mstz commited on
Commit
2ac3764
1 Parent(s): ee06934

Upload 3 files

Browse files
Files changed (3) hide show
  1. README.md +28 -1
  2. arrhythmia.data +0 -0
  3. arrhythmia.py +354 -0
README.md CHANGED
@@ -1,3 +1,30 @@
1
  ---
2
- license: cc
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language:
3
+ - en
4
+ tags:
5
+ - arrhythmia
6
+ - tabular_classification
7
+ - multiclass_classification
8
+ pretty_name: Arrhythmia
9
+ size_categories:
10
+ - 100<n<1K
11
+ task_categories: # Full list at https://github.com/huggingface/hub-docs/blob/main/js/src/lib/interfaces/Types.ts
12
+ - tabular-classification
13
+ configs:
14
+ - arrhytmia
15
  ---
16
+ # Arrhythmia
17
+ The [Arrhythmia dataset](https://archive.ics.uci.edu/ml/datasets/Arrhythmia) from the [UCI ML repository](https://archive.ics.uci.edu/ml/datasets).
18
+
19
+ # Configurations and tasks
20
+ | **Configuration** | **Task** | Description |
21
+ |-------------------|---------------------------|---------------------------------------------------------------|
22
+ | arrhytmia | Multiclass classification | Classify the arrhytmia type of the patient. |
23
+
24
+ # Usage
25
+ ```
26
+ from datasets import load_dataset
27
+ from sklearn.tree import DecisionTreeClassifier
28
+
29
+ dataset = load_dataset("mstz/arrhythmia", "arrhytmia")["train"]
30
+ ```
arrhythmia.data ADDED
The diff for this file is too large to render. See raw diff
 
arrhythmia.py ADDED
@@ -0,0 +1,354 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from typing import List
2
+ from functools import partial
3
+
4
+ import datasets
5
+
6
+ import pandas
7
+
8
+
9
+ VERSION = datasets.Version("1.0.0")
10
+
11
+ DESCRIPTION = "Arrhytmia dataset from the UCI ML repository."
12
+ _HOMEPAGE = "https://archive-beta.ics.uci.edu/dataset/5/arrhythmia"
13
+ _URLS = ("https://archive-beta.ics.uci.edu/dataset/5/arrhythmia")
14
+ _CITATION = """
15
+ @misc{misc_arrhythmia_5,
16
+ author = {Guvenir,H., Acar,Burak & Muderrisoglu,Haldun},
17
+ title = {{Arrhythmia}},
18
+ year = {1998},
19
+ howpublished = {UCI Machine Learning Repository},
20
+ note = {{DOI}: \\url{10.24432/C5BS32}}
21
+ }"""
22
+
23
+ # Dataset info
24
+ urls_per_split = {
25
+ "train": "https://huggingface.co/datasets/mstz/adult/raw/main/arrhythmia.data"
26
+ }
27
+ features_types_per_config = {
28
+ "arrhythmia": {
29
+ "age": datasets.Value("int8"),
30
+ "is_female": datasets.Value("bool"),
31
+ "height": datasets.Value("float64"),
32
+ "qrs_duration": datasets.Value("float64"),
33
+ "p_r_interval": datasets.Value("float64"),
34
+ "q_t_interval": datasets.Value("float64"),
35
+ "t_interval": datasets.Value("float64"),
36
+ "p_interval": datasets.Value("float64"),
37
+ "qrs": datasets.Value("float64"),
38
+ "t": datasets.Value("float64"),
39
+ "p": datasets.Value("float64"),
40
+ "qrst": datasets.Value("float64"),
41
+ "j": datasets.Value("float64"),
42
+ "heart_rate": datasets.Value("float64"),
43
+ "average_witdth_of_wave_q_on_DI": datasets.Value("float64"),
44
+ "average_witdth_of_wave_r_on_DI": datasets.Value("float64"),
45
+ "average_witdth_of_wave_s_on_DI": datasets.Value("float64"),
46
