README.md

metadata

license: mit
task_categories:
  - other
tags:
  - blockchain
  - proof-of-useful-work
  - np-complete
  - computational-complexity
  - consensus
  - cryptography
  - distributed-systems
  - research
pretty_name: COINjecture NP Solutions v2
size_categories:
  - 1K<n<10K
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/*.jsonl

🔬 COINjecture NP Solutions Dataset v2

Institutional-Grade Blockchain Research Data

A comprehensive, real-time dataset of NP-complete problem solutions generated through Proof-of-Useful-Work (PoUW) blockchain consensus

Overview • Data Schema • Metrics Categories • Usage • Citation

---

📋 Overview

This dataset contains institutional-grade metrics from the COINjecture Network B blockchain, which implements a novel Proof-of-Useful-Work (PoUW) consensus mechanism. Unlike traditional Proof-of-Work systems that compute arbitrary hashes, COINjecture miners solve genuine NP-complete computational problems, producing verifiable solutions with real-world applicability.

Key Characteristics

Property	Value
Network	COINjecture Network B (Fresh Genesis)
Genesis Hash	4a80254b4a48e867b57399469b0a1fbaba8848e8ac738587b55ebf6e6b8c4b23
Data Version	v3.0 (Institutional Grade)
Problem Types	SAT, SubsetSum, TSP
Update Frequency	Every 10 blocks (10 seconds)
Metrics Per Record	54+ fields
Format	JSON Lines (.jsonl)

Research Applications

• Computational Complexity: Empirical analysis of NP-complete problem hardness
• Algorithm Performance: Solve/verify time distributions across problem types
• Distributed Systems: Consensus metrics and network propagation analysis
• Energy Research: Computational efficiency and resource utilization studies
• Cryptographic Analysis: Hash function behavior and difficulty adjustment

---

📊 Data Schema

Each record represents a block in the COINjecture blockchain containing a solved NP-complete problem instance.

Core Fields

Field	Type	Description
block_height	uint64	Sequential block number in the canonical chain
block_hash	string	SHA-256 hash of the block header (hex-encoded)
prev_block_hash	string	Hash of the parent block (enables chain traversal)
timestamp	string	ISO 8601 timestamp of block creation
problem_type	string	NP-complete problem class: SAT, SubsetSum, or TSP

Problem Instance Fields

Field	Type	Description
problem_instance	object	Serialized problem definition (varies by type)
solution	object	Verified solution to the problem instance
problem_size	uint32	Instance complexity metric (variables, nodes, etc.)
is_satisfiable	boolean	For SAT: whether a satisfying assignment exists

---

📈 Metrics Categories

⏱️ Timing Metrics (Microsecond Precision)

High-resolution timing data for performance analysis:

Field	Type	Unit	Description
solve_time_us	uint64	μs	Time to find the solution
verify_time_us	uint64	μs	Time to verify solution correctness
block_time_seconds	float64	s	Total block production time
mining_attempts	uint64	count	Hash attempts before valid block found

💾 Memory Metrics

Resource utilization during computation:

Field	Type	Unit	Description
solve_memory_bytes	uint64	bytes	Peak memory during solve phase
verify_memory_bytes	uint64	bytes	Peak memory during verification
peak_memory_bytes	uint64	bytes	Maximum memory allocation

🌐 Network Metrics

Distributed system behavior:

Field	Type	Unit	Description
peer_count	uint32	count	Connected peers at block time
propagation_time_ms	uint64	ms	Block propagation latency
sync_lag_blocks	int64	blocks	Distance from network tip

⛏️ Mining Metrics

Consensus and difficulty data:

Field	Type	Description
difficulty_target	string	Current difficulty target (hex)
nonce	uint64	Winning nonce value
hash_rate_estimate	float64	Estimated network hash rate (H/s)
mined_locally	boolean	Whether this node mined the block

🔗 Chain Metrics

Blockchain state information:

Field	Type	Description
chain_work	string	Cumulative proof-of-work score
transaction_count	uint32	Transactions in block
block_size_bytes	uint64	Serialized block size

💰 Economic Metrics

Token economics data:

Field	Type	Unit	Description
block_reward	uint64	tokens	Mining reward for this block
total_fees	uint64	tokens	Transaction fees collected

🖥️ Hardware Context

Node environment information for reproducibility:

Field	Type	Description
cpu_model	string	Processor model identifier
cpu_cores	uint32	Physical CPU cores
cpu_threads	uint32	Logical CPU threads
ram_total_bytes	uint64	Total system RAM
os_info	string	Operating system details

🏷️ Provenance Metadata

Data lineage and quality indicators:

Field	Type	Description
node_version	string	Software version that produced this record
node_id	string	Unique node identifier (anonymized)
data_version	string	Schema version (currently v3.0)
measurement_confidence	float64	Data quality score (0.0-1.0)

---

🔬 Problem Types

SAT (Boolean Satisfiability)

The canonical NP-complete problem. Given a Boolean formula in CNF, find a satisfying assignment or prove none exists.

