Shankhdhar
commited on
Commit
•
5944e36
1
Parent(s):
ec34615
Saving weights and logs of step 100
Browse files- .gitattributes +3 -0
- Train.tsv +3 -0
- Untitled.ipynb +356 -134
- Val.tsv +0 -0
- events.out.tfevents.1626357469.t1v-n-1926f308-w-0.95268.3.v2 +2 -2
- events.out.tfevents.1626357733.t1v-n-1926f308-w-0.96764.3.v2 +2 -2
- events.out.tfevents.1626358530.t1v-n-1926f308-w-0.97646.3.v2 +3 -0
- events.out.tfevents.1626359662.t1v-n-1926f308-w-0.99131.3.v2 +3 -0
- events.out.tfevents.1626359960.t1v-n-1926f308-w-0.100528.3.v2 +3 -0
- events.out.tfevents.1626447876.t1v-n-1926f308-w-0.111464.3.v2 +3 -0
- events.out.tfevents.1626451747.t1v-n-1926f308-w-0.115707.3.v2 +3 -0
- events.out.tfevents.1626451941.t1v-n-1926f308-w-0.117207.3.v2 +3 -0
- flax_model.msgpack +1 -1
- flowmasteri.txt +0 -0
- run.sh +7 -7
- run_clm_flax.py +5 -3
- text_collection/text_collection.py +104 -0
- text_collection/text_collection.py.lock +0 -0
- train/train2.txt +3 -0
- train/val2.txt +0 -0
- val/val2.txt +0 -0
- wutang.txt +0 -0
- ye.txt +0 -0
- zenske.txt +0 -0
.gitattributes
CHANGED
@@ -15,3 +15,6 @@
|
|
15 |
*.pt filter=lfs diff=lfs merge=lfs -text
|
16 |
*.pth filter=lfs diff=lfs merge=lfs -text
|
17 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
15 |
*.pt filter=lfs diff=lfs merge=lfs -text
|
16 |
*.pth filter=lfs diff=lfs merge=lfs -text
|
17 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
18 |
+
Train.tsv filter=lfs diff=lfs merge=lfs -text
|
19 |
+
train/train2.tsv filter=lfs diff=lfs merge=lfs -text
|
20 |
+
train/train2.txt filter=lfs diff=lfs merge=lfs -text
|
Train.tsv
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:60458763e1bfeeeb3239b0a1b58bbdb329b50dffdf5508500266350461fc6bdf
|
3 |
+
size 19817903
|
Untitled.ipynb
CHANGED
@@ -2,17 +2,19 @@
|
|
2 |
"cells": [
|
3 |
{
|
4 |
"cell_type": "code",
|
5 |
-
"execution_count":
|
6 |
"id": "81fd300c",
|
7 |
"metadata": {},
|
8 |
"outputs": [],
|
9 |
"source": [
|
10 |
-
"import pandas as pd\n"
|
|
|
|
|
11 |
]
|
12 |
},
|
13 |
{
|
14 |
"cell_type": "code",
|
15 |
-
"execution_count":
|
16 |
"id": "1237ddf7",
|
17 |
"metadata": {},
|
18 |
"outputs": [
|
@@ -20,20 +22,33 @@
|
|
20 |
"name": "stdout",
|
21 |
"output_type": "stream",
|
22 |
"text": [
|
23 |
-
"
|
24 |
]
|
25 |
}
|
26 |
],
|
27 |
"source": [
|
28 |
"with open(\"Lilgpt.txt\",'r') as file:\n",
|
29 |
" data = file.read()\n",
|
30 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
31 |
"print(len(List))"
|
32 |
]
|
33 |
},
|
34 |
{
|
35 |
"cell_type": "code",
|
36 |
-
"execution_count":
|
37 |
"id": "18b93c2e",
|
38 |
"metadata": {},
|
39 |
"outputs": [
|
@@ -120,7 +135,7 @@
|
|
120 |
},
|
121 |
{
|
122 |
"cell_type": "code",
|
123 |
-
"execution_count":
|
124 |
"id": "9c7698db",
|
125 |
"metadata": {},
|
126 |
"outputs": [
|
@@ -128,11 +143,8 @@
|
|
128 |
"name": "stdout",
|
129 |
"output_type": "stream",
|
130 |
"text": [
|
131 |
-
"\n",
|
132 |
-
"<BOS>\n",
|
133 |
-
"3 Headed Goat[Intro]\n",
|
134 |
"(Aviator)\n",
|
135 |
-
"\n",
|
136 |
"[Chorus: Lil Baby]\n",
|
137 |
"These ain't no Guess jeans\n",
|
138 |
"I dropped out of school, I'm still good at math, but, nigga, don't test me\n",
|
@@ -143,7 +155,6 @@
|
|
143 |
"He say I'm hard and he say I'm garbage, I'm rich regardless\n",
|
144 |
"We in Miami in the middle of the winter, and we on them jet skis\n",
|
145 |
"If we in Atlanta, I'm runnin' the 'Cat and I'm workin' the red key\n",
|
146 |
-
"\n",
|
147 |
"[Verse 1: Lil Durk]\n",
|
148 |
"I cannot mention my homies inside of my song 'cause I know they be trappin' a lot\n",
|
149 |
"I can't keep takin' these pills, when I'm in the trenches, they say I be cappin' a lot\n",
|
@@ -160,7 +171,6 @@
|
|
160 |
"Only twenty-five, livin' like a boss, ridin' 'round with a chauffeur\n",
|
161 |
"I don't sell drugs, still be paranoid, keep lookin' over my shoulder\n",
|
162 |
"Niggas lyin' like I'm stealin' swag, boy, that's my shit like I wrote it\n",
|
163 |
-
"\n",
|
164 |
"[Verse 3: Polo G]\n",
|
165 |
"Uh\n",
|
166 |
"These rappers really nice as hell\n",
|
@@ -196,22 +206,24 @@
|
|
196 |
"Play like I'm dumb, as soon as it pop, I'm goin' retarded\n",
|
197 |
"He say I'm hard and he say I'm garbage, I'm rich regardless\n",
|
198 |
"We in Miami in the middle of the winter, and we on them jet skis\n",
|
199 |
-
"If we in Atlanta, I'm runnin' the 'Cat and I'm workin' the red key35EmbedShare URLCopyEmbedCopy
|
200 |
-
"<EOS>\n"
|
201 |
]
|
202 |
}
|
203 |
],
|
204 |
"source": [
|
205 |
"NewList = []\n",
|
206 |
"for l in List:\n",
|
207 |
-
" n = l
|
|
|
|
|
|
|
208 |
" NewList.append(n)\n",
|
209 |
"print(NewList[1])"
|
210 |
]
