feat(readme): updated readme
Browse files
README.md
CHANGED
|
@@ -1,33 +1,51 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
|
|
|
|
|
|
| 3 |
base_model: Dmitriy007/rugpt2_gen_news
|
| 4 |
tags:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
-
datasets:
|
| 7 |
-
- baneks
|
| 8 |
model-index:
|
| 9 |
-
- name: clown
|
| 10 |
results: []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
-
|
| 14 |
-
|
|
|
|
| 15 |
|
| 16 |
# clown
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Dmitriy007/rugpt2_gen_news]
|
|
|
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
| 22 |
-
|
| 23 |
|
| 24 |
## Intended uses & limitations
|
| 25 |
|
| 26 |
-
|
| 27 |
|
| 28 |
## Training and evaluation data
|
| 29 |
|
| 30 |
-
|
| 31 |
|
| 32 |
## Training procedure
|
| 33 |
|
|
@@ -47,7 +65,9 @@ The following hyperparameters were used during training:
|
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
-
|
|
|
|
|
|
|
| 51 |
|
| 52 |
### Framework versions
|
| 53 |
|
|
@@ -55,3 +75,6 @@ The following hyperparameters were used during training:
|
|
| 55 |
- Pytorch 2.1.0
|
| 56 |
- Datasets 2.12.0
|
| 57 |
- Tokenizers 0.14.1
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
language:
|
| 3 |
+
- ru
|
| 4 |
+
- en
|
| 5 |
+
license: apache-2.0
|
| 6 |
base_model: Dmitriy007/rugpt2_gen_news
|
| 7 |
tags:
|
| 8 |
+
- not-for-all-audiences
|
| 9 |
+
- art
|
| 10 |
+
- humour
|
| 11 |
+
- jokes
|
| 12 |
- generated_from_trainer
|
|
|
|
|
|
|
| 13 |
model-index:
|
| 14 |
+
- name: zeio/clown
|
| 15 |
results: []
|
| 16 |
+
datasets:
|
| 17 |
+
- zeio/baneks
|
| 18 |
+
metrics:
|
| 19 |
+
- loss
|
| 20 |
+
widget:
|
| 21 |
+
- text: 'Купил мужик шляпу'
|
| 22 |
+
example_title: hat
|
| 23 |
+
- text: 'Пришла бабка к врачу'
|
| 24 |
+
example_title: doctor
|
| 25 |
+
- text: 'Нашел мужик подкову'
|
| 26 |
+
example_title: horseshoe
|
| 27 |
---
|
| 28 |
|
| 29 |
+
<p align="center">
|
| 30 |
+
<img src="https://i.ibb.co/QbQZ8Gz/clown-logo.png"/>
|
| 31 |
+
</p>
|
| 32 |
|
| 33 |
# clown
|
| 34 |
|
| 35 |
+
This model is a fine-tuned version of [Dmitriy007/rugpt2_gen_news][base] on the [baneks][dataset] dataset for 1 epoch. It achieved `2.0760` loss during training.
|
| 36 |
+
Model evaluation has not been performed.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
| 40 |
+
The model is a fine-tuned variant of the [Dmitriy007/rugpt2_gen_news][base] architecture with causal language modeling head.
|
| 41 |
|
| 42 |
## Intended uses & limitations
|
| 43 |
|
| 44 |
+
The model should be used for studying abilities of natural language models to generate jokes.
|
| 45 |
|
| 46 |
## Training and evaluation data
|
| 47 |
|
| 48 |
+
The model is trained on a list of anecdotes pulled from a few vk communities (see [baneks][baneks] dataset for more details).
|
| 49 |
|
| 50 |
## Training procedure
|
| 51 |
|
|
|
|
| 65 |
|
| 66 |
### Training results
|
| 67 |
|
| 68 |
+
| Train Loss | Epoch |
|
| 69 |
+
|:----------:|:-----:|
|
| 70 |
+
| 2.0760 | 10 |
|
| 71 |
|
| 72 |
### Framework versions
|
| 73 |
|
|
|
|
| 75 |
- Pytorch 2.1.0
|
| 76 |
- Datasets 2.12.0
|
| 77 |
- Tokenizers 0.14.1
|
| 78 |
+
|
| 79 |
+
[base]: https://huggingface.co/Dmitriy007/rugpt2_gen_news
|
| 80 |
+
[dataset]: https://huggingface.co/datasets/zeio/baneks
|