+ "average_witdth_of_wave_rprime_on_DI": datasets.Value("float64"),
47
+ "average_witdth_of_wave_sprime_on_DI": datasets.Value("float64"),
48
+ "number_of_deflections_on_DI": datasets.Value("float64"),
49
+ "type_of_ragged_r_wave_on_DI": datasets.Value("float64"),
50
+ "type_of_diphasic_derivation_of_r_wave_on_DI": datasets.Value("float64"),
51
+ "type_of_ragged_p_wave_on_DI": datasets.Value("float64"),
52
+ "type_of_diphasic_derivation_of_p_wave_on_DI": datasets.Value("float64"),
53
+ "type_of_ragged_t_wave_on_DI": datasets.Value("float64"),
54
+ "type_of_diphasic_derivation_of_t_wave_on_DI": datasets.Value("float64"),
55
+ "average_witdth_of_wave_q_on_DII": datasets.Value("float64"),
56
+ "average_witdth_of_wave_r_on_DII": datasets.Value("float64"),
57
+ "average_witdth_of_wave_s_on_DII": datasets.Value("float64"),
58
+ "average_witdth_of_wave_rprime_on_DII": datasets.Value("float64"),
59
+ "average_witdth_of_wave_sprime_on_DII": datasets.Value("float64"),
60
+ "number_of_deflections_on_DII": datasets.Value("float64"),
61
+ "type_of_ragged_r_wave_on_DII": datasets.Value("float64"),
62
+ "type_of_diphasic_derivation_of_r_wave_on_DII": datasets.Value("float64"),
63
+ "type_of_ragged_p_wave_on_DII": datasets.Value("float64"),
64
+ "type_of_diphasic_derivation_of_p_wave_on_DII": datasets.Value("float64"),
65
+ "type_of_ragged_t_wave_on_DII": datasets.Value("float64"),
66
+ "type_of_diphasic_derivation_of_t_wave_on_DII": datasets.Value("float64"),
67
+ "average_witdth_of_wave_q_on_DIII": datasets.Value("float64"),
68
+ "average_witdth_of_wave_r_on_DIII": datasets.Value("float64"),
69
+ "average_witdth_of_wave_s_on_DIII": datasets.Value("float64"),
70
+ "average_witdth_of_wave_rprime_on_DIII": datasets.Value("float64"),
71
+ "average_witdth_of_wave_sprime_on_DIII": datasets.Value("float64"),
72
+ "number_of_deflections_on_DIII": datasets.Value("float64"),
73
+ "type_of_ragged_r_wave_on_DIII": datasets.Value("float64"),
74
+ "type_of_diphasic_derivation_of_r_wave_on_DIII": datasets.Value("float64"),
75
+ "type_of_ragged_p_wave_on_DIII": datasets.Value("float64"),
76
+ "type_of_diphasic_derivation_of_p_wave_on_DIII": datasets.Value("float64"),
77
+ "type_of_ragged_t_wave_on_DIII": datasets.Value("float64"),
78
+ "type_of_diphasic_derivation_of_t_wave_on_DIII": datasets.Value("float64"),
79
+ "average_witdth_of_wave_q_on_AVR": datasets.Value("float64"),
80
+ "average_witdth_of_wave_r_on_AVR": datasets.Value("float64"),
81
+ "average_witdth_of_wave_s_on_AVR": datasets.Value("float64"),
82
+ "average_witdth_of_wave_rprime_on_AVR": datasets.Value("float64"),
83
+ "average_witdth_of_wave_sprime_on_AVR": datasets.Value("float64"),
84
+ "number_of_deflections_on_AVR": datasets.Value("float64"),
85
+ "type_of_ragged_r_wave_on_AVR": datasets.Value("float64"),
86
+ "type_of_diphasic_derivation_of_r_wave_on_AVR": datasets.Value("float64"),
87
+ "type_of_ragged_p_wave_on_AVR": datasets.Value("float64"),
88
+ "type_of_diphasic_derivation_of_p_wave_on_AVR": datasets.Value("float64"),
89
+ "type_of_ragged_t_wave_on_AVR": datasets.Value("float64"),
90
+ "type_of_diphasic_derivation_of_t_wave_on_AVR": datasets.Value("float64"),
91
