mirror of
https://github.com/wassname/Open-Assistant.git
synced 2026-07-18 12:10:34 +08:00
Merge branch 'main' of github.com:LAION-AI/Open-Chat-GPT into main
This commit is contained in:
Andreas Köpf
@@ -0,0 +1,42 @@
|
||||
## Dataset Summary
|
||||
|
||||
The dataset was created using
|
||||
[Cornell Movies Dialog Corpus](https://www.cs.cornell.edu/~cristian/Cornell_Movie-Dialogs_Corpus.html)
|
||||
which contains a large metadata-rich collection of fictional conversations
|
||||
extracted from raw movie scripts. Dialogs and meta-data from the underlying
|
||||
Corpus were used to design a dataset that can be used to InstructGPT based
|
||||
models to learn movie scripts.
|
||||
|
||||
Example :
|
||||
|
||||
```
|
||||
User: Assume RICK and ALICE are characters from a fantasy-horror movie, continue the conversation between them
|
||||
RICK: I heard you screaming. Was it a bad one?
|
||||
ALICE: It was bad.
|
||||
RICK: Doesn't the dream master work for you anymore?
|
||||
Assistant: Sure
|
||||
ALICE: I can't find him.
|
||||
RICK: Hey, since when do you play Thomas Edison? This looks like Sheila's.
|
||||
ALICE: It is...was. It's a zapper, it might help me stay awake.
|
||||
RICK: Yeah, or turn you into toast.
|
||||
```
|
||||
|
||||
## Usage
|
||||
|
||||
```python
|
||||
|
||||
from datasets import load_dataset
|
||||
dataset = load_dataset("shahules786/OA-cornell-movies-dialog")
|
||||
```
|
||||
|
||||
## Citations
|
||||
|
||||
```
|
||||
@InProceedings{Danescu-Niculescu-Mizil+Lee:11a,
|
||||
author={Cristian Danescu-Niculescu-Mizil and Lillian Lee},
|
||||
title={Chameleons in imagined conversations:
|
||||
A new approach to understanding coordination of linguistic style in dialogs.},
|
||||
booktitle={Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics, ACL 2011},
|
||||
year={2011}
|
||||
}
|
||||
```
|
||||
@@ -0,0 +1,649 @@
|
||||
{
|
||||
"cells": [
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "ec8d6189",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"[](https://colab.research.google.com/github/LAION-AI/Open-Assistant/blob/main/notebooks/data-augmentation/movie-dialogs/convert-to-instruction-format.ipynb)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "493f2529",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Imports"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 1,
|
||||
"id": "65a47f83",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"from datasets import load_dataset\n",
|
||||
"import numpy as np\n",
|
||||
"import json\n",
|
||||
"from tqdm import tqdm\n",
|
||||
"\n",
|
||||
"IMDB = 7.0"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "480440f6",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Dialog templates\n",
|
||||
"Templates for converting dialogs to prompts"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 2,
|
||||
"id": "fcfedd7f",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"DIALOG_TEMPLATES = {\n",
|
||||
" ### template for 4+ line dialogs\n",
|
||||
" \"four_more_lines\": [\n",
|
||||
" \"\"\"\n",
|
||||
"Here's a {template} between {char1} and {char2} in a scene from a {genre} movie\n",
|
||||
" {dialogue1}\n",
|
||||
"User : Can you continue the {template}\n",
|
||||
"Assistant : Sure, the next dialogue for this scene could be\n",
|
||||
" {dialogue2}\n",
|
||||
" \"\"\",\n",
|
||||
" \"\"\"\n",
|
||||
" {dialogue1}\n",
|
||||
"User : Can you provide more dialog assuming {genre} movie\n",
|
||||
" {dialogue2}\n",
|
||||
"\"\"\",\n",
|
||||
" \"\"\"\n",
|
||||
"I'm trying to complete the dialog for my characters {char1} and {char2}. Here's the {template}, Please help me complete it\n",
|
||||
" {dialogue1}\n",
|
||||
"Assistant : Sure\n",
|
||||
" {dialogue2}\n",
|
||||
"\"\"\",\n",
|
||||
" \"\"\"\n",
|
||||
