Datasets:

ThGaskin
/

Migration_flows

Unnamed: 0 int64 0 763k	kind stringclasses 2 values	source stringclasses 5 values	fold stringclasses 5 values	year float64 1.99k 2.02k	orig stringclasses 231 values	dest stringclasses 231 values	x float64 0 985k	y float64 0 1.04M
0	Train	Quantmig	Fold 1	2,009	AUT	BEL	265.8	435.84283
1	Train	Quantmig	Fold 1	2,009	AUT	BGR	1,156.9	563.4307
2	Train	Quantmig	Fold 1	2,009	AUT	CHE	2,455.6	1,784.3075
3	Train	Quantmig	Fold 1	2,009	AUT	CYP	32	59.130985
4	Train	Quantmig	Fold 1	2,009	AUT	DEU	14,245.8	14,956.544
5	Train	Quantmig	Fold 1	2,009	AUT	DNK	224.1	306.25037
6	Train	Quantmig	Fold 1	2,009	AUT	EST	29.5	67.82613
7	Train	Quantmig	Fold 1	2,009	AUT	FIN	108.1	184.8558
8	Train	Quantmig	Fold 1	2,009	AUT	HRV	1,281.3	895.652
9	Train	Quantmig	Fold 1	2,009	AUT	HUN	3,455.3	4,189.3896
10	Train	Quantmig	Fold 1	2,009	AUT	IRL	166.8	326.12753
11	Train	Quantmig	Fold 1	2,009	AUT	ISL	29.1	28.900314
12	Train	Quantmig	Fold 1	2,009	AUT	ITA	1,440.4	1,716.5468
13	Train	Quantmig	Fold 1	2,009	AUT	LIE	125.1	210.239
14	Train	Quantmig	Fold 1	2,009	AUT	LTU	50.9	174.46758
15	Train	Quantmig	Fold 1	2,009	AUT	LVA	57.6	123.04902
16	Train	Quantmig	Fold 1	2,009	AUT	MLT	24.2	27.652607
17	Train	Quantmig	Fold 1	2,009	AUT	NOR	157.4	258.94217
18	Train	Quantmig	Fold 1	2,009	AUT	POL	2,971.5	4,684.278
19	Train	Quantmig	Fold 1	2,009	AUT	PRT	282.5	194.7011
20	Train	Quantmig	Fold 1	2,009	AUT	ROU	3,842.4	6,312.44
21	Train	Quantmig	Fold 1	2,009	AUT	SVK	2,312.1	1,359.0314
22	Train	Quantmig	Fold 1	2,009	AUT	SVN	542.7	714.98425
23	Train	Quantmig	Fold 1	2,009	AUT	SWE	454.9	660.6436
24	Train	Quantmig	Fold 1	2,009	BEL	AUT	344.3	506.532
25	Train	Quantmig	Fold 1	2,009	BEL	CHE	1,002.1	1,438.7233
26	Train	Quantmig	Fold 1	2,009	BEL	CZE	312.5	389.7095
27	Train	Quantmig	Fold 1	2,009	BEL	DEU	3,479.5	4,232.1577
28	Train	Quantmig	Fold 1	2,009	BEL	DNK	334.7	433.15054
29	Train	Quantmig	Fold 1	2,009	BEL	ESP	3,335.3	3,303.9502
30	Train	Quantmig	Fold 1	2,009	BEL	FIN	254.8	232.67583
31	Train	Quantmig	Fold 1	2,009	BEL	FRA	13,451.2	12,100.666
32	Train	Quantmig	Fold 1	2,009	BEL	HRV	103	162.0624
33	Train	Quantmig	Fold 1	2,009	BEL	HUN	439.8	559.78595
34	Train	Quantmig	Fold 1	2,009	BEL	ITA	2,030.5	2,938.1177
35	Train	Quantmig	Fold 1	2,009	BEL	LTU	81.4	131.0801
36	Train	Quantmig	Fold 1	2,009	BEL	LUX	1,636.3	2,181.4026
37	Train	Quantmig	Fold 1	2,009	BEL	LVA	99.6	82.30625
38	Train	Quantmig	Fold 1	2,009	BEL	NOR	294.1	370.45218
39	Train	Quantmig	Fold 1	2,009	BEL	PRT	1,416.7	823.8594
40	Train	Quantmig	Fold 1	2,009	BEL	ROU	2,042	749.09564
41	Train	Quantmig	Fold 1	2,009	BEL	SVK	315.6	246.88145
42	Train	Quantmig	Fold 1	2,009	BEL	SVN	107.7	70.244316
43	Train	Quantmig	Fold 1	2,009	BEL	SWE	579.1	589.0935
44	Train	Quantmig	Fold 1	2,009	BGR	AUT	2,611.9	2,791.4573
45	Train	Quantmig	Fold 1	2,009	BGR	BEL	2,704.3	2,969.419
46	Train	Quantmig	Fold 1	2,009	BGR	CHE	450.6	617.9871
