jlpan commited on
Commit
d50d161
·
1 Parent(s): 4ef90a1

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -35
README.md CHANGED
@@ -6,7 +6,6 @@ tags:
6
  model-index:
7
  - name: starcoder-finetuned-test_newSnippet
8
  results: []
9
- library_name: peft
10
  ---
11
 
12
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [bigcode/starcoder](https://huggingface.co/bigcode/starcoder) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 0.1728
20
 
21
  ## Model description
22
 
@@ -35,7 +34,7 @@ More information needed
35
  ### Training hyperparameters
36
 
37
  The following hyperparameters were used during training:
38
- - learning_rate: 8e-05
39
  - train_batch_size: 32
40
  - eval_batch_size: 32
41
  - seed: 42
@@ -43,47 +42,28 @@ The following hyperparameters were used during training:
43
  - total_train_batch_size: 256
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: cosine
46
- - lr_scheduler_warmup_steps: 80
47
- - training_steps: 800
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss |
52
  |:-------------:|:-----:|:----:|:---------------:|
53
- | 4.5787 | 0.06 | 50 | 0.3535 |
54
- | 0.236 | 0.12 | 100 | 0.1948 |
55
- | 0.1862 | 1.01 | 150 | 0.1847 |
56
- | 0.1866 | 1.07 | 200 | 0.1808 |
57
- | 0.1838 | 1.13 | 250 | 0.1794 |
58
- | 0.1718 | 2.02 | 300 | 0.1772 |
59
- | 0.1796 | 2.08 | 350 | 0.1761 |
60
- | 0.178 | 2.14 | 400 | 0.1762 |
61
- | 0.1666 | 3.03 | 450 | 0.1743 |
62
- | 0.1772 | 3.09 | 500 | 0.1739 |
63
- | 0.1739 | 3.15 | 550 | 0.1746 |
64
- | 0.1652 | 4.04 | 600 | 0.1731 |
65
- | 0.1755 | 4.1 | 650 | 0.1731 |
66
- | 0.1706 | 4.16 | 700 | 0.1735 |
67
- | 0.1668 | 5.04 | 750 | 0.1728 |
68
- | 0.1747 | 5.11 | 800 | 0.1728 |
69
 
70
 
71
  ### Framework versions
72
 
73
- - PEFT 0.5.0.dev0
74
- - PEFT 0.5.0.dev0
75
- - PEFT 0.5.0.dev0
76
- - PEFT 0.5.0.dev0
77
- - PEFT 0.5.0.dev0
78
- - PEFT 0.5.0.dev0
79
- - PEFT 0.5.0.dev0
80
- - PEFT 0.5.0.dev0
81
- - PEFT 0.5.0.dev0
82
- - PEFT 0.5.0.dev0
83
- - PEFT 0.5.0.dev0
84
- - PEFT 0.5.0.dev0
85
- - PEFT 0.5.0.dev0
86
- - PEFT 0.5.0.dev0
87
  - Transformers 4.32.0.dev0
88
  - Pytorch 2.0.1+cu117
89
  - Datasets 2.12.0
 
6
  model-index:
7
  - name: starcoder-finetuned-test_newSnippet
8
  results: []
 
9
  ---
10
 
11
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
15
 
16
  This model is a fine-tuned version of [bigcode/starcoder](https://huggingface.co/bigcode/starcoder) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.1863
19
 
20
  ## Model description
21
 
 
34
  ### Training hyperparameters
35
 
36
  The following hyperparameters were used during training:
37
+ - learning_rate: 5e-05
38
  - train_batch_size: 32
39
  - eval_batch_size: 32
40
  - seed: 42
 
42
  - total_train_batch_size: 256
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: cosine
45
+ - lr_scheduler_warmup_steps: 25
46
+ - training_steps: 275
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
+ | 6.5337 | 0.09 | 25 | 0.7416 |
53
+ | 0.4715 | 0.18 | 50 | 0.2515 |
54
+ | 0.2329 | 0.27 | 75 | 0.2060 |
55
+ | 0.2093 | 0.36 | 100 | 0.1973 |
56
+ | 0.1994 | 0.45 | 125 | 0.1935 |
57
+ | 0.1836 | 1.03 | 150 | 0.1893 |
58
+ | 0.1912 | 1.12 | 175 | 0.1877 |
59
+ | 0.1947 | 1.21 | 200 | 0.1870 |
60
+ | 0.194 | 1.3 | 225 | 0.1865 |
61
+ | 0.1908 | 1.39 | 250 | 0.1863 |
62
+ | 0.1845 | 1.48 | 275 | 0.1863 |
 
 
 
 
 
63
 
64
 
65
  ### Framework versions
66
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
67
  - Transformers 4.32.0.dev0
68
  - Pytorch 2.0.1+cu117
69
  - Datasets 2.12.0