update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,229 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- yelp_review_full
|
7 |
+
metrics:
|
8 |
+
- accuracy
|
9 |
+
model-index:
|
10 |
+
- name: YELP_DistilBERT_5E
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: Text Classification
|
14 |
+
type: text-classification
|
15 |
+
dataset:
|
16 |
+
name: yelp_review_full
|
17 |
+
type: yelp_review_full
|
18 |
+
config: yelp_review_full
|
19 |
+
split: train
|
20 |
+
args: yelp_review_full
|
21 |
+
metrics:
|
22 |
+
- name: Accuracy
|
23 |
+
type: accuracy
|
24 |
+
value: 0.9666666666666667
|
25 |
+
---
|
26 |
+
|
27 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
28 |
+
should probably proofread and complete it, then remove this comment. -->
|
29 |
+
|
30 |
+
# YELP_DistilBERT_5E
|
31 |
+
|
32 |
+
This model is a fine-tuned version of [distilbert-base-cased](https://huggingface.co/distilbert-base-cased) on the yelp_review_full dataset.
|
33 |
+
It achieves the following results on the evaluation set:
|
34 |
+
- Loss: 0.1557
|
35 |
+
- Accuracy: 0.9667
|
36 |
+
|
37 |
+
## Model description
|
38 |
+
|
39 |
+
More information needed
|
40 |
+
|
41 |
+
## Intended uses & limitations
|
42 |
+
|
43 |
+
More information needed
|
44 |
+
|
45 |
+
## Training and evaluation data
|
46 |
+
|
47 |
+
More information needed
|
48 |
+
|
49 |
+
## Training procedure
|
50 |
+
|
51 |
+
### Training hyperparameters
|
52 |
+
|
53 |
+
The following hyperparameters were used during training:
|
54 |
+
- learning_rate: 1e-05
|
55 |
+
- train_batch_size: 16
|
56 |
+
- eval_batch_size: 8
|
57 |
+
- seed: 42
|
58 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
59 |
+
- lr_scheduler_type: linear
|
60 |
+
- num_epochs: 5
|
61 |
+
|
62 |
+
### Training results
|
63 |
+
|
64 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
65 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
66 |
+
| 0.6211 | 0.03 | 50 | 0.3873 | 0.8933 |
|
67 |
+
| 0.3252 | 0.06 | 100 | 0.2181 | 0.92 |
|
68 |
+
| 0.2241 | 0.1 | 150 | 0.1850 | 0.94 |
|
69 |
+
| 0.2645 | 0.13 | 200 | 0.1514 | 0.9467 |
|
70 |
+
| 0.2094 | 0.16 | 250 | 0.1850 | 0.92 |
|
71 |
+
| 0.2693 | 0.19 | 300 | 0.1504 | 0.9467 |
|
72 |
+
| 0.2524 | 0.22 | 350 | 0.1479 | 0.96 |
|
73 |
+
| 0.2538 | 0.26 | 400 | 0.1375 | 0.94 |
|
74 |
+
| 0.1937 | 0.29 | 450 | 0.1204 | 0.9467 |
|
75 |
+
| 0.1692 | 0.32 | 500 | 0.1396 | 0.9533 |
|
76 |
+
| 0.1987 | 0.35 | 550 | 0.1151 | 0.94 |
|
77 |
+
| 0.207 | 0.38 | 600 | 0.1705 | 0.94 |
|
78 |
+
| 0.2135 | 0.42 | 650 | 0.1189 | 0.9467 |
