You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
RAGnosis is a lightweight framework, built with [LlamaIndex](https://llamaindex.ai), that allows you to evaluate the performance of LLMs and retrieval models in RAG frameworks with your documents. It can be used as an application (thanks to [FastAPI](https://fastapi.tiangolo.com/) + [Gradio](https://gradio.app)) running locally on your machine, or as a python package.
14
+
**diRAGnosis** is a lightweight framework, built with [LlamaIndex](https://llamaindex.ai), that allows you to evaluate the performance of LLMs and retrieval models in RAG frameworks with your documents. It can be used as an application (thanks to [FastAPI](https://fastapi.tiangolo.com/) + [Gradio](https://gradio.app)) running locally on your machine, or as a python package.
15
15
16
16
## Installation and usage
17
17
@@ -20,8 +20,8 @@ RAGnosis is a lightweight framework, built with [LlamaIndex](https://llamaindex.
@@ -51,7 +51,7 @@ You will see the application running on http://localhost:8000/dashboard and you
51
51
52
52
> _Required: [Docker](https://docs.docker.com/desktop/), [docker compose](https://docs.docker.com/compose/) and [conda](https://anaconda.org/anaconda/conda)_
53
53
54
-
- Set up RAGnosis app using the dedicated script:
54
+
- Set up diRAGnosis app using the dedicated script:
55
55
56
56
```bash
57
57
# For MacOs/Linux users
@@ -79,16 +79,16 @@ You will see the application running on http://localhost:8000/dashboard and you
79
79
80
80
### As a python package
81
81
82
-
As a python package, you will be able to install RAGnosis using `pip`:
82
+
As a python package, you will be able to install diRAGnosis using `pip`:
83
83
84
84
```bash
85
-
pip install ragnosis
85
+
pip install diragnosis
86
86
```
87
87
88
-
Once you have installed it, you can import the four functions ([detailed in the dedicated reference file](https://github.com/AstraBert/RAGnosis/tree/main/REFERENCE.md)) available for RAGnosis like this:
88
+
Once you have installed it, you can import the four functions ([detailed in the dedicated reference file](https://github.com/AstraBert/diRAGnosis/tree/main/REFERENCE.md)) available for diRAGnosis like this:
RAGnosis takes care of the evaluation of LLM and retrieval model performance on your documents in a completely automated way:
133
+
diRAGnosis takes care of the evaluation of LLM and retrieval model performance on your documents in a completely automated way:
134
134
135
135
- Once your documents are uploaded, they are converted into a synthetic question dataset (either for Retrieval Augmented Generation or for retrieval only) by an LLM of your choice
136
136
- The documents are also chunked and uploaded to a vector database served by [Qdrant](https://qdrant.tech) - you can choose a semantic search only or an hybrid search setting
@@ -140,8 +140,8 @@ RAGnosis takes care of the evaluation of LLM and retrieval model performance on
140
140
141
141
## Contributing
142
142
143
-
Contributions are always welcome! Follow the contributions guidelines reported [here](https://github.com/AstraBert/RAGnosis/tree/main/CONTRIBUTING.md).
143
+
Contributions are always welcome! Follow the contributions guidelines reported [here](https://github.com/AstraBert/diRAGnosis/tree/main/CONTRIBUTING.md).
144
144
145
145
## License and rights of usage
146
146
147
-
The software is provided under MIT [license](https://github.com/AstraBert/RAGnosis/tree/main/LICENSE).
147
+
The software is provided under MIT [license](https://github.com/AstraBert/diRAGnosis/tree/main/LICENSE).
Copy file name to clipboardexpand all lines: REFERENCE.md
+2-2
Original file line number
Diff line number
Diff line change
@@ -1,6 +1,6 @@
1
-
# Evaluation Framework Reference
1
+
# diRAGnosis reference
2
2
3
-
This document provides a reference for the functions contained in the `evaluation.py` file. These functions are used for generating question datasets, evaluating language learning models (LLMs), and evaluating retrieval performance.
