cag / README.md
gigabrain's picture
Update README.md
d10d92e
|
Raw History Blame Contribute Delete
3.06 kB
---
language: en
thumbnail: http://www.huggingtweets.com/doveywan-irenezhao_-layahheilpern/1668969714119/predictions.png
tags:
- huggingtweets
widget:
- text: "My dream is"
---
<div class="inline-flex flex-col" style="line-height: 1.5;">
<div class="flex">
<div
style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url(&#39;https://pbs.twimg.com/profile_images/1592314373317558274/kWBIBveR_400x400.jpg&#39;)">
</div>
<div
style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url(&#39;https://pbs.twimg.com/profile_images/1569305276343369729/9tyrIoYq_400x400.jpg&#39;)">
</div>
<div
style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url(&#39;https://pbs.twimg.com/profile_images/1423875044598456321/SVjwd6Bb_400x400.jpg&#39;)">
</div>
</div>
<div style="text-align: center; font-size: 16px; font-weight: 800">Layah Heilpern & Dovey "Rug The Fiat" Wan & Irene Zhao</div>
<div style="text-align: center; font-size: 14px;">@doveywan-irenezhao_-layahheilpern</div>
</div>
## How does it work?
The model uses the following pipeline.
![pipeline](https://github.com/borisdayma/huggingtweets/blob/master/img/pipeline.png?raw=true)
## Training data
The model was trained on tweets from Layah Heilpern & Dovey "Rug The Fiat" Wan & Irene Zhao.
| Data | Layah Heilpern | Dovey "Rug The Fiat" Wan | Irene Zhao |
| --- | --- | --- | --- |
| Tweets downloaded | 3249 | 3247 | 1945 |
| Retweets | 115 | 310 | 223 |
| Short tweets | 1453 | 269 | 417 |
| Tweets kept | 1681 | 2668 | 1305 |
[Explore the data](https://wandb.ai/wandb/huggingtweets/runs/38f27zgg/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
## Training procedure
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @doveywan-irenezhao_-layahheilpern's tweets.
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/zek1fxw0) for full transparency and reproducibility.
At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/zek1fxw0/artifacts) is logged and versioned.
## How to use
You can use this model directly with a pipeline for text generation:
```python
from transformers import pipeline
generator = pipeline('text-generation',
model='huggingtweets/cag')
generator("In crypto, ", num_return_sequences=5)
```
## Limitations and bias
The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias).
In addition, the data present in the user's tweets further affects the text generated by the model.
## About
*Built by Gigabrain*