
+ " vector_search_scalar_quantized_index_name,\n",
+ ")\n",
+ "setup_vector_search_index(\n",
+ " wiki_data_collection,\n",
+ " vector_index_definition_binary_quantized,\n",
+ " vector_search_binary_quantized_index_name,\n",
+ ")\n",
+ "setup_vector_search_index(\n",
+ " wiki_data_collection,\n",
+ " vector_index_definition_float32_ann,\n",
+ " vector_search_float32_ann_index_name,\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {
+ "executionInfo": {
+ "elapsed": 47,
+ "status": "ok",
+ "timestamp": 1756744594454,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "G-YgBPNOKwF-"
+ },
+ "outputs": [],
+ "source": [
+ "def get_embedding(text, task_prefix=\"document\"):\n",
+ " \"\"\"Fetch embedding for a given text using Voyage AI.\"\"\"\n",
+ " if not text.strip():\n",
+ " print(\"Empty text provided for embedding.\")\n",
+ " return []\n",
+ " result = voyage_client.embed([text], model=\"voyage-3-large\", input_type=task_prefix)\n",
+ " return result.embeddings[0]\n",
+ "\n",
+ "def custom_vector_search(\n",
+ " user_query,\n",
+ " collection,\n",
+ " embedding_path,\n",
+ " vector_search_index_name=\"vector_index\",\n",
+ " top_k=5,\n",
+ " num_candidates=25,\n",
+ " use_full_precision=False,\n",
+ "):\n",
+ " \"\"\"Perform vector search on a MongoDB collection using specified index.\"\"\"\n",
+ " # Generate embedding for the user query\n",
+ " query_embedding = get_embedding(user_query, task_prefix=\"query\")\n",
+ "\n",
+ " if query_embedding is None or not query_embedding:\n",
+ " return {\n",
+ " \"error\": \"Invalid query or embedding generation failed.\",\n",
+ " \"execution_time_ms\": None,\n",
+ " \"results\": [],\n",
+ " }\n",
+ "\n",
+ " # Define the vector search stage\n",
+ " vector_search_stage = {\n",
+ " \"$vectorSearch\": {\n",
+ " \"index\": vector_search_index_name,\n",
+ " \"queryVector\": query_embedding,\n",
+ " \"path\": embedding_path,\n",
+ " \"limit\": top_k,\n",
+ " }\n",
+ " }\n",
+ "\n",
+ " # Configure search precision approach\n",
+ " if not use_full_precision:\n",
+ " # For approximate nearest neighbor (ANN) search\n",
+ " vector_search_stage[\"$vectorSearch\"][\"numCandidates\"] = num_candidates\n",
+ " else:\n",
+ " # For exact nearest neighbor (ENN) search\n",
+ " vector_search_stage[\"$vectorSearch\"][\"exact\"] = True\n",
+ "\n",
+ " # Project stage to fetch desired fields and vector search score\n",
+ " project_stage = {\n",
+ " \"$project\": {\n",
+ " \"_id\": 0,\n",
+ " \"title\": 1,\n",
+ " \"text\": 1,\n",
+ " \"wiki_id\": 1,\n",
+ " \"url\": 1,\n",
+ " \"score\": {\"$meta\": \"vectorSearchScore\"},\n",
+ " }\n",
+ " }\n",
+ "\n",
+ " # Define the aggregate pipeline\n",
+ " pipeline = [vector_search_stage, project_stage]\n",
+ "\n",
+ " try:\n",
+ " # Execute the explain command to measure latency\n",
+ " explain_result = collection.database.command(\n",
+ " \"explain\",\n",
+ " {\"aggregate\": collection.name, \"pipeline\": pipeline, \"cursor\": {}},\n",
+ " verbosity=\"executionStats\",\n",
+ " )\n",
+ "\n",
+ " # Extract the execution time\n",
+ " vector_search_explain = explain_result[\"stages\"][0][\"$vectorSearch\"]\n",
+ " execution_time_ms = vector_search_explain[\"explain\"][\"query\"][\"stats\"][\"context\"][\"millisElapsed\"]\n",
+ "\n",
+ " # Execute the actual aggregate query\n",
+ " results = list(collection.aggregate(pipeline))\n",
+ "\n",
+ " return {\n",
+ " \"results\": results,\n",
+ " \"execution_time_ms\": execution_time_ms,\n",
+ " }\n",
+ " except Exception as e:\n",
+ " print(f\"Error during vector search: {e}\")\n",
+ " return {\n",
+ " \"error\": str(e),\n",
+ " \"execution_time_ms\": None,\n",
+ " \"results\": [],\n",
+ " }"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 175
+ },
+ "executionInfo": {
+ "elapsed": 1420,
+ "status": "ok",
+ "timestamp": 1756744627479,
+ "user": {
+ "displayName": "Javier Armendariz",
+ "userId": "14823381200026660254"
+ },
+ "user_tz": 360
+ },
+ "id": "8qxNXCRdbJKD",
+ "outputId": "81d84363-b33f-4485-d16b-5025fc1a7be6"
+ },
+ "outputs": [
+ {
+ "data": {
+ "application/vnd.google.colaboratory.intrinsic+json": {
+ "summary": "{\n \"name\": \"results_df\",\n \"rows\": 4,\n \"fields\": [\n {\n \"column\": \"precision\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [\n \"_scalar_\",\n \"Float32_ENN\",\n \"_float32_ann\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"top_k\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 5,\n \"max\": 5,\n \"num_unique_values\": 1,\n \"samples\": [\n 5\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"num_candidates\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 25,\n \"max\": 25,\n \"num_unique_values\": 1,\n \"samples\": [\n 25\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"latency_ms\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 91.0287034707709,\n \"min\": 1.72466,\n \"max\": 205.778084,\n \"num_unique_values\": 4,\n \"samples\": [\n 205.778084\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"results\",\n \"properties\": {\n \"dtype\": \"object\",\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}",
+ "type": "dataframe",
+ "variable_name": "results_df"
+ },
+ "text/html": [
+ "\n",
+ " | \n", + " | precision | \n", + "top_k | \n", + "num_candidates | \n", + "latency_ms | \n", + "results | \n", + "
|---|---|---|---|---|---|
| 0 | \n", + "_float32_ann | \n", + "5 | \n", + "25 | \n", + "86.283681 | \n", + "{'title': 'Facebook', 'text': 'Data is read fr... | \n", + "
| 1 | \n", + "_scalar_ | \n", + "5 | \n", + "25 | \n", + "205.778084 | \n", + "{'title': 'Facebook', 'text': 'Data is read fr... | \n", + "
| 2 | \n", + "_binary_ | \n", + "5 | \n", + "25 | \n", + "26.408544 | \n", + "{'title': 'Facebook', 'text': 'Data is read fr... | \n", + "
| 3 | \n", + "Float32_ENN | \n", + "5 | \n", + "25 | \n", + "1.724660 | \n", + "{'title': 'Facebook', 'text': 'Data is read fr... | \n", + "