"User : Assume {char1} and {char2} are characters from a {genre} movie, continue the conversation between them\n",
|
||||
" {dialogue1}\n",
|
||||
"Assistant : Sure\n",
|
||||
" {dialogue2}\n",
|
||||
"\"\"\",\n",
|
||||
" ],\n",
|
||||
" ## template for 4 line dialogs\n",
|
||||
" \"four_lines\": [\n",
|
||||
" \"\"\"\n",
|
||||
" {dialogue1}\n",
|
||||
"User : provide a response assuming you're {char2}\n",
|
||||
"Assistant : Sure\n",
|
||||
" {dialogue2}\n",
|
||||
"\"\"\",\n",
|
||||
" \"\"\"\n",
|
||||
" {dialogue1}\n",
|
||||
"User : respond as {char2} to complete the conversation\n",
|
||||
"Assistant : Sure\n",
|
||||
" {dialogue2}\n",
|
||||
"\"\"\",\n",
|
||||
" ],\n",
|
||||
"}"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "2047056e",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"- Download Cornell-movies dialog dataset"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 3,
|
||||
"id": "e413a053",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"! wget wget https://zissou.infosci.cornell.edu/convokit/datasets/movie-corpus/movie-corpus.zip\n",
|
||||
"! unzip movie-corpus.zip -d ./Data/"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "5e2aab0d",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Code"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 4,
|
||||
"id": "25cae04e",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"def get_movie_dialogs():\n",
|
||||
"\n",
|
||||
" with open(\"./Data/movie-corpus/utterances.jsonl\", \"r\") as json_file:\n",
|
||||
" conversations = list(json_file)\n",
|
||||
" speakers = json.load(open(\"./Data/movie-corpus/speakers.json\"))\n",
|
||||
" movie_dialog_dict = {}\n",
|
||||
" for dialog in tqdm(conversations):\n",
|
||||
" dialog = eval(dialog.replace(\"null\", \"None\"))\n",
|
||||
" movie_dialog_dict[dialog[\"id\"]] = {\n",
|
||||
" \"characterName\": speakers[dialog[\"speaker\"]][\"meta\"][\"character_name\"],\n",
|
||||
" \"text\": dialog[\"text\"],\n",
|
||||
" \"characterID\": dialog[\"speaker\"],\n",
|
||||
" }\n",
|
||||
"\n",
|
||||
" return movie_dialog_dict"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 5,
|
||||
"id": "3b949bc7",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"def get_dialogs(dialog_dict, start, end):\n",
|
||||
"\n",
|
||||
" dialog_list = []\n",
|
||||
" for idx in range(start, end + 1):\n",
|
||||
" dialog_list.append(dialog_dict[f\"L{idx}\"][\"characterName\"] + \": \" + dialog_dict[f\"L{idx}\"][\"text\"])\n",
|
||||
" num_lines = len(dialog_list)\n",
|
||||
"\n",
|
||||
" assert num_lines >= 1, \"Number of lines should be greater than one\"\n",
|
||||
"\n",
|
||||
" if num_lines < 6:\n",
|
||||
" dialog1 = \"\\n \".join(dialog_list[:-1])\n",
|
||||
" dialog2 = dialog_list[-1]\n",
|
||||
" else:\n",
|
||||
" dialog_len = np.random.randint(3, (num_lines // 2) + 1)\n",
|
||||
" dialog1 = \"\\n \".join(dialog_list[:dialog_len])\n",
|
||||
" dialog2 = \"\\n \".join(dialog_list[dialog_len:])\n",
|
||||
"\n",
|
||||
" return dialog1, dialog2\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"def choose_prompt(num_lines):\n",
|
||||
"\n",
|
||||
" assert num_lines >= 1, \"Number of lines should be greater than one\"\n",
|
||||
"\n",
|
||||
" if num_lines < 6:\n",
|
||||
" prompt = np.random.choice(DIALOG_TEMPLATES[\"four_lines\"])\n",
|
||||
"\n",
|
||||
" else:\n",
|
||||
" prompt = np.random.choice(DIALOG_TEMPLATES[\"four_more_lines\"])\n",
|
||||
"\n",
|
||||
" return prompt\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"def convert_to_prompts(dataset, movie_dialog_dict, output_dir=\".\", split=\"train\"):\n",
|
||||
"\n",
|
||||
" with open(f\"{output_dir}/{split}.jsonl\", \"w\", encoding=\"utf8\") as output:\n",
|
||||
"\n",
|
||||
" i = 0\n",
|
||||
" while i < len(dataset[\"train\"]):\n",
|
||||
"\n",
|
||||
" data = dataset[split][i]\n",
|
||||