|
211 |
},
|
212 |
{
|
213 |
"cell_type": "code",
|
214 |
-
"execution_count":
|
215 |
"id": "f31e28d8",
|
216 |
"metadata": {},
|
217 |
"outputs": [
|
@@ -219,21 +231,21 @@
|
|
219 |
"name": "stdout",
|
220 |
"output_type": "stream",
|
221 |
"text": [
|
222 |
-
"
|
223 |
-
"
|
224 |
]
|
225 |
}
|
226 |
],
|
227 |
"source": [
|
|
|
228 |
"counter = 0\n",
|
229 |
"List_val = []\n",
|
230 |
"List_train = []\n",
|
231 |
-
"for l in
|
232 |
-
"
|
233 |
-
"
|
234 |
-
" List_val.append(n)\n",
|
235 |
" else:\n",
|
236 |
-
" List_train.append(
|
237 |
" counter += 1\n",
|
238 |
"print(len(List_val))\n",
|
239 |
"print(len(List_train))"
|
@@ -241,7 +253,7 @@
|
|
241 |
},
|
242 |
{
|
243 |
"cell_type": "code",
|
244 |
-
"execution_count":
|
245 |
"id": "9efd0b25",
|
246 |
"metadata": {},
|
247 |
"outputs": [
|
@@ -249,165 +261,375 @@
|
|
249 |
"name": "stdout",
|
250 |
"output_type": "stream",
|
251 |
"text": [
|
252 |
-
"<BOS
|
253 |
-
"
|
254 |
-
"
|
255 |
-
"
|
256 |
-
"
|
257 |
-
"
|
258 |
-
"
|
259 |
-
"
|
260 |
-
"
|
261 |
-
"
|
262 |
-
"
|
263 |
-
"\n",
|
264 |
-
"
|
265 |
-
"
|
266 |
-
"
|
267 |
-
"
|
268 |
-
"
|
269 |
-
"
|
270 |
-
"I
|
271 |
-
"
|
272 |
-
"
|
273 |
-
"
|
274 |
-
"
|
275 |
-
"
|
276 |
-
"
|
277 |
-
"
|
278 |
-
"
|
279 |
-
"
|
280 |
-
"
|
281 |
-
"
|
282 |
-
"
|
283 |
-
"
|
284 |
-
"
|
285 |
-
"
|
286 |
-
"
|
287 |
-
"
|
288 |
-
"
|
289 |
-
"
|
290 |
-
"My
|
291 |
-
"
|
292 |
-
"
|
293 |
-
"I
|
294 |
-
"
|
295 |
-
"
|
296 |
-
"
|
297 |
-
"
|
298 |
-
"
|
299 |
-
"
|
300 |
-
"
|
301 |
-
"
|
302 |
-
"
|
303 |
-
"
|
304 |
-
"
|
305 |
-
"
|
306 |
-
"
|
307 |
-
"I
|
308 |
-
"
|
309 |
-
"
|
310 |
-
"Shorty my baby, my everything, she the truth\n",
|
311 |
-
"Together we cool, me and her can't lose\n",
|
312 |
-
"Keep 'em on their feet, baby, I know they so confused\n",
|
313 |
-
"Shorty my Beyoncé\n",
|
314 |
-
"Durk and DeJ, Durk and DeJ, Durk and DeJ\n",
|
315 |
-
"Shorty my Beyoncé\n",
|
316 |
-
"Durk and DeJ, Durk and DeJ, Durk and DeJ\n",
|
317 |
-
"My Beyoncé\n",
|
318 |
-
"\n",
|
319 |
-
"[Verse 3: Lil Durk]\n",
|
320 |
-
"You and I\n",
|
321 |
-
"White dress, flowers, and a suit and tie\n",
|
322 |
-
"Me and you like Bonnie and Clyde\n",
|
323 |
-
"No beat the case, we're do or die\n",
|
324 |
-
"Who am I to say you ain't natural?\n",
|
325 |
-
"Your haters my haters, ain't switchin' up, baby, I got you\n",
|
326 |
-
"I'm with her like a tattoo\n",
|
327 |
-
"The way you wear that dress, they gon' attack you\n",
|
328 |
-
"The way you look at me, baby, I got you\n",
|
329 |
-
"I hit it from the front, I like the back too\n",
|
330 |
-
"She say, \"Lay down so I can ride you\"\n",
|
331 |
-
"I know that she fiending\n",
|
332 |
-
"She scratchin' my back, I like how she screamin'\n",
|
333 |
-
"I fuck her and leave her, she fiending\n",
|
334 |
-
"Shawty my Beyoncé\n",
|
335 |
-
"[Chorus: Lil Durk]\n",
|
336 |
-
"Ooh, I like the way she move\n",
|
337 |
-
"Shorty my baby, my everything, she the truth\n",
|
338 |
-
"Together we cool, me and her can't lose\n",
|
339 |
-
"Keep 'em on their feet, baby, I know they so confused\n",
|
340 |
-
"Shorty my Beyoncé\n",
|
341 |
-
"Durk and DeJ, Durk and DeJ, Durk and DeJ\n",
|
342 |
-
"Shorty my Beyoncé\n",
|
343 |
-
"Durk and DeJ, Durk and DeJ, Durk and DeJ\n",
|
344 |
-
"My Beyoncé\n",
|
345 |
-
"\n",
|
346 |
-
"[Outro: Lil Durk]\n",
|
347 |
-
"Ooh, ooh\n",
|
348 |
-
"Durk and DeJ, Durk and DeJ, Durk and DeJ\n",
|
349 |
-
"Ooh, ooh43EmbedShare URLCopyEmbedCopy\n",
|
350 |
-
"<EOS>\n"
|
351 |
]
|
352 |
}
|
353 |
],
|
354 |
"source": [
|
355 |
-
"print(
|
356 |
]
|
357 |
},
|
358 |
{
|
359 |
"cell_type": "code",
|
360 |
-
"execution_count":
|
361 |
"id": "5dfecab1",
|
362 |
"metadata": {},
|
363 |
"outputs": [],
|
364 |
"source": [
|
365 |
"val_set =List_val[0]\n",
|
366 |
"for i in range(1,len(List_val)):\n",
|
367 |
-
" val_set = val_set+List_val[i]"
|
368 |
]
|
369 |
},
|
370 |
{
|
371 |
"cell_type": "code",
|
372 |
-
"execution_count":
|
373 |
"id": "3a895d2f",
|
374 |
"metadata": {},
|
375 |
"outputs": [],
|
376 |
"source": [
|
377 |
"train_set =List_train[0]\n",
|
378 |
"for i in range(1,len(List_train)):\n",
|
379 |
-
" train_set = train_set+List_train[i]"
|
380 |
]
|
381 |
},
|
382 |
{
|
383 |
"cell_type": "code",
|
384 |
-
"execution_count":
|
385 |
"id": "74ea6efc",
|
386 |
"metadata": {},
|
387 |
"outputs": [],
|
388 |
"source": [
|
389 |
-
"file1 = open(\"
|
390 |
"file1.write(train_set)\n",
|
391 |
"file1.close()"
|
392 |
]
|
393 |
},
|
394 |
{
|
395 |
"cell_type": "code",
|
396 |
-
"execution_count":
|
397 |
"id": "3416e6da",
|
398 |
"metadata": {},
|
399 |
"outputs": [],
