alexandreacff
commited on
Commit
•
8087594
1
Parent(s):
8baafd4
End of training
Browse files- README.md +3 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/May09_16-59-59_5c7bc4386fe7/events.out.tfevents.1715275977.5c7bc4386fe7.1797.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +363 -0
README.md
CHANGED
@@ -2,6 +2,7 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: asapp/sew-mid-100k
|
4 |
tags:
|
|
|
5 |
- generated_from_trainer
|
6 |
metrics:
|
7 |
- accuracy
|
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# sew-ft-fake-detection
|
17 |
|
18 |
-
This model is a fine-tuned version of [asapp/sew-mid-100k](https://huggingface.co/asapp/sew-mid-100k) on
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss: 0.
|
21 |
- Accuracy: 0.7439
|
22 |
|
23 |
## Model description
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: asapp/sew-mid-100k
|
4 |
tags:
|
5 |
+
- audio-classification
|
6 |
- generated_from_trainer
|
7 |
metrics:
|
8 |
- accuracy
|
|
|
16 |
|
17 |
# sew-ft-fake-detection
|
18 |
|
19 |
+
This model is a fine-tuned version of [asapp/sew-mid-100k](https://huggingface.co/asapp/sew-mid-100k) on the alexandreacff/kaggle-fake-detection dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 0.6232
|
22 |
- Accuracy: 0.7439
|
23 |
|
24 |
## Model description
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 9.850746268656717,
|
3 |
+
"eval_accuracy": 0.7439252336448599,
|
4 |
+
"eval_loss": 0.623187780380249,
|
5 |
+
"eval_runtime": 38.2416,
|
6 |
+
"eval_samples_per_second": 27.98,
|
7 |
+
"eval_steps_per_second": 1.752,
|
8 |
+
"total_flos": 7.07452399466208e+17,
|
9 |
+
"train_loss": 0.5083336223255505,
|
10 |
+
"train_runtime": 1835.4297,
|
11 |
+
"train_samples_per_second": 23.302,
|
12 |
+
"train_steps_per_second": 0.18
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 9.850746268656717,
|
3 |
+
"eval_accuracy": 0.7439252336448599,
|
4 |
+
"eval_loss": 0.623187780380249,
|
5 |
+
"eval_runtime": 38.2416,
|
6 |
+
"eval_samples_per_second": 27.98,
|
7 |
+
"eval_steps_per_second": 1.752
|
8 |
+
}
|
runs/May09_16-59-59_5c7bc4386fe7/events.out.tfevents.1715275977.5c7bc4386fe7.1797.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b4a29843ffe2c9d503136f1e5057965e0b89876de4350c48afc3bb3caf91ecfe
|
3 |
+
size 363
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 9.850746268656717,
|
3 |
+
"total_flos": 7.07452399466208e+17,
|
4 |
+
"train_loss": 0.5083336223255505,
|
5 |
+
"train_runtime": 1835.4297,
|
6 |
+
"train_samples_per_second": 23.302,
|
7 |
+
"train_steps_per_second": 0.18
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,363 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.7439252336448599,
|
3 |
+
"best_model_checkpoint": "sew-ft-fake-detection/checkpoint-301",
|
4 |
+
"epoch": 9.850746268656717,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 330,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.29850746268656714,
|
13 |
+
"grad_norm": 1.730605125427246,
|
14 |
+
"learning_rate": 9.090909090909091e-06,
|
15 |
+
"loss": 0.6755,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.5970149253731343,
|
20 |
+
"grad_norm": 0.684485673904419,
|
21 |
+
"learning_rate": 1.8181818181818182e-05,
|
22 |
+
"loss": 0.6506,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.8955223880597015,
|
27 |
+
"grad_norm": 0.7370636463165283,
|
28 |
+
"learning_rate": 2.7272727272727273e-05,
|
29 |
+
