File size: 216,229 Bytes
bb3cf7a |
1 |
{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[],"gpuType":"A100","mount_file_id":"1Nxm-c8F7b7OL2R8Gz8TiintK04dAbLsJ","authorship_tag":"ABX9TyNClFBqrW6BhTcxW+7/L2Le"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"accelerator":"GPU","widgets":{"application/vnd.jupyter.widget-state+json":{"3e5152e0342147ec8ef57fc25d051b98":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_4ea3e8b8db12482ca116967c87ca3b00","IPY_MODEL_fa75b9900597448381c077ab54b27163","IPY_MODEL_75ff198870684610b364ac07680ac806"],"layout":"IPY_MODEL_1a4e80cc2b3849cdae00da8a9a8fdce4"}},"4ea3e8b8db12482ca116967c87ca3b00":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_aadc780c97e54833b5d957df1f7fa4ed","placeholder":"","style":"IPY_MODEL_54ca56dacd784ba4abd42474b855cdee","value":"modules.json: 100%"}},"fa75b9900597448381c077ab54b27163":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_a6bfe88742484d629f2a8107c89bb93d","max":349,"min":0,"orientation":"horizontal","style":"IPY_MODEL_53a54f9fc61942ef9cdb8f70d30708da","value":349}},"75ff198870684610b364ac07680ac806":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_4e543694f4694eed9fe5170fc1150a4f","placeholder":"","style":"IPY_MODEL_d12a293b1ba443cb9e7af103d0a1c61d","value":" 349/349 [00:00<00:00, 48.4kB/s]"}},"1a4e80cc2b3849cdae00da8a9a8fdce4":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"aadc780c97e54833b5d957df1f7fa4ed":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"54ca56dacd784ba4abd42474b855cdee":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"a6bfe88742484d629f2a8107c89bb93d":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"53a54f9fc61942ef9cdb8f70d30708da":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"4e543694f4694eed9fe5170fc1150a4f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"d12a293b1ba443cb9e7af103d0a1c61d":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"83e088c53e384a2eb052b857ba07e93d":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_ed5a4804e48b4cc09cf10968e04b7a4e","IPY_MODEL_f778592ce7e946efa3370e93c7bd9f87","IPY_MODEL_3b44932062274b6f8fddf0f40c8367e5"],"layout":"IPY_MODEL_bd34cda3e45a4604b2457e5ba6509862"}},"ed5a4804e48b4cc09cf10968e04b7a4e":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_0302305a32894e3aaa98c27e741a534e","placeholder":"","style":"IPY_MODEL_6f3faf488288410f8aed0f16a0950ba8","value":"config_sentence_transformers.json: 100%"}},"f778592ce7e946efa3370e93c7bd9f87":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_abb8f95f1c764c8c8988949633af0b11","max":124,"min":0,"orientation":"horizontal","style":"IPY_MODEL_ed358615afc94433821da06cf1f89ec7","value":124}},"3b44932062274b6f8fddf0f40c8367e5":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_8b4bf1e4d5544baf903ec7bb4a419e5f","placeholder":"","style":"IPY_MODEL_7dfdd78645db4896b4b6c5faa068e706","value":" 124/124 [00:00<00:00, 17.2kB/s]"}},"bd34cda3e45a4604b2457e5ba6509862":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"0302305a32894e3aaa98c27e741a534e":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"6f3faf488288410f8aed0f16a0950ba8":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"abb8f95f1c764c8c8988949633af0b11":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"ed358615afc94433821da06cf1f89ec7":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"8b4bf1e4d5544baf903ec7bb4a419e5f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"7dfdd78645db4896b4b6c5faa068e706":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"9f0b3bb9fe8246d0a015b09f83a6f69f":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_587344bf71ce46fb95cb5fc5ad2231d2","IPY_MODEL_554635ced6254ed4b5f2d95ca31041ba","IPY_MODEL_6252867ee31c4672a42441defef185c5"],"layout":"IPY_MODEL_ecf64f158f3e4869896c58efc71aee5d"}},"587344bf71ce46fb95cb5fc5ad2231d2":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_fffa43683f2e4eb691b5cdadf68a0122","placeholder":"","style":"IPY_MODEL_06034f9366074dc7ab227a41a45f4432","value":"README.md: "}},"554635ced6254ed4b5f2d95ca31041ba":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_8158f8488ce441149724a581c7cb1ea6","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_d8941216bf0d4e628798025d22bc6b53","value":1}},"6252867ee31c4672a42441defef185c5":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_8279de4f60a3484baae4c15e89345731","placeholder":"","style":"IPY_MODEL_57b306baba344afca11aa5240f479b78","value":" 94.6k/? [00:00<00:00, 11.3MB/s]"}},"ecf64f158f3e4869896c58efc71aee5d":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"fffa43683f2e4eb691b5cdadf68a0122":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"06034f9366074dc7ab227a41a45f4432":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"8158f8488ce441149724a581c7cb1ea6":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":"20px"}},"d8941216bf0d4e628798025d22bc6b53":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"8279de4f60a3484baae4c15e89345731":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"57b306baba344afca11aa5240f479b78":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"379903f6a51c4ad3ac85fbbf87c8cc55":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_28df7838b47a4c80828c220612c7e0f5","IPY_MODEL_d0c89eb784944402a3544b2eb55f320b","IPY_MODEL_3665d48b3177448e87b89390bc2b203f"],"layout":"IPY_MODEL_eeb0139000354daca0941e61c9c643f0"}},"28df7838b47a4c80828c220612c7e0f5":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_325381ad60ec43cc9b5751962e9c2d4f","placeholder":"","style":"IPY_MODEL_719a0527b42348978e6ce4ca198d9a2d","value":"sentence_bert_config.json: 100%"}},"d0c89eb784944402a3544b2eb55f320b":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_451543a8df6d4f009249624853d462ec","max":52,"min":0,"orientation":"horizontal","style":"IPY_MODEL_6dc9bee4fdb04e94944a1233209d0cbc","value":52}},"3665d48b3177448e87b89390bc2b203f":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_c9b0cc9725614591a713e667b6141b82","placeholder":"","style":"IPY_MODEL_5997f651f25a45fcbcae8c83c04318d8","value":" 52.0/52.0 [00:00<00:00, 7.38kB/s]"}},"eeb0139000354daca0941e61c9c643f0":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"325381ad60ec43cc9b5751962e9c2d4f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"719a0527b42348978e6ce4ca198d9a2d":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"451543a8df6d4f009249624853d462ec":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"6dc9bee4fdb04e94944a1233209d0cbc":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"c9b0cc9725614591a713e667b6141b82":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"5997f651f25a45fcbcae8c83c04318d8":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"dfd9dee097ff448897403622b98197a2":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_99db021ef8ec43868d9ec00aa792cea3","IPY_MODEL_781dfc174e1949f49038bfae99006393","IPY_MODEL_5266d924e6f34db498e52d4d0619a40c"],"layout":"IPY_MODEL_b1e7ac1904ed4f28a54c175cf0128259"}},"99db021ef8ec43868d9ec00aa792cea3":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_dccc8bbb86b94f829847303aeba2cf87","placeholder":"","style":"IPY_MODEL_7fbf0efcc06f415c81daab914ced1177","value":"config.json: 100%"}},"781dfc174e1949f49038bfae99006393":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_b9746207d1f14dff9355a6f540976143","max":779,"min":0,"orientation":"horizontal","style":"IPY_MODEL_014d1738b79d4e74a3cdebca3a346b39","value":779}},"5266d924e6f34db498e52d4d0619a40c":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_b673f6d547ce4c26bc38ad0122501811","placeholder":"","style":"IPY_MODEL_dcf24466e66c425b87cc6d1c505deccc","value":" 779/779 [00:00<00:00, 