@arizeai/phoenix-mcp

{ "cells": [ { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from datasets import load_dataset" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "https://huggingface.co/datasets/nvidia/ChatRAG-Bench/viewer/hybridial" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "name = \"hybridial\"\n", "df = load_dataset(\"nvidia/ChatRAG-Bench\", name)[\"test\"].to_pandas()\n", "print(df.messages.apply(len).max())\n", "print(df.ctxs.apply(len).max())\n", "df.loc[\n", " (df.messages.apply(len) >= df.messages.apply(len).max() // 2)\n", " & (df.ctxs.apply(len) >= df.ctxs.apply(len).max() // 2)\n", "].sample(10, random_state=42).to_csv(f\"{name}_samples.csv.gz\", index=False)\n", "df.sample(10)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "https://huggingface.co/datasets/nvidia/ChatRAG-Bench/viewer/sqa" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "name = \"sqa\"\n", "df = load_dataset(\"nvidia/ChatRAG-Bench\", name)[\"test\"].to_pandas()\n", "print(df.messages.apply(len).max())\n", "df.loc[df.messages.apply(len) >= df.messages.apply(len).max() // 2].sample(\n", " 100, random_state=42\n", ").to_csv(f\"{name}_samples.csv.gz\", index=False)\n", "df.sample(10)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "https://huggingface.co/datasets/nvidia/ChatRAG-Bench/viewer/doqa_cooking" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "name = \"doqa_cooking\"\n", "df = load_dataset(\"nvidia/ChatRAG-Bench\", name)[\"test\"].to_pandas()\n", "print(df.messages.apply(len).max())\n", "df.loc[df.messages.apply(len) >= df.messages.apply(len).max() // 2].sample(\n", " 100, random_state=42\n", ").to_csv(f\"{name}_samples.csv.gz\", index=False)\n", "df.sample(10)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "https://huggingface.co/datasets/nvidia/ChatRAG-Bench/viewer/doqa_travel" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "name = \"doqa_travel\"\n", "df = load_dataset(\"nvidia/ChatRAG-Bench\", name)[\"test\"].to_pandas()\n", "print(df.messages.apply(len).max())\n", "df.loc[df.messages.apply(len) >= df.messages.apply(len).max() // 2].sample(\n", " 100, random_state=42\n", ").to_csv(f\"{name}_samples.csv.gz\", index=False)\n", "df.sample(10)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "name = \"doqa_movies\"\n", "df = load_dataset(\"nvidia/ChatRAG-Bench\", name)[\"test\"].to_pandas()\n", "print(df.messages.apply(len).max())\n", "df.loc[df.messages.apply(len) >= df.messages.apply(len).max() // 2].sample(\n", " 100, random_state=42\n", ").to_csv(f\"{name}_samples.csv.gz\", index=False)\n", "df.sample(10)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "https://huggingface.co/datasets/nvidia/ChatQA-Training-Data/viewer/synthetic_convqa" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "name = \"synthetic_convqa\"\n", "df = load_dataset(\"nvidia/ChatQA-Training-Data\", name)[\"train\"].to_pandas()\n", "print(df.messages.apply(len).max())\n", "df.loc[df.messages.apply(len) >= df.messages.apply(len).max() // 2].sample(\n", " 100, random_state=42\n", ").to_csv(f\"{name}_samples.csv.gz\", index=False)\n", "df.sample(10)" ] } ], "metadata": { "language_info": { "name": "python" } }, "nbformat": 4, "nbformat_minor": 2 }