Datasets:
emrgnt-cmplxty
commited on
Commit
•
753ff5e
1
Parent(s):
f288e10
Update README.md
Browse files
README.md
CHANGED
@@ -5,7 +5,7 @@ size_categories:
|
|
5 |
- 1B<n<10B
|
6 |
task_categories:
|
7 |
- text-generation
|
8 |
-
pretty_name:
|
9 |
configs:
|
10 |
- config_name: default
|
11 |
data_files:
|
@@ -38,9 +38,9 @@ dataset_info:
|
|
38 |
|
39 |
### Getting Started
|
40 |
|
41 |
-
The
|
42 |
|
43 |
-
To access and utilize the
|
44 |
|
45 |
```python
|
46 |
from datasets import load_dataset
|
@@ -53,11 +53,11 @@ ds = load_dataset("SciPhi/OpenWebSearch-V1", data_files="arxiv/*", streaming=Tru
|
|
53 |
|
54 |
---
|
55 |
|
56 |
-
A full set of scripts to recreate the dataset from scratch can be found [here](https://github.com/SciPhi/
|
57 |
|
58 |
### Dataset Summary
|
59 |
|
60 |
-
|
61 |
|
62 |
|
63 |
| Dataset | Token Count |
|
@@ -89,13 +89,22 @@ The raw dataset structure is as follows:
|
|
89 |
}
|
90 |
```
|
91 |
|
92 |
-
The indexed dataset is structured as a qdrant database dump, each entry has meta data {"url", "vector"}.
|
93 |
|
94 |
## Dataset Creation
|
95 |
|
96 |
-
This dataset was created
|
97 |
|
98 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
99 |
|
100 |
### Source Data
|
101 |
|
|
|
5 |
- 1B<n<10B
|
6 |
task_categories:
|
7 |
- text-generation
|
8 |
+
pretty_name: AgentSearch-V1
|
9 |
configs:
|
10 |
- config_name: default
|
11 |
data_files:
|
|
|
38 |
|
39 |
### Getting Started
|
40 |
|
41 |
+
The AgentSearch-V1 dataset includes over one billion embeddings sourced from over 50 million high-quality documents. This extensive collection encompasses the majority of content from sources like Arxiv, Wikipedia, Project Gutenberg, and includes quality-filtered CC data.
|
42 |
|
43 |
+
To access and utilize the AgentSearch-V1 dataset, you can stream it via HuggingFace with the following Python code:
|
44 |
|
45 |
```python
|
46 |
from datasets import load_dataset
|
|
|
53 |
|
54 |
---
|
55 |
|
56 |
+
A full set of scripts to recreate the dataset from scratch can be found [here](https://github.com/SciPhi/agent-search).
|
57 |
|
58 |
### Dataset Summary
|
59 |
|
60 |
+
We take a similar approach to RedPajama-v1 and divide AgentSearch into a number of categories.
|
61 |
|
62 |
|
63 |
| Dataset | Token Count |
|
|
|
89 |
}
|
90 |
```
|
91 |
|
92 |
+
The indexed dataset can be downloaded directly and is structured as a qdrant database dump, each entry has meta data {"url", "vector"}. In addition, there is a corresponding sqlite dataset which contains the mapping from urls onto embeddings, text chunks, and other metadata.
|
93 |
|
94 |
## Dataset Creation
|
95 |
|
96 |
+
This dataset was created as a step towards making humanities most important knowledge locally searchable and LLM optimal. It was created by filtering, cleaning, and augmenting locally publicly available datasets.
|
97 |
|
98 |
+
To cite our work, please use the following:
|
99 |
+
|
100 |
+
```
|
101 |
+
@software{SciPhi2023AgentSearch,
|
102 |
+
author = {SciPhi},
|
103 |
+
title = {AgentSearch [ΨΦ]: A Comprehensive Agent-First Framework and Dataset for Webscale Search},
|
104 |
+
year = {2023},
|
105 |
+
url = {https://github.com/SciPhi-AI/agent-search}
|
106 |
+
}
|
107 |
+
```
|
108 |
|
109 |
### Source Data
|
110 |
|