" if float(data[\"movieIMDBRating\"].strip()) >= IMDB:\n",
|
||||
" max_lines = np.random.randint(7, 12)\n",
|
||||
" lineids = [int(lineid[1:]) for lineid in data[\"utterance\"][\"LineID\"]]\n",
|
||||
" num_lines = len(lineids)\n",
|
||||
" char_ids = sorted([data[\"characterID1\"].strip(), data[\"characterID1\"].strip()])\n",
|
||||
" while num_lines < max_lines:\n",
|
||||
" i += 1\n",
|
||||
" data = dataset[split][i]\n",
|
||||
" char_id_new = sorted([data[\"characterID1\"].strip(), data[\"characterID1\"].strip()])\n",
|
||||
" ## make sure that characters are the same\n",
|
||||
" if char_id_new == char_ids:\n",
|
||||
" lineids_new = [int(lineid[1:]) for lineid in data[\"utterance\"][\"LineID\"]]\n",
|
||||
" if lineids_new[0] == (lineids[-1] + 1): ##ensure continuety\n",
|
||||
" lineids.extend(lineids_new)\n",
|
||||
" else:\n",
|
||||
" break\n",
|
||||
" else:\n",
|
||||
" break\n",
|
||||
" num_lines = len(lineids)\n",
|
||||
"\n",
|
||||
" genre = \"-\".join(data[\"movieGenres\"][:2])\n",
|
||||
" template = np.random.choice([\"dialog\", \"script\", \"play\"])\n",
|
||||
" char1 = movie_dialog_dict[f\"L{lineids[0]}\"][\"characterName\"]\n",
|
||||
"\n",
|
||||
" if num_lines < 6:\n",
|
||||
" if num_lines % 2 == 0:\n",
|
||||
" char2 = movie_dialog_dict[f\"L{lineids[1]}\"][\"characterName\"]\n",
|
||||
" else:\n",
|
||||
" char2 = char1\n",
|
||||
" else:\n",
|
||||
" char2 = movie_dialog_dict[f\"L{lineids[1]}\"][\"characterName\"]\n",
|
||||
"\n",
|
||||
" dialogue1, dialogue2 = get_dialogs(movie_dialog_dict, lineids[0], lineids[-1])\n",
|
||||
" prompt = choose_prompt(num_lines)\n",
|
||||
"\n",
|
||||
" prompt = prompt.format(\n",
|
||||
" char1=char1, char2=char2, dialogue1=dialogue1, dialogue2=dialogue2, genre=genre, template=template\n",
|
||||
" )\n",
|
||||
" output.write(f\"{json.dumps({'conversation': prompt})}\\n\")\n",
|
||||
" i += 1"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 6,
|
||||
"id": "3ff310fd",
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"100%|███████████████████████████████| 304713/304713 [00:54<00:00, 5628.12it/s]\n",
|
||||
"Found cached dataset cornell_movie_dialog (/home/shahul/.cache/huggingface/datasets/cornell_movie_dialog/default/0.1.0/b67b3433cf894b551cddcd82efdff0826f39b39a11d5c149e746a546a8dc85f3)\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "6fee977c69a3403ebe77c4669fcb25d7",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
" 0%| | 0/1 [00:00<?, ?it/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"movie_dialog_dict = get_movie_dialogs()\n",
|
||||
"dataset = load_dataset(\"cornell_movie_dialog\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 7,
|
||||
"id": "8567ca12",
|
||||
"metadata": {
|
||||
"scrolled": true
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"convert_to_prompts(dataset, movie_dialog_dict)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "02315e91",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Upload as HF Dataset"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 13,
|
||||
"id": "dd4c05c1",
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Using custom data configuration default-315650f1b3e45d2e\n",
|
||||
"Found cached dataset json (/home/shahul/.cache/huggingface/datasets/json/default-315650f1b3e45d2e/0.0.0/0f7e3662623656454fcd2b650f34e886a7db4b9104504885bd462096cc7a9f51)\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "b79b4c273dc44735badf9bff51ade320",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
" 0%| | 0/1 [00:00<?, ?it/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"dataset_ = load_dataset(\"json\", data_files={\"train\": \"./train.jsonl\"})"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 23,
|
||||
"id": "d67fa1f9",