"loss": 0.6344,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.9850746268656716,
|
34 |
+
"eval_accuracy": 0.6504672897196262,
|
35 |
+
"eval_loss": 0.6394948959350586,
|
36 |
+
"eval_runtime": 30.1661,
|
37 |
+
"eval_samples_per_second": 35.47,
|
38 |
+
"eval_steps_per_second": 2.221,
|
39 |
+
"step": 33
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 1.1940298507462686,
|
43 |
+
"grad_norm": 1.093192458152771,
|
44 |
+
"learning_rate": 2.9292929292929294e-05,
|
45 |
+
"loss": 0.635,
|
46 |
+
"step": 40
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 1.4925373134328357,
|
50 |
+
"grad_norm": 0.6133630275726318,
|
51 |
+
"learning_rate": 2.8282828282828285e-05,
|
52 |
+
"loss": 0.6023,
|
53 |
+
"step": 50
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 1.7910447761194028,
|
57 |
+
"grad_norm": 0.9280603528022766,
|
58 |
+
"learning_rate": 2.7272727272727273e-05,
|
59 |
+
"loss": 0.6157,
|
60 |
+
"step": 60
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 2.0,
|
64 |
+
"eval_accuracy": 0.5214953271028038,
|
65 |
+
"eval_loss": 0.6703016757965088,
|
66 |
+
"eval_runtime": 34.578,
|
67 |
+
"eval_samples_per_second": 30.945,
|
68 |
+
"eval_steps_per_second": 1.938,
|
69 |
+
"step": 67
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 2.08955223880597,
|
73 |
+
"grad_norm": 0.9497929215431213,
|
74 |
+
"learning_rate": 2.6262626262626265e-05,
|
75 |
+
"loss": 0.5853,
|
76 |
+
"step": 70
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 2.388059701492537,
|
80 |
+
"grad_norm": 1.5462061166763306,
|
81 |
+
"learning_rate": 2.5252525252525256e-05,
|
82 |
+
"loss": 0.5659,
|
83 |
+
"step": 80
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 2.6865671641791042,
|
87 |
+
"grad_norm": 1.5534844398498535,
|
88 |
+
"learning_rate": 2.4242424242424244e-05,
|
89 |
+
"loss": 0.5512,
|
90 |
+
"step": 90
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"epoch": 2.9850746268656714,
|
94 |
+
"grad_norm": 1.4848833084106445,
|
95 |
+
"learning_rate": 2.3232323232323235e-05,
|
96 |
+
"loss": 0.5572,
|
97 |
+
"step": 100
|
98 |
+
},
|
99 |
+
{
|
100 |
+
"epoch": 2.9850746268656714,
|
101 |
+
"eval_accuracy": 0.7130841121495327,
|
102 |
+
"eval_loss": 0.561539351940155,
|
103 |
+
"eval_runtime": 36.5879,
|
104 |
+
"eval_samples_per_second": 29.245,
|
105 |
+
"eval_steps_per_second": 1.831,
|
106 |
+
"step": 100
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 3.283582089552239,
|
110 |
+
"grad_norm": 2.040893316268921,
|
111 |
+
"learning_rate": 2.222222222222222e-05,
|
112 |
+
"loss": 0.5456,
|
113 |
+
"step": 110
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 3.582089552238806,
|
117 |
+
"grad_norm": 1.084596037864685,
|
118 |
+
"learning_rate": 2.121212121212121e-05,
|
119 |
+
"loss": 0.5214,
|
120 |
+
"step": 120
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 3.8805970149253732,
|
124 |
+
"grad_norm": 1.547035574913025,
|
125 |
+
"learning_rate": 2.0202020202020203e-05,
|
126 |
+
"loss": 0.4968,
|
127 |
+
"step": 130
|
128 |
+
},
|
129 |
+
{
|
130 |
+
"epoch": 4.0,
|
131 |
+
"eval_accuracy": 0.694392523364486,
|
132 |
+
"eval_loss": 0.8148747086524963,
|
133 |
+
"eval_runtime": 39.5102,
|
134 |
+
"eval_samples_per_second": 27.082,
|
135 |
+
"eval_steps_per_second": 1.696,
|
136 |
+
"step": 134
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 4.17910447761194,
|
140 |
+
"grad_norm": 1.8584659099578857,
|
141 |
+
"learning_rate": 1.919191919191919e-05,