|
400 |
"source": [
|
401 |
-
"file2 = open(\"
|
402 |
"file2.write(val_set)\n",
|
403 |
"file2.close()"
|
404 |
]
|
405 |
},
|
406 |
{
|
407 |
"cell_type": "code",
|
408 |
-
"execution_count":
|
409 |
"id": "1bd0ca86",
|
410 |
"metadata": {},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
411 |
"outputs": [],
|
412 |
"source": []
|
413 |
}
|
|
|
2 |
"cells": [
|
3 |
{
|
4 |
"cell_type": "code",
|
5 |
+
"execution_count": 1,
|
6 |
"id": "81fd300c",
|
7 |
"metadata": {},
|
8 |
"outputs": [],
|
9 |
"source": [
|
10 |
+
"import pandas as pd\n",
|
11 |
+
"import os\n",
|
12 |
+
"import random"
|
13 |
]
|
14 |
},
|
15 |
{
|
16 |
"cell_type": "code",
|
17 |
+
"execution_count": 2,
|
18 |
"id": "1237ddf7",
|
19 |
"metadata": {},
|
20 |
"outputs": [
|
|
|
22 |
"name": "stdout",
|
23 |
"output_type": "stream",
|
24 |
"text": [
|
25 |
+
"8500\n"
|
26 |
]
|
27 |
}
|
28 |
],
|
29 |
"source": [
|
30 |
"with open(\"Lilgpt.txt\",'r') as file:\n",
|
31 |
" data = file.read()\n",
|
32 |
+
"List1 = data.split(\"<EOS>\")\n",
|
33 |
+
"with open(\"ye.txt\",'r') as file:\n",
|
34 |
+
" data = file.read()\n",
|
35 |
+
"List2 = data.split(\"<EOS>\")\n",
|
36 |
+
"with open(\"zenske.txt\",'r') as file:\n",
|
37 |
+
" data = file.read()\n",
|
38 |
+
"List3 = data.split(\"<EOS>\")\n",
|
39 |
+
"with open(\"flowmasteri.txt\",'r') as file:\n",
|
40 |
+
" data = file.read()\n",
|
41 |
+
"List4 = data.split(\"<EOS>\")\n",
|
42 |
+
"with open(\"wutang.txt\",'r') as file:\n",
|
43 |
+
" data = file.read()\n",
|
44 |
+
"List5 = data.split(\"<EOS>\")\n",
|
45 |
+
"List = List1+List2+List3+List4+List5\n",
|
46 |
"print(len(List))"
|
47 |
]
|
48 |
},
|
49 |
{
|
50 |
"cell_type": "code",
|
51 |
+
"execution_count": 3,
|
52 |
"id": "18b93c2e",
|
53 |
"metadata": {},
|
54 |
"outputs": [
|
|
|
135 |
},
|
136 |
{
|
137 |
"cell_type": "code",
|
138 |
+
"execution_count": 4,
|
139 |
"id": "9c7698db",
|
140 |
"metadata": {},
|
141 |
"outputs": [
|
|
|
143 |
"name": "stdout",
|
144 |
"output_type": "stream",
|
145 |
"text": [
|
146 |
+
"<BOS>3 Headed Goat[Intro]\n",
|
|
|
|
|
147 |
"(Aviator)\n",
|
|
|
148 |
"[Chorus: Lil Baby]\n",
|
149 |
"These ain't no Guess jeans\n",
|
150 |
"I dropped out of school, I'm still good at math, but, nigga, don't test me\n",
|
|
|
155 |
"He say I'm hard and he say I'm garbage, I'm rich regardless\n",
|
156 |
"We in Miami in the middle of the winter, and we on them jet skis\n",
|
157 |
"If we in Atlanta, I'm runnin' the 'Cat and I'm workin' the red key\n",
|
|
|
158 |
"[Verse 1: Lil Durk]\n",
|
159 |
"I cannot mention my homies inside of my song 'cause I know they be trappin' a lot\n",
|
160 |
"I can't keep takin' these pills, when I'm in the trenches, they say I be cappin' a lot\n",
|
|
|
171 |
"Only twenty-five, livin' like a boss, ridin' 'round with a chauffeur\n",
|
172 |
"I don't sell drugs, still be paranoid, keep lookin' over my shoulder\n",
|
173 |
"Niggas lyin' like I'm stealin' swag, boy, that's my shit like I wrote it\n",
|
|
|
174 |
"[Verse 3: Polo G]\n",
|
175 |
"Uh\n",
|
176 |
"These rappers really nice as hell\n",
|
|
|
206 |
"Play like I'm dumb, as soon as it pop, I'm goin' retarded\n",
|
207 |
"He say I'm hard and he say I'm garbage, I'm rich regardless\n",
|
208 |
"We in Miami in the middle of the winter, and we on them jet skis\n",
|
209 |
+
"If we in Atlanta, I'm runnin' the 'Cat and I'm workin' the red key35EmbedShare URLCopyEmbedCopy<EOS>\n"
|
|
|
210 |
]
|
211 |
}
|
212 |
],
|
213 |
"source": [
|
214 |
"NewList = []\n",
|
215 |
"for l in List:\n",
|
216 |
+
" n = l\n",
|
217 |
+
" n = n.replace(\"<BOS>\",\"\")\n",
|
218 |
+
" n = os.linesep.join([s for s in n.splitlines() if s])\n",
|
219 |
+
" n = \"<BOS>\"+ n + \"<EOS>\"\n",
|
220 |
" NewList.append(n)\n",
|
221 |
"print(NewList[1])"
|
222 |
]
|
223 |
},
|
224 |
{
|
225 |
"cell_type": "code",
|
226 |
+
"execution_count": 5,
|
227 |
"id": "f31e28d8",
|
228 |
"metadata": {},
|
229 |
"outputs": [
|
|
|
231 |
"name": "stdout",
|
232 |
"output_type": "stream",
|
233 |
"text": [
|
234 |
+
"800\n",
|
235 |
+
"7700\n"
|
236 |
]
|
237 |
}
|
238 |
],
|
239 |
"source": [
|
240 |
+
"random.shuffle(NewList)\n",
|
241 |
"counter = 0\n",
|
242 |
"List_val = []\n",
|
243 |
"List_train = []\n",
|
244 |
+
"for l in NewList:\n",
|
245 |
+
" if counter<800:\n",
|
246 |
+
" List_val.append(l)\n",
|
|
|
247 |
" else:\n",
|
248 |
+
" List_train.append(l)\n",
|
249 |
" counter += 1\n",
|
250 |
"print(len(List_val))\n",
|
251 |
"print(len(List_train))"
|
|
|
253 |
},
|
254 |
{
|
255 |
"cell_type": "code",
|
256 |
+
"execution_count": 6,
|
257 |
"id": "9efd0b25",
|
258 |
"metadata": {},
|
259 |
