{"id":23678446,"url":"https://github.com/hermaeus1618/patternrecognition","last_synced_at":"2025-12-28T07:30:13.541Z","repository":{"id":267008373,"uuid":"900012978","full_name":"Hermaeus1618/PatternRecognition","owner":"Hermaeus1618","description":"Stock Pattern Recognition Dataset","archived":false,"fork":false,"pushed_at":"2024-12-07T16:39:27.000Z","size":6107,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2024-12-29T15:20:50.557Z","etag":null,"topics":["data-analysis","machine-learning","pattern-recognition","python","stock-market"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Hermaeus1618.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-12-07T16:20:49.000Z","updated_at":"2024-12-07T16:39:30.000Z","dependencies_parsed_at":"2024-12-07T17:32:18.123Z","dependency_job_id":null,"html_url":"https://github.com/Hermaeus1618/PatternRecognition","commit_stats":null,"previous_names":["hermaeus1618/patternrecognition"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Hermaeus1618%2FPatternRecognition","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Hermaeus1618%2FPatternRecognition/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Hermaeus1618%2FPatternRecognition/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Hermaeus1618%2FPatternRecognition/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Hermaeus1618","download_url":"https://codeload.github.com/Hermaeus1618/PatternRecognition/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":239718370,"owners_count":19685725,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["data-analysis","machine-learning","pattern-recognition","python","stock-market"],"created_at":"2024-12-29T15:20:09.226Z","updated_at":"2025-12-28T07:30:13.509Z","avatar_url":"https://github.com/Hermaeus1618.png","language":null,"readme":"# STOCK PATTERN RECOGNITION DATASET\n\nHere is the complete dataset in ZIP Format.\n\nUnfortunately the code cannot be provided as the model was a corporate project,\nbut the training data is gathered form public source and can be distributed.\n\nThis file contains several files which are numpy 1-D array in binary format.\nThese files are generated from `numpy.ndarray.tobytes()` to directly write inside ZIP stream.\n\nFile includes 37 different patterns although number of dataset per pattern are variable.\nRecommended dataset size for better training is \u003e300.\n\nValues are seprated by \"\\_GAP\\_\" string which can be easily converted back into list by `pd.Series().str.split()` like:\n``` python\nwith zipfile.ZipFile(DATASET.zfs, \"r\") as ZFILE:\n    PATTERNDF=pd.Series([F.filename for F in ZFILE.infolist()])\nPATTERNDF=PATTERNDF.str.split(\"_GAP_\", expand=True)\n```\nNaming convention for Numpy files:\n`IOC_GAP_CupAndHandle_GAP_HOUR1_GAP_1722503700_GAP_1725272100`\nSTOCK_GAP_PATTERN_GAP_TIMEFRAME_GAP_START_GAP_END\n\n- IOC: Name of stock.\n- CupAndHandle: Name of pattern.\n- HOUR1: Timeframe in which the pattern is formed.\n- 1722503700: Timestamp of pattern start.\n- 1725272100: Timestamp of pattern end.\n\n__Some points about dataset:__\nArrays are 1-Dimensional consist of price of stock which is forming the pattern.\nDatasets are variable in length.\nPrice data has to be normalized before training.\nNormalization can be as simple as `ARRAY/ARRAY.mean()`, avoid using complex normalization algorithms.\n\n__Some about training:__\nDue to variable length model need to include `LSTM` layer with `Embedding` layer as input.\n\nValues other than \u003cPATTERN\u003e are useless for training purpose, these are included to avoid naming conflit in the data generation algorithm.\nAfter which you can filter the pattern (or patterns for categorical classification) from dataframe.\n\nFor training a binary classification model use 1:1 ratio for both dataset.\n\nCan expect to get ocassional updates for **DATASET.zfs** in near future.\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhermaeus1618%2Fpatternrecognition","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhermaeus1618%2Fpatternrecognition","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhermaeus1618%2Fpatternrecognition/lists"}