Spaces:
Running
Running
Update LiveRAG_README.md
Browse files- LiveRAG_README.md +19 -10
LiveRAG_README.md
CHANGED
@@ -1,23 +1,32 @@
|
|
1 |
# π LiveRAG Challenge Operational Instructions
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
---
|
3 |
## π£ LiveRAG Challenge Step-by-Step Guide
|
4 |
|
5 |
1. Accessing DataMorgana and Falcon3-10B-Instruct via AI71-Platform\
|
6 |
-
1.1
|
7 |
-
|
8 |
-
1.2
|
9 |
-
|
10 |
-
1.3 Use your AI71-Platform key for programatic access to
|
11 |
-
1.4 Use your AI71-Platform key for programatic access to Falcon3-10B-Instruct (see [AI71 documentation](https://platform.ai71.ai/documentation))\
|
12 |
**Remark:** Accessing Falcon via the AI71-Platform API is optional and recommended for a quick start. However, for a robust solution, suitable for the live event, it is recommended to set up your own Falcon3-10B-Instruct instance
|
13 |
|
14 |
-
|
15 |
2.1 To use the pre-built indices follow this [sample notebook](Operational_Instructions/Indices_Usage_Examples_for_LiveRAG.ipynb)\
|
16 |
2.2 To build your own **OpenSearch** index, create or use your existing AWS account and apply the provided AWS credits to save costs (see [here](Operational_Instructions/AWS_and_Pinecone_Accounts_for_LiveRAG.md))\
|
17 |
2.3 To build your own **Pinecone** index, create or use your existing Pinecone account and apply the provided Pinecone credits to save costs (see [here](Operational_Instructions/Pinecone_for_LiveRAG.md))
|
18 |
-
|
19 |
-
|
20 |
-
3.1 Use your AWS
|
21 |
3.2 Build your own indices or use the pre-built Pinecone and OpenSearch indices\
|
22 |
3.3 Use the DataMorgana [API](Operational_Instructions/DM_API_usage_example.ipynb) to get training Q&A pairs\
|
23 |
3.4 Follow the [evaluation guidelines](Operational_Instructions/Evaluation_Guidelines_for_LiveRAG.md) to evaluate your RAG system
|
|
|
1 |
# π LiveRAG Challenge Operational Instructions
|
2 |
+
|
3 |
+
|
4 |
+
## π£ LiveRAG Challenge Onboarding
|
5 |
+
|
6 |
+
|
7 |
+
1. Accessing DataMorgana and Falcon3-10B-Instruct via AI71-Platform\
|
8 |
+
1.1 Register to [AI71-Platform](https://platform.ai71.ai/register) and send the email address used for registration to sigir2025-liverag-tech@tii.ae
|
9 |
+
1.2 After you're notified that you are approved, login to your [AI71-Platform account](https://platform.ai71.ai)\
|
10 |
+
|
11 |
+
|
12 |
---
|
13 |
## π£ LiveRAG Challenge Step-by-Step Guide
|
14 |
|
15 |
1. Accessing DataMorgana and Falcon3-10B-Instruct via AI71-Platform\
|
16 |
+
1.1 Login to your [AI71-Platform account](https://platform.ai71.ai)\
|
17 |
+
1.1.1 Access the DataMorgana [Sandbox](https://platform.ai71.ai/playground) and [try it](Operational_Instructions/DM_Sandbox.md) to get your first Q&A\
|
18 |
+
1.1.2 Create your AI71-Platform [API key](https://platform.ai71.ai/api-keys)\
|
19 |
+
1.2 Use your AI71-Platform key for programatic access to DataMorgana bulk requests generation via its [API](Operational_Instructions/DM_API_usage_example.ipynb)\
|
20 |
+
1.3 Use your AI71-Platform key for programatic access to Falcon3-10B-Instruct (see [AI71 documentation](https://platform.ai71.ai/documentation))\
|
|
|
21 |
**Remark:** Accessing Falcon via the AI71-Platform API is optional and recommended for a quick start. However, for a robust solution, suitable for the live event, it is recommended to set up your own Falcon3-10B-Instruct instance
|
22 |
|
23 |
+
2. Using [Pinecone](https://www.pinecone.io/) and [OpenSearch](https://opensearch.org/) pre-built indices of [FineWeb Sample-10BT](https://huggingface.co/datasets/HuggingFaceFW/fineweb/viewer/sample-10BT) and building your own indices\
|
24 |
2.1 To use the pre-built indices follow this [sample notebook](Operational_Instructions/Indices_Usage_Examples_for_LiveRAG.ipynb)\
|
25 |
2.2 To build your own **OpenSearch** index, create or use your existing AWS account and apply the provided AWS credits to save costs (see [here](Operational_Instructions/AWS_and_Pinecone_Accounts_for_LiveRAG.md))\
|
26 |
2.3 To build your own **Pinecone** index, create or use your existing Pinecone account and apply the provided Pinecone credits to save costs (see [here](Operational_Instructions/Pinecone_for_LiveRAG.md))
|
27 |
+
|
28 |
+
3. Building, training, and evaluating your RAG system\
|
29 |
+
3.1 Use your AWS team account to build your RAG system using Falcon3-10B-Instruct\
|
30 |
3.2 Build your own indices or use the pre-built Pinecone and OpenSearch indices\
|
31 |
3.3 Use the DataMorgana [API](Operational_Instructions/DM_API_usage_example.ipynb) to get training Q&A pairs\
|
32 |
3.4 Follow the [evaluation guidelines](Operational_Instructions/Evaluation_Guidelines_for_LiveRAG.md) to evaluate your RAG system
|