"outputs": [
|
|
|
261 |
"name": "stdout",
|
262 |
"output_type": "stream",
|
263 |
"text": [
|
264 |
+
"<BOS>PainI finally moved out my momma house\n",
|
265 |
+
"Got a happy home\n",
|
266 |
+
"Only thing fucked up, Daddy gone (Gone forever man)\n",
|
267 |
+
"Listen up this is real shit (Real shit)\n",
|
268 |
+
"It's fucked up when u got nobody to share it with (Fucked up)\n",
|
269 |
+
"Give my last so my grandma can see me now (See me now)\n",
|
270 |
+
"Smilin tellin everybody bout her grandbaby (her grandbaby)\n",
|
271 |
+
"T-Lady hold on all I got is you (Mama)\n",
|
272 |
+
"It's fucked up what I done put me and my mama through\n",
|
273 |
+
"What's happenin Ivory I know you lookin down watchin me\n",
|
274 |
+
"Everyday I got yo kids fresh from head to feet (All of em)\n",
|
275 |
+
"Wish I can put u in the bentley with the brown [? ]\n",
|
276 |
+
"I wish we never would've went out to that club\n",
|
277 |
+
"On the road to riches mane lost a flock of niggas\n",
|
278 |
+
"[? ] a whole block of niggas\n",
|
279 |
+
"Niggas bite the hand that feed em I done seen it all\n",
|
280 |
+
"Jus got a call... Slo dead... Not my fuckin dawg\n",
|
281 |
+
"I never thought this pain last this many years\n",
|
282 |
+
"I... I never thought this pain last this many years\n",
|
283 |
+
"Still in tears\n",
|
284 |
+
"I... I never thought this pain last this many years\n",
|
285 |
+
"I lost my daddy in a heartbeat\n",
|
286 |
+
"Real talk eyes turnin right in front me\n",
|
287 |
+
"Ever since I saw that shit... it hunt me\n",
|
288 |
+
"Pray for me cause I shake in my fuckin sleep\n",
|
289 |
+
"Wish I could[? ] shoot the 50 nigga\n",
|
290 |
+
"I shoulda put him under my wing and I miss him nigga\n",
|
291 |
+
"I told [? ] way back stop ridin with ya pack\n",
|
292 |
+
"He ain't listen\n",
|
293 |
+
"And now my nigga right back in penitentury he don't listen\n",
|
294 |
+
"I lost Ivory and it fucked me up\n",
|
295 |
+
"My whole life changed\n",
|
296 |
+
"Nigga put me up in beast mode\n",
|
297 |
+
"My heart cold\n",
|
298 |
+
"And as the years role\n",
|
299 |
+
"I wish that he could have everything I got\n",
|
300 |
+
"And a lil mo'\n",
|
301 |
+
"Yo kids love me nigga\n",
|
302 |
+
"My kids and yo kids call each other sister and brothers\n",
|
303 |
+
"You forever my lil hustler\n",
|
304 |
+
"Just saw yo gangsta at the car lot nigga and a tear dropped nigga\n",
|
305 |
+
"I never thought this pain last this many years\n",
|
306 |
+
"I... I never thought this pain last this many years\n",
|
307 |
+
"Still in tears\n",
|
308 |
+
"I... I never thought this pain last this many years\n",
|
309 |
+
"Seem like it was yesterday\n",
|
310 |
+
"The mo money I get this shit hurt\n",
|
311 |
+
"I thought that would take the stress away\n",
|
312 |
+
"To see u smilin with yo chain countin money nigga\n",
|
313 |
+
"When Trina pulled you on stage you was stuntin nigga\n",
|
314 |
+
"But through it all mane you kno how we rock nigga\n",
|
315 |
+
"From the yo to the block to the hospital\n",
|
316 |
+
"Never again would I have another big dawg\n",
|
317 |
+
"But ya mama and kids I got em all but I\n",
|
318 |
+
"I never thought this pain last this many years\n",
|
319 |
+
"I... I never thought this pain last this many years\n",
|
320 |
+
"Still in tears\n",
|
321 |
+
"I... I never thought this pain last this many yea2EmbedShare URLCopyEmbedCopy<EOS>\n"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
322 |
]
|
323 |
}
|
324 |
],
|
325 |
"source": [
|
326 |
+
"print(List_train[1])"
|
327 |
]
|
328 |
},
|
329 |
{
|
330 |
"cell_type": "code",
|
331 |
+
"execution_count": 7,
|
332 |
"id": "5dfecab1",
|
333 |
"metadata": {},
|
334 |
"outputs": [],
|
335 |
"source": [
|
336 |
"val_set =List_val[0]\n",
|
337 |
"for i in range(1,len(List_val)):\n",
|
338 |
+
" val_set = val_set+\"\\n\\n\"+ List_val[i]"
|
339 |
]
|
340 |
},
|
341 |
{
|
342 |
"cell_type": "code",
|
343 |
+
"execution_count": 8,
|
344 |
"id": "3a895d2f",
|
345 |
"metadata": {},
|
346 |
"outputs": [],
|
347 |
"source": [
|
348 |
"train_set =List_train[0]\n",
|
349 |
"for i in range(1,len(List_train)):\n",
|
350 |
+
" train_set = train_set+\"\\n\\n\"+List_train[i]"
|
351 |
]
|
352 |
},
|
353 |
{
|
354 |
"cell_type": "code",
|
355 |
+
"execution_count": 9,
|
356 |
"id": "74ea6efc",
|
357 |
"metadata": {},
|
358 |
"outputs": [],
|
359 |
"source": [
|
360 |
+
"file1 = open(\"train2.txt\",\"w+\")\n",
|
361 |
"file1.write(train_set)\n",
|
362 |
"file1.close()"
|
363 |
]
|
364 |
},
|
365 |
{
|
366 |
"cell_type": "code",
|
367 |
+
"execution_count": 10,
|
368 |
"id": "3416e6da",
|
369 |
"metadata": {},
