Skip to content

Wfh/descriptive names #79

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 8 commits into from
Nov 22, 2023
Merged

Wfh/descriptive names #79

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
ff421c0
Update name
hinthornw Nov 22, 2023
43405e2
Updat comp techniques
hinthornw Nov 22, 2023
0f13d3a
Update noteboo
hinthornw Nov 22, 2023
b7e2824
update
hinthornw Nov 22, 2023
8388f07
Merge branch 'main' into wfh/descriptive_names
hinthornw Nov 22, 2023
34884d5
merge
hinthornw Nov 22, 2023
c32190c
update notebook
hinthornw Nov 22, 2023
1275074
lint
hinthornw Nov 22, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
88 changes: 74 additions & 14 deletions docs/source/notebooks/retrieval/comparing_techniques.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,6 @@
"# Evaluating RAG Architectures on Benchmark Tasks\n",
"\n",
"\n",
"\n",
"#### Introduction\n",
"\n",
"If you ever wanted to compare different approaches to Q&A over docs, you'll find this notebook helpful to get started evaluating different configurations and common RAG architectures on benchmark tasks. The goal is to make it easy for you to experiment with different techniques, understand their tradeoffs, and make informed decisions for your specific use case.\n",
Expand Down Expand Up @@ -73,6 +72,18 @@
"os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\""
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import uuid\n",
"\n",
"# Generate a unique run ID for these experiments\n",
"run_uid = uuid.uuid4().hex[:6]"
]
},
{
"cell_type": "markdown",
"id": "2e8a666d-8bf5-4bfd-8b20-8b7defdb8cd5",
Expand Down Expand Up @@ -267,6 +278,12 @@
" dataset_name=langchain_docs.name,\n",
" llm_or_chain_factory=partial(chain_factory, retriever, llm=llm),\n",
" evaluation=RAG_EVALUATION,\n",
" project_name=f\"claude-2 qa-chain simple-index {run_uid}\",\n",
" project_metadata={\n",
" \"index_method\": \"basic\",\n",
" \"embedding_model\": \"thenlper/gte-base\",\n",
" \"llm\": \"claude-2\",\n",
" },\n",
" verbose=True,\n",
")"
]
Expand Down Expand Up @@ -338,8 +355,16 @@
"source": [
"chunked_results = client.run_on_dataset(\n",
" dataset_name=langchain_docs.name,\n",
" llm_or_chain_factory=partial(chain_factory, retriever, llm=llm),\n",
" llm_or_chain_factory=partial(chain_factory, chunked_retriever, llm=llm),\n",
" evaluation=RAG_EVALUATION,\n",
" project_name=f\"claude-2 qa-chain chunked {run_uid}\",\n",
" project_metadata={\n",
" \"index_method\": \"basic\",\n",
" \"chunk_size\": 4000,\n",
" \"chunk_overlap\": 200,\n",
" \"embedding_model\": \"thenlper/gte-base\",\n",
" \"llm\": \"claude-2\",\n",
" },\n",
" verbose=True,\n",
")"
]
Expand Down Expand Up @@ -399,6 +424,12 @@
" dataset_name=langchain_docs.name,\n",
" llm_or_chain_factory=partial(chain_factory, parent_doc_retriever, llm=llm),\n",
" evaluation=RAG_EVALUATION,\n",
" project_name=f\"claude-2 qa-chain parent-doc {run_uid}\",\n",
" project_metadata={\n",
" \"index_method\": \"parent-doc\",\n",
" \"embedding_model\": \"thenlper/gte-base\",\n",
" \"llm\": \"claude-2\",\n",
" },\n",
" verbose=True,\n",
")"
]
Expand Down Expand Up @@ -447,9 +478,15 @@
"source": [
"hyde_test_run = client.run_on_dataset(\n",
" dataset_name=langchain_docs.name,\n",
