Tristan Thrush commited on
Commit
2a0f457
1 Parent(s): 6d4b9bc

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -10
README.md CHANGED
@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 0.6912
20
- - Accuracy: 0.57
21
 
22
  ## Model description
23
 
@@ -37,12 +37,14 @@ More information needed
37
 
38
  The following hyperparameters were used during training:
39
  - learning_rate: 2e-05
40
- - train_batch_size: 16
41
- - eval_batch_size: 16
42
  - seed: 42
43
  - distributed_type: multi-GPU
44
- - gradient_accumulation_steps: 4
 
45
  - total_train_batch_size: 64
 
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - num_epochs: 5
@@ -51,11 +53,11 @@ The following hyperparameters were used during training:
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
53
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
54
- | No log | 0.57 | 1 | 0.6911 | 0.58 |
55
- | No log | 1.57 | 2 | 0.6912 | 0.57 |
56
- | No log | 2.57 | 3 | 0.6912 | 0.57 |
57
- | No log | 3.57 | 4 | 0.6912 | 0.57 |
58
- | No log | 4.57 | 5 | 0.6912 | 0.57 |
59
 
60
 
61
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.7376
20
+ - Accuracy: 0.6020
21
 
22
  ## Model description
23
 
 
37
 
38
  The following hyperparameters were used during training:
39
  - learning_rate: 2e-05
40
+ - train_batch_size: 8
41
+ - eval_batch_size: 8
42
  - seed: 42
43
  - distributed_type: multi-GPU
44
+ - num_devices: 4
45
+ - gradient_accumulation_steps: 2
46
  - total_train_batch_size: 64
47
+ - total_eval_batch_size: 32
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
  - num_epochs: 5
 
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
55
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
56
+ | 0.6521 | 1.0 | 1451 | 0.6670 | 0.6037 |
57
+ | 0.6101 | 2.0 | 2902 | 0.6763 | 0.6022 |
58
+ | 0.5772 | 3.0 | 4353 | 0.7034 | 0.6026 |
59
+ | 0.5503 | 4.0 | 5804 | 0.7215 | 0.6024 |
60
+ | 0.5347 | 5.0 | 7255 | 0.7376 | 0.6020 |
61
 
62
 
63
  ### Framework versions