|
370 |
"outputs": [],
|
371 |
"source": [
|
372 |
+
"file2 = open(\"val2.txt\",\"w+\")\n",
|
373 |
"file2.write(val_set)\n",
|
374 |
"file2.close()"
|
375 |
]
|
376 |
},
|
377 |
{
|
378 |
"cell_type": "code",
|
379 |
+
"execution_count": 15,
|
380 |
"id": "1bd0ca86",
|
381 |
"metadata": {},
|
382 |
+
"outputs": [
|
383 |
+
{
|
384 |
+
"data": {
|
385 |
+
"text/html": [
|
386 |
+
"<div>\n",
|
387 |
+
"<style scoped>\n",
|
388 |
+
" .dataframe tbody tr th:only-of-type {\n",
|
389 |
+
" vertical-align: middle;\n",
|
390 |
+
" }\n",
|
391 |
+
"\n",
|
392 |
+
" .dataframe tbody tr th {\n",
|
393 |
+
" vertical-align: top;\n",
|
394 |
+
" }\n",
|
395 |
+
"\n",
|
396 |
+
" .dataframe thead th {\n",
|
397 |
+
" text-align: right;\n",
|
398 |
+
" }\n",
|
399 |
+
"</style>\n",
|
400 |
+
"<table border=\"1\" class=\"dataframe\">\n",
|
401 |
+
" <thead>\n",
|
402 |
+
" <tr style=\"text-align: right;\">\n",
|
403 |
+
" <th></th>\n",
|
404 |
+
" <th>Songs</th>\n",
|
405 |
+
" </tr>\n",
|
406 |
+
" </thead>\n",
|
407 |
+
" <tbody>\n",
|
408 |
+
" <tr>\n",
|
409 |
+
" <th>0</th>\n",
|
410 |
+
" <td><BOS>Playa Hater[Intro: The Notorious B.I.G.]\\...</td>\n",
|
411 |
+
" </tr>\n",
|
412 |
+
" <tr>\n",
|
413 |
+
" <th>1</th>\n",
|
414 |
+
" <td><BOS>PainI finally moved out my momma house\\nG...</td>\n",
|
415 |
+
" </tr>\n",
|
416 |
+
" <tr>\n",
|
417 |
+
" <th>2</th>\n",
|
418 |
+
" <td><BOS>I Don’t Do Much[Flip talking]\\nCounting m...</td>\n",
|
419 |
+
" </tr>\n",
|
420 |
+
" <tr>\n",
|
421 |
+
" <th>3</th>\n",
|
422 |
+
" <td><BOS>Fuck the Weatherman[Intro]\\nLike seriousl...</td>\n",
|
423 |
+
" </tr>\n",
|
424 |
+
" <tr>\n",
|
425 |
+
" <th>4</th>\n",
|
426 |
+
" <td><BOS>It’s Kim Bitches (Get That Money)[Intro]\\...</td>\n",
|
427 |
+
" </tr>\n",
|
428 |
+
" </tbody>\n",
|
429 |
+
"</table>\n",
|
430 |
+
"</div>"
|
431 |
+
],
|
432 |
+
"text/plain": [
|
433 |
+
" Songs\n",
|
434 |
+
"0 <BOS>Playa Hater[Intro: The Notorious B.I.G.]\\...\n",
|
435 |
+
"1 <BOS>PainI finally moved out my momma house\\nG...\n",
|
436 |
+
"2 <BOS>I Don’t Do Much[Flip talking]\\nCounting m...\n",
|
437 |
+
"3 <BOS>Fuck the Weatherman[Intro]\\nLike seriousl...\n",
|
438 |
+
"4 <BOS>It’s Kim Bitches (Get That Money)[Intro]\\..."
|
439 |
+
]
|
440 |
+
},
|
441 |
+
"execution_count": 15,
|
442 |
+
"metadata": {},
|
443 |
+
"output_type": "execute_result"
|
444 |
+
}
|
445 |
+
],
|
446 |
+
"source": [
|
447 |
+
"df_train = pd.DataFrame(List_train,columns = [\"Songs\"])\n",
|
448 |
+
"df_train.head()"
|
449 |
+
]
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"cell_type": "code",
|
453 |
+
"execution_count": 17,
|
454 |
+
"id": "f78a2abc",
|
455 |
+
"metadata": {},
|
456 |
+
"outputs": [
|
457 |
+
{
|
458 |
+
"data": {
|
459 |
+
"text/html": [
|
460 |
+
"<div>\n",
|
461 |
+
"<style scoped>\n",
|
462 |
+
" .dataframe tbody tr th:only-of-type {\n",
|
463 |
+
" vertical-align: middle;\n",
|
464 |
+
" }\n",
|
465 |
+
"\n",
|
466 |
+
" .dataframe tbody tr th {\n",
|
467 |
+
" vertical-align: top;\n",
|
468 |
+
" }\n",
|
469 |
+
"\n",
|
470 |
+
" .dataframe thead th {\n",
|
471 |
+
" text-align: right;\n",
|
472 |
+
" }\n",
|
473 |
+
"</style>\n",
|
474 |
+
"<table border=\"1\" class=\"dataframe\">\n",
|
475 |
+
" <thead>\n",
|
476 |
+
" <tr style=\"text-align: right;\">\n",
|
477 |
+
" <th></th>\n",
|
478 |
+
" <th>Songs</th>\n",
|
479 |
+
" </tr>\n",
|
480 |
+
" </thead>\n",
|
481 |
+
" <tbody>\n",
|
482 |
+
" <tr>\n",
|
483 |
+
" <th>0</th>\n",
|
484 |
+
" <td><BOS>Straight Gutta[Verse 1: Streetlife]\\nI’m ...</td>\n",
|
485 |
+
" </tr>\n",
|
486 |
+
" <tr>\n",
|
487 |
+
" <th>1</th>\n",
|
488 |
+
" <td><BOS>Georgia... Bush / Weezy’z Ambitionz[Part ...</td>\n",
|
489 |
+
" </tr>\n",
|
490 |
+
" <tr>\n",
|
491 |
+
" <th>2</th>\n",
|
492 |
+
" <td><BOS>1991 Freestyle[Intro: Ol' Dirty Bastard]\\...</td>\n",
|
493 |
+
" </tr>\n",
|
494 |
+
" <tr>\n",
|
495 |
+
" <th>3</th>\n",
|
496 |
+
" <td><BOS>The Wolf* iTunes bonus track\\n[RZA]\\nWatc...</td>\n",
|
497 |
+
" </tr>\n",
|
498 |
+
" <tr>\n",
|
499 |
+
" <th>4</th>\n",
|
500 |
+
" <td><BOS>Never Mind[Pre-Chorus]\\nI'm a pimp under ...</td>\n",
|
501 |
+
" </tr>\n",
|
502 |
+
" </tbody>\n",
|
503 |
+
"</table>\n",
|
504 |
+
"</div>"
|
505 |
+
],
|
506 |
+
"text/plain": [
|
507 |
+
" Songs\n",
|
508 |
+
"0 <BOS>Straight Gutta[Verse 1: Streetlife]\\nI’m ...\n",
|
509 |
+
"1 <BOS>Georgia... Bush / Weezy’z Ambitionz[Part ...\n",
|
510 |
+
"2 <BOS>1991 Freestyle[Intro: Ol' Dirty Bastard]\\...\n",
|
511 |
+
"3 <BOS>The Wolf* iTunes bonus track\\n[RZA]\\nWatc...\n",
|
512 |
+
"4 <BOS>Never Mind[Pre-Chorus]\\nI'm a pimp under ..."
