README.md

---
dataset_name: "FryAI Pie – AI Knowledge Graph Dataset (Preview)"
pretty_name: "FryAI Pie Knowledge Graph (Preview)"
license: "apache-2.0"

tags:
  - ai
  - rag
  - knowledge-graph
  - embeddings
  - semantic-search
  - llm
  - machine-learning
  - dataset
  - education

size_categories:
  - n<1K

task_categories:
  - text-retrieval
  - feature-extraction
  - question-answering

task_ids:
  - document-retrieval
  - utterance-retrieval

homepage: "https://automatekc.gumroad.com/l/lghqtv"
repository: "https://huggingface.co/datasets/Lucasautomatekc/fryai-pie-knowledge-graph"
language:
  - en
annotations_creators:
  - machine-generated
---


# FryAI Pie – AI Knowledge Graph Dataset (Preview)
### Structured AI explanations, topic clusters, embeddings, and RAG‑ready data

This repository contains a **free preview** of the FryAI Pie Knowledge Graph Dataset — a structured, developer‑friendly dataset designed for:

- RAG pipelines  
- Semantic search  
- AI education tools  
- Topic clustering  
- Knowledge graph exploration  
- Embedding‑based retrieval  
- LLM fine‑tuning and evaluation  

The **full dataset** (500+ explanations, full graph, full embeddings, full RAG chunks) is available on Gumroad.

---

## 🔗 Full Dataset (Paid)
Get the complete dataset here:  
**https://automatekc.gumroad.com/l/lghqtv**

Includes:

- 500+ human‑readable explanations  
- Full topic/category/cluster metadata  
- Full knowledge graph (hundreds of nodes + edges)  
- Full RAG chunks  
- Full embeddings (26MB)  
- Clean JSON structure  
- Ready for production RAG systems  

---

## 🔗 API Access (RapidAPI)
Prefer API access instead of downloading files?  
Use the hosted API version:  
**https://rapidapi.com/lucasautomatekc/api/fryai-knowledge-graph-api**

---

# 📦 What’s Included in This Preview
This HuggingFace repo includes **small, safe sample files** that demonstrate the dataset structure without revealing the full paid content.

### Included (Free Preview)
- `sample_topics.json`  
- `sample_clusters.json`  
- `sample_embeddings.json`  
- `sample_graph.json`  
- `sample_rag_chunks.json`  

### Not Included (Paid Only)
- Full articles  
- Full topic/category metadata  
- Full clusters  
- Full knowledge graph  
- Full embeddings  
- Full RAG chunks  
- Full ZIP dataset  

---

# 🧠 Dataset Structure (Preview)

### Topics (sample)
```
{
  "ai-chips-memory-requirements": {
    "slug": "ai-chips-memory-requirements",
    "question": "How do AI chip memory requirements work?"
  },
  "agentic-ai": {
    "slug": "agentic-ai",
    "question": "How does agentic AI work?"
  }
}
```

---

### Clusters (sample)
```
{
  "artificial-intelligence": {
    "slug": "artificial-intelligence",
    "name": "Artificial Intelligence Topic Cluster",
    "articles": ["what-is-ai-liability"]
  }
}
```

---

### Knowledge Graph (sample)
```
[
  {
    "id": "topic:ai-chips-memory-requirements",
    "type": "topic",
    "slug": "ai-chips-memory-requirements",
    "edges": [
      { "target": "article:how-does-ai-chips-memory-requirements-work", "type": "explains" }
    ]
  }
]
```

---

### Embeddings (sample)
```
[
  {
    "id": "topic:ai-chips-memory-requirements",
    "type": "topic",
    "slug": "ai-chips-memory-requirements",
    "vector": [0.0313, -0.0237, 0.0314, -0.0207, 0.0215]
  }
]
```

---

### RAG Chunks (sample)
```
[
  {
    "id": "chunk:ai-chips-memory-requirements-1",
    "articleSlug": "how-does-ai-chips-memory-requirements-work",
    "chunkIndex": 0,
    "text": "Sample text explaining how AI chip memory requirements work...",
    "vector": [0.0123, -0.0044, 0.0099, 0.0021, -0.0033]
  }
]
```

---

# 🧪 Example Usage (Python)

### Load topics
```python
import json

with open("sample_topics.json") as f:
    topics = json.load(f)

print(topics.keys())
```

---

### Load embeddings
```python
import json
import numpy as np

with open("sample_embeddings.json") as f:
    data = json.load(f)

vec = np.array(data[0]["vector"])
print(vec.shape)
```

---

### Simple semantic search (preview)
```python
import numpy as np

def cosine_similarity(a, b):
    return np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b))

query = np.array([0.01, -0.02, 0.03, -0.01, 0.02])
scores = []

for item in data:
    sim = cosine_similarity(query, np.array(item["vector"]))
    scores.append((item["slug"], sim))

sorted(scores, key=lambda x: x[1], reverse=True)
```

---

### RAG Chunk Example
```python
with open("sample_rag_chunks.json") as f:
    chunks = json.load(f)

for c in chunks:
    print(c["articleSlug"], c["text"])
```

---

# 🏗️ Intended Use Cases
- RAG pipelines  
- Semantic search  
- AI education tools  
- Topic clustering  
- Knowledge graph exploration  
- Embedding‑based retrieval  
- LLM evaluation  
- Fine‑tuning datasets  
- AI explainability tools  

---

# 🏷️ Tags
```
ai
rag
knowledge-graph
embeddings
semantic-search
llm
machine-learning
dataset
education
```

---

# 📄 License
**Preview License:** apache-2.0  
Full dataset license included in the Gumroad package.

---

# 🙌 Support the Project
If you find this dataset useful, consider grabbing the full version or starring the repo.  
Your support helps expand the dataset with new topics, clusters, and explanations.
```