Datasets:
Agregar sección dataset
Browse files
README.md
CHANGED
@@ -34,6 +34,240 @@ task_ids:
|
|
34 |
En esta actividad intentaremos resolver un problema de clasificación multimodal. En un problema de clasificación multimodal, cada pieza de información viene en diferentes representaciones (imágenes, texto, audios, etc) y la idea es determinar cómo usar esos datos para un problema de clasificación.
|
35 |
En este caso trabajaremos con un dataset que contiene datos sobre especies de pájaros.
|
36 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
37 |
## Problema
|
38 |
|
39 |
El problema consiste en entrenar un modelo que clasifique instancias del dataset CUB de la mejor manera posible. Algunas preguntas que podrían guiar nuestro desarrollo son:
|
|
|
34 |
En esta actividad intentaremos resolver un problema de clasificación multimodal. En un problema de clasificación multimodal, cada pieza de información viene en diferentes representaciones (imágenes, texto, audios, etc) y la idea es determinar cómo usar esos datos para un problema de clasificación.
|
35 |
En este caso trabajaremos con un dataset que contiene datos sobre especies de pájaros.
|
36 |
|
37 |
+
## Dataset
|
38 |
+
|
39 |
+
|
40 |
+
### Data Instances
|
41 |
+
|
42 |
+
Una muestra del _dataset_ se encuentra a continuación:
|
43 |
+
|
44 |
+
```
|
45 |
+
{'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=334x500 at 0x7F59DE348AF0>,
|
46 |
+
'description': 'this bird has a short orange bill, white breast and body and white eyes.\na medium sized bird with a orange bill and a black crown and white eyes\nthis white-breasted bird has a short, squat, orange bill, a black head and wings, and small white eyes above a white stripe.\nthis bird has a white breast, a black head, a short red beak, and webbed feet.\nthis bird is white with black on its neck and has a long, pointy beak.\nthis bird has wings that are black and has a white belly\nthis bird has wings that are black and has a long bill\nthis is a medium sized bird, with a white belly, and a grey head and wings, with a short yellow bill.\nthis bird is white and gray in color, and has a bright orange beak.\nthis bird has a blunt orange beak with mostly black above the neck, the belly is solid white.\n',
|
47 |
+
'label': 6,
|
48 |
+
'file_name': 'Parakeet_Auklet_0048_795980.jpg'}
|
49 |
+
```
|
50 |
+
|
51 |
+
### Data Fields
|
52 |
+
|
53 |
+
Cada instancia de datos tiene los siguientes campos:
|
54 |
+
|
55 |
+
- `image`: imagen RGB de un pájaro
|
56 |
+
- `description`: texto con 10 descripciones del pájaro en la foto, cada descripción esta separado por un salto de linea (i.e. `\n`)
|
57 |
+
- `label`: un número entero que representa el id de la especie a la que pertenece el pájaro
|
58 |
+
<details>
|
59 |
+
<summary>Id2String</summary>
|
60 |
+
```txt
|
61 |
+
1 001.Black_footed_Albatross
|
62 |
+
2 002.Laysan_Albatross
|
63 |
+
3 003.Sooty_Albatross
|
64 |
+
4 004.Groove_billed_Ani
|
65 |
+
5 005.Crested_Auklet
|
66 |
+
6 006.Least_Auklet
|
67 |
+
7 007.Parakeet_Auklet
|
68 |
+
8 008.Rhinoceros_Auklet
|
69 |
+
9 009.Brewer_Blackbird
|
70 |
+
10 010.Red_winged_Blackbird
|
71 |
+
11 011.Rusty_Blackbird
|
72 |
+
12 012.Yellow_headed_Blackbird
|
73 |
+
13 013.Bobolink
|
74 |
+
14 014.Indigo_Bunting
|
75 |
+
15 015.Lazuli_Bunting
|
76 |
+
16 016.Painted_Bunting
|
77 |
+
17 017.Cardinal
