|
--- |
|
license: cc-by-sa-4.0 |
|
task_categories: |
|
- text-generation |
|
- conversational |
|
language: |
|
- en |
|
size_categories: |
|
- 10K<n<100K |
|
tags: |
|
- cartoons |
|
--- |
|
## <h1>Spongebob Transcripts Dataset π§½</h1> |
|
The Spongebob Transcripts Dataset is a collection of transcripts from the beloved animated television series, Spongebob Squarepants. This dataset includes information on each line of dialogue spoken by a character, including the character's name, their replica, and the episode ID. |
|
|
|
The number of characters in the dataset: **68** |
|
|
|
Total number of words in the dataset: **~56,000 words**, **~2800 rows**, **last updated to Season 1 Episode 30** |
|
|
|
## <h3>Dataset Overview π</h3> |
|
|
|
|Column | Description | |
|
|------------|-------------------------------------| |
|
|**Speaker** | The character speaking the dialogue.| |
|
|**Replica** | The line of dialogue spoken. | |
|
|**EP_ID** | The episode ID of the transcript. | |
|
|
|
## <h3>System Replicasπ</h3> |
|
The system replicas describe the actions and events that occur in each episode. These replicas are written in a specific format, using brackets to indicate actions and events. |
|
|
|
**<h5>Replica Format</h5>** |
|
|
|
`{system} : [The episode opens with a bubble transition, and we see a coral reef under the sea. The camera zooms to initiate parallax scrolling, which reveals the city of Bikini Bottom. It continues zooming to show a brown rock, a Moai head, and a pineapple, which each contain inhabitants.]` |
|
|
|
## <h3>Sample Data π¬</h3> |
|
|
|
|Speaker |Replica |EP_ID | |
|
|---------------|--------------------------------------------------------------------------------------------------------------------------------------------------|-------| |
|
|**Spongebob** | I just met this girl. She wears a hat full of... air. |s1e3_22| |
|
|**Patrick** | Do you mean she puts on "airs"? |s1e3_23| |
|
|**Spongebob** | I guess so. |s1e3_24| |
|
|**Patrick** | That's just fancy talk. If you wanna be fancy, hold your pinky up like this. The higher you hold it, the fancier you are. |s1e3_25| |
|
|
|
## <h3>π Interactions with Dataset</h3> |
|
|
|
**<h5>Using Pandas to filter rows</h5>** |
|
|
|
1. To find all rows with a specific ep_id, you can use the following code: |
|
|
|
``` |
|
import pandas as pd |
|
#Read the CSV file into a Pandas DataFrame |
|
df = pd.read_csv('dataset.csv') |
|
#Define the ep_id you want to filter by |
|
ep_id = 's1e2' |
|
#Filter the DataFrame to get rows with an ep_id that starts with the defined ep_id |
|
filtered_df = df[df['ep_id'].str.startswith(ep_id)] |
|
#Print the filtered DataFrame |
|
print(filtered_df) |
|
``` |
|
|
|
2. To find rows where a specific character says a specific word or phrase, you can use the following code: |
|
|
|
``` |
|
#Filter the DataFrame to get rows where a specific character says a specific word or phrase |
|
speaker = 'SpongeBob' |
|
word_or_phrase = 'jellyfish' |
|
filtered_df = df[df['speaker'] == speaker] |
|
filtered_df = filtered_df[filtered_df['replica'].str.contains(word_or_phrase)] |
|
#Print the filtered DataFrame |
|
print(filtered_df) |
|
``` |
|
|
|
You can replace `SpongeBob` and `jellyfish` with any other speaker and word/phrase that you want to filter by. |
|
|
|
## <h3>Data Sources π</h3> |
|
The transcripts were sourced *Encyclopedia SpongeBobia*. |
|
|
|
## <h3>Potential Uses π§</h3> |
|
This Dataset could be used for a variety of natural language processing (NLP) tasks, including dialogue generation. It could also be used for educational purposes, such as studying the language and communication styles of different characters. |