cassiebuhler commited on
Commit
4a9207f
·
1 Parent(s): 899836e

fixed 2020 and state data

Browse files
Files changed (1) hide show
  1. preprocess.ipynb +9 -124
preprocess.ipynb CHANGED
@@ -10,7 +10,7 @@
10
  },
11
  {
12
  "cell_type": "code",
13
- "execution_count": 1,
14
  "id": "6452373c-d10f-472c-9174-cd05a3363587",
15
  "metadata": {},
16
  "outputs": [],
@@ -30,7 +30,7 @@
30
  },
31
  {
32
  "cell_type": "code",
33
- "execution_count": 2,
34
  "id": "dd9ffc64-b4cf-4e5b-9c96-703e91a77837",
35
  "metadata": {},
36
  "outputs": [],
@@ -74,7 +74,7 @@
74
  },
75
  {
76
  "cell_type": "code",
77
- "execution_count": 3,
78
  "id": "8fc40e96-fffd-4b23-9963-c931fdce96f6",
79
  "metadata": {},
80
  "outputs": [],
@@ -105,21 +105,10 @@
105
  },
106
  {
107
  "cell_type": "code",
108
- "execution_count": 4,
109
  "id": "b1cd8f44-57fa-49a8-b438-f9e4aab747c5",
110
  "metadata": {},
111
- "outputs": [
112
- {
113
- "name": "stdout",
114
- "output_type": "stream",
115
- "text": [
116
- "The token has not been saved to the git credentials helper. Pass `add_to_git_credential=True` in this function directly or `--add-to-git-credential` if using via `huggingface-cli` if you want to set the git credential as well.\n",
117
- "Token is valid (permission: write).\n",
118
- "Your token has been saved to /home/rstudio/.cache/huggingface/token\n",
119
- "Login successful\n"
120
- ]
121
- }
122
- ],
123
  "source": [
124
  "import subprocess\n",
125
  "import os\n",
@@ -165,85 +154,10 @@
165
  },
166
  {
167
  "cell_type": "code",
168
- "execution_count": 5,
169
  "id": "7061577e-0632-4395-8ac5-241a1fab53b0",
170
  "metadata": {},
171
- "outputs": [
172
- {
173
- "data": {
174
- "application/vnd.jupyter.widget-view+json": {
175
- "model_id": "341b737abd4c4fceae622965133c09a2",
176
- "version_major": 2,
177
- "version_minor": 0
178
- },
179
- "text/plain": [
180
- "FloatProgress(value=0.0, layout=Layout(width='auto'), style=ProgressStyle(bar_color='black'))"
181
- ]
182
- },
183
- "metadata": {},
184
- "output_type": "display_data"
185
- },
186
- {
187
- "data": {
188
- "application/vnd.jupyter.widget-view+json": {
189
- "model_id": "859a3d133ac1491f8841d61bc2da25bf",
190
- "version_major": 2,
191
- "version_minor": 0
192
- },
193
- "text/plain": [
194
- "FloatProgress(value=0.0, layout=Layout(width='auto'), style=ProgressStyle(bar_color='black'))"
195
- ]
196
- },
197
- "metadata": {},
198
- "output_type": "display_data"
199
- },
200
- {
201
- "data": {
202
- "application/vnd.jupyter.widget-view+json": {
203
- "model_id": "b0e18752ba4143bf88c8b447e521e8a6",
204
- "version_major": 2,
205
- "version_minor": 0
206
- },
207
- "text/plain": [
208
- "FloatProgress(value=0.0, layout=Layout(width='auto'), style=ProgressStyle(bar_color='black'))"
209
- ]
210
- },
211
- "metadata": {},
212
- "output_type": "display_data"
213
- },
214
- {
215
- "name": "stderr",
216
- "output_type": "stream",
217
- "text": [
218
- "1576 features, 14118861 bytes of geometry and attributes, 45035 bytes of string pool, 0 bytes of vertices, 0 bytes of nodes\n",
219
- "Choosing a maxzoom of -z3 for features typically 91177 feet (27791 meters) apart, and at least 11358 feet (3462 meters) apart\n",
220
- "Choosing a maxzoom of -z10 for resolution of about 308 feet (93 meters) within features\n",
221
- " 99.9% 10/261/364 \n",
222
- " 100.0% 10/55/448 \r"
223
- ]
224
- },
225
- {
226
- "name": "stdout",
227
- "output_type": "stream",
228
- "text": [
229
- "Successfully generated PMTiles file: votes.pmtiles\n"
230
- ]
231
- },
232
- {
233
- "data": {
234
- "application/vnd.jupyter.widget-view+json": {
235
- "model_id": "17feadfbbbc54429b56aa85a8087310b",
236
- "version_major": 2,
237
- "version_minor": 0
238
- },
239
- "text/plain": [
240
- "votes.pmtiles: 0%| | 0.00/4.08M [00:00<?, ?B/s]"
241
- ]
242
- },
243
- "metadata": {},
244
- "output_type": "display_data"
245
- }
246
- ],
247
  "source": [
248
  "gdf_state = votes.filter(_.jurisdiction == 'State').execute().set_crs(\"EPSG:4326\")\n",
249
  "gdf_state.to_file(\"votes_state.geojson\")\n",
@@ -260,39 +174,10 @@
260
  },
261
  {
262
  "cell_type": "code",
263
- "execution_count": 6,
264
  "id": "f2979624-bcdf-4a8a-899a-c22fc3cdaf0e",
265
  "metadata": {},
266
- "outputs": [
267
- {
268
- "data": {
269
- "application/vnd.jupyter.widget-view+json": {
270
- "model_id": "cc5d666af6bd4f9da53710f207f6103b",
271
- "version_major": 2,
272
- "version_minor": 0
273
- },
274
- "text/plain": [
275
- "FloatProgress(value=0.0, layout=Layout(width='auto'), style=ProgressStyle(bar_color='black'))"
276
- ]
277
- },
278
- "metadata": {},
279
- "output_type": "display_data"
280
- },
281
- {
282
- "data": {
283
- "application/vnd.jupyter.widget-view+json": {
284
- "model_id": "9a0c8f8fb5bd4486af8220e633cd9c97",
285
- "version_major": 2,
286
- "version_minor": 0
287
- },
288
- "text/plain": [
289
- "votes.parquet: 0%| | 0.00/31.7M [00:00<?, ?B/s]"
290
- ]
291
- },
292
- "metadata": {},
293
- "output_type": "display_data"
294
- }
295
- ],
296
  "source": [
297
  "# save as parquet\n",
298
  "votes.execute().set_crs(\"EPSG:4326\").to_parquet(\"votes.parquet\")\n",
 