+ "average_witdth_of_wave_q_on_AVL": datasets.Value("float64"),
92
+ "average_witdth_of_wave_r_on_AVL": datasets.Value("float64"),
93
+ "average_witdth_of_wave_s_on_AVL": datasets.Value("float64"),
94
+ "average_witdth_of_wave_rprime_on_AVL": datasets.Value("float64"),
95
+ "average_witdth_of_wave_sprime_on_AVL": datasets.Value("float64"),
96
+ "number_of_deflections_on_AVL": datasets.Value("float64"),
97
+ "type_of_ragged_r_wave_on_AVL": datasets.Value("float64"),
98
+ "type_of_diphasic_derivation_of_r_wave_on_AVL": datasets.Value("float64"),
99
+ "type_of_ragged_p_wave_on_AVL": datasets.Value("float64"),
100
+ "type_of_diphasic_derivation_of_p_wave_on_AVL": datasets.Value("float64"),
101
+ "type_of_ragged_t_wave_on_AVL": datasets.Value("float64"),
102
+ "type_of_diphasic_derivation_of_t_wave_on_AVL": datasets.Value("float64"),
103
+ "average_witdth_of_wave_q_on_AVF": datasets.Value("float64"),
104
+ "average_witdth_of_wave_r_on_AVF": datasets.Value("float64"),
105
+ "average_witdth_of_wave_s_on_AVF": datasets.Value("float64"),
106
+ "average_witdth_of_wave_rprime_on_AVF": datasets.Value("float64"),
107
+ "average_witdth_of_wave_sprime_on_AVF": datasets.Value("float64"),
108
+ "number_of_deflections_on_AVF": datasets.Value("float64"),
109
+ "type_of_ragged_r_wave_on_AVF": datasets.Value("float64"),
110
+ "type_of_diphasic_derivation_of_r_wave_on_AVF": datasets.Value("float64"),
111
+ "type_of_ragged_p_wave_on_AVF": datasets.Value("float64"),
112
+ "type_of_diphasic_derivation_of_p_wave_on_AVF": datasets.Value("float64"),
113
+ "type_of_ragged_t_wave_on_AVF": datasets.Value("float64"),
114
+ "type_of_diphasic_derivation_of_t_wave_on_AVF": datasets.Value("float64"),
115
+ "average_witdth_of_wave_q_on_V1": datasets.Value("float64"),
116
+ "average_witdth_of_wave_r_on_V1": datasets.Value("float64"),
117
+ "average_witdth_of_wave_s_on_V1": datasets.Value("float64"),
118
+ "average_witdth_of_wave_rprime_on_V1": datasets.Value("float64"),
119
+ "average_witdth_of_wave_sprime_on_V1": datasets.Value("float64"),
120
+ "number_of_deflections_on_V1": datasets.Value("float64"),
121
+ "type_of_ragged_r_wave_on_V1": datasets.Value("float64"),
122
+ "type_of_diphasic_derivation_of_r_wave_on_V1": datasets.Value("float64"),
123
+ "type_of_ragged_p_wave_on_V1": datasets.Value("float64"),
124
+ "type_of_diphasic_derivation_of_p_wave_on_V1": datasets.Value("float64"),
125
+ "type_of_ragged_t_wave_on_V1": datasets.Value("float64"),
126
+ "type_of_diphasic_derivation_of_t_wave_on_V1": datasets.Value("float64"),
127
+ "average_witdth_of_wave_q_on_V2": datasets.Value("float64"),
128
+ "average_witdth_of_wave_r_on_V2": datasets.Value("float64"),
129
+ "average_witdth_of_wave_s_on_V2": datasets.Value("float64"),
130
+ "average_witdth_of_wave_rprime_on_V2": datasets.Value("float64"),
131
+ "average_witdth_of_wave_sprime_on_V2": datasets.Value("float64"),
132
+ "number_of_deflections_on_V2": datasets.Value("float64"),
133
+ "type_of_ragged_r_wave_on_V2": datasets.Value("float64"),
134
+ "type_of_diphasic_derivation_of_r_wave_on_V2": datasets.Value("float64"),
135
+ "type_of_ragged_p_wave_on_V2": datasets.Value("float64"),
136