117kB/s]"}},"b1e7ac1904ed4f28a54c175cf0128259":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"dccc8bbb86b94f829847303aeba2cf87":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"7fbf0efcc06f415c81daab914ced1177":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"b9746207d1f14dff9355a6f540976143":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"014d1738b79d4e74a3cdebca3a346b39":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"b673f6d547ce4c26bc38ad0122501811":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"dcf24466e66c425b87cc6d1c505deccc":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"7e3980af11f74ccf9e0007d36255adab":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_2890d1edc1ff47e3bc2628d6e6ae2002","IPY_MODEL_4afb99107f3c4a0a9e9d699a45cdf9c7","IPY_MODEL_6677622b5750424aa95a557299f5a57d"],"layout":"IPY_MODEL_516874da8f2f4aab8a760f1d5c12f47f"}},"2890d1edc1ff47e3bc2628d6e6ae2002":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_26d3ebf3f49743d0b9e76388861c2d1b","placeholder":"","style":"IPY_MODEL_3c72c470222c4b1988dac81c0b7ce221","value":"model.safetensors: 100%"}},"4afb99107f3c4a0a9e9d699a45cdf9c7":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_9452098cfb6b42b6b12eaedd2b3bc695","max":1340616616,"min":0,"orientation":"horizontal","style":"IPY_MODEL_76d7366f16bb4b9e831fda22fcb115d0","value":1340616616}},"6677622b5750424aa95a557299f5a57d":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_9379699bc0394891b0123acb6c676573","placeholder":"","style":"IPY_MODEL_aec28defc8d64c62823523c7b4f53cb2","value":" 1.34G/1.34G [00:04<00:00, 779MB/s]"}},"516874da8f2f4aab8a760f1d5c12f47f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"26d3ebf3f49743d0b9e76388861c2d1b":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"3c72c470222c4b1988dac81c0b7ce221":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"9452098cfb6b42b6b12eaedd2b3bc695":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"76d7366f16bb4b9e831fda22fcb115d0":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"9379699bc0394891b0123acb6c676573":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"aec28defc8d64c62823523c7b4f53cb2":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"74976e0346e7445e96af5eb8b1d7bdf3":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_9d2922d6a6c34301ae0f83643b32eb26","IPY_MODEL_eb9ab50511e14a76a0f32b72c33f1788","IPY_MODEL_2939d352038e45c5908b091adf763e4f"],"layout":"IPY_MODEL_2ec494b3ac8f4656875efdb1d55950e5"}},"9d2922d6a6c34301ae0f83643b32eb26":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_9279de5d15ed42b6a612ae7b10425cf1","placeholder":"","style":"IPY_MODEL_09b1487d037b499eb62776a4f75b87dd","value":"tokenizer_config.json: 100%"}},"eb9ab50511e14a76a0f32b72c33f1788":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_884bc2bad1414739b1a111c875604bf7","max":366,"min":0,"orientation":"horizontal","style":"IPY_MODEL_0d724e40d0b54778927bb5e7cda34f7a","value":366}},"2939d352038e45c5908b091adf763e4f":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_44f3881d4f3e4619b230df5523bbc0c9","placeholder":"","style":"IPY_MODEL_b1c5358892454738a5f2e922a8f2ad48","value":" 366/366 [00:00<00:00, 51.1kB/s]"}},"2ec494b3ac8f4656875efdb1d55950e5":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"9279de5d15ed42b6a612ae7b10425cf1":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"09b1487d037b499eb62776a4f75b87dd":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"884bc2bad1414739b1a111c875604bf7":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"0d724e40d0b54778927bb5e7cda34f7a":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"44f3881d4f3e4619b230df5523bbc0c9":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"b1c5358892454738a5f2e922a8f2ad48":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"752e7b570dd146749162eeee13785c38":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_471adb58375c4b769792835f43ce59b7","IPY_MODEL_68341620497e45ad974dcea16db71968","IPY_MODEL_5dd76602baf049ef8900395b5465b76f"],"layout":"IPY_MODEL_3f6d150b5b664f69bfcadee012e40c5d"}},"471adb58375c4b769792835f43ce59b7":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_e574e8789e1d4d39901ee0291aa45583","placeholder":"","style":"IPY_MODEL_61eb6966f88f4fc5a539e41881b01548","value":"vocab.txt: "}},"68341620497e45ad974dcea16db71968":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_1a1c16a041ad4b8b909660d8543f3d19","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_672cffa2e6d54d8e9076845474e3601e","value":1}},"5dd76602baf049ef8900395b5465b76f":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_bcca69c2407e457da130d864cb4a2606","placeholder":"","style":"IPY_MODEL_bfbb39a165e74248b0fdf4d12a245ecb","value":" 232k/? [00:00<00:00, 20.6MB/s]"}},"3f6d150b5b664f69bfcadee012e40c5d":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"e574e8789e1d4d39901ee0291aa45583":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"61eb6966f88f4fc5a539e41881b01548":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"1a1c16a041ad4b8b909660d8543f3d19":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":"20px"}},"672cffa2e6d54d8e9076845474e3601e":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"bcca69c2407e457da130d864cb4a2606":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"bfbb39a165e74248b0fdf4d12a245ecb":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"0dc6672bc0304e65a85495a7e655a756":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_c8c336f6d6e54036b390a82d5bbb342d","IPY_MODEL_9e5be51dc927416d948fb053e3ad0c06","IPY_MODEL_062c27293e6b43b1bb7d4788225f23ac"],"layout":"IPY_MODEL_4d49ea92d59b4913b036e60db152c2c5"}},"c8c336f6d6e54036b390a82d5bbb342d":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_ab7c1a28f9a2464485c081140821fda8","placeholder":"","style":"IPY_MODEL_da5810877aed41a4b863b49e9eddc33e","value":"tokenizer.json: "}},"9e5be51dc927416d948fb053e3ad0c06":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_abc66d323fcf43669f6253af60199001","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_f0a32c67e78c45f0950a367ce1c36fa3","value":1}},"062c27293e6b43b1bb7d4788225f23ac":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_84334da14fe746f99fa6c8ed6736dc1f","placeholder":"","style":"IPY_MODEL_18d0b69fb388496e98a095c54ed8f396","value":" 711k/? [00:00<00:00, 51.0MB/s]"}},"4d49ea92d59b4913b036e60db152c2c5":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"ab7c1a28f9a2464485c081140821fda8":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"da5810877aed41a4b863b49e9eddc33e":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"abc66d323fcf43669f6253af60199001":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":"20px"}},"f0a32c67e78c45f0950a367ce1c36fa3":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"84334da14fe746f99fa6c8ed6736dc1f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"18d0b69fb388496e98a095c54ed8f396":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"71c049a6b87e49c88e39a3ae50cb8db4":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_1d85c3519e12402a8dab82a62a5054b0","IPY_MODEL_e33110ab2a294bd7a5daba032bc9c005","IPY_MODEL_48e5e07211414b418137898981ec464d"],"layout":"IPY_MODEL_82d73c5e7a274957bdda9e073b41f364"}},"1d85c3519e12402a8dab82a62a5054b0":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_f52ca7c881a44ac795fdb4f3a643cbf5","placeholder":"","style":"IPY_MODEL_78ba42d49a694542a2ca4051e3036284","value":"special_tokens_map.json: 100%"}},"e33110ab2a294bd7a5daba032bc9c005":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_defd2965ccee448897678e03a2cb0f08","max":125,"min":0,"orientation":"horizontal","style":"IPY_MODEL_884b33a4e5ad4d098759cb9c1b44de75","value":125}},"48e5e07211414b418137898981ec464d":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_ca5fe127bae74c61b34384ac0f817d1b","placeholder":"","style":"IPY_MODEL_34892ebb1e9b44f89508d4d86c7f2c3c","value":" 125/125 [00:00<00:00, 