47	Train	Quantmig	Fold 1	2,009	BGR	CYP	869.4	508.87448
48	Train	Quantmig	Fold 1	2,009	BGR	CZE	989.7	573.93286
49	Train	Quantmig	Fold 1	2,009	BGR	DEU	17,283.7	16,525.758
50	Train	Quantmig	Fold 1	2,009	BGR	DNK	888.8	703.31934
51	Train	Quantmig	Fold 1	2,009	BGR	ESP	8,061.3	8,649.957
52	Train	Quantmig	Fold 1	2,009	BGR	EST	37.5	79.947044
53	Train	Quantmig	Fold 1	2,009	BGR	FIN	169.4	368.64505
54	Train	Quantmig	Fold 1	2,009	BGR	FRA	1,249.4	1,601.6003
55	Train	Quantmig	Fold 1	2,009	BGR	ISL	16.2	47.795727
56	Train	Quantmig	Fold 1	2,009	BGR	ITA	8,151.7	9,011.945
57	Train	Quantmig	Fold 1	2,009	BGR	LIE	0.7	4.092217
58	Train	Quantmig	Fold 1	2,009	BGR	LTU	31.2	56.235683
59	Train	Quantmig	Fold 1	2,009	BGR	LUX	46.8	84.96943
60	Train	Quantmig	Fold 1	2,009	BGR	LVA	67.9	38.10982
61	Train	Quantmig	Fold 1	2,009	BGR	MLT	61.7	39.94115
62	Train	Quantmig	Fold 1	2,009	BGR	NOR	529.6	544.4056
63	Train	Quantmig	Fold 1	2,009	BGR	POL	1,073.3	411.24857
64	Train	Quantmig	Fold 1	2,009	BGR	PRT	1,084	364.2148
65	Train	Quantmig	Fold 1	2,009	BGR	ROU	649.4	603.29626
66	Train	Quantmig	Fold 1	2,009	BGR	SVK	390.1	506.8726
67	Train	Quantmig	Fold 1	2,009	BGR	SVN	567.7	411.62726
68	Train	Quantmig	Fold 1	2,009	BGR	SWE	807.8	1,406.6085
69	Train	Quantmig	Fold 1	2,009	CHE	AUT	1,567.5	1,341.4968
70	Train	Quantmig	Fold 1	2,009	CHE	CYP	56.6	33.382317
71	Train	Quantmig	Fold 1	2,009	CHE	CZE	419	505.45453
72	Train	Quantmig	Fold 1	2,009	CHE	DEU	12,743.1	11,751.437
73	Train	Quantmig	Fold 1	2,009	CHE	DNK	371.3	456.94766
74	Train	Quantmig	Fold 1	2,009	CHE	ESP	2,992	3,646.418
75	Train	Quantmig	Fold 1	2,009	CHE	EST	33.9	44.834362
76	Train	Quantmig	Fold 1	2,009	CHE	FIN	305.5	231.0143
77	Train	Quantmig	Fold 1	2,009	CHE	FRA	9,023.8	6,001.576
78	Train	Quantmig	Fold 1	2,009	CHE	GRC	449.9	486.70853
79	Train	Quantmig	Fold 1	2,009	CHE	HRV	381.9	245.16962
80	Train	Quantmig	Fold 1	2,009	CHE	HUN	804.6	714.8738
81	Train	Quantmig	Fold 1	2,009	CHE	IRL	261.9	453.17676
82	Train	Quantmig	Fold 1	2,009	CHE	ISL	32.2	41.248158
83	Train	Quantmig	Fold 1	2,009	CHE	ITA	6,416	7,508.8467
84	Train	Quantmig	Fold 1	2,009	CHE	LIE	396	409.04602
85	Train	Quantmig	Fold 1	2,009	CHE	LUX	207.3	293.88428
86	Train	Quantmig	Fold 1	2,009	CHE	LVA	69.3	95.96142
87	Train	Quantmig	Fold 1	2,009	CHE	MLT	56.2	41.802433
88	Train	Quantmig	Fold 1	2,009	CHE	NOR	272	404.20352
89	Train	Quantmig	Fold 1	2,009	CHE	POL	1,110.4	1,274.6232
90	Train	Quantmig	Fold 1	2,009	CHE	PRT	4,736.2	4,144.034
91	Train	Quantmig	Fold 1	2,009	CHE	ROU	578.5	751.373
92	Train	Quantmig	Fold 1	2,009	CHE	SVK	501.8	306.014
93	Train	Quantmig	Fold 1	2,009	CHE	SVN	147.3	356.01648
94	Train	Quantmig	Fold 1	2,009	CYP	AUT	36.5	45.612373
95	Train	Quantmig	Fold 1	2,009	CYP	BGR	448.7	222.50162
96	Train	Quantmig	Fold 1	2,009	CYP	CHE	58	67.4531
97	Train	Quantmig	Fold 1	2,009	CYP	CZE	60.2	52.05166
98	Train	Quantmig	Fold 1	2,009	CYP	DEU	502.6	374.4708
99	Train	Quantmig	Fold 1	2,009	CYP	DNK	33.8	38.884377