" llm_or_chain_factory=partial(chain_factory, retriever),\n",
" llm_or_chain_factory=partial(chain_factory, retriever=retriever, llm=llm),\n",
" evaluation=RAG_EVALUATION,\n",
" verbose=True,\n",
" project_name=f\"claude-2 qa-chain HyDE {run_uid}\",\n",
" project_metadata={\n",
" \"index_method\": \"HyDE\",\n",
" \"embedding_model\": \"thenlper/gte-base\",\n",
" \"llm\": \"claude-2\",\n",
" },\n",
")"
]
},
Expand Down Expand Up @@ -515,6 +552,12 @@
" dataset_name=langchain_docs.name,\n",
" llm_or_chain_factory=partial(chain_factory, openai_retriever),\n",
" evaluation=RAG_EVALUATION,\n",
" project_name=f\"claude-2 qa-chain oai-emb basic {run_uid}\",\n",
" project_metadata={\n",
" \"index_method\": \"basic\",\n",
" \"embedding_model\": \"openai/text-embedding-ada-002\",\n",
" \"llm\": \"claude-2\",\n",
" },\n",
" verbose=True,\n",
")"
]
Expand Down Expand Up @@ -559,16 +602,6 @@
"ollama = ChatOllama(model=\"yarn-llama2:7b-128k\")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "256b75fe-9f0c-4820-8e6c-9c87df39f0a7",
"metadata": {},
"outputs": [],
"source": [
"ollama.invoke(\"Hi there\")"
]
},
{
"cell_type": "code",
"execution_count": null,
Expand All @@ -593,6 +626,12 @@
" dataset_name=langchain_docs.name,\n",
" llm_or_chain_factory=partial(chain_factory, llm=ollama, retriever=retriever),\n",
" evaluation=RAG_EVALUATION,\n",
" project_name=f\"yarn-llama2:7b-128k qa-chain basic {run_uid}\",\n",
" project_metadata={\n",
" \"index_method\": \"basic\",\n",
" \"embedding_model\": \"thenlper/gte-base\",\n",
" \"llm\": \"ollama/yarn-llama2:7b-128k\",\n",
" },\n",
" verbose=True,\n",
")"
]
Expand Down Expand Up @@ -649,9 +688,16 @@
"claude_simple_prompt_test_run = client.run_on_dataset(\n",
" dataset_name=langchain_docs.name,\n",
" llm_or_chain_factory=partial(\n",
" chain_factory, response_generator=generator, retriever=retriever\n",
" chain_factory, response_generator=generator, retriever=retriever, llm=llm\n",
" ),\n",
" evaluation=RAG_EVALUATION,\n",
" project_name=f\"claude-2 qa-chain basic rag-simple {run_uid}\",\n",
" project_metadata={\n",
" \"index_method\": \"basic\",\n",
" \"embedding_model\": \"thenlper/gte-base\",\n",
" \"prompt\": \"wfh/rag-simple\",\n",
" \"llm\": \"claude-2\",\n",
" },\n",
" verbose=True,\n",
")"
]
Expand Down Expand Up @@ -771,6 +817,13 @@
" dataset_name=langchain_docs.name,\n",
" llm_or_chain_factory=agent_executor,\n",
" evaluation=RAG_EVALUATION,\n",
" project_name=f\"oai-functions basic rag-simple {run_uid}\",\n",
" project_metadata={\n",
" \"index_method\": \"basic\",\n",
" \"embedding_model\": \"thenlper/gte-base\",\n",
" \"llm\": \"gpt-4-1106-preview\",\n",
" \"architecture\": \"oai-functions-agent\",\n",
" },\n",
" verbose=True,\n",
")"
]
Expand Down Expand Up @@ -839,6 +892,13 @@
" dataset_name=langchain_docs.name,\n",
" llm_or_chain_factory=assistant_exector,\n",
" evaluation=RAG_EVALUATION,\n",
" project_name=f\"oai-assistant basic rag-simple {run_uid}\",\n",
" project_metadata={\n",
" \"index_method\": \"basic\",\n",
" \"embedding_model\": \"thenlper/gte-base\",\n",
" \"llm\": \"gpt-4-1106-preview\",\n",
" \"architecture\": \"oai-assistant\",\n",
" },\n",
" verbose=True,\n",
")"
]
Expand Down
Loading