{
  "problem_type": "SAT",
  "problem_instance": {
    "num_variables": 50,
    "num_clauses": 215,
    "clauses": [[1, -3, 5], [-2, 4], ...]
  },
  "solution": {
    "satisfiable": true,
    "assignment": [true, false, true, ...]
  }
}

SubsetSum

Given a set of integers and a target sum, find a subset that sums to the target.

{
  "problem_type": "SubsetSum",
  "problem_instance": {
    "set": [3, 7, 1, 8, -2, 4],
    "target": 12
  },
  "solution": {
    "subset_indices": [1, 3, 5]
  }
}

TSP (Traveling Salesman Problem)

Find the shortest Hamiltonian cycle through all vertices in a weighted graph.

{
  "problem_type": "TSP",
  "problem_instance": {
    "num_cities": 20,
    "distances": [[0, 10, 15], [10, 0, 20], ...]
  },
  "solution": {
    "tour": [0, 3, 1, 4, 2, 0],
    "total_distance": 97
  }
}

---

📖 Usage

Loading with Hugging Face Datasets

from datasets import load_dataset

# Load the complete dataset
dataset = load_dataset("COINjecture/NP_Solutions_v2")

# Access records
for record in dataset["train"]:
    print(f"Block {record['block_height']}: {record['problem_type']}")
    print(f"  Solve time: {record['solve_time_us']}μs")
    print(f"  CPU: {record['cpu_model']}")

Loading Raw JSONL

import json
from pathlib import Path

records = []
for jsonl_file in Path("data").glob("*.jsonl"):
    with open(jsonl_file) as f:
        for line in f:
            records.append(json.loads(line))

print(f"Loaded {len(records)} records")

Filtering by Problem Type

sat_problems = dataset["train"].filter(
    lambda x: x["problem_type"] == "SAT"
)
print(f"SAT problems: {len(sat_problems)}")

Performance Analysis Example

import pandas as pd

# Convert to DataFrame for analysis
df = pd.DataFrame(dataset["train"])

# Analyze solve times by problem type
stats = df.groupby("problem_type")["solve_time_us"].agg(["mean", "std", "min", "max"])
print(stats)

# Hardware comparison
hardware_stats = df.groupby("cpu_model")["solve_time_us"].mean()
print(hardware_stats)

---

📊 Data Quality

Verification Standards

All data in this dataset meets the following quality criteria:

Standard	Description
Cryptographic Integrity	Every block hash is verified against the chain
Solution Validity	All NP-complete solutions are independently verified
Timing Accuracy	Microsecond-precision timestamps from monotonic clocks
Hardware Attribution	Full system context for reproducibility
Chain Continuity	prev_block_hash enables complete chain reconstruction

Data Versioning

Version	Release	Changes
v3.0	Nov 2024	Institutional-grade: 54+ fields, hardware context, chain linkage
v2.0	Oct 2024	Added timing metrics, energy estimates
v1.0	Sep 2024	Initial release: basic problem/solution data

---

🔄 Update Frequency

This dataset receives real-time updates approximately every 10 blocks (~10 seconds of blockchain time). New JSONL files are appended as blocks are mined on the COINjecture Network B.

Data Pipeline Architecture

⛓️ CONSENSUS LAYER 🌱 Genesis (Block 0) │ ┌───────┼───────┐ ▼ ▼ ▼ 🧮SAT 📊Sum 🗺️TSP │ │ │ └───────┴───────┘ │ ▼	🌐 P2P NETWORK 📡 Node 1 ◄────► 📡 Node 2 │ │ └────────┬────────┘ ▼ 💬 Gossipsub │ ▼
📈 METRICS ENGINE ⏱️Timing 💾Memory 🖥️Hardware 🌐Network │ │ │ │ └─────────┴──────────┴──────────┘ │ 54+ metrics/block │ ▼	🎯 DATA OUTPUT 📦 Buffer (10 blocks) │ every ~10 seconds │ ▼ 🤗 HuggingFace v2 │ ▼ 🔌 Datasets API

🔬 RESEARCH APPLICATIONS

🤖 Machine Learning	📊 Performance Analysis	🔐 Cryptography Research
Training data	Solve time analysis	Hash function studies
Benchmarking	Hardware comparisons	Difficulty research

---

Data flows from NP-complete problem solving → metrics collection → real-time research availability

---

📜 Citation

If you use this dataset in your research, please cite:

@dataset{coinjecture_np_solutions_v2,
  title={COINjecture NP Solutions Dataset v2},
  author={{COINjecture Network Contributors}},
  year={2024},
  publisher={Hugging Face},
  url={https://huggingface.co/datasets/COINjecture/NP_Solutions_v2},
  note={Institutional-grade blockchain research data from Proof-of-Useful-Work consensus}
}

---

📄 License

This dataset is released under the MIT License. You are free to use, modify, and distribute this data for any purpose, including commercial applications.

---

🔗 Related Resources

Resource	Link
Legacy Dataset	COINjecture/NP_Solutions
Source Code	GitHub
Network Explorer	Coming Soon
Technical Whitepaper	Coming Soon

---

🤝 Contributing

We welcome contributions to improve data quality and documentation. Please open an issue or pull request on our GitHub repository.

---

Built with 💎 by the COINjecture Network

Transforming computational waste into useful work