|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
- gpt2 |
|
- generation |
|
model-index: |
|
- name: resumes_model |
|
results: [] |
|
datasets: |
|
- mpuig/job-experience |
|
widget: |
|
- text: As a Software Developer, I |
|
example_title: Software Developer |
|
- text: As a Software Architect, I |
|
example_title: Software Architect |
|
- text: As a web developer, I |
|
example_title: Web Developer |
|
|
|
language: |
|
- en |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# Model Card for mpuig/job-experience |
|
|
|
This model is a fine-tuned version of [GPT-2](https://huggingface.co/gpt2) to generate fake job experience descriptions. |
|
|
|
While this may not have practical applications in the real world, it served as a valuable learning experience for understanding the process of fine-tuning a language learning model. Through this repository, I hope to share my insights and findings on the capabilities and limitations of GPT-2 in generating job experiences. |
|
|
|
The goal was to obtain a model where, starting with a sentence like "As a Software Engineer, I ", the model generates a complete new sentence related to the job title ("Software Engineer") like: |
|
|
|
"_As a software architect, I coordinated with the Marketing department to identify problems encountered and provide solutions to resolve them._" |
|
|
|
- **Resources for more information:** More information needed |
|
- [GitHub Repo](https://github.com/mpuig/gpt2-fine-tuning/) |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 5e-05 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3.0 |
|
|
|
### Training results |
|
|
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.25.1 |
|
- Pytorch 1.13.0+cu116 |
|
- Datasets 2.8.0 |
|
- Tokenizers 0.13.2 |