+ "type_of_diphasic_derivation_of_p_wave_on_V2": datasets.Value("float64"),
137
+ "type_of_ragged_t_wave_on_V2": datasets.Value("float64"),
138
+ "type_of_diphasic_derivation_of_t_wave_on_V2": datasets.Value("float64"),
139
+ "average_witdth_of_wave_q_on_V3": datasets.Value("float64"),
140
+ "average_witdth_of_wave_r_on_V3": datasets.Value("float64"),
141
+ "average_witdth_of_wave_s_on_V3": datasets.Value("float64"),
142
+ "average_witdth_of_wave_rprime_on_V3": datasets.Value("float64"),
143
+ "average_witdth_of_wave_sprime_on_V3": datasets.Value("float64"),
144
+ "number_of_deflections_on_V3": datasets.Value("float64"),
145
+ "type_of_ragged_r_wave_on_V3": datasets.Value("float64"),
146
+ "type_of_diphasic_derivation_of_r_wave_on_V3": datasets.Value("float64"),
147
+ "type_of_ragged_p_wave_on_V3": datasets.Value("float64"),
148
+ "type_of_diphasic_derivation_of_p_wave_on_V3": datasets.Value("float64"),
149
+ "type_of_ragged_t_wave_on_V3": datasets.Value("float64"),
150
+ "type_of_diphasic_derivation_of_t_wave_on_V3": datasets.Value("float64"),
151
+ "average_witdth_of_wave_q_on_V4": datasets.Value("float64"),
152
+ "average_witdth_of_wave_r_on_V4": datasets.Value("float64"),
153
+ "average_witdth_of_wave_s_on_V4": datasets.Value("float64"),
154
+ "average_witdth_of_wave_rprime_on_V4": datasets.Value("float64"),
155
+ "average_witdth_of_wave_sprime_on_V4": datasets.Value("float64"),
156
+ "number_of_deflections_on_V4": datasets.Value("float64"),
157
+ "type_of_ragged_r_wave_on_V4": datasets.Value("float64"),
158
+ "type_of_diphasic_derivation_of_r_wave_on_V4": datasets.Value("float64"),
159
+ "type_of_ragged_p_wave_on_V4": datasets.Value("float64"),
160
+ "type_of_diphasic_derivation_of_p_wave_on_V4": datasets.Value("float64"),
161
+ "type_of_ragged_t_wave_on_V4": datasets.Value("float64"),
162
+ "type_of_diphasic_derivation_of_t_wave_on_V4": datasets.Value("float64"),
163
+ "average_witdth_of_wave_q_on_V5": datasets.Value("float64"),
164
+ "average_witdth_of_wave_r_on_V5": datasets.Value("float64"),
165
+ "average_witdth_of_wave_s_on_V5": datasets.Value("float64"),
166
+ "average_witdth_of_wave_rprime_on_V5": datasets.Value("float64"),
167
+ "average_witdth_of_wave_sprime_on_V5": datasets.Value("float64"),
168
+ "number_of_deflections_on_V5": datasets.Value("float64"),
169
+ "type_of_ragged_r_wave_on_V5": datasets.Value("float64"),
170
+ "type_of_diphasic_derivation_of_r_wave_on_V5": datasets.Value("float64"),
171
+ "type_of_ragged_p_wave_on_V5": datasets.Value("float64"),
172
+ "type_of_diphasic_derivation_of_p_wave_on_V5": datasets.Value("float64"),
173
+ "type_of_ragged_t_wave_on_V5": datasets.Value("float64"),
174
+ "type_of_diphasic_derivation_of_t_wave_on_V5": datasets.Value("float64"),
175
+ "average_witdth_of_wave_q_on_V6": datasets.Value("float64"),
176
+ "average_witdth_of_wave_r_on_V6": datasets.Value("float64"),
177
+ "average_witdth_of_wave_s_on_V6": datasets.Value("float64"),
178
+ "average_witdth_of_wave_rprime_on_V6": datasets.Value("float64"),
179
+ "average_witdth_of_wave_sprime_on_V6": datasets.Value("float64"),
180
+ "number_of_deflections_on_V6": datasets.Value("float64"),
181
+ "type_of_ragged_r_wave_on_V6": datasets.Value("float64"),
