AlekseyKorshuk
commited on
Commit
•
3062cd9
1
Parent(s):
7c8145a
huggingartists
Browse files- README.md +3 -3
- config.json +1 -1
- evaluation.txt +1 -1
- flax_model.msgpack +1 -1
- optimizer.pt +2 -2
- pytorch_model.bin +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- tokenizer_config.json +1 -1
- trainer_state.json +171 -7
- training_args.bin +1 -1
README.md
CHANGED
@@ -45,15 +45,15 @@ from datasets import load_dataset
|
|
45 |
dataset = load_dataset("huggingartists/rihanna")
|
46 |
```
|
47 |
|
48 |
-
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/
|
49 |
|
50 |
## Training procedure
|
51 |
|
52 |
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Rihanna's lyrics.
|
53 |
|
54 |
-
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/
|
55 |
|
56 |
-
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/
|
57 |
|
58 |
## How to use
|
59 |
|
|
|
45 |
dataset = load_dataset("huggingartists/rihanna")
|
46 |
```
|
47 |
|
48 |
+
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/3g4a2qa9/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
|
49 |
|
50 |
## Training procedure
|
51 |
|
52 |
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Rihanna's lyrics.
|
53 |
|
54 |
+
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/3du2fhxz) for full transparency and reproducibility.
|
55 |
|
56 |
+
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/3du2fhxz/artifacts) is logged and versioned.
|
57 |
|
58 |
## How to use
|
59 |
|
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"activation_function": "gelu_new",
|
4 |
"architectures": [
|
5 |
"GPT2LMHeadModel"
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "huggingartists/rihanna",
|
3 |
"activation_function": "gelu_new",
|
4 |
"architectures": [
|
5 |
"GPT2LMHeadModel"
|
evaluation.txt
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"eval_loss": 2.
|
|
|
1 |
+
{"eval_loss": 2.237405776977539, "eval_runtime": 8.4305, "eval_samples_per_second": 22.181, "eval_steps_per_second": 2.847, "epoch": 2.0}
|
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:fd6be841db360da1a00ad23a76226c5b19e42247bc72575b241a6f446840b95d
|
3 |
size 497764120
|
optimizer.pt
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 sha256:7a71909dbe3a3827eb2d96903ec0cd4f6475fbd143d3adcebee21e68eb480324
|
3 |
+
size 995604017
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 510403817
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c5c96ee233c3281090b842a4a11ff662f1751f775bd8bd50db7037dbc4bdf1d8
|
3 |
size 510403817
|
rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:26ebd8f09a6c1db84aa3aaeb1ea1345bcb3c2ae6c124b74ff19c3f608acd9e3c
|
3 |
size 14503
|
scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 623
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0b2179c852b9a3c2fc23aeed095aeb8cf7d63d1da7d9c6eca4435ee27bc43703
|
3 |
size 623
|
tokenizer_config.json
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "
|
|
|
1 |
+
{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "huggingartists/rihanna", "tokenizer_class": "GPT2Tokenizer"}
|
trainer_state.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"best_metric": 2.
|
3 |
-
"best_model_checkpoint": "output/rihanna/checkpoint-
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -170,11 +170,175 @@
|
|
170 |
"eval_samples_per_second": 22.593,
|
171 |
"eval_steps_per_second": 2.855,
|
172 |
"step": 131
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
173 |
}
|
174 |
],
|
175 |
-
"max_steps":
|
176 |
-
"num_train_epochs":
|
177 |
-
"total_flos":
|
178 |
"trial_name": null,
|
179 |
"trial_params": null
|
180 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 2.237405776977539,
|
3 |
+
"best_model_checkpoint": "output/rihanna/checkpoint-262",
|
4 |
+
"epoch": 2.0,
|
5 |
+
"global_step": 262,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
170 |
"eval_samples_per_second": 22.593,
|
171 |
"eval_steps_per_second": 2.855,
|
172 |
"step": 131
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"epoch": 1.03,
|
176 |
+
"learning_rate": 3.153829445781081e-07,
|
177 |
+
"loss": 2.3816,
|
178 |
+
"step": 135
|
179 |
+
},
|
180 |
+
{
|
181 |
+
"epoch": 1.07,
|
182 |
+
"learning_rate": 1.5916577414195624e-06,
|
183 |
+
"loss": 2.1418,
|
184 |
+
"step": 140
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 1.11,
