the-embedding-archives.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-09-22T13:16:46.731681Z","iopub.execute_input":"2023-09-22T13:16:46.732266Z","iopub.status.idle":"2023-09-22T13:16:46.737903Z","shell.execute_reply.started":"2023-09-22T13:16:46.732232Z","shell.execute_reply":"2023-09-22T13:16:46.737111Z"},"trusted":true},"execution_count":3,"outputs":[]},{"cell_type":"code","source":"!pip install cohere","metadata":{"execution":{"iopub.status.busy":"2023-09-22T13:17:07.166452Z","iopub.execute_input":"2023-09-22T13:17:07.167488Z","iopub.status.idle":"2023-09-22T13:17:21.709528Z","shell.execute_reply.started":"2023-09-22T13:17:07.167448Z","shell.execute_reply":"2023-09-22T13:17:21.708116Z"},"trusted":true},"execution_count":5,"outputs":[{"name":"stdout","text":"Collecting cohere\n  Downloading cohere-4.27-py3-none-any.whl (47 kB)\n\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m47.6/47.6 kB\u001b[0m \u001b[31m3.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n\u001b[?25hRequirement already satisfied: aiohttp<4.0,>=3.0 in /opt/conda/lib/python3.10/site-packages (from cohere) (3.8.4)\nRequirement already satisfied: backoff<3.0,>=2.0 in /opt/conda/lib/python3.10/site-packages (from cohere) (2.2.1)\nCollecting fastavro==1.8.2 (from cohere)\n  Downloading fastavro-1.8.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.7 MB)\n\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.7/2.7 MB\u001b[0m \u001b[31m36.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n\u001b[?25hRequirement already satisfied: importlib_metadata<7.0,>=6.0 in /opt/conda/lib/python3.10/site-packages (from cohere) (6.7.0)\nRequirement already satisfied: requests<3.0.0,>=2.25.0 in /opt/conda/lib/python3.10/site-packages (from cohere) (2.31.0)\nRequirement already satisfied: urllib3<3,>=1.26 in /opt/conda/lib/python3.10/site-packages (from cohere) (1.26.15)\nRequirement already satisfied: attrs>=17.3.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp<4.0,>=3.0->cohere) (23.1.0)\nRequirement already satisfied: charset-normalizer<4.0,>=2.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp<4.0,>=3.0->cohere) (3.1.0)\nRequirement already satisfied: multidict<7.0,>=4.5 in /opt/conda/lib/python3.10/site-packages (from aiohttp<4.0,>=3.0->cohere) (6.0.4)\nRequirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /opt/conda/lib/python3.10/site-packages (from aiohttp<4.0,>=3.0->cohere) (4.0.2)\nRequirement already satisfied: yarl<2.0,>=1.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp<4.0,>=3.0->cohere) (1.9.2)\nRequirement already satisfied: frozenlist>=1.1.1 in /opt/conda/lib/python3.10/site-packages (from aiohttp<4.0,>=3.0->cohere) (1.3.3)\nRequirement already satisfied: aiosignal>=1.1.2 in /opt/conda/lib/python3.10/site-packages (from aiohttp<4.0,>=3.0->cohere) (1.3.1)\nRequirement already satisfied: zipp>=0.5 in /opt/conda/lib/python3.10/site-packages (from importlib_metadata<7.0,>=6.0->cohere) (3.15.0)\nRequirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.10/site-packages (from requests<3.0.0,>=2.25.0->cohere) (3.4)\nRequirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.10/site-packages (from requests<3.0.0,>=2.25.0->cohere) (2023.7.22)\nInstalling collected packages: fastavro, cohere\n  Attempting uninstall: fastavro\n    Found existing installation: fastavro 1.7.4\n    Uninstalling fastavro-1.7.4:\n      Successfully uninstalled fastavro-1.7.4\n\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\napache-beam 2.46.0 requires dill<0.3.2,>=0.3.1.1, but you have dill 0.3.7 which is incompatible.\u001b[0m\u001b[31m\n\u001b[0mSuccessfully installed cohere-4.27 fastavro-1.8.2\n","output_type":"stream"}]},{"cell_type":"code","source":"#Run: pip install cohere datasets\nfrom datasets import load_dataset\nimport torch\nimport cohere\n\nco = cohere.Client(f\"4ndsM9x1OQVOmoWWKgHzmCM8Ku1FETkEDZlaQn4H\")  # Add your cohere API key from www.cohere.com\n\n#Load at max 1000 documents + embeddings\nmax_docs = 1000\ndocs_stream = load_dataset(f\"Cohere/wikipedia-22-12-en-embeddings\", split=\"train\", streaming=True)","metadata":{"execution":{"iopub.status.busy":"2023-09-22T13:24:33.771429Z","iopub.execute_input":"2023-09-22T13:24:33.772273Z","iopub.status.idle":"2023-09-22T13:24:34.944067Z","shell.execute_reply.started":"2023-09-22T13:24:33.772204Z","shell.execute_reply":"2023-09-22T13:24:34.942963Z"},"trusted":true},"execution_count":11,"outputs":[{"output_type":"display_data","data":{"text/plain":"Downloading:   0%|          | 0.00/1.29k [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"707d9df62cdd4e0ca4a5025c82c68499"}},"metadata":{}}]},{"cell_type":"code","source":"docs = []\ndoc_embeddings = []\n\nfor doc in docs_stream:\n    docs.append(doc)\n    doc_embeddings.append(doc['emb'])\n    if len(docs) >= max_docs:\n        break\n\ndoc_embeddings = torch.tensor(doc_embeddings)\n\nquery = 'what is Data Science?'\nresponse = co.embed(texts=[query], model='multilingual-22-12')\nquery_embedding = response.embeddings \nquery_embedding = torch.tensor(query_embedding)\n\n# Compute dot score between query embedding and document embeddings\ndot_scores = torch.mm(query_embedding, doc_embeddings.transpose(0, 1))\ntop_k = torch.topk(dot_scores, k=3)\n\n# Print results\nprint(\"Query:\", query)\nfor doc_id in top_k.indices[0].tolist():\n    print(docs[doc_id]['title'])\n    print(docs[doc_id]['text'], \"\\n\")","metadata":{"execution":{"iopub.status.busy":"2023-09-22T13:24:42.801946Z","iopub.execute_input":"2023-09-22T13:24:42.802319Z","iopub.status.idle":"2023-09-22T13:24:50.509079Z","shell.execute_reply.started":"2023-09-22T13:24:42.802291Z","shell.execute_reply":"2023-09-22T13:24:50.508200Z"},"trusted":true},"execution_count":12,"outputs":[{"name":"stdout","text":"Query: what is Data Science?\nMicrosoft Office\nWhile Microsoft removed Data Analyzer, FrontPage, Vizact, and Schedule+ from Office 2007; they also added Communicator, Groove, SharePoint Designer, and Office Customization Tool (OCT) to the suite. \n\nWikipedia\nA working group led by Peter Stone (formed as a part of the Stanford-based project One Hundred Year Study on Artificial Intelligence) in its report called Wikipedia \"the best-known example of crowdsourcing... that far exceeds traditionally-compiled information sources, such as encyclopedias and dictionaries, in scale and depth\". \n\nWikipedia\nStudies related to Wikipedia have been using machine learning and artificial intelligence to support various operations. One of the most important areas—automatic detection of vandalism and data quality assessment in Wikipedia. \n\n","output_type":"stream"}]}]}