|
79 |
+
| 0.1847 | 0.45 | 700 | 0.1315 | 0.9533 |
|
80 |
+
| 0.169 | 0.48 | 750 | 0.1407 | 0.9533 |
|
81 |
+
| 0.1767 | 0.51 | 800 | 0.1675 | 0.9333 |
|
82 |
+
| 0.1899 | 0.54 | 850 | 0.0913 | 0.9467 |
|
83 |
+
| 0.1641 | 0.58 | 900 | 0.0954 | 0.96 |
|
84 |
+
| 0.1765 | 0.61 | 950 | 0.1237 | 0.9467 |
|
85 |
+
| 0.1663 | 0.64 | 1000 | 0.1029 | 0.9533 |
|
86 |
+
| 0.1238 | 0.67 | 1050 | 0.1267 | 0.96 |
|
87 |
+
| 0.2087 | 0.7 | 1100 | 0.1111 | 0.96 |
|
88 |
+
| 0.1354 | 0.74 | 1150 | 0.0916 | 0.9667 |
|
89 |
+
| 0.1937 | 0.77 | 1200 | 0.1059 | 0.96 |
|
90 |
+
| 0.2216 | 0.8 | 1250 | 0.1049 | 0.9467 |
|
91 |
+
| 0.1788 | 0.83 | 1300 | 0.1472 | 0.94 |
|
92 |
+
| 0.2138 | 0.86 | 1350 | 0.1234 | 0.9467 |
|
93 |
+
| 0.1555 | 0.9 | 1400 | 0.1386 | 0.94 |
|
94 |
+
| 0.1583 | 0.93 | 1450 | 0.1642 | 0.9467 |
|
95 |
+
| 0.1525 | 0.96 | 1500 | 0.1571 | 0.94 |
|
96 |
+
| 0.2049 | 0.99 | 1550 | 0.1257 | 0.9333 |
|
97 |
+
| 0.1266 | 1.02 | 1600 | 0.1677 | 0.94 |
|
98 |
+
| 0.1282 | 1.06 | 1650 | 0.1307 | 0.9533 |
|
99 |
+
| 0.1007 | 1.09 | 1700 | 0.1375 | 0.9533 |
|
100 |
+
| 0.0991 | 1.12 | 1750 | 0.1513 | 0.9533 |
|
101 |
+
| 0.1211 | 1.15 | 1800 | 0.1229 | 0.9667 |
|
102 |
+
| 0.1833 | 1.18 | 1850 | 0.1105 | 0.9733 |
|
103 |
+
| 0.1596 | 1.22 | 1900 | 0.1279 | 0.9533 |
|
104 |
+
| 0.1172 | 1.25 | 1950 | 0.1124 | 0.96 |
|
105 |
+
| 0.1137 | 1.28 | 2000 | 0.1407 | 0.9467 |
|
106 |
+
| 0.1135 | 1.31 | 2050 | 0.1377 | 0.96 |
|
107 |
+
| 0.096 | 1.34 | 2100 | 0.1022 | 0.9667 |
|
108 |
+
| 0.1203 | 1.38 | 2150 | 0.1719 | 0.9467 |
|
109 |
+
| 0.1289 | 1.41 | 2200 | 0.1254 | 0.9667 |
|
110 |
+
| 0.1392 | 1.44 | 2250 | 0.1086 | 0.9667 |
|
111 |
+
| 0.1319 | 1.47 | 2300 | 0.1511 | 0.9467 |
|
112 |
+
| 0.1161 | 1.5 | 2350 | 0.1758 | 0.9467 |
|
113 |
+
| 0.1402 | 1.54 | 2400 | 0.1369 | 0.96 |
|
114 |
+
| 0.1433 | 1.57 | 2450 | 0.1495 | 0.9667 |
|
115 |
+
| 0.1882 | 1.6 | 2500 | 0.1186 | 0.9467 |
|
116 |
+
| 0.1474 | 1.63 | 2550 | 0.1249 | 0.9533 |
|
117 |
+
| 0.0937 | 1.66 | 2600 | 0.1390 | 0.96 |
|
118 |
+
| 0.1231 | 1.7 | 2650 | 0.1467 | 0.96 |
|
119 |
+
| 0.1485 | 1.73 | 2700 | 0.1602 | 0.9533 |
|
120 |
+
| 0.1683 | 1.76 | 2750 | 0.1884 | 0.9533 |
|
121 |
+
| 0.1141 | 1.79 | 2800 | 0.1634 | 0.96 |
|
122 |
+
| 0.1351 | 1.82 | 2850 | 0.1212 | 0.9733 |
|
123 |
+
| 0.1298 | 1.86 | 2900 | 0.1224 | 0.96 |
|
124 |
+
| 0.1616 | 1.89 | 2950 | 0.1241 | 0.96 |
|
125 |
+
| 0.1159 | 1.92 | 3000 | 0.1532 | 0.9533 |
|
126 |
+
| 0.1101 | 1.95 | 3050 | 0.1105 | 0.96 |
|
127 |
+
| 0.0779 | 1.98 | 3100 | 0.1334 | 0.9533 |
|
128 |
+
