Getting Started with chronos-lab¶

This tutorial introduces the core concepts of chronos-lab.

What You'll Learn¶

• Fetching OHLCV data from Yahoo Finance and Intrinio
• Persisting data with ArcticDB for high-performance storage
• Working with datasets for organizing securities metadata
• Storing visualizations and other files

Setup¶

Install chronos-lab with all required extras for this tutorial:

In [ ]:

!pip install chronos-lab[yfinance,intrinio,arcticdb,analysis,visualization]

!pip install chronos-lab[yfinance,intrinio,arcticdb,analysis,visualization]

Understanding Configuration¶

On first import, chronos-lab automatically creates a configuration file at ~/.chronos_lab/.env. Let's examine its default settings:

In [1]:

import chronos_lab
!cat ~/.chronos_lab/.env

import chronos_lab !cat ~/.chronos_lab/.env

# Intrinio API Settings
#INTRINIO_API_KEY=

# Datatset Settings
DATASET_LOCAL_PATH=~/.chronos_lab/datasets
#DATASET_DDB_TABLE_NAME=
#DATASET_DDB_MAP='{
#    "ddb_watchlist": {
#        "pk": "map#ibpm#watchlist",
#        "sk": "name"
#    },
#    "ddb_securities_intrinio": {
#        "pk": "map#intrinio#securities"
#    },
#    "ddb_ohlcv_anomalies": {
#        "pk": "chronos_lab#ohlcv_anomalies"
#    }
#}'

# Store Settings
STORE_LOCAL_PATH=~/.chronos_lab/store
#STORE_S3_BUCKET=

# ArcticDB Settings
ARCTICDB_LOCAL_PATH=~/.chronos_lab/arcticdb
ARCTICDB_DEFAULT_LIBRARY_NAME=uscomp
#ARCTICDB_S3_BUCKET=

# Logging
LOG_LEVEL=WARNING

✓ Chronos Lab: Created config at /Users/vitalik/.chronos_lab/.env

Key settings:

• DATASET_LOCAL_PATH: Where JSON datasets are stored (securities lists, metadata)
• STORE_LOCAL_PATH: Where files (charts, reports) are saved
• ARCTICDB_LOCAL_PATH: High-performance time series database location
• INTRINIO_API_KEY: Your Intrinio API key (optional, for Intrinio data)

Now let's import the library:

In [1]:

from chronos_lab.sources import ohlcv_from_yfinance, ohlcv_from_arcticdb, from_dataset, ohlcv_from_intrinio, securities_from_intrinio
from chronos_lab.storage import ohlcv_to_arcticdb, to_dataset, to_store
import random

from chronos_lab.sources import ohlcv_from_yfinance, ohlcv_from_arcticdb, from_dataset, ohlcv_from_intrinio, securities_from_intrinio from chronos_lab.storage import ohlcv_to_arcticdb, to_dataset, to_store import random

Working with Yahoo Finance Data¶

Yahoo Finance provides free access to historical price data for thousands of securities. Let's start with a simple example.

Fetching OHLCV Data¶

Download 6 months of daily OHLCV (Open, High, Low, Close, Volume) data for a few tech stocks:

In [2]:

symbols = ["AAPL", "MSFT", "GOOGL", "NVDA", "META", "TSLA", "AMZN"]

ohlcv = ohlcv_from_yfinance(symbols=symbols, period='6mo')

print(f"Downloaded data for {len(ohlcv.index.get_level_values('symbol').unique())} symbols")
print(f"Date range: {ohlcv.index.get_level_values('date').min()} to {ohlcv.index.get_level_values('date').max()}")
ohlcv.head(10)

symbols = ["AAPL", "MSFT", "GOOGL", "NVDA", "META", "TSLA", "AMZN"] ohlcv = ohlcv_from_yfinance(symbols=symbols, period='6mo') print(f"Downloaded data for {len(ohlcv.index.get_level_values('symbol').unique())} symbols") print(f"Date range: {ohlcv.index.get_level_values('date').min()} to {ohlcv.index.get_level_values('date').max()}") ohlcv.head(10)