182
+ "type_of_diphasic_derivation_of_r_wave_on_V6": datasets.Value("float64"),
183
+ "type_of_ragged_p_wave_on_V6": datasets.Value("float64"),
184
+ "type_of_diphasic_derivation_of_p_wave_on_V6": datasets.Value("float64"),
185
+ "type_of_ragged_t_wave_on_V6": datasets.Value("float64"),
186
+ "type_of_diphasic_derivation_of_t_wave_on_V6": datasets.Value("float64"),
187
+ "amplitude_of_jj_wave_on_DI": datasets.Value("float64"),
188
+ "amplitude_of_q_wave_on_DI": datasets.Value("float64"),
189
+ "amplitude_of_r_wave_on_DI": datasets.Value("float64"),
190
+ "amplitude_of_s_wave_on_DI": datasets.Value("float64"),
191
+ "amplitude_of_rprime_wave_on_DI": datasets.Value("float64"),
192
+ "amplitude_of_sprime_wave_on_DI": datasets.Value("float64"),
193
+ "amplitude_of_p_wave_on_DI": datasets.Value("float64"),
194
+ "amplitude_of_t_wave_on_DI": datasets.Value("float64"),
195
+ "qrsa_on_DI": datasets.Value("float64"),
196
+ "qrsta_on_DI": datasets.Value("float64"),
197
+ "amplitude_of_jj_wave_on_DII": datasets.Value("float64"),
198
+ "amplitude_of_q_wave_on_DII": datasets.Value("float64"),
199
+ "amplitude_of_r_wave_on_DII": datasets.Value("float64"),
200
+ "amplitude_of_s_wave_on_DII": datasets.Value("float64"),
201
+ "amplitude_of_rprime_wave_on_DII": datasets.Value("float64"),
202
+ "amplitude_of_sprime_wave_on_DII": datasets.Value("float64"),
203
+ "amplitude_of_p_wave_on_DII": datasets.Value("float64"),
204
+ "amplitude_of_t_wave_on_DII": datasets.Value("float64"),
205
+ "qrsa_on_DII": datasets.Value("float64"),
206
+ "qrsta_on_DII": datasets.Value("float64"),
207
+ "amplitude_of_jj_wave_on_DIII": datasets.Value("float64"),
208
+ "amplitude_of_q_wave_on_DIII": datasets.Value("float64"),
209
+ "amplitude_of_r_wave_on_DIII": datasets.Value("float64"),
210
+ "amplitude_of_s_wave_on_DIII": datasets.Value("float64"),
211
+ "amplitude_of_rprime_wave_on_DIII": datasets.Value("float64"),
212
+ "amplitude_of_sprime_wave_on_DIII": datasets.Value("float64"),
213
+ "amplitude_of_p_wave_on_DIII": datasets.Value("float64"),
214
+ "amplitude_of_t_wave_on_DIII": datasets.Value("float64"),
215
+ "qrsa_on_DIII": datasets.Value("float64"),
216
+ "qrsta_on_DIII": datasets.Value("float64"),
217
+ "amplitude_of_jj_wave_on_AVR": datasets.Value("float64"),
218
+ "amplitude_of_q_wave_on_AVR": datasets.Value("float64"),
219
+ "amplitude_of_r_wave_on_AVR": datasets.Value("float64"),
220
+ "amplitude_of_s_wave_on_AVR": datasets.Value("float64"),
221
+ "amplitude_of_rprime_wave_on_AVR": datasets.Value("float64"),
222
+ "amplitude_of_sprime_wave_on_AVR": datasets.Value("float64"),
223
+ "amplitude_of_p_wave_on_AVR": datasets.Value("float64"),
224
+ "amplitude_of_t_wave_on_AVR": datasets.Value("float64"),
225
+ "qrsa_on_AVR": datasets.Value("float64"),
226
+ "qrsta_on_AVR": datasets.Value("float64"),
227
+ "amplitude_of_jj_wave_on_AVL": datasets.Value("float64"),
228
+ "amplitude_of_q_wave_on_AVL": datasets.Value("float64"),
229
+ "amplitude_of_r_wave_on_AVL": datasets.Value("float64"),
230
+ "amplitude_of_s_wave_on_AVL": datasets.Value("float64"),
231
+ "amplitude_of_rprime_wave_on_AVL": datasets.Value("float64"),
