Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/mrseanryan/gpt-summarizer
Summarize text using ChatGPT or a local LLM, with support for multiple large text files, PDF files and translation.
https://github.com/mrseanryan/gpt-summarizer
chatgpt nlp pdf pdf-summarizer summarization
Last synced: 7 days ago
JSON representation
Summarize text using ChatGPT or a local LLM, with support for multiple large text files, PDF files and translation.
- Host: GitHub
- URL: https://github.com/mrseanryan/gpt-summarizer
- Owner: mrseanryan
- License: mit
- Created: 2023-07-17T13:34:31.000Z (over 1 year ago)
- Default Branch: master
- Last Pushed: 2024-07-16T11:41:55.000Z (4 months ago)
- Last Synced: 2024-07-16T14:28:13.805Z (4 months ago)
- Topics: chatgpt, nlp, pdf, pdf-summarizer, summarization
- Language: Python
- Homepage: https://medium.com/p/15f843b1ea15
- Size: 764 KB
- Stars: 6
- Watchers: 3
- Forks: 0
- Open Issues: 3
-
Metadata Files:
- Readme: README.gpu.md
- License: LICENSE
Awesome Lists containing this project
README
### Using GPU with the local model
If you have an NVIDIA graphics card, then you can run part or all of the model (depending on the card's RAM) on the GPU,
which has much higher level of parallelism than the typical CPU.
Required:
- latest NVIDIA graphic driver
- up to date version of CUDA
#### Tip: if you get errors when running the model, like this:
```
>> Cuda error: no kernel image is available for execution on the device
```
Then recommend to build ctransformers locally.
This is actually quite simple:
```
pip3 uninstall ctransformers
pip3 install ctransformers --no-binary ctransformers # use --no-binary to force a local build. This ensures that the local version of CUDA and NVIDIA graphics driver will be used.
```
You can tweak the settings in `config.py`.
# More details
## local via ctransformers
- [ctransformers](https://github.com/marella/ctransformers).
## local via pytorch
- [pytorch](https://pytorch.org/get-started/locally/)
- [Building pytorch from scratch](https://github.com/pytorch/pytorch#from-source)