Data Exports

This page explains how to export the required datasets from Odoo for ML training.

Revenue Dataset

File: data/amazon_revenue_forecasting.csv

How to export from Odoo:

1. Go to Sales → Orders → Orders
2. Set the filter to confirmed orders (state = sale)
3. Click Action → Export
4. Select the fields listed below
5. Export as CSV

Required fields:

Odoo Field	CSV Column	Description
sale.order.line.product_uom_qty	quantity_sold	Units sold per line
sale.order.line.price_unit	price	Unit selling price
sale.order.line.discount	discount_percent	Line discount %
sale.order.date_order	month, quarter	Order date (parsed)
sale.order.line.margin / price_subtotal	gross_margin_pct	Margin ratio
(computed)	seasonal_index	Seasonal index from date
sale.order.line.price_subtotal	net_revenue	Target variable

---

Churn Dataset

File: data/amazon_churn_prediction.csv

How to export from Odoo:

1. Go to Sales → Orders → Customers or CRM → Customers
2. Use a pivot/analysis view or custom report
3. The dataset requires per-customer aggregation — Odoo's built-in export may need to be supplemented with a custom report or Python script using the XML-RPC API

Required fields:

CSV Column	Description	Source
days_since_last_order	Days since most recent order	Computed from date_order
total_orders	Lifetime order count	COUNT(sale.order) per partner
avg_days_between_orders	Average order cadence	Computed
churn_risk_score	Pre-computed risk signal	From CRM or custom field
total_spent	Lifetime revenue	SUM(amount_total)
avg_order_value	Average order size	AVG(amount_total)
churned	Target variable (0/1)	Manual label or inferred from inactivity

Churn label convention: A customer is marked as churned (churned = 1) if they have had no orders in the last 90 days.

---

Growth Dataset

File: data/amazon_growth_projection.csv

How to export from Odoo:

1. Go to Sales → Reporting → Sales Analysis
2. Group by month or quarter
3. Export revenue totals per period

Required fields:

CSV Column	Description
period	Month or quarter label
total_revenue	Revenue for the period
order_count	Number of orders
new_customers	New partners acquired
avg_order_value	Average order value for the period
revenue_growth_3m	Target variable — 3-month forward revenue

---

Automating Exports (Python + XML-RPC)

For automated, scheduled exports, use the Odoo XML-RPC API:

import xmlrpc.client
import pandas as pd
from datetime import datetime, timedelta
 
URL     = "https://your-odoo.com"
DB      = "your_db"
USER    = "admin@company.com"
API_KEY = "your_key"
 
common  = xmlrpc.client.ServerProxy(f"{URL}/xmlrpc/2/common")
uid     = common.authenticate(DB, USER, API_KEY, {})
models  = xmlrpc.client.ServerProxy(f"{URL}/xmlrpc/2/object")
 
# Pull all confirmed order lines from the last 2 years
cutoff  = (datetime.now() - timedelta(days=730)).strftime("%Y-%m-%d")
lines   = models.execute_kw(
    DB, uid, API_KEY,
    'sale.order.line', 'search_read',
    [[['order_id.state', '=', 'sale'],
      ['order_id.date_order', '>=', cutoff]]],
    {
        'fields': [
            'product_uom_qty', 'price_unit', 'discount',
            'price_subtotal', 'order_id'
        ],
        'limit': 50000
    }
)
 
df = pd.DataFrame(lines)
df.to_csv("data/amazon_revenue_forecasting.csv", index=False)
print(f"Exported {len(df)} order lines.")