gmakstutis/cognee
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
cognee/notebooks/full_run.ipynb
Vasilije bb679c2dd7
Improve processing, update networkx client, and Neo4j, and dspy (#69) 
* Update cognify and the networkx client to prepare for running in Neo4j

* Fix for openai model

* Add the fix to the infra so that the models can be passed to the library. Enable llm_provider to be passed.

* Auto graph generation now works with neo4j

* Added fixes for both neo4j and networkx

* Explicitly name semantic node connections

* Added updated docs, readme, chunkers and updates to cognify

* Make docs build trigger only when changes on it happen

* Update docs, test git actions

* Separate cognify logic into tasks

* Introduce dspy knowledge graph extraction

---------
Co-authored-by: Boris Arzentar <borisarzentar@gmail.com>
2024-04-20 19:05:40 +02:00

864 lines
56 KiB
Text
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

{
"cells": [
{
"cell_type": "code",
"execution_count": null,
"id": "38135bf7",
"metadata": {},
"outputs": [],
"source": [
"from os import path\n",
"import cognee\n",
"import dspy\n",
"from dspy.datasets import HotPotQA\n",
"\n",
"data_directory_path = path.abspath(\"../.data\")\n",
"cognee.config.data_root_directory(data_directory_path)\n",
"\n",
"cognee_directory_path = path.abspath(\"../.cognee_system\")\n",
"cognee.config.system_root_directory(cognee_directory_path)\n",
"\n",
"await cognee.prune.prune_system()\n",
"\n",
"colbertv2_wiki17_abstracts = dspy.ColBERTv2(url = \"http://20.102.90.50:2017/wiki17_abstracts\")\n",
"dspy.configure(rm = colbertv2_wiki17_abstracts)\n",
"\n",
"dataset = HotPotQA(\n",
" train_seed = 1,\n",
" train_size = 10,\n",
" eval_seed = 2023,\n",
" dev_size = 0,\n",
" test_size = 0,\n",
" keep_details = True,\n",
")\n",
"\n",
"texts_to_add = []\n",
"\n",
"for train_case in dataset.train:\n",
" train_case_text = \"\\r\\n\".join(\" \".join(context_sentences) for context_sentences in train_case.get(\"context\")[\"sentences\"])\n",
"\n",
" texts_to_add.append(train_case_text)\n",
"\n",
"dataset_name = \"train_dataset\"\n",
"await cognee.add(texts_to_add, dataset_name)\n"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "44603a2a",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
" from .autonotebook import tqdm as notebook_tqdm\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"['train_dataset']\n",
"10\n"
]
}
],
"source": [
"from os import path\n",
"import cognee\n",
"\n",
"data_directory_path = path.abspath(\"../.data\")\n",
"cognee.config.data_root_directory(data_directory_path)\n",
"\n",
"cognee_directory_path = path.abspath(\"../.cognee_system\")\n",
"cognee.config.system_root_directory(cognee_directory_path)\n",
"\n",
"print(cognee.datasets.list_datasets())\n",
"\n",
"train_dataset = cognee.datasets.query_data('train_dataset')\n",
"print(len(train_dataset))"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "65bfaf09",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:NetworkXAdapter:File /Users/borisarzentar/Projects/Topoteretes/cognee/.cognee_system/databases/cognee_graph.pkl not found. Initializing an empty graph.<frozen abc>:123: ResourceWarning: unclosed <socket.socket fd=76, family=2, type=1, proto=6, laddr=('192.168.9.243', 63064), raddr=('35.245.15.233', 6333)>\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=76 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 67806.30it/s]\n",
"WARNING:NetworkXAdapter:File /Users/borisarzentar/Projects/Topoteretes/cognee/.cognee_system/databases/cognee_graph.pkl not found. Initializing an empty graph.WARNING:NetworkXAdapter:File /Users/borisarzentar/Projects/Topoteretes/cognee/.cognee_system/databases/cognee_graph.pkl not found. Initializing an empty graph."
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Processing document (faf8441f25bf58d5a55a2821e5e40a9e).\n",
"Processing document (0a4c678b77a75927992f72b81fda2398).\n",
"Processing document (ae7c88b9432a5f51a512885da6f2d1c6).\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/networkx/classes/reportviews.py:921: ResourceWarning: unclosed <ssl.SSLSocket fd=79, family=2, type=1, proto=0, laddr=('192.168.9.243', 63068), raddr=('35.245.15.233', 6333)>\n",