[*********************100%***********************]  7 of 7 completed

Downloaded data for 7 symbols
Date range: 2025-08-04 00:00:00+00:00 to 2026-02-02 00:00:00+00:00

Out[2]:

		open	high	low	close	volume
date	symbol
2025-08-04 00:00:00+00:00	AAPL	204.080327	207.443257	201.256270	202.922775	75109300
2025-08-05 00:00:00+00:00	AAPL	202.972655	204.908581	201.735270	202.493668	44155100
2025-08-06 00:00:00+00:00	AAPL	205.197986	214.927501	205.158061	212.801971	108483100
2025-08-07 00:00:00+00:00	AAPL	218.420141	220.386003	216.124970	219.567719	90224800
2025-08-08 00:00:00+00:00	AAPL	220.366045	230.514676	218.789363	228.868149	113854000
2025-08-11 00:00:00+00:00	AAPL	227.699265	229.337676	224.542322	226.959976	61806100
2025-08-12 00:00:00+00:00	AAPL	227.789171	230.576477	226.850094	229.427582	55626200
2025-08-13 00:00:00+00:00	AAPL	230.846229	234.772415	230.206834	233.104034	69878500
2025-08-14 00:00:00+00:00	AAPL	233.833325	234.892296	230.626442	232.554565	51916300
2025-08-15 00:00:00+00:00	AAPL	233.773377	234.053105	229.117886	231.365707	56038700

The returned DataFrame has a MultiIndex (date, symbol) with columns for open, high, low, close, and volume. All timestamps are UTC timezone-aware.

Storing Data to ArcticDB¶

ArcticDB provides high-performance storage for time series data. Let's persist our downloaded data:

In [3]:

result = ohlcv_to_arcticdb(ohlcv=ohlcv, backend='LMDB')

if result['statusCode'] == 0:
    print(f"✓ Data was successfully stored to ArcticDB")

result = ohlcv_to_arcticdb(ohlcv=ohlcv, backend='LMDB') if result['statusCode'] == 0: print(f"✓ Data was successfully stored to ArcticDB")

✓ Data was successfully stored to ArcticDB

Let's examine where the data is stored locally:

In [4]:

!ls -lh ~/.chronos_lab/arcticdb/

!ls -lh ~/.chronos_lab/arcticdb/

total 0
drwxr-xr-x@ 4 vitalik  staff   128B Jan 27 11:38 _arctic_cfg
drwxr-xr-x@ 4 vitalik  staff   128B Jan 27 11:38 uscomp

ArcticDB uses LMDB for local storage, providing fast read/write operations and automatic compression. "uscomp" is the default library name.

Reading Data from ArcticDB¶

Retrieve stored data with optional date range filtering:

In [4]:

# Read all data for specific symbols
ohlcv_df = ohlcv_from_arcticdb(symbols=["AAPL", "MSFT"])

print(f"Retrieved {len(ohlcv_df)} rows")
ohlcv_df.tail(5)

# Read all data for specific symbols ohlcv_df = ohlcv_from_arcticdb(symbols=["AAPL", "MSFT"]) print(f"Retrieved {len(ohlcv_df)} rows") ohlcv_df.tail(5)

Retrieved 252 rows

Out[4]:

		open	high	low	close	volume
date	symbol
2026-01-27 00:00:00+00:00	MSFT	473.700012	482.869995	473.160004	480.579987	29213900
2026-01-28 00:00:00+00:00	MSFT	483.209991	483.739990	478.000000	481.630005	36875400
2026-01-29 00:00:00+00:00	MSFT	439.989990	442.500000	421.019989	433.500000	128855300
2026-01-30 00:00:00+00:00	MSFT	439.170013	439.600006	426.450012	430.290009	58470400
2026-02-02 00:00:00+00:00	MSFT	430.255005	430.380005	424.519989	429.570007	7936885

In [5]:

# Read specific date range
ohlcv_dates_df = ohlcv_from_arcticdb(symbols=['AAPL', 'MSFT'], start_date='2026-01-21', end_date='2026-01-22')

print(f"Retrieved {len(ohlcv_dates_df)} rows")
ohlcv_dates_df.tail(5)