10
  },
11
  {
12
  "cell_type": "code",
13
+ "execution_count": null,
14
  "id": "6452373c-d10f-472c-9174-cd05a3363587",
15
  "metadata": {},
16
  "outputs": [],
 
30
  },
31
  {
32
  "cell_type": "code",
33
+ "execution_count": null,
34
  "id": "dd9ffc64-b4cf-4e5b-9c96-703e91a77837",
35
  "metadata": {},
36
  "outputs": [],
 
74
  },
75
  {
76
  "cell_type": "code",
77
+ "execution_count": null,
78
  "id": "8fc40e96-fffd-4b23-9963-c931fdce96f6",
79
  "metadata": {},
80
  "outputs": [],
 
105
  },
106
  {
107
  "cell_type": "code",
108
+ "execution_count": null,
109
  "id": "b1cd8f44-57fa-49a8-b438-f9e4aab747c5",
110
  "metadata": {},
111
+ "outputs": [],
 
 
 
 
 
 
 
 
 
 
 
112
  "source": [
113
  "import subprocess\n",
114
  "import os\n",
 
154
  },
155
  {
156
  "cell_type": "code",
157
+ "execution_count": null,
158
  "id": "7061577e-0632-4395-8ac5-241a1fab53b0",
159
  "metadata": {},
160
+ "outputs": [],
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
161
  "source": [
162
  "gdf_state = votes.filter(_.jurisdiction == 'State').execute().set_crs(\"EPSG:4326\")\n",
163
  "gdf_state.to_file(\"votes_state.geojson\")\n",
 
174
  },
175
  {
176
  "cell_type": "code",
177
+ "execution_count": null,
178
  "id": "f2979624-bcdf-4a8a-899a-c22fc3cdaf0e",
179
  "metadata": {},
180
+ "outputs": [],
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
181
  "source": [
182
  "# save as parquet\n",
183
  "votes.execute().set_crs(\"EPSG:4326\").to_parquet(\"votes.parquet\")\n",