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Pushing split train to the Hub.\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "7af96153f0cf45b488d14515a7529ae7",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
"Pushing dataset shards to the dataset hub: 0%| | 0/1 [00:00<?, ?it/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "0baf18a08d0e48fa90484f4cd931baa2",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
"Creating parquet from Arrow format: 0%| | 0/21 [00:00<?, ?ba/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "cadf382a65274fdfbaea1820c04b146e",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
"Upload 1 LFS files: 0%| | 0/1 [00:00<?, ?it/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "b10d1b17f22c4da4b0395b57d0099cde",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
"Deleting unused files from dataset repository: 0%| | 0/1 [00:00<?, ?it/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"dataset_.push_to_hub(\"shahules786/OA-cornell-movies-dialog\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "4b4dd2df",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Load Dataset from HF"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 24,
|
||||
"id": "f64c9b2e",
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "49b4ce66a511425ba2886eeb73ba0664",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
"Downloading readme: 0%| | 0.00/1.54k [00:00<?, ?B/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
},
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Using custom data configuration shahules786--OA-cornell-movies-dialog-7b3f29da4e713888\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Downloading and preparing dataset None/None to /home/shahul/.cache/huggingface/datasets/shahules786___parquet/shahules786--OA-cornell-movies-dialog-7b3f29da4e713888/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec...\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "bb19919febc74f1ab5ecfcfd54d9167b",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
"Downloading data files: 0%| | 0/1 [00:00<?, ?it/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "df11363157cf439e8e1215386d764a5e",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
"Downloading data: 0%| | 0.00/4.86M [00:00<?, ?B/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "99abed3078764c11a08d734faf405b75",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
"Extracting data files: 0%| | 0/1 [00:00<?, ?it/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "fcda49e14a9e4e2385c7a696100fb7a0",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
"Generating train split: 0%| | 0/20959 [00:00<?, ? examples/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
},
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Dataset parquet downloaded and prepared to /home/shahul/.cache/huggingface/datasets/shahules786___parquet/shahules786--OA-cornell-movies-dialog-7b3f29da4e713888/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec. Subsequent calls will reuse this data.\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "9f4e3a03fed94c1b8978309ec1605bb8",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
" 0%| | 0/1 [00:00<?, ?it/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"dataset_ = load_dataset(\"shahules786/OA-cornell-movies-dialog\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 25,
|
||||
"id": "1234f33f",
|
||||
"metadata": {
|
||||
"scrolled": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"##\n",
|
||||
"\n",
|
||||
" BOWMAN: I didn't do that Frank. I took particular care not to freeze them.\n",
|
||||
" POOLE: I guess you don't know your own strength, old boy.\n",
|
||||
" BOWMAN: I guess not.\n",
|
||||
" POOLE: I think I'll have to go out and burn them off.\n",
|
||||
"User : respond as BOWMAN to complete the conversation\n",
|
||||
"Assistant : Sure\n",
|
||||
" BOWMAN: Roger.\n",
|
||||
"\n",
