metadata
dataset_info:
features:
- name: ticker
dtype: string
- name: prompt
dtype: string
- name: text
dtype: string
- name: url
dtype: string
- name: result_1
dtype: string
- name: result_1_bin
dtype: int64
- name: relevance
dtype: string
- name: token_count
dtype: int64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 2284852
num_examples: 3600
- name: val
num_bytes: 128718
num_examples: 200
- name: test
num_bytes: 128847
num_examples: 200
download_size: 965881
dataset_size: 2542417
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: val
path: data/val-*
- split: test
path: data/test-*
INFO
A random selection of tweets for the purpose of fine-tuning a LLM to predict stock price movement the day after the tweets.
Source koen430/preprocessed_stock_twitter
More info will follow soon