3
+
This document provides a reference for the functions contained in the `evaluation` module of the `diRAGnosis` package. These functions are used for generating question datasets, evaluating language learning models (LLMs), and evaluating retrieval performance.
RAGnosis is a lightweight framework, built with [LlamaIndex](https://llamaindex.ai), that allows you to evaluate the performance of LLMs and retrieval models in RAG frameworks with your documents. It can be used as an application (thanks to [FastAPI](https://fastapi.tiangolo.com/) + [Gradio](https://gradio.app)) running locally on your machine, or as a python package.
14
+
**diRAGnosis** is a lightweight framework, built with [LlamaIndex](https://llamaindex.ai), that allows you to evaluate the performance of LLMs and retrieval models in RAG frameworks with your documents. It can be used as an application (thanks to [FastAPI](https://fastapi.tiangolo.com/) + [Gradio](https://gradio.app)) running locally on your machine, or as a python package.
15
15
16
16
## Installation and usage
17
17
@@ -20,8 +20,8 @@ RAGnosis is a lightweight framework, built with [LlamaIndex](https://llamaindex.
@@ -51,7 +51,7 @@ You will see the application running on http://localhost:8000/dashboard and you
51
51
52
52
> _Required: [Docker](https://docs.docker.com/desktop/), [docker compose](https://docs.docker.com/compose/) and [conda](https://anaconda.org/anaconda/conda)_
53
53
54
-
- Set up RAGnosis app using the dedicated script:
54
+
- Set up diRAGnosis app using the dedicated script:
55
55
56
56
```bash
57
57
# For MacOs/Linux users
@@ -79,16 +79,16 @@ You will see the application running on http://localhost:8000/dashboard and you
79
79
80
80
### As a python package
81
81
82
-
As a python package, you will be able to install RAGnosis using `pip`:
82
+
As a python package, you will be able to install diRAGnosis using `pip`:
83
83
84
84
```bash
85
-
pip install ragnosis
85
+
pip install diRAGnosis
86
86
```
87
87
88
-
Once you have installed it, you can import the four functions ([detailed in the dedicated reference file](https://github.com/AstraBert/RAGnosis/tree/main/REFERENCE.md)) available for RAGnosis like this:
88
+
Once you have installed it, you can import the four functions ([detailed in the dedicated reference file](https://github.com/AstraBert/diRAGnosis/tree/main/REFERENCE.md)) available for diRAGnosis like this:
RAGnosis takes care of the evaluation of LLM and retrieval model performance on your documents in a completely automated way:
133
+
diRAGnosis takes care of the evaluation of LLM and retrieval model performance on your documents in a completely automated way:
134
134
135
135
- Once your documents are uploaded, they are converted into a synthetic question dataset (either for Retrieval Augmented Generation or for retrieval only) by an LLM of your choice
136
136
- The documents are also chunked and uploaded to a vector database served by [Qdrant](https://qdrant.tech) - you can choose a semantic search only or an hybrid search setting
@@ -140,9 +140,8 @@ RAGnosis takes care of the evaluation of LLM and retrieval model performance on
140
140
141
141
## Contributing
142
142
143
-
Contributions are always welcome! Follow the contributions guidelines reported [here](https://github.com/AstraBert/RAGnosis/tree/main/CONTRIBUTING.md).
143
+
Contributions are always welcome! Follow the contributions guidelines reported [here](https://github.com/AstraBert/diRAGnosis/tree/main/CONTRIBUTING.md).
144
144
145
145
## License and rights of usage
146
146
147
-
The software is provided under MIT [license](https://github.com/AstraBert/RAGnosis/tree/main/LICENSE).
148
-
147
+
The software is provided under MIT [license](https://github.com/AstraBert/diRAGnosis/tree/main/LICENSE).
0 commit comments