15.7kB/s]"}},"82d73c5e7a274957bdda9e073b41f364":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"f52ca7c881a44ac795fdb4f3a643cbf5":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"78ba42d49a694542a2ca4051e3036284":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"defd2965ccee448897678e03a2cb0f08":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"884b33a4e5ad4d098759cb9c1b44de75":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"ca5fe127bae74c61b34384ac0f817d1b":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"34892ebb1e9b44f89508d4d86c7f2c3c":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"659a3956fb6b47c094b91b0915a182cf":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_4ad3a0f3e4d449878ac78524349305be","IPY_MODEL_ffa98713c3134ee98def68e49d17db55","IPY_MODEL_c235e2c14acb4f6fb981698ef7b7d10f"],"layout":"IPY_MODEL_54af40221e16408d922cb301b71f9135"}},"4ad3a0f3e4d449878ac78524349305be":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_d0cb4f189f9a4f29b12391de532b6c3a","placeholder":"","style":"IPY_MODEL_d128d109a9ca41c3b8c118705d5dd19e","value":"config.json: 100%"}},"ffa98713c3134ee98def68e49d17db55":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_a5f41095d2b2444f82328edc5282270b","max":191,"min":0,"orientation":"horizontal","style":"IPY_MODEL_ab6fcebe7ca74226acb91d161c508f63","value":191}},"c235e2c14acb4f6fb981698ef7b7d10f":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_c84d64e9c66e45e19fed4c48e849a86b","placeholder":"","style":"IPY_MODEL_c1dad8b695f14042a2bab52431030434","value":" 191/191 [00:00<00:00, 26.5kB/s]"}},"54af40221e16408d922cb301b71f9135":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"d0cb4f189f9a4f29b12391de532b6c3a":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"d128d109a9ca41c3b8c118705d5dd19e":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"a5f41095d2b2444f82328edc5282270b":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"ab6fcebe7ca74226acb91d161c508f63":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"c84d64e9c66e45e19fed4c48e849a86b":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"c1dad8b695f14042a2bab52431030434":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"85cefa93669c484aa1a0cd2ad15cbdbf":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_7521f78c662941a58dd04e0e662092c6","IPY_MODEL_3fe19361b7b945c39d1d1c08a72a8de4","IPY_MODEL_096001b8bfa746e781dd058ff77392af"],"layout":"IPY_MODEL_2e545288effe458597352b129414608d"}},"7521f78c662941a58dd04e0e662092c6":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_a04f5b58f1834b79aefab5cc799ff05b","placeholder":"","style":"IPY_MODEL_499320899a0646b29362d75e9c470d4f","value":"Batches: 100%"}},"3fe19361b7b945c39d1d1c08a72a8de4":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_829abcc9f4f74780a291fd1c288fe4b1","max":949,"min":0,"orientation":"horizontal","style":"IPY_MODEL_ff34c10b6e0f4c2c961f0b275ec91bb5","value":949}},"096001b8bfa746e781dd058ff77392af":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_f4869e024876443e8afd530536776233","placeholder":"","style":"IPY_MODEL_78a489f8f9be468bb19c987cd63a8872","value":" 949/949 [03:05<00:00, 23.38it/s]"}},"2e545288effe458597352b129414608d":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"a04f5b58f1834b79aefab5cc799ff05b":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"499320899a0646b29362d75e9c470d4f":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"829abcc9f4f74780a291fd1c288fe4b1":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"ff34c10b6e0f4c2c961f0b275ec91bb5":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"f4869e024876443e8afd530536776233":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"78a489f8f9be468bb19c987cd63a8872":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}}}}},"cells":[{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"4qwgcNYJj44r","executionInfo":{"status":"ok","timestamp":1764001709510,"user_tz":300,"elapsed":20263,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"c513a5d3-0af5-4542-910a-cf7865719da1"},"outputs":[{"output_type":"stream","name":"stdout","text":["Archive: /content/ufo_enriched_v2.zip\n"," inflating: /content/ufo_enriched_v2.jsonl \n"]}],"source":["!unzip /content/ufo_enriched_v2.zip -d /content/"]},{"cell_type":"code","source":["#!