" self._report = lambda n, nbr, k, dd: (n, nbr, k, dd)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/threading.py:910: ResourceWarning: unclosed <socket.socket fd=79, family=2, type=1, proto=6, laddr=('192.168.9.243', 63070), raddr=('35.245.15.233', 6333)>\n",
" self._invoke_excepthook = _make_invoke_excepthook()\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=79 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 78085.45it/s]\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/pydantic/_internal/_core_utils.py:197: ResourceWarning: unclosed <ssl.SSLSocket fd=79, family=2, type=1, proto=0, laddr=('192.168.9.243', 63072), raddr=('35.245.15.233', 6333)>\n",
" return f(schema, self._walk)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=75 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 118866.91it/s]\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/pydantic/_internal/_core_utils.py:200: ResourceWarning: unclosed <ssl.SSLSocket fd=75, family=2, type=1, proto=0, laddr=('192.168.9.243', 63073), raddr=('35.245.15.233', 6333)>\n",
" schema = self._schema_type_to_method[schema['type']](schema.copy(), f)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 75282.38it/s]\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/pydantic/_internal/_core_utils.py:200: ResourceWarning: unclosed <ssl.SSLSocket fd=75, family=2, type=1, proto=0, laddr=('192.168.9.243', 63074), raddr=('35.245.15.233', 6333)>\n",
" schema = self._schema_type_to_method[schema['type']](schema.copy(), f)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Document (DOCUMENT__faf8441f25bf58d5a55a2821e5e40a9e) classified.\n",
"Document (DOCUMENT__0a4c678b77a75927992f72b81fda2398) classified.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=75 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Document (DOCUMENT__ae7c88b9432a5f51a512885da6f2d1c6) classified.\n",
"Document (DOCUMENT__faf8441f25bf58d5a55a2821e5e40a9e) summarized.\n",
"Document (DOCUMENT__ae7c88b9432a5f51a512885da6f2d1c6) summarized.\n",
"Document (DOCUMENT__0a4c678b77a75927992f72b81fda2398) summarized.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"[nltk_data] Downloading package stopwords to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package stopwords is already up-to-date!\n",
"[nltk_data] Downloading package stopwords to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package stopwords is already up-to-date!\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=85 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=88 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=83 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=81 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=79 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=84 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=82 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/ipykernel/iostream.py:720: ResourceWarning: unclosed <socket.socket fd=75, family=2, type=1, proto=6, laddr=('192.168.9.243', 63090), raddr=('35.245.15.233', 6333)>\n",
" with self._buffer_lock:\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 61941.20it/s]\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/networkx/classes/reportviews.py:947: ResourceWarning: unclosed <ssl.SSLSocket fd=75, family=2, type=1, proto=0, laddr=('192.168.9.243', 63092), raddr=('35.245.15.233', 6333)>\n",
" return (\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"[nltk_data] Downloading package stopwords to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package stopwords is already up-to-date!\n",
"[nltk_data] Downloading package stopwords to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package stopwords is already up-to-date!\n",
"[nltk_data] Downloading package stopwords to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package stopwords is already up-to-date!\n",
"[nltk_data] Downloading package stopwords to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package stopwords is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 111212.61it/s]\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 120823.57it/s]\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/pydantic/main.py:175: ResourceWarning: unclosed <socket.socket fd=81, family=2, type=1, proto=6, laddr=('192.168.9.243', 63094), raddr=('35.245.15.233', 6333)>\n",
