ptran74 commited on
Commit
179083e
1 Parent(s): 931a1da

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -6
README.md CHANGED
@@ -19,6 +19,8 @@ It achieves the following results on the evaluation set:
19
  - Exact: 63.9013
20
  - F1: 72.1497
21
 
 
 
22
  ### Before fine-tuning:
23
 
24
  ```
@@ -48,19 +50,20 @@ DatasetDict({
48
  })
49
  ```
50
 
51
- ## Model description
52
-
53
- More information needed
54
-
55
  ## Intended uses & limitations
56
 
57
- More information needed
58
 
59
  ## Training and evaluation data
60
 
61
- More information needed
 
 
62
 
63
  ## Training procedure
 
 
 
64
 
65
  ### Training hyperparameters
66
 
 
19
  - Exact: 63.9013
20
  - F1: 72.1497
21
 
22
+ ## More accurate metrics:
23
+
24
  ### Before fine-tuning:
25
 
26
  ```
 
50
  })
51
  ```
52
 
 
 
 
 
53
  ## Intended uses & limitations
54
 
55
+ This model is fine-tuned to answer questions from the DSPFirst textbook. I'm not really sure what I am doing so you should review it before using it.
56
 
57
  ## Training and evaluation data
58
 
59
+ - `batch_size` of 6 results in 14.82 GB VRAM
60
+ - 4.52 GB RAM
61
+ - 30% of the total questions is dedicated for evaluating.
62
 
63
  ## Training procedure
64
+ - The model was trained from Google Colab
65
+ - Utilizes Tesla P100 16GB, took 3.8 hours to train
66
+ - `load_best_model_at_end` is enabled in TrainingArguments
67
 
68
  ### Training hyperparameters
69