232
+ "amplitude_of_sprime_wave_on_AVL": datasets.Value("float64"),
233
+ "amplitude_of_p_wave_on_AVL": datasets.Value("float64"),
234
+ "amplitude_of_t_wave_on_AVL": datasets.Value("float64"),
235
+ "qrsa_on_AVL": datasets.Value("float64"),
236
+ "qrsta_on_AVL": datasets.Value("float64"),
237
+ "amplitude_of_jj_wave_on_AVF": datasets.Value("float64"),
238
+ "amplitude_of_q_wave_on_AVF": datasets.Value("float64"),
239
+ "amplitude_of_r_wave_on_AVF": datasets.Value("float64"),
240
+ "amplitude_of_s_wave_on_AVF": datasets.Value("float64"),
241
+ "amplitude_of_rprime_wave_on_AVF": datasets.Value("float64"),
242
+ "amplitude_of_sprime_wave_on_AVF": datasets.Value("float64"),
243
+ "amplitude_of_p_wave_on_AVF": datasets.Value("float64"),
244
+ "amplitude_of_t_wave_on_AVF": datasets.Value("float64"),
245
+ "qrsa_on_AVF": datasets.Value("float64"),
246
+ "qrsta_on_AVF": datasets.Value("float64"),
247
+ "amplitude_of_jj_wave_on_V1": datasets.Value("float64"),
248
+ "amplitude_of_q_wave_on_V1": datasets.Value("float64"),
249
+ "amplitude_of_r_wave_on_V1": datasets.Value("float64"),
250
+ "amplitude_of_s_wave_on_V1": datasets.Value("float64"),
251
+ "amplitude_of_rprime_wave_on_V1": datasets.Value("float64"),
252
+ "amplitude_of_sprime_wave_on_V1": datasets.Value("float64"),
253
+ "amplitude_of_p_wave_on_V1": datasets.Value("float64"),
254
+ "amplitude_of_t_wave_on_V1": datasets.Value("float64"),
255
+ "qrsa_on_V1": datasets.Value("float64"),
256
+ "qrsta_on_V1": datasets.Value("float64"),
257
+ "amplitude_of_jj_wave_on_V2": datasets.Value("float64"),
258
+ "amplitude_of_q_wave_on_V2": datasets.Value("float64"),
259
+ "amplitude_of_r_wave_on_V2": datasets.Value("float64"),
260
+ "amplitude_of_s_wave_on_V2": datasets.Value("float64"),
261
+ "amplitude_of_rprime_wave_on_V2": datasets.Value("float64"),
262
+ "amplitude_of_sprime_wave_on_V2": datasets.Value("float64"),
263
+ "amplitude_of_p_wave_on_V2": datasets.Value("float64"),
264
+ "amplitude_of_t_wave_on_V2": datasets.Value("float64"),
265
+ "qrsa_on_V2": datasets.Value("float64"),
266
+ "qrsta_on_V2": datasets.Value("float64"),
267
+ "amplitude_of_jj_wave_on_V3": datasets.Value("float64"),
268
+ "amplitude_of_q_wave_on_V3": datasets.Value("float64"),
269
+ "amplitude_of_r_wave_on_V3": datasets.Value("float64"),
270
+ "amplitude_of_s_wave_on_V3": datasets.Value("float64"),
271
+ "amplitude_of_rprime_wave_on_V3": datasets.Value("float64"),
272
+ "amplitude_of_sprime_wave_on_V3": datasets.Value("float64"),
273
+ "amplitude_of_p_wave_on_V3": datasets.Value("float64"),
274
+ "amplitude_of_t_wave_on_V3": datasets.Value("float64"),
275
+ "qrsa_on_V3": datasets.Value("float64"),
276
+ "qrsta_on_V3": datasets.Value("float64"),
277
+ "amplitude_of_jj_wave_on_V4": datasets.Value("float64"),
278
+ "amplitude_of_q_wave_on_V4": datasets.Value("float64"),
279
+ "amplitude_of_r_wave_on_V4": datasets.Value("float64"),
280
+ "amplitude_of_s_wave_on_V4": datasets.Value("float64"),
281
+ "amplitude_of_rprime_wave_on_V4": datasets.Value("float64"),
282
+ "amplitude_of_sprime_wave_on_V4": datasets.Value("float64"),
283
+ "amplitude_of_p_wave_on_V4": datasets.Value("float64"),
284