| 0.1427 | 2.02 | 3150 | 0.1026 | 0.9733 |
|
129 |
+
| 0.0673 | 2.05 | 3200 | 0.1231 | 0.96 |
|
130 |
+
| 0.0901 | 2.08 | 3250 | 0.1077 | 0.9733 |
|
131 |
+
| 0.0532 | 2.11 | 3300 | 0.1385 | 0.9467 |
|
132 |
+
| 0.0984 | 2.14 | 3350 | 0.1432 | 0.9467 |
|
133 |
+
| 0.1006 | 2.18 | 3400 | 0.1183 | 0.9667 |
|
134 |
+
| 0.067 | 2.21 | 3450 | 0.1533 | 0.9533 |
|
135 |
+
| 0.0901 | 2.24 | 3500 | 0.1314 | 0.9733 |
|
136 |
+
| 0.0644 | 2.27 | 3550 | 0.1354 | 0.9667 |
|
137 |
+
| 0.076 | 2.3 | 3600 | 0.1548 | 0.96 |
|
138 |
+
| 0.0932 | 2.34 | 3650 | 0.1624 | 0.9667 |
|
139 |
+
| 0.0777 | 2.37 | 3700 | 0.1878 | 0.9533 |
|
140 |
+
| 0.106 | 2.4 | 3750 | 0.1721 | 0.96 |
|
141 |
+
| 0.0621 | 2.43 | 3800 | 0.1470 | 0.9667 |
|
142 |
+
| 0.0919 | 2.46 | 3850 | 0.1478 | 0.96 |
|
143 |
+
| 0.091 | 2.5 | 3900 | 0.1371 | 0.9667 |
|
144 |
+
| 0.0912 | 2.53 | 3950 | 0.1467 | 0.9667 |
|
145 |
+
| 0.0775 | 2.56 | 4000 | 0.1289 | 0.9733 |
|
146 |
+
| 0.1053 | 2.59 | 4050 | 0.1107 | 0.9733 |
|
147 |
+
| 0.063 | 2.62 | 4100 | 0.1031 | 0.9733 |
|
148 |
+
| 0.0859 | 2.66 | 4150 | 0.0953 | 0.98 |
|
149 |
+
| 0.084 | 2.69 | 4200 | 0.1216 | 0.9733 |
|
150 |
+
| 0.1215 | 2.72 | 4250 | 0.1025 | 0.9733 |
|
151 |
+
| 0.0675 | 2.75 | 4300 | 0.0992 | 0.9667 |
|
152 |
+
| 0.0608 | 2.78 | 4350 | 0.1288 | 0.96 |
|
153 |
+
| 0.0965 | 2.82 | 4400 | 0.1179 | 0.9667 |
|
154 |
+
| 0.061 | 2.85 | 4450 | 0.1178 | 0.9733 |
|
155 |
+
| 0.0821 | 2.88 | 4500 | 0.1188 | 0.9733 |
|
156 |
+
| 0.0802 | 2.91 | 4550 | 0.1423 | 0.9667 |
|
157 |
+
| 0.0901 | 2.94 | 4600 | 0.1367 | 0.96 |
|
158 |
+
| 0.1069 | 2.98 | 4650 | 0.1118 | 0.9733 |
|
159 |
+
| 0.0653 | 3.01 | 4700 | 0.1359 | 0.9533 |
|
160 |
+
| 0.0577 | 3.04 | 4750 | 0.1046 | 0.9667 |
|
161 |
+
| 0.0467 | 3.07 | 4800 | 0.1366 | 0.96 |
|
162 |
+
| 0.041 | 3.1 | 4850 | 0.1276 | 0.9667 |
|
163 |
+
| 0.0585 | 3.13 | 4900 | 0.1426 | 0.9667 |
|
164 |
+
| 0.0635 | 3.17 | 4950 | 0.1571 | 0.96 |
|
165 |
+
| 0.0395 | 3.2 | 5000 | 0.1527 | 0.96 |
|
166 |
+
| 0.034 | 3.23 | 5050 | 0.1323 | 0.9667 |
|
167 |
+
| 0.0405 | 3.26 | 5100 | 0.1377 | 0.96 |
|
168 |
+
| 0.0306 | 3.29 | 5150 | 0.1526 | 0.9667 |
|
169 |
+
| 0.0471 | 3.33 | 5200 | 0.1419 | 0.9667 |
|
170 |
+
| 0.0646 | 3.36 | 5250 | 0.1459 | 0.9667 |
|
171 |
+
| 0.0508 | 3.39 | 5300 | 0.1312 | 0.9667 |
|
172 |
+
| 0.0593 | 3.42 | 5350 | 0.1483 | 0.96 |
|
173 |
+
| 0.05 | 3.45 | 5400 | 0.1076 | 0.9733 |
|
174 |
+
| 0.0559 | 3.49 | 5450 | 0.1412 | 0.9667 |
|
175 |
+
| 0.0614 | 3.52 | 5500 | 0.1597 | 0.9667 |
|
176 |
+
| 0.0691 | 3.55 | 5550 | 0.1656 | 0.96 |
|