|
513 |
+
]
|
514 |
+
},
|
515 |
+
"execution_count": 17,
|
516 |
+
"metadata": {},
|
517 |
+
"output_type": "execute_result"
|
518 |
+
}
|
519 |
+
],
|
520 |
+
"source": [
|
521 |
+
"df_val = pd.DataFrame(List_val,columns = [\"Songs\"])\n",
|
522 |
+
"df_val.head()"
|
523 |
+
]
|
524 |
+
},
|
525 |
+
{
|
526 |
+
"cell_type": "code",
|
527 |
+
"execution_count": 18,
|
528 |
+
"id": "240e7d42",
|
529 |
+
"metadata": {},
|
530 |
+
"outputs": [],
|
531 |
+
"source": [
|
532 |
+
"df_train.to_csv(\"Train.tsv\",sep=\"\\t\",index = False)"
|
533 |
+
]
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"cell_type": "code",
|
537 |
+
"execution_count": 19,
|
538 |
+
"id": "5899b85b",
|
539 |
+
"metadata": {},
|
540 |
+
"outputs": [],
|
541 |
+
"source": [
|
542 |
+
"df_val.to_csv(\"Val.tsv\",sep=\"\\t\",index = False)"
|
543 |
+
]
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"cell_type": "code",
|
547 |
+
"execution_count": 20,
|
548 |
+
"id": "c0dd966c",
|
549 |
+
"metadata": {},
|
550 |
+
"outputs": [],
|
551 |
+
"source": [
|
552 |
+
"df = pd.read_csv(\"Train.tsv\",sep = \"\\t\")"
|
553 |
+
]
|
554 |
+
},
|
555 |
+
{
|
556 |
+
"cell_type": "code",
|
557 |
+
"execution_count": 21,
|
558 |
+
"id": "922af4c8",
|
559 |
+
"metadata": {},
|
560 |
+
"outputs": [
|
561 |
+
{
|
562 |
+
"data": {
|
563 |
+
"text/html": [
|
564 |
+
"<div>\n",
|
565 |
+
"<style scoped>\n",
|
566 |
+
" .dataframe tbody tr th:only-of-type {\n",
|
567 |
+
" vertical-align: middle;\n",
|
568 |
+
" }\n",
|
569 |
+
"\n",
|
570 |
+
" .dataframe tbody tr th {\n",
|
571 |
+
" vertical-align: top;\n",
|
572 |
+
" }\n",
|
573 |
+
"\n",
|
574 |
+
" .dataframe thead th {\n",
|
575 |
+
" text-align: right;\n",
|
576 |
+
" }\n",
|
577 |
+
"</style>\n",
|
578 |
+
"<table border=\"1\" class=\"dataframe\">\n",
|
579 |
+
" <thead>\n",
|
580 |
+
" <tr style=\"text-align: right;\">\n",
|
581 |
+
" <th></th>\n",
|
582 |
+
" <th>Songs</th>\n",
|
583 |
+
" </tr>\n",
|
584 |
+
" </thead>\n",
|
585 |
+
" <tbody>\n",
|
586 |
+
" <tr>\n",
|
587 |
+
" <th>0</th>\n",
|
588 |
+
" <td><BOS>Playa Hater[Intro: The Notorious B.I.G.]\\...</td>\n",
|
589 |
+
" </tr>\n",
|
590 |
+
" <tr>\n",
|
591 |
+
" <th>1</th>\n",
|
592 |
+
" <td><BOS>PainI finally moved out my momma house\\nG...</td>\n",
|
593 |
+
" </tr>\n",
|
594 |
+
" <tr>\n",
|
595 |
+
" <th>2</th>\n",
|
596 |
+
" <td><BOS>I Don’t Do Much[Flip talking]\\nCounting m...</td>\n",
|
597 |
+
" </tr>\n",
|
598 |
+
" <tr>\n",
|
599 |
+
" <th>3</th>\n",
|
600 |
+
" <td><BOS>Fuck the Weatherman[Intro]\\nLike seriousl...</td>\n",
|
601 |
+
" </tr>\n",
|
602 |
+
" <tr>\n",
|
603 |
+
" <th>4</th>\n",
|
604 |
+
" <td><BOS>It’s Kim Bitches (Get That Money)[Intro]\\...</td>\n",
|
605 |
+
" </tr>\n",
|
606 |
+
" </tbody>\n",
|
607 |
+
"</table>\n",
|
608 |
+
"</div>"
|
609 |
+
],
|
610 |
+
"text/plain": [
|
611 |
+
" Songs\n",
|
612 |
+
"0 <BOS>Playa Hater[Intro: The Notorious B.I.G.]\\...\n",
|
613 |
+
"1 <BOS>PainI finally moved out my momma house\\nG...\n",
|
614 |
+
"2 <BOS>I Don’t Do Much[Flip talking]\\nCounting m...\n",
|
615 |
+
"3 <BOS>Fuck the Weatherman[Intro]\\nLike seriousl...\n",
|
616 |
+
"4 <BOS>It’s Kim Bitches (Get That Money)[Intro]\\..."