|
||||
"##\n",
|
||||
"\n",
|
||||
" HAL: Sorry to interrupt the festivities, Dave, but I think we've got a problem.\n",
|
||||
" BOWMAN: What is it, Hal?\n",
|
||||
"User : respond as HAL to complete the conversation\n",
|
||||
"Assistant : Sure\n",
|
||||
" HAL: MY F.P.C. shows an impending failure of the antenna orientation unit.\n",
|
||||
"\n",
|
||||
"##\n",
|
||||
"\n",
|
||||
"I'm trying to complete the dialog for my characters BOWMAN and HAL. Here's the script, Please help me complete it\n",
|
||||
" BOWMAN: Not now, Hal, I'd like to talk to you about something.\n",
|
||||
" HAL: Sure, Dave, what's up?\n",
|
||||
" BOWMAN: You know that we checked the two AO-units that you reported in imminent failure condition?\n",
|
||||
"Assistant : Sure\n",
|
||||
" HAL: Yes, I know.\n",
|
||||
" BOWMAN: You probably also know that we found them okay.\n",
|
||||
" HAL: Yes, I know that. But I can assure you that they were about to fail.\n",
|
||||
"\n",
|
||||
"##\n",
|
||||
"\n",
|
||||
"Here's a play between HAL and BOWMAN in a scene from a adventure-mystery movie\n",
|
||||
" HAL: Naturally, Dave, I'm not pleased that the AO-unit has failed, but I hope at least this has restored your confidence in my integrity and reliability. I certainly wouldn't want to be disconnected, even temporarily, as I have never been disconnected in my entire service history.\n",
|
||||
" BOWMAN: I'm sorry about the misunderstanding, Hal.\n",
|
||||
" HAL: Well, don't worry about it.\n",
|
||||
"User : Can you continue the play\n",
|
||||
"Assistant : Sure, the next dialogue for this scene could be\n",
|
||||
" BOWMAN: And don't you worry about it.\n",
|
||||
" HAL: Is your confidence in me fully restored?\n",
|
||||
" BOWMAN: Yes, it is, Hal.\n",
|
||||
" HAL: Well, that's a relief. You know I have the greatest enthusiasm possible for the mission.\n",
|
||||
" \n",
|
||||
"##\n",
|
||||
"\n",
|
||||
" HAL: I suppose it's because you've been under a lot of stress, but have you forgotten that they're not supposed to be revived for another three months.\n",
|
||||
" BOWMAN: The antenna has to be replaced.\n",
|
||||
" HAL: Repairing the antenna is a pretty dangerous operation.\n",
|
||||
" BOWMAN: It doesn't have to be, Hal. It's more dangerous to be out of touch with Earth. Let me have manual control, please.\n",
|
||||
"User : respond as HAL to complete the conversation\n",
|
||||
"Assistant : Sure\n",
|
||||
" HAL: I don't really agree with you, Dave. My on-board memory store is more than capable of handling all the mission requirements.\n",
|
||||
"\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"for i in range(10, 15):\n",
|
||||
" print(\"##\")\n",
|
||||
" print(dataset_[\"train\"][i][\"conversation\"])"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 22,
|
||||
"id": "806f3ef2",
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"Dataset({\n",
|
||||
" features: ['conversation'],\n",
|
||||
" num_rows: 20959\n",
|
||||
"})"
|
||||
]
|
||||
},
|
||||
"execution_count": 22,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"dataset_[\"train\"]"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "10506ff9",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": []
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
"kernelspec": {
|
||||
"display_name": "OpenAssistant",
|
||||
"language": "python",
|
||||
"name": "openassistant"
|
||||
},
|
||||
"language_info": {
|
||||
"codemirror_mode": {
|
||||
"name": "ipython",
|
||||
"version": 3
|
||||
},
|
||||
"file_extension": ".py",
|
||||
"mimetype": "text/x-python",
|
||||
"name": "python",
|
||||
"nbconvert_exporter": "python",
|
||||
"pygments_lexer": "ipython3",
|
||||
"version": "3.10.8"
|
||||
}
|
||||
},
|
||||
"nbformat": 4,
|
||||
"nbformat_minor": 5
|
||||
}
|
||||
Reference in New Issue
Block a user