# Read specific date range ohlcv_dates_df = ohlcv_from_arcticdb(symbols=['AAPL', 'MSFT'], start_date='2026-01-21', end_date='2026-01-22') print(f"Retrieved {len(ohlcv_dates_df)} rows") ohlcv_dates_df.tail(5)

Retrieved 4 rows

Out[5]:

		open	high	low	close	volume
date	symbol
2026-01-21 00:00:00+00:00	AAPL	248.699997	251.559998	245.179993	247.649994	54641700
2026-01-22 00:00:00+00:00	AAPL	249.199997	251.000000	248.149994	248.350006	39708300
2026-01-21 00:00:00+00:00	MSFT	452.600006	452.690002	438.679993	444.109985	37980500
2026-01-22 00:00:00+00:00	MSFT	447.619995	452.839996	444.700012	451.140015	25349400

You can also filter by period, limit which columns are returned, and pivot the data for easier analysis:

In [6]:

# Get last 30 days as dictionary format
ohlcv_df_pivoted = ohlcv_from_arcticdb(
    symbols=["NVDA", "TSLA"],
    period="30d",
    columns=["close"],
    pivot=True
)
ohlcv_df_pivoted.tail()

# Get last 30 days as dictionary format ohlcv_df_pivoted = ohlcv_from_arcticdb( symbols=["NVDA", "TSLA"], period="30d", columns=["close"], pivot=True ) ohlcv_df_pivoted.tail()

Out[6]:

	close
symbol	NVDA	TSLA
date
2026-01-27 00:00:00+00:00	188.520004	430.899994
2026-01-28 00:00:00+00:00	191.520004	431.459991
2026-01-29 00:00:00+00:00	192.509995	416.559998
2026-01-30 00:00:00+00:00	191.130005	430.410004
2026-02-02 00:00:00+00:00	188.199905	418.450012

Working with Datasets¶

Datasets are JSON files that store structured metadata like securities lists, watchlists, or configuration data. They're stored in ~/.chronos_lab/datasets/.

Using the Example Dataset¶

chronos-lab includes an example dataset with 50 major US stocks. Let's explore it:

In [7]:

# Load the example dataset as a DataFrame
example_securities = from_dataset(dataset_name='example')

print(f"Example dataset contains {len(example_securities)} securities")
print(f"\nFirst 10 securities:")
example_securities.head(10)

# Load the example dataset as a DataFrame example_securities = from_dataset(dataset_name='example') print(f"Example dataset contains {len(example_securities)} securities") print(f"\nFirst 10 securities:") example_securities.head(10)

Example dataset contains 50 securities

First 10 securities:

Out[7]:

	weight
AAPL	0.02
MSFT	0.02
AMZN	0.02
GOOGL	0.02
META	0.02
NVDA	0.02
MCD	0.02
JPM	0.02
JNJ	0.02
V	0.02

Let's see where this dataset is stored:

In [9]:

!cat ~/.chronos_lab/datasets/example.json | head -n 20

!cat ~/.chronos_lab/datasets/example.json | head -n 20

{
  "AAPL": {"weight": 0.02},
  "MSFT": {"weight": 0.02},
  "AMZN": {"weight": 0.02},
  "GOOGL": {"weight": 0.02},
  "META": {"weight": 0.02},
  "NVDA": {"weight": 0.02},
  "MCD": {"weight": 0.02},
  "JPM": {"weight": 0.02},
  "JNJ": {"weight": 0.02},
  "V": {"weight": 0.02},
  "PG": {"weight": 0.02},
  "XOM": {"weight": 0.02},
  "UNH": {"weight": 0.02},
  "MA": {"weight": 0.02},
  "HD": {"weight": 0.02},
  "LLY": {"weight": 0.02},
  "ABBV": {"weight": 0.02},
  "PFE": {"weight": 0.02},
  "AVGO": {"weight": 0.02},

Datasets are simple JSON files with a key-value structure. The keys become the DataFrame index when loaded.