End of preview. Expand in Data Studio

Deep learning four decades of human migration: datasets

This repository contains all migration flow estimates associated with the paper "Deep learning four decades of human migration." Evaluation code, training data, trained neural networks, and smaller flow datasets are available in the main GitHub repository, which also provides detailed instructions on data sourcing. Due to file size limits, the larger datasets are archived here. The repository contains three folders:

Estimates

This folder contains all the migration estimates. Data is available in both NetCDF (.nc) and CSV (.csv) formats. The NetCDF format is more compact and pre-indexed, making it suitable for large files. In Python, datasets can be opened as xarray.Dataset objects, enabling coordinate-based data selection.

Each dataset uses the following coordinate conventions:

Year: 1990–2023
Birth ISO: Country of birth (UN ISO3)
Origin ISO: Country of origin (UN ISO3)
Destination ISO: Destination country (UN ISO3)
Country ISO: Used for net migration data (UN ISO3)

The following data files are provided:

T.nc: Full table of flows disaggregated by country of birth. Dimensions: Year, Birth ISO, Origin ISO, Destination ISO
flows.nc: Total origin-destination flows (equivalent to T summed over Birth ISO). Dimensions: Year, Origin ISO, Destination ISO
net_migration.nc: Net migration data by country. Dimensions: Year, Country ISO
stocks.nc: Stock estimates for each country pair. Dimensions: Year, Origin ISO (corresponding to Birth ISO), Destination ISO
test_flows.nc: Flow estimates on a randomly selected set of test edges, used for model validation

Additionally, two CSV files are provided for convenience:

mig_unilateral.csv: Unilateral migration estimates per country, comprising:
- imm: Total immigration flows
- emi: Total emigration flows
- net: Net migration
- imm_pop: Total immigrant population (non-native-born)
- emi_pop: Total emigrant population (living abroad)
mig_bilateral.csv: Bilateral flow data, comprising:
- mig_prev: Total origin-destination flows
- mig_brth: Total birth-destination flows, where Origin ISO reflects place of birth

Each dataset includes a mean variable (mean estimate) and a std variable (standard deviation of the estimate).

An ISO3 conversion table is also provided.

Data

The Data contains all the data used to train, evaluate, and test the neural network. It is stored thematically in different folders, and most folders again contains its own README file to further explain the specific sources and imputation methods. All data is given both as a .csv file and a .nc file, and follows the ISO3-naming convention outlined in the main README.

Training_data

This folder contains all the tensors used to train the neural network. All data is given as a PyTorch tensor (.pt) and can be loaded using torch.load(). The folder contains targets, weights, masks, input covariates (scaled and unscaled), and the edge indices of each input. See the folder README for further details.

Net migration (`Net_migration`)

This folder contains net migration data, sourced from national statistical offices, together with a list of sources and the UN WPP net migration figures.

GDP indicators (`GDP_data`)

This folder contains data on GDP/capita, GDP growth, nominal GDP, and other GDP-related indicators for all countries and years included in the training period.