/usr/bin/env python3\n","# -*- coding: utf-8 -*-\n","\"\"\"\n","Step 3 — BGE-Large embeddings for UFO corpus (A100-friendly)\n","\n","Inputs\n","------\n","- /content/ufo_enriched_v2.jsonl\n"," * First line: metadata header (from merge pipeline)\n"," * Remaining lines: enriched UFO events with:\n"," - provenance fields (raw_snapshot, raw_hash, lineage, etc.)\n"," - macro context\n"," - moon metrics\n"," - geomagnetic latitude\n"," - nearest airport\n","\n","Outputs\n","-------\n","- /content/ufo_bge_large_embeddings_v1.npz\n"," * ufo_uuid: np.array of shape (N,), dtype=object (event IDs)\n"," * embeddings: np.array of shape (N, 1024), dtype=float32\n","\n","- /content/ufo_bge_large_embeddings_v1.meta.json\n"," * JSON with dataset + model + pipeline metadata\n","\n","Notes\n","-----\n","- Only the WITNESS TEXT is embedded:\n"," * primary: full_text\n"," * fallback: summary\n","- Macro / moon / geomag / airports are NOT embedded here.\n","- Final “single file” with semantic dedupe will be built later\n"," by combining this NPZ with the enriched JSONL into\n"," ufo_semantic_v1.jsonl.\n","\"\"\"\n","\n","import json\n","import os\n","from typing import List, Tuple\n","\n","import numpy as np\n","\n","from sentence_transformers import SentenceTransformer\n","import torch\n","\n","\n","# -------------------------------------------------------------------\n","# CONFIG\n","# -------------------------------------------------------------------\n","ENRICHED_PATH = \"/content/ufo_enriched_v2.jsonl\"\n","OUT_NPZ = \"/content/ufo_bge_large_embeddings_v1.npz\"\n","OUT_META = \"/content/ufo_bge_large_embeddings_v1.meta.json\"\n","\n","EMBED_MODEL_NAME = \"BAAI/bge-large-en-v1.5\"\n","BATCH_SIZE = 256 # Safe on A100, still OK on CPU if needed\n","\n","\n","# -------------------------------------------------------------------\n","# UTIL: Build text to embed for a row (witness text only)\n","# -------------------------------------------------------------------\n","def build_embed_text(row: dict) -> str:\n"," \"\"\"\n"," Choose the field used for semantic embedding.\n","\n"," Priority:\n"," 1) full_text (already your cleaned, merged narrative)\n"," 2) summary\n"," 3) fallback to \"\" (still encodable, but low information)\n"," \"\"\"\n"," for key in (\"full_text\", \"summary\"):\n"," val = row.get(key)\n"," if isinstance(val, str) and val.strip():\n"," return val.strip()\n"," return \"\"\n","\n","\n","# -------------------------------------------------------------------\n","# LOAD DATA: split header vs. event rows\n","# -------------------------------------------------------------------\n","def load_enriched_events(path: str) -> Tuple[dict, List[str], List[str]]:\n"," \"\"\"\n"," Reads ufo_enriched_v2.jsonl and returns:\n"," - header_obj: metadata header (first line with 'dataset' key)\n"," - uuids: list of ufo_uuid strings\n"," - texts: list of strings used for embedding (witness text)\n"," \"\"\"\n"," header_obj = None\n"," uuids: List[str] = []\n"," texts: List[str] = []\n","\n"," total_lines = 0\n"," events = 0\n","\n"," print(f\"Loading enriched JSONL from: {path}\")\n"," with open(path, \"r\", encoding=\"utf-8\") as f:\n"," for line in f:\n"," total_lines += 1\n"," line = line.strip()\n"," if not line:\n"," continue\n","\n"," row = json.loads(line)\n","\n"," # Treat the first line with 'dataset' as the header\n"," if header_obj is None and \"dataset\" in row and \"merge_version\" in row:\n"," header_obj = row\n"," continue\n","\n"," # All event rows MUST have ufo_uuid (from your merge pipeline)\n"," ufo_id = row.get(\"ufo_uuid\")\n"," if not isinstance(ufo_id, str):\n"," # If somehow no UUID, skip this row from embedding.