" self.__pydantic_validator__.validate_python(data, self_instance=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/pydantic/main.py:175: ResourceWarning: unclosed <ssl.SSLSocket fd=79, family=2, type=1, proto=0, laddr=('192.168.9.243', 63096), raddr=('35.245.15.233', 6333)>\n",
" self.__pydantic_validator__.validate_python(data, self_instance=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/networkx/classes/reportviews.py:947: ResourceWarning: unclosed <ssl.SSLSocket fd=79, family=2, type=1, proto=0, laddr=('192.168.9.243', 63097), raddr=('35.245.15.233', 6333)>\n",
" return (\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 57456.22it/s]\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"[nltk_data] Downloading package vader_lexicon to\n",
"[nltk_data] /Users/borisarzentar/nltk_data...\n",
"[nltk_data] Package vader_lexicon is already up-to-date!\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/networkx/classes/reportviews.py:947: ResourceWarning: unclosed <ssl.SSLSocket fd=81, family=2, type=1, proto=0, laddr=('192.168.9.243', 63099), raddr=('35.245.15.233', 6333)>\n",
" return (\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 64245.36it/s]\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 116972.62it/s]\n"
]
},
{
"data": {
"text/plain": [
"<networkx.classes.multidigraph.MultiDiGraph at 0x296f97350>"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from os import path\n",
"import cognee\n",
"\n",
"await cognee.prune.prune_system()\n",
"\n",
"data_directory_path = path.abspath(\"../.data\")\n",
"cognee.config.data_root_directory(data_directory_path)\n",
"\n",
"cognee_directory_path = path.abspath(\"../.cognee_system\")\n",
"cognee.config.system_root_directory(cognee_directory_path)\n",
"\n",
"await cognee.cognify('train_dataset')"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "a514cf38",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/json/decoder.py:353: ResourceWarning: unclosed <socket.socket fd=79, family=2, type=1, proto=6, laddr=('192.168.9.243', 63138), raddr=('35.245.15.233', 6333)>\n",
" obj, end = self.scan_once(s, idx)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/json/decoder.py:353: ResourceWarning: unclosed <socket.socket fd=81, family=2, type=1, proto=6, laddr=('192.168.9.243', 63139), raddr=('35.245.15.233', 6333)>\n",
" obj, end = self.scan_once(s, idx)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/json/decoder.py:353: ResourceWarning: unclosed <socket.socket fd=88, family=2, type=1, proto=6, laddr=('192.168.9.243', 63140), raddr=('35.245.15.233', 6333)>\n",
" obj, end = self.scan_once(s, idx)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=88 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/json/decoder.py:353: ResourceWarning: unclosed <socket.socket fd=89, family=2, type=1, proto=6, laddr=('192.168.9.243', 63141), raddr=('35.245.15.233', 6333)>\n",
" obj, end = self.scan_once(s, idx)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=89 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Graph is visualized at: https://hub.graphistry.com/graph/graph.html?dataset=bbc22f0135744f678f1bbc590b2501d8&type=arrow&viztoken=380021af-aef1-4173-8ced-ce8adf956812&usertag=993172cb-pygraphistry-0.33.7&splashAfter=1713632217&info=true\n"
]
}
],
"source": [
"import networkx as nx\n",
"import pandas as pd\n",
"import graphistry\n",
"from cognee.config import Config\n",
"from cognee.utils import render_graph\n",
"from cognee.infrastructure.databases.graph.get_graph_client import get_graph_client, GraphDBType\n",
"\n",
"config = Config()\n",
"config.load()\n",
"\n",
"graphistry.register(\n",
" api = 3,\n",
" username = config.graphistry_username,\n",
" password = config.graphistry_password\n",
")\n",
"\n",
"graph_client = await get_graph_client(GraphDBType.NETWORKX, \"cognee_graph.pkl\")\n",
"graph = graph_client.graph\n",
"\n",
"await render_graph(graph)\n",
"\n",
"# edges = nx.to_pandas_edgelist(graph)\n",
"\n",
"# nodes_data = [{\n",
"# \"id\": node_id,\n",
"# \"label\": node[\"name\"] if \"name\" in node else node_id,\n",
"# } for (node_id, node) in graph.nodes(data = True)]\n",
"\n",
"# nodes = pd.DataFrame(nodes_data)\n",
"\n",
