AlekseyKorshuk
commited on
Commit
·
fac7d7c
1
Parent(s):
2a429ab
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,94 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: other
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- accuracy
|
7 |
+
model-index:
|
8 |
+
- name: amazon-reviews-input-output-6.7b
|
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 |
+
# amazon-reviews-input-output-6.7b
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [facebook/opt-6.7b](https://huggingface.co/facebook/opt-6.7b) on an unknown dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 2.8574
|
20 |
+
- Accuracy: 0.0388
|
21 |
+
|
22 |
+
## Model description
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Intended uses & limitations
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training and evaluation data
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training procedure
|
35 |
+
|
36 |
+
### Training hyperparameters
|
37 |
+
|
38 |
+
The following hyperparameters were used during training:
|
39 |
+
- learning_rate: 5e-05
|
40 |
+
- train_batch_size: 8
|
41 |
+
- eval_batch_size: 8
|
42 |
+
- seed: 42
|
43 |
+
- distributed_type: multi-GPU
|
44 |
+
- num_devices: 8
|
45 |
+
- total_train_batch_size: 64
|
46 |
+
- total_eval_batch_size: 64
|
47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
+
- lr_scheduler_type: linear
|
49 |
+
- num_epochs: 2.0
|
50 |
+
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
54 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
55 |
+
| 2.9912 | 0.06 | 1 | 2.7441 | 0.0404 |
|
56 |
+
| 2.9329 | 0.12 | 2 | 2.7441 | 0.0404 |
|
57 |
+
| 2.9138 | 0.19 | 3 | 2.8262 | 0.0389 |
|
58 |
+
| 2.9395 | 0.25 | 4 | 2.8262 | 0.0389 |
|
59 |
+
| 2.9109 | 0.31 | 5 | 2.7949 | 0.0399 |
|
60 |
+
| 2.8394 | 0.38 | 6 | 2.7461 | 0.0403 |
|
61 |
+
| 2.9365 | 0.44 | 7 | 2.7207 | 0.0399 |
|
62 |
+
| 2.7588 | 0.5 | 8 | 2.7070 | 0.0403 |
|
63 |
+
| 2.9751 | 0.56 | 9 | 2.6816 | 0.0407 |
|
64 |
+
| 2.844 | 0.62 | 10 | 2.6738 | 0.0404 |
|
65 |
+
| 2.731 | 0.69 | 11 | 2.6680 | 0.0406 |
|
66 |
+
| 2.7434 | 0.75 | 12 | 2.6699 | 0.0404 |
|
67 |
+
| 2.9043 | 0.81 | 13 | 2.6855 | 0.0400 |
|
68 |
+
| 2.8564 | 0.88 | 14 | 2.6855 | 0.0400 |
|
69 |
+
| 2.8716 | 0.94 | 15 | 2.6855 | 0.0400 |
|
70 |
+
| 2.896 | 1.0 | 16 | 2.6953 | 0.0398 |
|
71 |
+
| 1.9858 | 1.06 | 17 | 2.7070 | 0.0400 |
|
72 |
+
| 2.0563 | 1.12 | 18 | 2.7285 | 0.0400 |
|
73 |
+
| 2.04 | 1.19 | 19 | 2.7676 | 0.0398 |
|
74 |
+
| 1.9885 | 1.25 | 20 | 2.7910 | 0.0396 |
|
75 |
+
| 2.09 | 1.31 | 21 | 2.7969 | 0.0393 |
|
76 |
+
| 2.059 | 1.38 | 22 | 2.8105 | 0.0395 |
|
77 |
+
| 2.0498 | 1.44 | 23 | 2.7930 | 0.0398 |
|
78 |
+
| 1.9568 | 1.5 | 24 | 2.7910 | 0.0401 |
|
79 |
+
| 2.1418 | 1.56 | 25 | 2.7930 | 0.0398 |
|
80 |
+
| 1.975 | 1.62 | 26 | 2.7930 | 0.0397 |
|
81 |
+
| 1.996 | 1.69 | 27 | 2.7949 | 0.0393 |
|
82 |
+
| 1.9617 | 1.75 | 28 | 2.8047 | 0.0392 |
|
83 |
+
| 2.2062 | 1.81 | 29 | 2.8145 | 0.0388 |
|
84 |
+
| 1.9929 | 1.88 | 30 | 2.8145 | 0.0386 |
|
85 |
+
| 1.9235 | 1.94 | 31 | 2.8281 | 0.0390 |
|
86 |
+
| 1.9127 | 2.0 | 32 | 2.8574 | 0.0388 |
|
87 |
+
|
88 |
+
|
89 |
+
### Framework versions
|
90 |
+
|
91 |
+
- Transformers 4.25.0.dev0
|
92 |
+
- Pytorch 1.12.1+cu113
|
93 |
+
- Datasets 2.3.2
|
94 |
+
- Tokenizers 0.12.1
|