File size: 2,673 Bytes
a9a7ec0
73641fe
97d82a1
a9a7ec0
 
 
 
97d82a1
a9a7ec0
 
73641fe
97d82a1
 
 
8969b91
97d82a1
 
 
 
 
8969b91
c07dcab
8969b91
97d82a1
c07dcab
97d82a1
 
8969b91
97d82a1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
73641fe
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
---
title: DATFID MASTER
emoji: 🤖
colorFrom: indigo
colorTo: pink
sdk: docker
pinned: false
short_description: DATFID - Secure and powerful forecasting SDK
---

# DATFID MASTER

**DATFID** delivers reliable and interpretable forecasting for any time series. Our models consistently rank at the top—see our website [datfid.com](https://datfid.com) for live results.

We achieved **state-of-the-art accuracy in the M5 Forecasting Competition**, beated all the benchmarks, forecasting hierarchical Walmart product sales over a 5-year period using advanced exogenous features and rigorous error metrics ([Kaggle M5 page](https://www.kaggle.com/c/m5-forecasting-accuracy)).

---

##  Why DATFID Stands Out

- **Professionally recognized benchmarking**  
  M5 is part of the renowned “M-Competitions” series, used worldwide to compare forecasting methods under realistic conditions. ([Makridakis Competitions, Wikipedia](https://en.wikipedia.org/wiki/Makridakis_Competitions)).

- **Hierarchical & intermittent series**  
  DATFID handles complex retail data with product-store hierarchies, zero-heavy series, pricing, calendar events, and promotional features—all core challenges of the M5 setup ([Background & Organization of the M5 Competition](https://www.sciencedirect.com/science/article/pii/S0169207021001187)).

- **Robust accuracy**  
  Scoring well on 42,840 series across multiple aggregation levels is a testament to our method’s strength—especially on challenging intermittent product-level data.

- **Fast, scalable, and interpretable**  
  Unlike “black-box” approaches, we prioritize transparency and efficiency without sacrificing accuracy.

---

##  Getting Started

### 1. Request a Token

To access the API, request your personal DATFID token via **admin@datfid.com**.

### 2. Install the SDK

```bash
pip install -i https://test.pypi.org/simple/ datfid
```

### 3. Example Usage

```python
import pandas as pd
from datfid.client import DATFIDClient

# Initialize with your DATFID token
client = DATFIDClient(token="dt+YOUR_TOKEN_HERE")

# Fit the model
fit_result = client.fit_model(
    df=my_dataframe,
    id_col="SKU_ID",
    time_col="Date",
    y="Sales",
    lagged_features={"Sales": 2, "Promo": 1},
    current_features=["Price", "Holiday"],
    filter_by_significance=True,
    meanvar_test=True
)

# Forecast
forecast_df = client.forecast_model(df_forecast=my_forecast_dataframe)
print(forecast_df.head())

# Check token validity
client.secure_ping()
```

---

##  Contact & Info

- Website: [datfid.com](https://datfid.com)  
- Get your token: **admin@datfid.com**

---

Happy forecasting with confidence!  
— The DATFID Team