|
617 |
+
]
|
618 |
+
},
|
619 |
+
"execution_count": 21,
|
620 |
+
"metadata": {},
|
621 |
+
"output_type": "execute_result"
|
622 |
+
}
|
623 |
+
],
|
624 |
+
"source": [
|
625 |
+
"df.head()"
|
626 |
+
]
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"cell_type": "code",
|
630 |
+
"execution_count": null,
|
631 |
+
"id": "be9f103f",
|
632 |
+
"metadata": {},
|
633 |
"outputs": [],
|
634 |
"source": []
|
635 |
}
|
Val.tsv
ADDED
The diff for this file is too large to render.
See raw diff
|
|
events.out.tfevents.1626357469.t1v-n-1926f308-w-0.95268.3.v2
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d2254be5c9709a35e6a20a7a795789ea4d605f4e2bde68aeda8de3b6623dbd7f
|
3 |
+
size 113336
|
events.out.tfevents.1626357733.t1v-n-1926f308-w-0.96764.3.v2
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5c5b373c809857af52071b4078af469dfd5781cc3edccee286293ccf860202be
|
3 |
+
size 7499
|
events.out.tfevents.1626358530.t1v-n-1926f308-w-0.97646.3.v2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:00566fffb65e756600911e0f37b177d585f68f4ad68d1a6c0a3530b59a66079c
|
3 |
+
size 113336
|
events.out.tfevents.1626359662.t1v-n-1926f308-w-0.99131.3.v2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:52dc8b0bbb392d47d3f14c2270a79387f36808449954df4e8b48828f69f720c8
|
3 |
+
size 40
|
events.out.tfevents.1626359960.t1v-n-1926f308-w-0.100528.3.v2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5b36a686716fbe468b3e656c14476497e60407502dd59b6819aa337f76722e0d
|
3 |
+
size 113336
|
events.out.tfevents.1626447876.t1v-n-1926f308-w-0.111464.3.v2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e57a9d1e97e640dbcc715deff93ca140880825c478cc2e3e93dc019996fb5082
|
3 |
+
size 40
|
events.out.tfevents.1626451747.t1v-n-1926f308-w-0.115707.3.v2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8da8c6d1c7dddaf6f0c6d63a47f39127cfe30dd180c6b96de36aabe985bcd54a
|
3 |
+
size 40
|
events.out.tfevents.1626451941.t1v-n-1926f308-w-0.117207.3.v2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f1d2fbd80ea41fd70f1baff6dc2dbb4107fe920c2bdecd42a07b86a1bad0b24e
|
3 |
+
size 44657
|
flax_model.msgpack
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 497764120
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:742e9e9fd4077234f5e0a5f68fd466c23b7ffae51c771abaafa5ae4d083edfca
|
3 |
size 497764120
|
flowmasteri.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
run.sh
CHANGED
@@ -4,18 +4,18 @@ python3 run_clm_flax.py \
|
|
4 |
--model_type="gpt2" \
|
5 |
--config_name="./" \
|
6 |
--tokenizer_name="./" \
|
7 |
-
--train_file="/home/anantshankhdhar/gpt2-rap-lyric-generator/
|
8 |
-
--validation_file="/home/anantshankhdhar/gpt2-rap-lyric-generator/
|
|
|
9 |
--do_train \
|
10 |
--do_eval \
|
11 |
-
--block_size="512" \
|
12 |
--per_device_train_batch_size="64" \
|
13 |
--per_device_eval_batch_size="32" \
|
14 |
-
--learning_rate="5e-
|
15 |
--adam_beta1="0.9" --adam_beta2="0.98" --weight_decay="0.01" \
|
16 |
--overwrite_output_dir \
|
17 |
--num_train_epochs="100" \
|
18 |
-
--logging_steps="
|
19 |
-
--save_steps="
|
20 |
-
--eval_steps="
|
21 |
--push_to_hub
|
|
|
4 |
--model_type="gpt2" \
|
5 |
--config_name="./" \
|
6 |
--tokenizer_name="./" \
|
7 |
+
--train_file="/home/anantshankhdhar/gpt2-rap-lyric-generator/Train.tsv" \
|
8 |
+
--validation_file="/home/anantshankhdhar/gpt2-rap-lyric-generator/Val.tsv" \
|
9 |
+
--block_size="512" \
|
10 |
--do_train \
|
11 |
--do_eval \
|
|
|
12 |
--per_device_train_batch_size="64" \
|
13 |
--per_device_eval_batch_size="32" \
|
14 |
+
--learning_rate="5e-4" \
|
15 |
--adam_beta1="0.9" --adam_beta2="0.98" --weight_decay="0.01" \
|
16 |
--overwrite_output_dir \
|
17 |
--num_train_epochs="100" \
|
18 |
+
--logging_steps="100" \
|
19 |
+
--save_steps="100" \
|
20 |
+
--eval_steps="100" \
|
21 |
--push_to_hub
|
run_clm_flax.py
CHANGED
@@ -161,10 +161,10 @@ class DataTrainingArguments:
|
|
161 |
else:
|
162 |
if self.train_file is not None:
|
163 |
extension = self.train_file.split(".")[-1]
|
164 |
-
assert extension in ["csv", "json", "txt"], "`train_file` should be a csv, a json or a txt file."
|
165 |
if self.validation_file is not None:
|
166 |
extension = self.validation_file.split(".")[-1]
|
167 |
-
assert extension in ["csv", "json", "txt"], "`validation_file` should be a csv, a json or a txt file."
|
168 |
|
169 |
|
170 |
class TrainState(train_state.TrainState):
|
@@ -306,7 +306,9 @@ def main():
|
|
306 |
extension = data_args.train_file.split(".")[-1]
|
307 |
if extension == "txt":
|
308 |
extension = "text"
|
309 |
-
|
|
|
|
|
310 |
# See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at
|
311 |
# https://huggingface.co/docs/datasets/loading_datasets.html.
|
312 |
|
|
|
161 |
else:
|
162 |
if self.train_file is not None:
|
163 |
extension = self.train_file.split(".")[-1]
|
164 |
+
assert extension in ["csv", "json", "txt","tsv"], "`train_file` should be a csv, a json or a txt file."