|
78 |
+
18 018.Spotted_Catbird
|
79 |
+
19 019.Gray_Catbird
|
80 |
+
20 020.Yellow_breasted_Chat
|
81 |
+
21 021.Eastern_Towhee
|
82 |
+
22 022.Chuck_will_Widow
|
83 |
+
23 023.Brandt_Cormorant
|
84 |
+
24 024.Red_faced_Cormorant
|
85 |
+
25 025.Pelagic_Cormorant
|
86 |
+
26 026.Bronzed_Cowbird
|
87 |
+
27 027.Shiny_Cowbird
|
88 |
+
28 028.Brown_Creeper
|
89 |
+
29 029.American_Crow
|
90 |
+
30 030.Fish_Crow
|
91 |
+
31 031.Black_billed_Cuckoo
|
92 |
+
32 032.Mangrove_Cuckoo
|
93 |
+
33 033.Yellow_billed_Cuckoo
|
94 |
+
34 034.Gray_crowned_Rosy_Finch
|
95 |
+
35 035.Purple_Finch
|
96 |
+
36 036.Northern_Flicker
|
97 |
+
37 037.Acadian_Flycatcher
|
98 |
+
38 038.Great_Crested_Flycatcher
|
99 |
+
39 039.Least_Flycatcher
|
100 |
+
40 040.Olive_sided_Flycatcher
|
101 |
+
41 041.Scissor_tailed_Flycatcher
|
102 |
+
42 042.Vermilion_Flycatcher
|
103 |
+
43 043.Yellow_bellied_Flycatcher
|
104 |
+
44 044.Frigatebird
|
105 |
+
45 045.Northern_Fulmar
|
106 |
+
46 046.Gadwall
|
107 |
+
47 047.American_Goldfinch
|
108 |
+
48 048.European_Goldfinch
|
109 |
+
49 049.Boat_tailed_Grackle
|
110 |
+
50 050.Eared_Grebe
|
111 |
+
51 051.Horned_Grebe
|
112 |
+
52 052.Pied_billed_Grebe
|
113 |
+
53 053.Western_Grebe
|
114 |
+
54 054.Blue_Grosbeak
|
115 |
+
55 055.Evening_Grosbeak
|
116 |
+
56 056.Pine_Grosbeak
|
117 |
+
57 057.Rose_breasted_Grosbeak
|
118 |
+
58 058.Pigeon_Guillemot
|
119 |
+
59 059.California_Gull
|
120 |
+
60 060.Glaucous_winged_Gull
|
121 |
+
61 061.Heermann_Gull
|
122 |
+
62 062.Herring_Gull
|
123 |
+
63 063.Ivory_Gull
|
124 |
+
64 064.Ring_billed_Gull
|
125 |
+
65 065.Slaty_backed_Gull
|
126 |
+
66 066.Western_Gull
|
127 |
+
67 067.Anna_Hummingbird
|
128 |
+
68 068.Ruby_throated_Hummingbird
|
129 |
+
69 069.Rufous_Hummingbird
|
130 |
+
70 070.Green_Violetear
|
131 |
+
71 071.Long_tailed_Jaeger
|
132 |
+
72 072.Pomarine_Jaeger
|
133 |
+
73 073.Blue_Jay
|
134 |
+
74 074.Florida_Jay
|
135 |
+
75 075.Green_Jay
|
136 |
+
76 076.Dark_eyed_Junco
|
137 |
+
77 077.Tropical_Kingbird
|
138 |
+
78 078.Gray_Kingbird
|
139 |
+
79 079.Belted_Kingfisher
|
140 |
+
80 080.Green_Kingfisher
|
141 |
+
81 081.Pied_Kingfisher
|
142 |
+
82 082.Ringed_Kingfisher
|
143 |
+
83 083.White_breasted_Kingfisher
|
144 |
+
84 084.Red_legged_Kittiwake
|
145 |
+
85 085.Horned_Lark
|
146 |
+
86 086.Pacific_Loon
|
147 |
+
87 087.Mallard
|
148 |
+
88 088.Western_Meadowlark
|
149 |
+
89 089.Hooded_Merganser
|
150 |
+
90 090.Red_breasted_Merganser
|
151 |
+
91 091.Mockingbird
|
152 |
+
92 092.Nighthawk
|
153 |
+
93 093.Clark_Nutcracker
|
154 |
+
94 094.White_breasted_Nuthatch
|
155 |
+
95 095.Baltimore_Oriole
|
156 |
+
96 096.Hooded_Oriole
|
157 |
+
97 097.Orchard_Oriole
|
158 |
+
98 098.Scott_Oriole
|
159 |
+
99 099.Ovenbird
|
160 |
+
100 100.Brown_Pelican
|
161 |
+
101 101.White_Pelican
|
162 |
+
102 102.Western_Wood_Pewee
|
163 |
+
103 103.Sayornis
|
164 |
+
104 104.American_Pipit
|
165 |
+
105 105.Whip_poor_Will
|
166 |
+
106 106.Horned_Puffin
|
167 |
+
107 107.Common_Raven
|
168 |
+
108 108.White_necked_Raven
|
169 |
+
109 109.American_Redstart
|
170 |
+
110 110.Geococcyx
|
171 |
+
111 111.Loggerhead_Shrike
|
172 |
+
112 112.Great_Grey_Shrike
|
173 |
+
113 113.Baird_Sparrow
|
174 |
+
114 114.Black_throated_Sparrow
|
175 |
+
115 115.Brewer_Sparrow
|
176 |
+
116 116.Chipping_Sparrow
|
177 |
+
117 117.Clay_colored_Sparrow
|
178 |