Creating custom Datasets¶

Define watchlists and store them in a Dataset for future use:

In [8]:

# Store custom watchlist
watchlist = {
    'tech_leaders': {
        'symbols': ['AAPL', 'MSFT', 'GOOGL', 'AMZN', 'META', 'NVDA'],
        'description': 'Large-cap technology companies',
        'created': '2026-01-27'
    },
    'dividend_stocks': {
        'symbols': ['JNJ', 'PG', 'KO', 'PEP', 'MCD'],
        'description': 'Dividend paying stocks',
        'created': '2026-01-27'
    }
}

result = to_dataset(
    dataset_name='watchlists',
    dataset=watchlist
)

# Store custom watchlist watchlist = { 'tech_leaders': { 'symbols': ['AAPL', 'MSFT', 'GOOGL', 'AMZN', 'META', 'NVDA'], 'description': 'Large-cap technology companies', 'created': '2026-01-27' }, 'dividend_stocks': { 'symbols': ['JNJ', 'PG', 'KO', 'PEP', 'MCD'], 'description': 'Dividend paying stocks', 'created': '2026-01-27' } } result = to_dataset( dataset_name='watchlists', dataset=watchlist )

Using Datasets to Drive Downloads¶

A common pattern is to maintain a dataset of securities and use it to drive bulk downloads. Let's randomly sample from the example dataset:

In [9]:

# Random sample of 15 symbols
random.seed(42)
sampled_symbols = random.sample(list(example_securities.index), 15)

print(f"Downloading data for: {', '.join(sampled_symbols)}")

# Fetch 3 months of data
sampled_ohlcv = ohlcv_from_yfinance(symbols=sampled_symbols, period='3mo')

# Store to ArcticDB
store_result = ohlcv_to_arcticdb(ohlcv=sampled_ohlcv)
if store_result['statusCode'] == 0:
    print(f"✓ Data was successfully stored to ArcticDB")

# Random sample of 15 symbols random.seed(42) sampled_symbols = random.sample(list(example_securities.index), 15) print(f"Downloading data for: {', '.join(sampled_symbols)}") # Fetch 3 months of data sampled_ohlcv = ohlcv_from_yfinance(symbols=sampled_symbols, period='3mo') # Store to ArcticDB store_result = ohlcv_to_arcticdb(ohlcv=sampled_ohlcv) if store_result['statusCode'] == 0: print(f"✓ Data was successfully stored to ArcticDB")

Downloading data for: IBM, JPM, MSFT, PFE, LLY, HD, JNJ, MCD, NEE, NVDA, AMGN, CRM, AMZN, MDT, NFLX

[*********************100%***********************]  15 of 15 completed

✓ Data was successfully stored to ArcticDB

Working with Intrinio¶

Intrinio provides institutional-grade financial data through a commercial API. This section requires an API key in your ~/.chronos_lab/.env file.

Note: If you don't have an Intrinio API key, you can skip this section or sign up at intrinio.com.

Fetching Securities List¶

Intrinio maintains comprehensive securities databases. Let's fetch a sample of US equities:

In [12]:

# Fetch securities metadata from Intrinio
try:
    securities_df = securities_from_intrinio()
    
    if securities_df is not None and len(securities_df) > 0:
        print(f"Fetched {len(securities_df)} securities from Intrinio")
        print(f"\nFirst 5 securities:")
        for idx, row in securities_df.head(5).iterrows():
            ticker = row.get('ticker', 'N/A')
            name = row.get('name', 'N/A')[:40]
            print(f"  {ticker:6} - {name}")
    else:
        securities_df = None
        print("No securities returned (check API key)")
        
except Exception as e:
    securities_df = None
    print(f"Intrinio API not configured or error occurred: {e}")
    print("Skipping Intrinio examples...")

# Fetch securities metadata from Intrinio try: securities_df = securities_from_intrinio() if securities_df is not None and len(securities_df) > 0: print(f"Fetched {len(securities_df)} securities from Intrinio") print(f"\nFirst 5 securities:") for idx, row in securities_df.head(5).iterrows(): ticker = row.get('ticker', 'N/A') name = row.get('name', 'N/A')[:40] print(f" {ticker:6} - {name}") else: securities_df = None print("No securities returned (check API key)") except Exception as e: securities_df = None print(f"Intrinio API not configured or error occurred: {e}") print("Skipping Intrinio examples...")