|
165 |
if self.validation_file is not None:
|
166 |
extension = self.validation_file.split(".")[-1]
|
167 |
+
assert extension in ["csv", "json", "txt","tsv"], "`validation_file` should be a csv, a json or a txt file."
|
168 |
|
169 |
|
170 |
class TrainState(train_state.TrainState):
|
|
|
306 |
extension = data_args.train_file.split(".")[-1]
|
307 |
if extension == "txt":
|
308 |
extension = "text"
|
309 |
+
if extension == "tsv":
|
310 |
+
extension = "csv"
|
311 |
+
dataset = load_dataset(extension, data_files=data_files, delimiter='\t',column_names = ["Songs"],cache_dir=model_args.cache_dir)
|
312 |
# See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at
|
313 |
# https://huggingface.co/docs/datasets/loading_datasets.html.
|
314 |
|
text_collection/text_collection.py
ADDED
@@ -0,0 +1,104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# coding=utf-8
|
2 |
+
# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
|
3 |
+
#
|
4 |
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
5 |
+
# you may not use this file except in compliance with the License.
|
6 |
+
# You may obtain a copy of the License at
|
7 |
+
#
|
8 |
+
# http://www.apache.org/licenses/LICENSE-2.0
|
9 |
+
#
|
10 |
+
# Unless required by applicable law or agreed to in writing, software
|
11 |
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
12 |
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
13 |
+
# See the License for the specific language governing permissions and
|
14 |
+
# limitations under the License.
|
15 |
+
"""Large-scale Indonesian Summarization Dataset"""
|
16 |
+
import glob
|
17 |
+
import json
|
18 |
+
import os
|
19 |
+
import re
|
20 |
+
from pathlib import Path
|
21 |
+
import datasets
|
22 |
+
logger = datasets.logging.get_logger(__name__)
|
23 |
+
_CITATION = """\
|
24 |
+
"""
|
25 |
+
_DESCRIPTION = """\
|
26 |
+
This module load text dataset from local directory. The text dataset should have the format like Oscar dataset
|
27 |
+
where each new entry is separated by empty lines.
|
28 |
+
"""
|
29 |
+
_HOMEPAGE = ""
|
30 |
+
_LICENSE = ""
|
31 |
+
class TextCollectionConfig(datasets.BuilderConfig):
|
32 |
+
"""BuilderConfig for TextCollection"""
|
33 |
+
def __init__(self, **kwargs):
|
34 |
+
"""BuilderConfig for TextCollection.
|
35 |
+
Args:
|
36 |
+
**kwargs: keyword arguments forwarded to super.
|
37 |
+
"""
|
38 |
+
super(TextCollectionConfig, self).__init__(**kwargs)
|
39 |
+
class TextCollection(datasets.GeneratorBasedBuilder):
|
40 |
+
VERSION = datasets.Version("1.0.0")
|
41 |
+
BUILDER_CONFIGS = [
|
42 |
+
TextCollectionConfig(
|
43 |
+
name="text_collection",
|
44 |
+
version=VERSION,
|
45 |
+
description="Id Collection dataset",
|
46 |
+
),
|
47 |
+
]
|
48 |
+
@property
|
49 |
+
def manual_download_instructions(self):
|
50 |
+
return """\
|
51 |
+
You need to manually collect text datasets in a directory. The text dataset can then be loaded
|
52 |
+
using the following command:
|
53 |
+
`datasets.load_dataset("text_collection", data_dir="<path/to/dataset>")`.
|
54 |
+
"""
|
55 |
+
def _info(self):
|
56 |
+
return datasets.DatasetInfo(
|
57 |
+
description=_DESCRIPTION,
|
58 |
+
features=datasets.Features({"id": datasets.Value("int64"), "text": datasets.Value("string")}),
|
59 |
+
supervised_keys=None,
|
60 |
+
homepage=_HOMEPAGE,
|
61 |
+
license=_LICENSE,
|
62 |
+
citation=_CITATION,
|
63 |
+
)
|
64 |
+
def _split_generators(self, dl_manager):
|
65 |
+
data_dir = os.path.abspath(os.path.expanduser(dl_manager.manual_dir))
|
66 |
+
print("# Data directory", data_dir)
|
67 |
+
if not os.path.exists(data_dir):
|
68 |
+
raise FileNotFoundError(
|
69 |
+
"{} does not exist. Make sure you insert a manual dir via `datasets.load_dataset('id_liputan6', "
|
70 |
+
"'canonical', data_dir=...)`. Manual download instructions:\n{}".format(
|
71 |
+
data_dir, self.manual_download_instructions
|
72 |
+
)
|
73 |
+
)
|
74 |
+
split_generators = [
|
75 |
+
datasets.SplitGenerator(
|
76 |
+
name=datasets.Split.TRAIN,
|
77 |
+
gen_kwargs={
|
78 |
+
"article_dir": os.path.join(data_dir, ""),
|
79 |
+
"split": "train",
|
80 |
+
},
|
81 |
+
)
|
82 |
+
]
|
83 |
+
return split_generators
|
84 |
+
def _generate_examples(self, article_dir, split):
|
85 |
+
logger.info("⏳ Generating %s examples from = %s", split, article_dir)
|
86 |
+
id_ = 0
|
87 |
+
current_lines = []
|
88 |
+
for path in sorted(glob.glob(os.path.join(article_dir, "**/*.txt"), recursive=True)):
|
89 |
+
with open(path, "r") as f:
|
90 |
+
print("# Reading", path)
|
91 |
+
for line in f:
|
92 |
+
if len(line.strip()) > 0:
|
93 |
+
current_lines.append(line)
|
94 |
+
elif current_lines:
|
95 |
+
feature = id_, {"id": id_, "text": "".join(current_lines).rstrip()}
|
96 |
+
yield feature
|
97 |
+
id_ += 1
|
98 |
+
current_lines = []
|
99 |
+
# last paragraph
|
100 |
+
if current_lines:
|
101 |
+
feature = id_, {"id": id_, "text": "".join(current_lines).rstrip()}
|
102 |
+
yield feature
|
103 |
+
id_ += 1
|
104 |
+
current_lines = []
|
text_collection/text_collection.py.lock
ADDED
File without changes
|
train/train2.txt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5874cf342a7153d6949d6ecdbe72cabfde2bede960a4409bd7f82e88d6d4ed0f
|
3 |
+
size 19715139
|
train/val2.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
val/val2.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
wutang.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
ye.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
zenske.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|