|
142 |
+
"loss": 0.4943,
|
143 |
+
"step": 140
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 4.477611940298507,
|
147 |
+
"grad_norm": 2.004117727279663,
|
148 |
+
"learning_rate": 1.8181818181818182e-05,
|
149 |
+
"loss": 0.5122,
|
150 |
+
"step": 150
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 4.776119402985074,
|
154 |
+
"grad_norm": 1.3472633361816406,
|
155 |
+
"learning_rate": 1.717171717171717e-05,
|
156 |
+
"loss": 0.4988,
|
157 |
+
"step": 160
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 4.985074626865671,
|
161 |
+
"eval_accuracy": 0.7037383177570093,
|
162 |
+
"eval_loss": 0.8099448680877686,
|
163 |
+
"eval_runtime": 35.842,
|
164 |
+
"eval_samples_per_second": 29.853,
|
165 |
+
"eval_steps_per_second": 1.869,
|
166 |
+
"step": 167
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 5.074626865671641,
|
170 |
+
"grad_norm": 1.908387303352356,
|
171 |
+
"learning_rate": 1.616161616161616e-05,
|
172 |
+
"loss": 0.4812,
|
173 |
+
"step": 170
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 5.373134328358209,
|
177 |
+
"grad_norm": 2.6505331993103027,
|
178 |
+
"learning_rate": 1.5151515151515153e-05,
|
179 |
+
"loss": 0.4671,
|
180 |
+
"step": 180
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 5.6716417910447765,
|
184 |
+
"grad_norm": 2.50083327293396,
|
185 |
+
"learning_rate": 1.4141414141414143e-05,
|
186 |
+
"loss": 0.4538,
|
187 |
+
"step": 190
|
188 |
+
},
|
189 |
+
{
|
190 |
+
"epoch": 5.970149253731344,
|
191 |
+
"grad_norm": 1.8395198583602905,
|
192 |
+
"learning_rate": 1.3131313131313132e-05,
|
193 |
+
"loss": 0.4756,
|
194 |
+
"step": 200
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 6.0,
|
198 |
+
"eval_accuracy": 0.7102803738317757,
|
199 |
+
"eval_loss": 0.8318859934806824,
|
200 |
+
"eval_runtime": 36.3806,
|
201 |
+
"eval_samples_per_second": 29.411,
|
202 |
+
"eval_steps_per_second": 1.842,
|
203 |
+
"step": 201
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 6.268656716417911,
|
207 |
+
"grad_norm": 1.486412763595581,
|
208 |
+
"learning_rate": 1.2121212121212122e-05,
|
209 |
+
"loss": 0.4576,
|
210 |
+
"step": 210
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 6.567164179104478,
|
214 |
+
"grad_norm": 1.7824214696884155,
|
215 |
+
"learning_rate": 1.111111111111111e-05,
|
216 |
+
"loss": 0.4462,
|
217 |
+
"step": 220
|
218 |
+
},
|
219 |
+
{
|
220 |
+
"epoch": 6.865671641791045,
|
221 |
+
"grad_norm": 2.2163608074188232,
|
222 |
+
"learning_rate": 1.0101010101010101e-05,
|
223 |
+
"loss": 0.4334,
|
224 |
+
"step": 230
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 6.985074626865671,
|
228 |
+
"eval_accuracy": 0.7336448598130841,
|
229 |
+
"eval_loss": 0.7316926717758179,
|
230 |
+
"eval_runtime": 38.478,
|
231 |
+
"eval_samples_per_second": 27.808,
|
232 |
+
"eval_steps_per_second": 1.741,
|
233 |
+
"step": 234
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 7.164179104477612,
|
237 |
+
"grad_norm": 2.735869884490967,
|
238 |
+
"learning_rate": 9.090909090909091e-06,
|
239 |
+
"loss": 0.4686,
|
240 |
+
"step": 240
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 7.462686567164179,
|
244 |
+
"grad_norm": 2.0291202068328857,
|
245 |
+
"learning_rate": 8.08080808080808e-06,
|
246 |
+
"loss": 0.4604,
|
247 |
+
"step": 250
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 7.7611940298507465,
|
251 |
+
"grad_norm": 1.6389554738998413,
|
252 |
+