"# plotter = graphistry.edges(edges, source = \"source\", destination = \"target\").nodes(nodes, \"id\")\n",
"\n",
"# plotter.bind(edge_title = \"relationship_name\", edge_label = \"relationship_name\", point_title = \"name\", point_label = \"name\")\n",
"# url = plotter.plot(render = False, as_files = True)\n",
"# print(f\"Graph is visualized at: {url}\")\n"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "e916c484",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 122845.72it/s]\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 88434.12it/s]\n",
"Fetching 7 files: 100%|██████████| 7/7 [00:00<00:00, 56461.78it/s]\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/pydantic/_internal/_core_utils.py:568: ResourceWarning: unclosed <socket.socket fd=79, family=2, type=1, proto=6, laddr=('192.168.9.243', 63135), raddr=('35.245.15.233', 6333)>\n",
" return _validate_core_schema(schema)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=79 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/Projects/Topoteretes/cognee/.venv/lib/python3.11/site-packages/pydantic/_internal/_core_utils.py:568: ResourceWarning: unclosed <socket.socket fd=81, family=2, type=1, proto=6, laddr=('192.168.9.243', 63136), raddr=('35.245.15.233', 6333)>\n",
" return _validate_core_schema(schema)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n",
"/Users/borisarzentar/.pyenv/versions/3.11.8/lib/python3.11/asyncio/selector_events.py:868: ResourceWarning: unclosed transport <_SelectorSocketTransport fd=81 read=idle write=<idle, bufsize=0>>\n",
" _warn(f\"unclosed transport {self!r}\", ResourceWarning, source=self)\n",
"ResourceWarning: Enable tracemalloc to get the object allocation traceback\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"True\n",
"Through the Window is an album by the American music project Prurient, the performing name of the artist Dominick Fernow. The three-song album was released on March 19, 2013 through the English label Blackest Ever Black. Though released in 2013, the tracks for \"Through the Window\" were recorded in October 2011 at the same time as Prurient's two Hydra Head Records releases — the studio album \"Bermuda Drain\" (2011) and the EP \"Time's Arrow\" (2011) — and were noted for musically showing more techno influences, akin to one of Fernow's other projects, Vatican Shadow.\n",
"Yes I Am is the fourth studio album by American singer-songwriter Melissa Etheridge, released by Island Records on September 21, 1993 (see 1993 in music). The title is generally thought to refer to Etheridge's recent coming out as a lesbian, confirming long-standing rumors about her personal life. This is the album that gave Etheridge national recognition. The rock ballad \"Come to My Window\" was the first single released from the album, which peaked at No. 25 on the \"Billboard\" Hot 100, and its video featured actress Juliette Lewis having a nervous breakdown. This single brought the album into the public consciousness and was quickly followed by \"I'm the Only One\", which became a major hit and reached No. 8 on the Hot 100, and \"If I Wanted To\", which hit No. 16.\n",
"Christian Joseph \"Chris\" Arena is an American singer-songwriter and producer whose work can be seen across television networks such as ABC, ABC Family, CBS, The CW and others as well as feature films. Arena was nominated for a 2016 Emmy Award for 'Outstanding Original Song in Daytime Drama' for his song 'Dreams' written for ABC's \"General Hospital\". Most recently, Arena's work can be seen on \"\" and \"Scream\" on MTV. Other tracks, \"Baby Fish\", \"Yes It Do\", \"Train\", \"City Inside Me\", and \"Closed Window\" were featured in Season 4 of ABC's \"Pretty Little Liars\". His song \"Dreams\" was used as a theme song on ABC's \"General Hospital\" in January 2015 as well as his latest song \"Nothing's Gonna Stop Us\", which was featured in TV Land's \"Younger\" in April 2015. His song \"For You\" was also picked for the romantic comedy/feature film, \"This Thing With Sarah\".\n",