177 |
+
| 0.0472 | 3.58 | 5600 | 0.1556 | 0.9667 |
|
178 |
+
| 0.055 | 3.61 | 5650 | 0.1347 | 0.9667 |
|
179 |
+
| 0.0564 | 3.65 | 5700 | 0.1424 | 0.96 |
|
180 |
+
| 0.0567 | 3.68 | 5750 | 0.1448 | 0.9733 |
|
181 |
+
| 0.0645 | 3.71 | 5800 | 0.1290 | 0.9667 |
|
182 |
+
| 0.0361 | 3.74 | 5850 | 0.1367 | 0.9667 |
|
183 |
+
| 0.0546 | 3.77 | 5900 | 0.1406 | 0.9667 |
|
184 |
+
| 0.043 | 3.81 | 5950 | 0.1337 | 0.96 |
|
185 |
+
| 0.0148 | 3.84 | 6000 | 0.1475 | 0.9533 |
|
186 |
+
| 0.0922 | 3.87 | 6050 | 0.1318 | 0.9733 |
|
187 |
+
| 0.0671 | 3.9 | 6100 | 0.1446 | 0.9733 |
|
188 |
+
| 0.0295 | 3.93 | 6150 | 0.1217 | 0.9733 |
|
189 |
+
| 0.0503 | 3.97 | 6200 | 0.1133 | 0.9733 |
|
190 |
+
| 0.0457 | 4.0 | 6250 | 0.1145 | 0.9733 |
|
191 |
+
| 0.0487 | 4.03 | 6300 | 0.1119 | 0.9733 |
|
192 |
+
| 0.0491 | 4.06 | 6350 | 0.1274 | 0.9667 |
|
193 |
+
| 0.0417 | 4.09 | 6400 | 0.1377 | 0.9733 |
|
194 |
+
| 0.0595 | 4.13 | 6450 | 0.1271 | 0.9733 |
|
195 |
+
| 0.035 | 4.16 | 6500 | 0.1183 | 0.9733 |
|
196 |
+
| 0.0482 | 4.19 | 6550 | 0.1153 | 0.9733 |
|
197 |
+
| 0.0196 | 4.22 | 6600 | 0.1388 | 0.9733 |
|
198 |
+
| 0.028 | 4.25 | 6650 | 0.1310 | 0.9733 |
|
199 |
+
| 0.0193 | 4.29 | 6700 | 0.1460 | 0.9667 |
|
200 |
+
| 0.0233 | 4.32 | 6750 | 0.1233 | 0.9733 |
|
201 |
+
| 0.0316 | 4.35 | 6800 | 0.1220 | 0.9667 |
|
202 |
+
| 0.0132 | 4.38 | 6850 | 0.1350 | 0.9533 |
|
203 |
+
| 0.0415 | 4.41 | 6900 | 0.1547 | 0.9667 |
|
204 |
+
| 0.0157 | 4.45 | 6950 | 0.1562 | 0.9667 |
|
205 |
+
| 0.0186 | 4.48 | 7000 | 0.1424 | 0.9667 |
|
206 |
+
| 0.0012 | 4.51 | 7050 | 0.1421 | 0.9667 |
|
207 |
+
| 0.0223 | 4.54 | 7100 | 0.1475 | 0.9733 |
|
208 |
+
| 0.0455 | 4.57 | 7150 | 0.1457 | 0.96 |
|
209 |
+
| 0.0571 | 4.61 | 7200 | 0.1559 | 0.9667 |
|
210 |
+
| 0.0305 | 4.64 | 7250 | 0.1614 | 0.9667 |
|
211 |
+
| 0.0457 | 4.67 | 7300 | 0.1691 | 0.9667 |
|
212 |
+
| 0.022 | 4.7 | 7350 | 0.1622 | 0.9667 |
|
213 |
+
| 0.0338 | 4.73 | 7400 | 0.1560 | 0.9667 |
|
214 |
+
| 0.0365 | 4.77 | 7450 | 0.1553 | 0.9667 |
|
215 |
+
| 0.025 | 4.8 | 7500 | 0.1512 | 0.9667 |
|
216 |
+
| 0.0441 | 4.83 | 7550 | 0.1550 | 0.9667 |
|
217 |
+
| 0.0363 | 4.86 | 7600 | 0.1564 | 0.9667 |
|
218 |
+
| 0.0188 | 4.89 | 7650 | 0.1553 | 0.9667 |
|
219 |
+
| 0.0427 | 4.93 | 7700 | 0.1572 | 0.9733 |
|
220 |
+
| 0.0362 | 4.96 | 7750 | 0.1568 | 0.9667 |
|
221 |
+
| 0.0115 | 4.99 | 7800 | 0.1557 | 0.9667 |
|
222 |
+
|
223 |
+
|
224 |
+
### Framework versions
|
225 |
+
|
226 |
+
- Transformers 4.24.0
|
227 |
+
- Pytorch 1.13.0
|
228 |
+
- Datasets 2.3.2
|
229 |
+
- Tokenizers 0.13.2
|