Fetched 4885 securities from Intrinio

First 5 securities:
  A      - Agilent Technologies Inc.
  AA     - Alcoa Corp
  AAL    - American Airlines Group Inc
  AAME   - Atlantic American Corp.
  AAOI   - Applied Optoelectronics Inc

Saving Securities as a Dataset¶

Let's save the securities list as a dataset for future use:

In [13]:

if securities_df is not None:
    # Save to dataset
    result = to_dataset(dataset_name='intrinio_securities', dataset=securities_df.to_dict(orient='index'))
    
    if result['statusCode'] == 0:
        print(f"✓ Dataset was successfully saved")
else:
    print("No securities to save (Intrinio not configured)")

if securities_df is not None: # Save to dataset result = to_dataset(dataset_name='intrinio_securities', dataset=securities_df.to_dict(orient='index')) if result['statusCode'] == 0: print(f"✓ Dataset was successfully saved") else: print("No securities to save (Intrinio not configured)")

✓ Dataset was successfully saved

Let's examine the saved dataset:

In [14]:

!cat ~/.chronos_lab/datasets/intrinio_securities.json 2>/dev/null | head -n 16 || echo "Dataset not created (Intrinio not configured)"

!cat ~/.chronos_lab/datasets/intrinio_securities.json 2>/dev/null | head -n 16 || echo "Dataset not created (Intrinio not configured)"

{
  "sec_NX6GzO": {
    "company_id": "com_NXnP4y",
    "exchange": "NYSE",
    "exchange_mic": "XNYS",
    "stock_exchange_id": "sxg_ozMr9y",
    "name": "Agilent Technologies Inc.",
    "code": "EQS",
    "currency": "USD",
    "ticker": "A",
    "composite_ticker": "A:US",
    "sec_figi": "BBG000C2V541",
    "composite_figi": "BBG000C2V3D6",
    "share_class_figi": "BBG001SCTQY4",
    "primary_listing": true
  },

Using Securities Dataset for OHLCV Download¶

Now use the saved dataset to drive OHLCV downloads from Intrinio:

In [16]:

if securities_df is not None:
    # Load our securities dataset
    intrinio_securities = from_dataset(dataset_name='intrinio_securities')
    
    # Sample 10 securities for download
    sample_sec_id = random.sample(list(intrinio_securities.index), min(10, len(intrinio_securities)))
    
    print(f"Downloading OHLCV data from Intrinio for {len(sample_sec_id)} securities...")
    print(f"Security IDs: {', '.join(sample_sec_id)}")
    
    try:
        intrinio_ohlcv_df = ohlcv_from_intrinio(
            symbols=sample_sec_id,
            period='3m'
        )
        
        if intrinio_ohlcv_df is not None and len(intrinio_ohlcv_df) > 0:
            print(f"\n✓ Downloaded {len(intrinio_ohlcv_df)} rows")
            
            # Store to a different library in ArcticDB
            store_result = ohlcv_to_arcticdb(ohlcv=intrinio_ohlcv_df,
                                             library_name='intrinio')
            if store_result['statusCode'] == 0:
                print(f"✓ Stored to ArcticDB")
            
            # Show sample
            print(f"\nSample data:")
            display(intrinio_ohlcv_df.head())
        else:
            print("No data returned from Intrinio")
    except Exception as e:
        print(f"Error fetching from Intrinio: {e}")
else:
    print("Skipping Intrinio OHLCV download (no securities available)")

