Update README.md
Browse files
README.md
CHANGED
@@ -5,28 +5,31 @@ tags:
|
|
5 |
model-index:
|
6 |
- name: distilgpt2-2k_clean_medical_articles_causal_language_model
|
7 |
results: []
|
|
|
|
|
|
|
|
|
8 |
---
|
9 |
|
10 |
-
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
11 |
-
should probably proofread and complete it, then remove this comment. -->
|
12 |
-
|
13 |
# distilgpt2-2k_clean_medical_articles_causal_language_model
|
14 |
|
15 |
-
This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2)
|
16 |
It achieves the following results on the evaluation set:
|
17 |
- Loss: 2.9268
|
18 |
|
19 |
## Model description
|
20 |
|
21 |
-
|
|
|
|
|
22 |
|
23 |
## Intended uses & limitations
|
24 |
|
25 |
-
|
26 |
|
27 |
## Training and evaluation data
|
28 |
|
29 |
-
|
30 |
|
31 |
## Training procedure
|
32 |
|
@@ -49,10 +52,11 @@ The following hyperparameters were used during training:
|
|
49 |
| 2.998 | 2.0 | 3982 | 2.9367 |
|
50 |
| 2.9484 | 3.0 | 5973 | 2.9268 |
|
51 |
|
|
|
52 |
|
53 |
### Framework versions
|
54 |
|
55 |
- Transformers 4.26.1
|
56 |
- Pytorch 1.12.1
|
57 |
- Datasets 2.9.0
|
58 |
-
- Tokenizers 0.12.1
|
|
|
5 |
model-index:
|
6 |
- name: distilgpt2-2k_clean_medical_articles_causal_language_model
|
7 |
results: []
|
8 |
+
language:
|
9 |
+
- en
|
10 |
+
metrics:
|
11 |
+
- perplexity
|
12 |
---
|
13 |
|
|
|
|
|
|
|
14 |
# distilgpt2-2k_clean_medical_articles_causal_language_model
|
15 |
|
16 |
+
This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2).
|
17 |
It achieves the following results on the evaluation set:
|
18 |
- Loss: 2.9268
|
19 |
|
20 |
## Model description
|
21 |
|
22 |
+
This is a causal language modeling project.
|
23 |
+
|
24 |
+
For more information on how it was created, check out the following link: https://github.com/DunnBC22/NLP_Projects/blob/main/Causal%20Language%20Modeling/2000%20Clean%20Medical%20Articles/2%2C000%20Clean%20Medical%20Articles%20-%20CLM.ipynb
|
25 |
|
26 |
## Intended uses & limitations
|
27 |
|
28 |
+
This model is intended to demonstrate my ability to solve a complex problem using technology.
|
29 |
|
30 |
## Training and evaluation data
|
31 |
|
32 |
+
Dataset Source: https://www.kaggle.com/datasets/trikialaaa/2k-clean-medical-articles-medicalnewstoday
|
33 |
|
34 |
## Training procedure
|
35 |
|
|
|
52 |
| 2.998 | 2.0 | 3982 | 2.9367 |
|
53 |
| 2.9484 | 3.0 | 5973 | 2.9268 |
|
54 |
|
55 |
+
Perplexity: 18.67
|
56 |
|
57 |
### Framework versions
|
58 |
|
59 |
- Transformers 4.26.1
|
60 |
- Pytorch 1.12.1
|
61 |
- Datasets 2.9.0
|
62 |
+
- Tokenizers 0.12.1
|