Update README.md
Browse files
README.md
CHANGED
|
@@ -44,7 +44,6 @@ Training is only done for a relatively small dataset and few epochs, and thus, t
|
|
| 44 |
Even with the correct output, the syntax of the model can be occasionally dubious.<br>
|
| 45 |
Model is not perfect and identifier renamings must be reviewed till performance in test settings is not evaluated.
|
| 46 |
|
| 47 |
-
[More Information Needed]
|
| 48 |
|
| 49 |
### Recommendations
|
| 50 |
|
|
@@ -60,7 +59,6 @@ Clone the repository and load model state dict using 'model_26_2'
|
|
| 60 |
|
| 61 |
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
| 62 |
Trained on a subset of a dataset of 1000 classes with 612 lines of code on average for 3 epochs and a Learning Rate of 2e-5.
|
| 63 |
-
[More Information Needed]
|
| 64 |
|
| 65 |
|
| 66 |
|
|
@@ -72,13 +70,11 @@ Perplexty of Base Model: 37580
|
|
| 72 |
Perplexity of Fine-tuned Model: 23
|
| 73 |
|
| 74 |
|
| 75 |
-
[More Information Needed]
|
| 76 |
|
| 77 |
#### Factors
|
| 78 |
|
| 79 |
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 80 |
|
| 81 |
-
[More Information Needed]
|
| 82 |
|
| 83 |
#### Metrics
|
| 84 |
|
|
@@ -88,7 +84,6 @@ Perplexity is used to evaluate the performance of the model. It judges how surpr
|
|
| 88 |
|
| 89 |
<!-- Relevant interpretability work for the model goes here -->
|
| 90 |
|
| 91 |
-
[More Information Needed]
|
| 92 |
|
| 93 |
|
| 94 |
|
|
|
|
| 44 |
Even with the correct output, the syntax of the model can be occasionally dubious.<br>
|
| 45 |
Model is not perfect and identifier renamings must be reviewed till performance in test settings is not evaluated.
|
| 46 |
|
|
|
|
| 47 |
|
| 48 |
### Recommendations
|
| 49 |
|
|
|
|
| 59 |
|
| 60 |
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
| 61 |
Trained on a subset of a dataset of 1000 classes with 612 lines of code on average for 3 epochs and a Learning Rate of 2e-5.
|
|
|
|
| 62 |
|
| 63 |
|
| 64 |
|
|
|
|
| 70 |
Perplexity of Fine-tuned Model: 23
|
| 71 |
|
| 72 |
|
|
|
|
| 73 |
|
| 74 |
#### Factors
|
| 75 |
|
| 76 |
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 77 |
|
|
|
|
| 78 |
|
| 79 |
#### Metrics
|
| 80 |
|
|
|
|
| 84 |
|
| 85 |
<!-- Relevant interpretability work for the model goes here -->
|
| 86 |
|
|
|
|
| 87 |
|
| 88 |
|
| 89 |
|