update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- natural_questions
|
7 |
+
model-index:
|
8 |
+
- name: toy-qa
|
9 |
+
results: []
|
10 |
+
---
|
11 |
+
|
12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
13 |
+
should probably proofread and complete it, then remove this comment. -->
|
14 |
+
|
15 |
+
# toy-qa
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the natural_questions dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.2284
|
20 |
+
|
21 |
+
## Model description
|
22 |
+
|
23 |
+
More information needed
|
24 |
+
|
25 |
+
## Intended uses & limitations
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Training and evaluation data
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training procedure
|
34 |
+
|
35 |
+
### Training hyperparameters
|
36 |
+
|
37 |
+
The following hyperparameters were used during training:
|
38 |
+
- learning_rate: 2e-05
|
39 |
+
- train_batch_size: 8
|
40 |
+
- eval_batch_size: 8
|
41 |
+
- seed: 42
|
42 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
43 |
+
- lr_scheduler_type: linear
|
44 |
+
- num_epochs: 50
|
45 |
+
|
46 |
+
### Training results
|
47 |
+
|
48 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
49 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
50 |
+
| 5.3253 | 1.0 | 14 | 3.3435 |
|
51 |
+
| 3.1799 | 2.0 | 28 | 0.6626 |
|
52 |
+
| 2.8858 | 3.0 | 42 | 0.7105 |
|
53 |
+
| 2.7921 | 4.0 | 56 | 0.5784 |
|
54 |
+
| 2.3648 | 5.0 | 70 | 0.5843 |
|
55 |
+
| 2.0496 | 6.0 | 84 | 0.3544 |
|
56 |
+
| 1.5883 | 7.0 | 98 | 0.3406 |
|
57 |
+
| 0.829 | 8.0 | 112 | 0.3717 |
|
58 |
+
| 0.5334 | 9.0 | 126 | 0.4049 |
|
59 |
+
| 0.4915 | 10.0 | 140 | 0.4101 |
|
60 |
+
| 0.57 | 11.0 | 154 | 0.3464 |
|
61 |
+
| 0.5129 | 12.0 | 168 | 0.3892 |
|
62 |
+
| 0.4334 | 13.0 | 182 | 0.2455 |
|
63 |
+
| 0.5406 | 14.0 | 196 | 0.2647 |
|
64 |
+
| 0.2201 | 15.0 | 210 | 0.2422 |
|
65 |
+
| 0.0917 | 16.0 | 224 | 0.2218 |
|
66 |
+
| 0.1753 | 17.0 | 238 | 0.2453 |
|
67 |
+
| 0.0231 | 18.0 | 252 | 0.2061 |
|
68 |
+
| 0.1107 | 19.0 | 266 | 0.2102 |
|
69 |
+
| 0.1182 | 20.0 | 280 | 0.2496 |
|
70 |
+
| 0.0829 | 21.0 | 294 | 0.2151 |
|
71 |
+
| 0.0775 | 22.0 | 308 | 0.2490 |
|
72 |
+
| 0.0067 | 23.0 | 322 | 0.2369 |
|
73 |
+
| 0.0058 | 24.0 | 336 | 0.2334 |
|
74 |
+
| 0.005 | 25.0 | 350 | 0.2288 |
|
75 |
+
| 0.0042 | 26.0 | 364 | 0.2148 |
|
76 |
+
| 0.0071 | 27.0 | 378 | 0.2582 |
|
77 |
+
| 0.0043 | 28.0 | 392 | 0.2232 |
|
78 |
+
| 0.0028 | 29.0 | 406 | 0.2170 |
|
79 |
+
| 0.0034 | 30.0 | 420 | 0.2290 |
|
80 |
+
| 0.0021 | 31.0 | 434 | 0.2299 |
|
81 |
+
| 0.0028 | 32.0 | 448 | 0.2214 |
|
82 |
+
| 0.0019 | 33.0 | 462 | 0.2312 |
|
83 |
+
| 0.002 | 34.0 | 476 | 0.2263 |
|
84 |
+
| 0.0024 | 35.0 | 490 | 0.2397 |
|
85 |
+
| 0.0019 | 36.0 | 504 | 0.2272 |
|
86 |
+
| 0.0087 | 37.0 | 518 | 0.2321 |
|
87 |
+
| 0.0016 | 38.0 | 532 | 0.2425 |
|
88 |
+
| 0.0016 | 39.0 | 546 | 0.2449 |
|
89 |
+
| 0.0014 | 40.0 | 560 | 0.2208 |
|
90 |
+
| 0.0013 | 41.0 | 574 | 0.2185 |
|
91 |
+
| 0.0013 | 42.0 | 588 | 0.2246 |
|
92 |
+
| 0.0016 | 43.0 | 602 | 0.2344 |
|
93 |
+
| 0.0016 | 44.0 | 616 | 0.2411 |
|
94 |
+
| 0.0015 | 45.0 | 630 | 0.2345 |
|
95 |
+
| 0.0015 | 46.0 | 644 | 0.2317 |
|
96 |
+
| 0.0014 | 47.0 | 658 | 0.2289 |
|
97 |
+
| 0.0011 | 48.0 | 672 | 0.2288 |
|
98 |
+
| 0.0012 | 49.0 | 686 | 0.2288 |
|
99 |
+
| 0.0013 | 50.0 | 700 | 0.2284 |
|
100 |
+
|
101 |
+
|
102 |
+
### Framework versions
|
103 |
+
|
104 |
+
- Transformers 4.17.0
|
105 |
+
- Pytorch 1.10.0
|
106 |
+
- Datasets 1.18.3
|
107 |
+
- Tokenizers 0.11.6
|