"learning_rate": 7.070707070707071e-06,
|
253 |
+
"loss": 0.4321,
|
254 |
+
"step": 260
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 8.0,
|
258 |
+
"eval_accuracy": 0.7383177570093458,
|
259 |
+
"eval_loss": 0.6548246145248413,
|
260 |
+
"eval_runtime": 40.7344,
|
261 |
+
"eval_samples_per_second": 26.268,
|
262 |
+
"eval_steps_per_second": 1.645,
|
263 |
+
"step": 268
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 8.059701492537313,
|
267 |
+
"grad_norm": 1.9984620809555054,
|
268 |
+
"learning_rate": 6.060606060606061e-06,
|
269 |
+
"loss": 0.4594,
|
270 |
+
"step": 270
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 8.35820895522388,
|
274 |
+
"grad_norm": 1.9911682605743408,
|
275 |
+
"learning_rate": 5.050505050505051e-06,
|
276 |
+
"loss": 0.4236,
|
277 |
+
"step": 280
|
278 |
+
},
|
279 |
+
{
|
280 |
+
"epoch": 8.656716417910447,
|
281 |
+
"grad_norm": 2.0912842750549316,
|
282 |
+
"learning_rate": 4.04040404040404e-06,
|
283 |
+
"loss": 0.4326,
|
284 |
+
"step": 290
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 8.955223880597014,
|
288 |
+
"grad_norm": 2.4890761375427246,
|
289 |
+
"learning_rate": 3.0303030303030305e-06,
|
290 |
+
"loss": 0.4436,
|
291 |
+
"step": 300
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 8.985074626865671,
|
295 |
+
"eval_accuracy": 0.7439252336448599,
|
296 |
+
"eval_loss": 0.623187780380249,
|
297 |
+
"eval_runtime": 33.9508,
|
298 |
+
"eval_samples_per_second": 31.516,
|
299 |
+
"eval_steps_per_second": 1.973,
|
300 |
+
"step": 301
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 9.253731343283581,
|
304 |
+
"grad_norm": 2.2211296558380127,
|
305 |
+
"learning_rate": 2.02020202020202e-06,
|
306 |
+
"loss": 0.438,
|
307 |
+
"step": 310
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 9.552238805970148,
|
311 |
+
"grad_norm": 1.8036954402923584,
|
312 |
+
"learning_rate": 1.01010101010101e-06,
|
313 |
+
"loss": 0.4104,
|
314 |
+
"step": 320
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 9.850746268656717,
|
318 |
+
"grad_norm": 1.8852014541625977,
|
319 |
+
"learning_rate": 0.0,
|
320 |
+
"loss": 0.4493,
|
321 |
+
"step": 330
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 9.850746268656717,
|
325 |
+
"eval_accuracy": 0.7439252336448599,
|
326 |
+
"eval_loss": 0.6278349757194519,
|
327 |
+
"eval_runtime": 37.2552,
|
328 |
+
"eval_samples_per_second": 28.721,
|
329 |
+
"eval_steps_per_second": 1.798,
|
330 |
+
"step": 330
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 9.850746268656717,
|
334 |
+
"step": 330,
|
335 |
+
"total_flos": 7.07452399466208e+17,
|
336 |
+
"train_loss": 0.5083336223255505,
|
337 |
+
"train_runtime": 1835.4297,
|
338 |
+
"train_samples_per_second": 23.302,
|
339 |
+
"train_steps_per_second": 0.18
|
340 |
+
}
|
341 |
+
],
|
342 |
+
"logging_steps": 10,
|
343 |
+
"max_steps": 330,
|
344 |
+
"num_input_tokens_seen": 0,
|
345 |
+
"num_train_epochs": 10,
|
346 |
+
"save_steps": 500,
|
347 |
+
"stateful_callbacks": {
|
348 |
+
"TrainerControl": {
|
349 |
+
"args": {
|
350 |
+
"should_epoch_stop": false,
|
351 |
+
"should_evaluate": false,
|
352 |
+
"should_log": false,
|
353 |
+
"should_save": true,
|
354 |
+
"should_training_stop": true
|
355 |
+
},
|
356 |
+
"attributes": {}
|
357 |
+
}
|
358 |
+
},
|
359 |
+
"total_flos": 7.07452399466208e+17,
|
360 |
+
"train_batch_size": 32,
|
361 |
+
"trial_name": null,
|
362 |
+
"trial_params": null
|
363 |
+
}
|