+ "amplitude_of_t_wave_on_V4": datasets.Value("float64"),
285
+ "qrsa_on_V4": datasets.Value("float64"),
286
+ "qrsta_on_V4": datasets.Value("float64"),
287
+ "amplitude_of_jj_wave_on_V5": datasets.Value("float64"),
288
+ "amplitude_of_q_wave_on_V5": datasets.Value("float64"),
289
+ "amplitude_of_r_wave_on_V5": datasets.Value("float64"),
290
+ "amplitude_of_s_wave_on_V5": datasets.Value("float64"),
291
+ "amplitude_of_rprime_wave_on_V5": datasets.Value("float64"),
292
+ "amplitude_of_sprime_wave_on_V5": datasets.Value("float64"),
293
+ "amplitude_of_p_wave_on_V5": datasets.Value("float64"),
294
+ "amplitude_of_t_wave_on_V5": datasets.Value("float64"),
295
+ "qrsa_on_V5": datasets.Value("float64"),
296
+ "qrsta_on_V5": datasets.Value("float64"),
297
+ "amplitude_of_jj_wave_on_V6": datasets.Value("float64"),
298
+ "amplitude_of_q_wave_on_V6": datasets.Value("float64"),
299
+ "amplitude_of_r_wave_on_V6": datasets.Value("float64"),
300
+ "amplitude_of_s_wave_on_V6": datasets.Value("float64"),
301
+ "amplitude_of_rprime_wave_on_V6": datasets.Value("float64"),
302
+ "amplitude_of_sprime_wave_on_V6": datasets.Value("float64"),
303
+ "amplitude_of_p_wave_on_V6": datasets.Value("float64"),
304
+ "amplitude_of_t_wave_on_V6": datasets.Value("float64"),
305
+ "qrsa_on_V6": datasets.Value("float64"),
306
+ "qrsta_on_V6": datasets.Value("float64"),
307
+ "type_of_arrhytmia": datasets.ClassLabel(num_classes=16)
308
+ }
309
+ }
310
+ features_per_config = {k: datasets.Features(features_types_per_config[k]) for k in features_types_per_config}
311
+
312
+
313
+ class ArrhythmiaConfig(datasets.BuilderConfig):
314
+ def __init__(self, **kwargs):
315
+ super(ArrhythmiaConfig, self).__init__(version=VERSION, **kwargs)
316
+ self.features = features_per_config[kwargs["name"]]
317
+
318
+
319
+ class Arrhythmia(datasets.GeneratorBasedBuilder):
320
+ # dataset versions
321
+ DEFAULT_CONFIG = "arrhytmia"
322
+ BUILDER_CONFIGS = [
323
+ ArrhythmiaConfig(name="arrhytmia",
324
+ description="Multiclass classification of arrhytmia type."),
325
+ ]
326
+
327
+
328
+ def _info(self):
329
+ info = datasets.DatasetInfo(description=DESCRIPTION, citation=_CITATION, homepage=_HOMEPAGE,
330
+ features=features_per_config[self.config.name])
331
+
332
+ return info
333
+
334
+ def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
335
+ downloads = dl_manager.download_and_extract(urls_per_split)
336
+
337
+ return [
338
+ datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloads["train"]})
339
+ ]
340
+
341
+ def _generate_examples(self, filepath: str):
342
+ data = pandas.read_csv(filepath)
343
+ data = self.preprocess(data, config=self.config.name)
344
+
345
+ for row_id, row in data.iterrows():
346
+ data_row = dict(row)
347
+
348
+ yield row_id, data_row
349
+
350
+ def preprocess(self, data: pandas.DataFrame, config: str = DEFAULT_CONFIG) -> pandas.DataFrame:
351
+ data = data[(data.t != "?") & (data.qrst != "?") & (data.p != "?") & (data.j != "?") & (data.heart_rate != "?")]
352
+ data = data.astype({"t": int, "qrst": int, "p": int, "j": int, "heart_rate": int})
353
+
354
+ return data[list(features_types_per_config["income"].keys())]