|
188 |
+
"learning_rate": 3.8302217539407e-06,
|
189 |
+
"loss": 2.1866,
|
190 |
+
"step": 145
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 1.15,
|
194 |
+
"learning_rate": 6.998927551907465e-06,
|
195 |
+
"loss": 2.3514,
|
196 |
+
"step": 150
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 1.18,
|
200 |
+
"learning_rate": 1.1052270183036815e-05,
|
201 |
+
"loss": 2.3302,
|
202 |
+
"step": 155
|
203 |
+
},
|
204 |
+
{
|
205 |
+
"epoch": 1.22,
|
206 |
+
"learning_rate": 1.5932040657672757e-05,
|
207 |
+
"loss": 2.0237,
|
208 |
+
"step": 160
|
209 |
+
},
|
210 |
+
{
|
211 |
+
"epoch": 1.26,
|
212 |
+
"learning_rate": 2.1568161872809022e-05,
|
213 |
+
"loss": 2.1338,
|
214 |
+
"step": 165
|
215 |
+
},
|
216 |
+
{
|
217 |
+
"epoch": 1.3,
|
218 |
+
"learning_rate": 2.7879694970972374e-05,
|
219 |
+
"loss": 2.1066,
|
220 |
+
"step": 170
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 1.34,
|
224 |
+
"learning_rate": 3.477600168191081e-05,
|
225 |
+
"loss": 2.1939,
|
226 |
+
"step": 175
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 1.37,
|
230 |
+
"learning_rate": 4.215804595500327e-05,
|
231 |
+
"loss": 2.1481,
|
232 |
+
"step": 180
|
233 |
+
},
|
234 |
+
{
|
235 |
+
"epoch": 1.41,
|
236 |
+
"learning_rate": 4.991981618998877e-05,
|
237 |
+
"loss": 2.4364,
|
238 |
+
"step": 185
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 1.45,
|
242 |
+
"learning_rate": 5.794984764173325e-05,
|
243 |
+
"loss": 2.1677,
|
244 |
+
"step": 190
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 1.49,
|
248 |
+
"learning_rate": 6.613282313617852e-05,
|
249 |
+
"loss": 2.3095,
|
250 |
+
"step": 195
|
251 |
+
},
|
252 |
+
{
|
253 |
+
"epoch": 1.53,
|
254 |
+
"learning_rate": 7.435122911001204e-05,
|
255 |
+
"loss": 2.4404,
|
256 |
+
"step": 200
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 1.56,
|
260 |
+
"learning_rate": 8.248704319210595e-05,
|
261 |
+
"loss": 2.0718,
|
262 |
+
"step": 205
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 1.6,
|
266 |
+
"learning_rate": 9.042342909181217e-05,
|
267 |
+
"loss": 2.2015,
|
268 |
+
"step": 210
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 1.64,
|
272 |
+
"learning_rate": 9.804641445426643e-05,
|
273 |
+
"loss": 2.1899,
|
274 |
+
"step": 215
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 1.68,
|
278 |
+
"learning_rate": 0.00010524652758746261,
|
279 |
+
"loss": 2.1465,
|
280 |
+
"step": 220
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 1.72,
|
284 |
+
"learning_rate": 0.00011192036955648884,
|
285 |
+
"loss": 2.3478,
|
286 |
+
"step": 225
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 1.76,
|
290 |
+
"learning_rate": 0.00011797209906849287,
|
291 |
+
"loss": 2.3704,
|
292 |
+
"step": 230
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 1.79,
|
296 |
+
"learning_rate": 0.0001233148088243337,
|
297 |
+
"loss": 2.2886,
|
298 |
+
"step": 235
|
299 |
+
},
|
300 |
+
{
|
301 |
+
"epoch": 1.83,
|
302 |
+
"learning_rate": 0.00012787177357149405,
|
303 |
+
"loss": 2.1345,
|
304 |
+
"step": 240
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 1.87,
|
308 |
+
"learning_rate": 0.00013157755193529395,
|
309 |
+
"loss": 2.0747,
|
310 |
+
"step": 245
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 1.91,
|
314 |
+
"learning_rate": 0.00013437892620529645,
|
315 |
+
"loss": 2.2118,
|
316 |
+
"step": 250
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 1.95,
|
320 |
+
"learning_rate": 0.000136235666580879,
|
321 |
+
"loss": 2.1172,
|
322 |
+
"step": 255
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 1.98,
|
326 |
+
"learning_rate": 0.00013712110890084145,
|
327 |
+
"loss": 2.2353,
|
328 |
+
"step": 260
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 2.0,
|
332 |
+
"eval_loss": 2.237405776977539,
|
333 |
+
"eval_runtime": 8.3034,
|
334 |
+
"eval_samples_per_second": 22.521,
|
335 |
+
"eval_steps_per_second": 2.89,
|
336 |
+
"step": 262
|
337 |
}
|
338 |
],
|
339 |
+
"max_steps": 262,
|
340 |
+
"num_train_epochs": 2,
|
341 |
+
"total_flos": 272919527424000.0,
|
342 |
"trial_name": null,
|
343 |
"trial_params": null
|
344 |
}
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 2671
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:74c710dd5a8d4b67cb50ad0843ddfa8eb650173d3491cfb493f1dda199d69582
|
3 |
size 2671
|