+
118 118.House_Sparrow
|
179 |
+
119 119.Field_Sparrow
|
180 |
+
120 120.Fox_Sparrow
|
181 |
+
121 121.Grasshopper_Sparrow
|
182 |
+
122 122.Harris_Sparrow
|
183 |
+
123 123.Henslow_Sparrow
|
184 |
+
124 124.Le_Conte_Sparrow
|
185 |
+
125 125.Lincoln_Sparrow
|
186 |
+
126 126.Nelson_Sharp_tailed_Sparrow
|
187 |
+
127 127.Savannah_Sparrow
|
188 |
+
128 128.Seaside_Sparrow
|
189 |
+
129 129.Song_Sparrow
|
190 |
+
130 130.Tree_Sparrow
|
191 |
+
131 131.Vesper_Sparrow
|
192 |
+
132 132.White_crowned_Sparrow
|
193 |
+
133 133.White_throated_Sparrow
|
194 |
+
134 134.Cape_Glossy_Starling
|
195 |
+
135 135.Bank_Swallow
|
196 |
+
136 136.Barn_Swallow
|
197 |
+
137 137.Cliff_Swallow
|
198 |
+
138 138.Tree_Swallow
|
199 |
+
139 139.Scarlet_Tanager
|
200 |
+
140 140.Summer_Tanager
|
201 |
+
141 141.Artic_Tern
|
202 |
+
142 142.Black_Tern
|
203 |
+
143 143.Caspian_Tern
|
204 |
+
144 144.Common_Tern
|
205 |
+
145 145.Elegant_Tern
|
206 |
+
146 146.Forsters_Tern
|
207 |
+
147 147.Least_Tern
|
208 |
+
148 148.Green_tailed_Towhee
|
209 |
+
149 149.Brown_Thrasher
|
210 |
+
150 150.Sage_Thrasher
|
211 |
+
151 151.Black_capped_Vireo
|
212 |
+
152 152.Blue_headed_Vireo
|
213 |
+
153 153.Philadelphia_Vireo
|
214 |
+
154 154.Red_eyed_Vireo
|
215 |
+
155 155.Warbling_Vireo
|
216 |
+
156 156.White_eyed_Vireo
|
217 |
+
157 157.Yellow_throated_Vireo
|
218 |
+
158 158.Bay_breasted_Warbler
|
219 |
+
159 159.Black_and_white_Warbler
|
220 |
+
160 160.Black_throated_Blue_Warbler
|
221 |
+
161 161.Blue_winged_Warbler
|
222 |
+
162 162.Canada_Warbler
|
223 |
+
163 163.Cape_May_Warbler
|
224 |
+
164 164.Cerulean_Warbler
|
225 |
+
165 165.Chestnut_sided_Warbler
|
226 |
+
166 166.Golden_winged_Warbler
|
227 |
+
167 167.Hooded_Warbler
|
228 |
+
168 168.Kentucky_Warbler
|
229 |
+
169 169.Magnolia_Warbler
|
230 |
+
170 170.Mourning_Warbler
|
231 |
+
171 171.Myrtle_Warbler
|
232 |
+
172 172.Nashville_Warbler
|
233 |
+
173 173.Orange_crowned_Warbler
|
234 |
+
174 174.Palm_Warbler
|
235 |
+
175 175.Pine_Warbler
|
236 |
+
176 176.Prairie_Warbler
|
237 |
+
177 177.Prothonotary_Warbler
|
238 |
+
178 178.Swainson_Warbler
|
239 |
+
179 179.Tennessee_Warbler
|
240 |
+
180 180.Wilson_Warbler
|
241 |
+
181 181.Worm_eating_Warbler
|
242 |
+
182 182.Yellow_Warbler
|
243 |
+
183 183.Northern_Waterthrush
|
244 |
+
184 184.Louisiana_Waterthrush
|
245 |
+
185 185.Bohemian_Waxwing
|
246 |
+
186 186.Cedar_Waxwing
|
247 |
+
187 187.American_Three_toed_Woodpecker
|
248 |
+
188 188.Pileated_Woodpecker
|
249 |
+
189 189.Red_bellied_Woodpecker
|
250 |
+
190 190.Red_cockaded_Woodpecker
|
251 |
+
191 191.Red_headed_Woodpecker
|
252 |
+
192 192.Downy_Woodpecker
|
253 |
+
193 193.Bewick_Wren
|
254 |
+
194 194.Cactus_Wren
|
255 |
+
195 195.Carolina_Wren
|
256 |
+
196 196.House_Wren
|
257 |
+
197 197.Marsh_Wren
|
258 |
+
198 198.Rock_Wren
|
259 |
+
199 199.Winter_Wren
|
260 |
+
200 200.Common_Yellowthroat
|
261 |
+
```
|
262 |
+
</details>
|
263 |
+
- `file_name`: nombre del archivo que tiene la imagen
|
264 |
+
|
265 |
+
### Data Splits
|
266 |
+
|
267 |
+
| |train| test|
|
268 |
+
|-------------|----:|----:|
|
269 |
+
|# of examples|5994|5794|
|
270 |
+
|
271 |
## Problema
|
272 |
|
273 |
El problema consiste en entrenar un modelo que clasifique instancias del dataset CUB de la mejor manera posible. Algunas preguntas que podrían guiar nuestro desarrollo son:
|