"Melissa Lou Etheridge (born May 29, 1961) is an American singer-songwriter, guitarist, and activist. Her self-titled debut album \"Melissa Etheridge\" was released in 1988 and became an underground success. The album peaked at No. 22 on the \"Billboard\" 200, and its lead single, \"Bring Me Some Water\", garnered Etheridge her first Grammy Award nomination for Best Rock Vocal Performance, Female. In 1993, Etheridge won her first Grammy award for her single \"Ain't It Heavy\" from her third album, \"Never Enough\". Later that year, she released what would become her mainstream breakthrough album, \"Yes I Am\". Its tracks \"I'm the Only One\" and \"Come to My Window\" both reached the top 30 in the United States, and the latter earned Etheridge her second Grammy award. \"Yes I Am\" peaked at No. 15 on the \"Billboard\" 200, and spent 138 weeks on the chart, earning a RIAA certification of 6x Platinum, her largest to date.\n",
"American singer-songwriter Bridgit Mendler has embarked on two concert tours, one of which have been worldwide. Her debut tour, \"\", started on August 25, 2012. Mendler played at state fairs and music festivals in the United States and Canada to promote her debut album \"Hello My Name Is...\" with the new songs. \"Rocks at My Window\" and \"Hold On for Dear Love\" were the only songs from the album not to be performed. She performing two covers songs in the shows, \"Animal\", originally by Neon Trees, and \"This Love, by Maroon 5. She also sang a song from the soundtrack of \"Lemonade Mouth\", \"Somebody\". On January 19, 2013 Mendler in New York from UNICEF charity. Her second concert, the Summer Tour, visited North America in the leg one and started in Iowa and ended in Washington. In 2014 Mendler announced the second leg of the tour as part of the release of her upcoming second album. The leg started in Charlottetown, Canada, on June 28, 2014.\n",
"John Townes Van Zandt (March 7, 1944 – January 1, 1997), best known as Townes Van Zandt, was an American singer songwriter. He is widely regarded for his poetic, often heroically sad songs. In 1983, six years after Emmylou Harris had first popularized it, Willie Nelson and Merle Haggard covered his song \"Pancho and Lefty\", scoring a number one hit on the Billboard country music charts. Much of his life was spent touring various dive bars, often living in cheap motel rooms and backwoods cabins. For much of the 1970s, he lived in a simple shack without electricity or a phone.\n",
"Native Window is the self-titled debut album of American progressive rock band Native Window. It was released on June 23, 2009. The tracks on the album are all original material composed by the members of the band. The album came about soon after the band formed; it was because Steve Walsh, the lead singer for Kansas, would not write any new material. Phil Ehart, Rich Williams, Billy Greer, and David Ragsdale got together to write new songs, and released them under the Native Window name.\n",
"At My Window is an album released by Folk/country singer-songwriter Townes Van Zandt in 1987. This was Van Zandt's first studio album in the nine years that followed 1978's \"Flyin' Shoes\", and his only studio album recorded in the 1980s. Although the songwriter had become less prolific, this release showed that the quality of his material remained high.\n",
"Love Letter for Fire is a collaborative album of duets by American singer-songwriter Sam Beam, commonly known as Iron & Wine, and American singer-songwriter Jesca Hoop, released on April 15, 2016 on Sub Pop.\n",
"Little Window is the debut album of American singer-songwriter Baby Dee. The album was released in 2002 on the Durtro label. It was produced, composed, and performed entirely by Dee.\n"
]
}
],
"source": [
"from os import path\n",
"import cognee\n",
"from cognee.modules.search.vector.search_similarity import search_similarity\n",
"\n",
"data_directory_path = path.abspath(\"../.data\")\n",
"cognee.config.data_root_directory(data_directory_path)\n",
"\n",
"cognee_directory_path = path.abspath(\"../.cognee_system\")\n",
"cognee.config.system_root_directory(cognee_directory_path)\n",
"\n",
"results = await search_similarity(\"At My Window was released by which American singer-songwriter?\")\n",
"\n",
"for result in results:\n",
" print(\"At My Window\" in result)\n",
" print(result)"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.8"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
Reference in a new issue View git blame Copy permalink