Gapless Network Data

Ethereum blockchain network metrics for financial time series forecasting.

Quick Start

import gapless_network_data as gmd

# Fetch latest blocks (recommended for live trading)
df = gmd.fetch_blocks(limit=1000)

# Compute block utilization (alpha feature #2)
df['utilization'] = df['gas_used'] / df['gas_limit']

# Date range query (inclusive [start, end])
# Returns all blocks from Jan 1 through Jan 31 (both dates included)
df = gmd.fetch_blocks(start='2024-01-01', end='2024-01-31')

# Start-only: all blocks from Jan 1 to most recent
df = gmd.fetch_blocks(start='2024-01-01')

# End-only: all blocks from genesis to Jan 31
df = gmd.fetch_blocks(end='2024-01-31')

# Same-day query (returns all blocks on March 13)
df = gmd.fetch_blocks(start='2024-03-13', end='2024-03-13')

# Second-precision query (explicit times supported)
df = gmd.fetch_blocks(start='2024-03-13 12:00:00', end='2024-03-13 12:05:00')

Installation

pip install gapless-network-data

AI Agent Discoverability

This package is optimized for Claude Code CLI and AI coding agents.

Probe Module

import gapless_network_data as gmd

# Get ranked features for ML pipelines
features = gmd.probe.get_alpha_features()
for f in features[:5]:
    print(f"{f.rank}. {f.name} ({f.importance})")

# Get protocol era boundaries
eras = gmd.probe.get_protocol_eras()

# Get setup workflow
workflow = gmd.probe.get_setup_workflow()

# Get example code
print(gmd.probe.get_quick_start())

Alpha Feature Rankings

AI agents should prioritize these features for ML pipelines:

Get rankings programmatically: gmd.probe.get_alpha_features()

Protocol Era Boundaries

Filter data appropriately based on protocol changes:

• EIP-1559 (block 12,965,000, Aug 2021): base_fee_per_gas introduced
• The Merge (block 15,537,394, Sep 2022): difficulty=0 forever
• EIP-4844 (block 19,426,587, Mar 2024): blob_gas fields introduced

Get eras programmatically: gmd.probe.get_protocol_eras()

API Reference

fetch_blocks()

gmd.fetch_blocks(
    start: str | None = None,     # ISO 8601 date (inclusive)
    end: str | None = None,       # ISO 8601 date (inclusive for date-only)
    limit: int | None = None,     # Max blocks (0 = empty DataFrame)
    include_deprecated: bool = False  # Include difficulty fields
) -> pd.DataFrame

Date Range Semantics (inclusive [start, end]):

• Date-only inputs include the entire day: end='2024-03-13' includes all of March 13
• Explicit times are preserved: end='2024-03-13 12:00:00' excludes blocks after noon
• Same-day queries work: start='2024-03-13', end='2024-03-13' returns all blocks on March 13

Parameter Requirements:

• At least one of start, end, or limit must be specified
• Empty strings ("") are rejected — use None to omit
• start must be ≤ end if both provided
• limit=0 returns empty DataFrame (0 rows, not entire blockchain)

Returns pandas DataFrame with columns:

• timestamp (datetime64[ns, UTC])
• number (uint64)
• gas_limit, gas_used, base_fee_per_gas, transaction_count, size (uint64)
• blob_gas_used, excess_blob_gas (Int64, nullable - pd.NA for pre-EIP4844)

Deprecated Fields

Excluded by default (use include_deprecated=True for pre-Merge analysis):

• difficulty: Always 0 post-Merge (Sep 2022)
• total_difficulty: Frozen post-Merge

Setup

Credentials via .env file (simplest), Doppler (recommended for teams), or environment variables.

Environment Variables

# Option 1: .env file (simplest for small teams)
# Create .env in your project root:
CLICKHOUSE_HOST_READONLY=<host>
CLICKHOUSE_USER_READONLY=<user>
CLICKHOUSE_PASSWORD_READONLY=<password>

# Option 2: Doppler (recommended for production)
doppler configure set token <token_from_1password>
doppler setup --project gapless-network-data --config prd

# Option 3: Environment variables
export CLICKHOUSE_HOST_READONLY=<host>
export CLICKHOUSE_USER_READONLY=<user>
export CLICKHOUSE_PASSWORD_READONLY=<password>

Get setup instructions: gmd.probe.get_setup_workflow()

Time Precision

• Timestamp storage: Millisecond precision (DateTime64(3))
• Block granularity: ~12 second intervals (Ethereum block time)
• Query precision: Second-level supported for start/end parameters

Supported timestamp formats:

Data Coverage

• Blocks: 23.87M Ethereum blocks (2015-2025)
• Update frequency: Real-time (~12 second intervals)
• Storage: ClickHouse Cloud (AWS)
• Deduplication: Automatic via ReplacingMergeTree

Exceptions

All exceptions include structured context (timestamp, endpoint, HTTP status):

Credential & Database:

• CredentialException: Credential resolution failed
• DatabaseException: ClickHouse query failed

Parameter Validation (fetch_blocks):

• ValueError: Empty string for start/end (use None to omit)
• ValueError: No parameters specified (must have start, end, or limit)
• ValueError: Reversed date range (start > end)

Feature Engineering Integration

Combine with OHLCV price data:

import gapless_crypto_data as gcd
import gapless_network_data as gmd

# Fetch both data sources
df_ohlcv = gcd.get_data(symbol="ETHUSDT", timeframe="1m", start_date="2024-01-01")
df_blocks = gmd.fetch_blocks(start="2024-01-01", end="2024-01-02")

# Temporal alignment (forward-fill prevents data leakage)
df_blocks_aligned = df_blocks.set_index('timestamp').reindex(
    df_ohlcv.index, method='ffill'
)

# Join and engineer features
df = df_ohlcv.join(df_blocks_aligned)
df['gas_pressure'] = df['base_fee_per_gas'] / df['base_fee_per_gas'].rolling(60).median()
df['block_utilization'] = df['gas_used'] / df['gas_limit']

Infrastructure (Reference)

Dual-pipeline architecture for production reliability:

Related Projects

• gapless-crypto-data - OHLCV data collection
• BigQuery Ethereum Dataset

Documentation

• Architecture Overview
• Data Format Specification

License

MIT License