if securities_df is not None: # Load our securities dataset intrinio_securities = from_dataset(dataset_name='intrinio_securities') # Sample 10 securities for download sample_sec_id = random.sample(list(intrinio_securities.index), min(10, len(intrinio_securities))) print(f"Downloading OHLCV data from Intrinio for {len(sample_sec_id)} securities...") print(f"Security IDs: {', '.join(sample_sec_id)}") try: intrinio_ohlcv_df = ohlcv_from_intrinio( symbols=sample_sec_id, period='3m' ) if intrinio_ohlcv_df is not None and len(intrinio_ohlcv_df) > 0: print(f"\n✓ Downloaded {len(intrinio_ohlcv_df)} rows") # Store to a different library in ArcticDB store_result = ohlcv_to_arcticdb(ohlcv=intrinio_ohlcv_df, library_name='intrinio') if store_result['statusCode'] == 0: print(f"✓ Stored to ArcticDB") # Show sample print(f"\nSample data:") display(intrinio_ohlcv_df.head()) else: print("No data returned from Intrinio") except Exception as e: print(f"Error fetching from Intrinio: {e}") else: print("Skipping Intrinio OHLCV download (no securities available)")

Downloading OHLCV data from Intrinio for 10 securities...
Security IDs: sec_gNqdJv, sec_gkkVxo, sec_qzEmnz, sec_XWMajy, sec_y3GKvd, sec_g4awwk, sec_XLBN6P, sec_gNQ70y, sec_gProVz, sec_gAAkVO

✓ Downloaded 590 rows
✓ Stored to ArcticDB

Sample data:

		intraperiod	frequency	open	high	low	close	volume	adj_open	adj_high	adj_low	adj_close	adj_volume	factor	split_ratio	dividend	change	percent_change	fifty_two_week_high	fifty_two_week_low	symbol
date	id
2025-12-03 00:00:00+00:00	sec_gNqdJv	False	daily	9.8800	9.9900	9.8800	9.8900	29630.0	9.8800	9.9900	9.8800	9.8900	29630.0	1.0	1.0	0.0	NaN	NaN	9.99	9.88	EVOX
2025-12-04 00:00:00+00:00	sec_gNqdJv	False	daily	9.9000	9.9100	9.9000	9.9000	180941.0	9.9000	9.9100	9.9000	9.9000	180941.0	1.0	1.0	0.0	0.01	0.001	9.99	9.88	EVOX
2025-12-05 00:00:00+00:00	sec_gNqdJv	False	daily	9.8999	9.8999	9.8999	9.8999	209.0	9.8999	9.8999	9.8999	9.8999	209.0	1.0	1.0	0.0	0.00	0.000	9.99	9.88	EVOX
2025-12-08 00:00:00+00:00	sec_gNqdJv	False	daily	9.9000	9.9100	9.8900	9.8900	3.0	9.9000	9.9100	9.8900	9.8900	3.0	1.0	1.0	0.0	-0.01	-0.001	9.99	9.88	EVOX
2025-12-10 00:00:00+00:00	sec_gNqdJv	False	daily	9.8900	9.9000	9.8800	9.8800	65787.0	9.8900	9.9000	9.8800	9.8800	65787.0	1.0	1.0	0.0	-0.01	-0.001	9.99	9.88	EVOX

In production environments, independent processes can write datasets to DynamoDB (use ddb_ prefix), and other processes can read them with from_dataset().

Visualization and Storage¶

chronos-lab provides convenient functions to store generated files like charts, reports, or any binary content.

Creating a Simple Chart¶

Let's create a basic candlestick-style chart using mplfinance with a Bloomberg-inspired theme:

In [10]:

import mplfinance as mpf
from chronos_lab.plot import bloomberg_style
import io

# Get data for a single symbol
symbol = "AAPL"
chart_data = ohlcv_from_arcticdb(symbols=[symbol], period="60d")

if len(chart_data) > 0:
    # Extract symbol data and prepare for mplfinance
    symbol_data = chart_data.xs(symbol, level='symbol')

    # Create the candlestick chart with volume
    fig, axes = mpf.plot(
        symbol_data,
        type='candle',
        style=bloomberg_style,
        volume=True,
        figsize=(12, 8),
        panel_ratios=(3, 1),
        returnfig=True,
        ylabel='Price',
        ylabel_lower='Volume',
        datetime_format='%Y-%m-%d'
    )

    # Adjust styling
    for ax in axes:
        ax.set_facecolor('#000000')
        ax.grid(True, which='major', linestyle=':', linewidth=0.8, color='#C0C0C0')

    fig.suptitle(
        f"{symbol} - Last 60 Days",
        color='white',
        fontsize=20,
        y=0.92
    )

    print(f"\nChart created for {symbol} using mplfinance")
else:
    print(f"No data available for {symbol}")

import mplfinance as mpf from chronos_lab.plot import bloomberg_style import io # Get data for a single symbol symbol = "AAPL" chart_data = ohlcv_from_arcticdb(symbols=[symbol], period="60d") if len(chart_data) > 0: # Extract symbol data and prepare for mplfinance symbol_data = chart_data.xs(symbol, level='symbol') # Create the candlestick chart with volume fig, axes = mpf.plot( symbol_data, type='candle', style=bloomberg_style, volume=True, figsize=(12, 8), panel_ratios=(3, 1), returnfig=True, ylabel='Price', ylabel_lower='Volume', datetime_format='%Y-%m-%d' ) # Adjust styling for ax in axes: ax.set_facecolor('#000000') ax.grid(True, which='major', linestyle=':', linewidth=0.8, color='#C0C0C0') fig.suptitle( f"{symbol} - Last 60 Days", color='white', fontsize=20, y=0.92 ) print(f"\nChart created for {symbol} using mplfinance") else: print(f"No data available for {symbol}")

Chart created for AAPL using mplfinance

Storing the Chart¶

Save the chart to local storage using to_store():

In [11]:

if len(chart_data) > 0:
    # Save figure to bytes buffer
    buf = io.BytesIO()
    fig.savefig(buf, format='png', dpi=150, bbox_inches='tight')
    buf.seek(0)
    chart_bytes = buf.read()
    buf.close()
    
    # Store using to_store()
    store_result = to_store(
        file_name=f"{symbol}_chart.png",
        content=chart_bytes,
        folder="charts"
    )
    
    if store_result.get('local_statusCode') == 0:
        print(f"✓ Chart saved to: {store_result['file_path']}")
        print(f"\nStorage directory contents:")
        !ls -lh ~/.chronos_lab/store/charts/

if len(chart_data) > 0: # Save figure to bytes buffer buf = io.BytesIO() fig.savefig(buf, format='png', dpi=150, bbox_inches='tight') buf.seek(0) chart_bytes = buf.read() buf.close() # Store using to_store() store_result = to_store( file_name=f"{symbol}_chart.png", content=chart_bytes, folder="charts" ) if store_result.get('local_statusCode') == 0: print(f"✓ Chart saved to: {store_result['file_path']}") print(f"\nStorage directory contents:") !ls -lh ~/.chronos_lab/store/charts/

✓ Chart saved to: /Users/vitalik/.chronos_lab/store/charts/AAPL_chart.png

Storage directory contents:
total 144
-rw-r--r--@ 1 vitalik  staff    70K Feb  2 10:15 AAPL_chart.png

Let's verify we can read the saved chart:

In [12]:

if len(chart_data) > 0 and store_result.get('local_statusCode') == 0:
    from IPython.display import Image, display
    
    display(Image(filename=store_result['file_path']))
    print(f"\nSuccessfully loaded chart from storage!")

if len(chart_data) > 0 and store_result.get('local_statusCode') == 0: from IPython.display import Image, display display(Image(filename=store_result['file_path'])) print(f"\nSuccessfully loaded chart from storage!")

Successfully loaded chart from storage!

Next Steps¶

You've learned the fundamentals of chronos-lab! Here's what to explore next:

Advanced Tutorials¶

• OHLCV Anomaly Detection: Detect unusual patterns in price data using machine learning

Key Concepts Recap¶

1. Data Flow: Fetch → Store → Retrieve

   • Use ohlcv_from_*() to fetch data
   • Use ohlcv_to_arcticdb() to store
   • Use ohlcv_from_arcticdb() to retrieve

2. Datasets: Organize securities metadata

   • Use from_dataset() to read
   • Use to_dataset() to write
   • Stored as JSON in ~/.chronos_lab/datasets/

3. Storage: Persist any file type

   • Use to_store() for charts, reports, etc.
   • Stored in ~/.chronos_lab/store/