\n"," continue\n","\n"," text = build_embed_text(row)\n"," uuids.append(ufo_id)\n"," texts.append(text)\n"," events += 1\n","\n"," print(f\"Total lines read: {total_lines}\")\n"," print(f\"Header present: {'yes' if header_obj is not None else 'no'}\")\n"," print(f\"Events for embed: {events}\")\n"," return header_obj, uuids, texts\n","\n","\n","# -------------------------------------------------------------------\n","# MAIN\n","# -------------------------------------------------------------------\n","def main():\n"," # --------------------------\n"," # 1) Load data (header + events)\n"," # --------------------------\n"," header_obj, uuids, texts = load_enriched_events(ENRICHED_PATH)\n"," if not uuids:\n"," raise RuntimeError(\"No events with ufo_uuid found – check input file path / format.\")\n","\n"," # --------------------------\n"," # 2) Load BGE-Large model\n"," # --------------------------\n"," device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n"," print(f\"\\nLoading embedding model: {EMBED_MODEL_NAME} on device: {device}\")\n"," model = SentenceTransformer(EMBED_MODEL_NAME, device=device)\n","\n"," # --------------------------\n"," # 3) Encode all texts in batches\n"," # --------------------------\n"," print(\"\\nEncoding texts with BGE-Large…\")\n"," # SentenceTransformer handles batching internally\n"," embeddings = model.encode(\n"," texts,\n"," batch_size=BATCH_SIZE,\n"," show_progress_bar=True,\n"," normalize_embeddings=True, # standard for BGE\n"," )\n","\n"," embeddings = np.asarray(embeddings, dtype=\"float32\")\n"," uuids_arr = np.asarray(uuids, dtype=object)\n","\n"," print(\"\\n=== EMBEDDING SUMMARY ===\")\n"," print(f\"Num events embedded: {len(uuids_arr):,}\")\n"," print(f\"Embedding shape: {embeddings.shape} (rows, dim)\")\n"," print(f\"Embedding dtype: {embeddings.dtype}\")\n","\n"," # --------------------------\n"," # 4) Save NPZ (UUIDs + embeddings)\n"," # --------------------------\n"," np.savez(\n"," OUT_NPZ,\n"," ufo_uuid=uuids_arr,\n"," embeddings=embeddings,\n"," )\n"," print(f\"\\nSaved embeddings NPZ → {OUT_NPZ}\")\n","\n"," # --------------------------\n"," # 5) Save meta JSON\n"," # --------------------------\n"," meta = {\n"," \"dataset\": header_obj.get(\"dataset\") if header_obj else \"UFO Witness Corpus v2\",\n"," \"stage\": \"bge_large_embeddings_v1\",\n"," \"source_enriched_path\": os.path.abspath(ENRICHED_PATH),\n"," \"output_npz\": os.path.abspath(OUT_NPZ),\n"," \"embedding_model\": EMBED_MODEL_NAME,\n"," \"embedding_dim\": int(embeddings.shape[1]),\n"," \"normalize_embeddings\": True,\n"," \"batch_size\": BATCH_SIZE,\n"," \"num_events\": int(embeddings.shape[0]),\n"," }\n","\n"," with open(OUT_META, \"w\", encoding=\"utf-8\") as f:\n"," json.dump(meta, f, ensure_ascii=False, indent=2)\n","\n"," print(f\"Saved meta JSON → {OUT_META}\")\n"," print(\"\\nStep 3 complete. Use this NPZ in Step 4 (semantic clustering/dedupe).\")\n","\n","\n","if __name__ == \"__main__\":\n"," main()\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":849,"referenced_widgets":["3e5152e0342147ec8ef57fc25d051b98","4ea3e8b8db12482ca116967c87ca3b00","fa75b9900597448381c077ab54b27163","75ff198870684610b364ac07680ac806","1a4e80cc2b3849cdae00da8a9a8fdce4","aadc780c97e54833b5d957df1f7fa4ed","54ca56dacd784ba4abd42474b855cdee","a6bfe88742484d629f2a8107c89bb93d","53a54f9fc61942ef9cdb8f70d30708da","4e543694f4694eed9fe5170fc1150a4f","d12a293b1ba443cb9e7af103d0a1c61d","83e088c53e384a2eb052b857ba07e93d","ed5a4804e48b4cc09cf10968e04b7a4e","f778592ce7e946efa3370e93c7bd9f87","3b44932062274b6f8fddf0f40c8367e5","bd34cda3e45a4604b2457e5ba6509862","0302305a32894e3aaa98c27e741a534e","6f3faf488288410f8aed0f16a0950ba8","abb8f95f1c764c8c8988949633af0b11","ed358615afc94433821da06cf1f89ec7","8b4bf1e4d5544baf903ec7bb4a419e5f","7dfdd78645db4896b4b6c5faa068e706","9f0b3bb9fe8246d0a015b09f83a6f69f","587344bf71ce46fb95cb5fc5ad2231d2","554635ced6254ed4b5f2d95ca31041ba","6252867ee31c4672a42441defef185c5","ecf64f158f3e4869896c58efc71aee5d","fffa43683f2e4eb691b5cdadf68a0122","06034f9366074dc7ab227a41a45f4432","8158f8488ce441149724a581c7cb1ea6","d8941216bf0d4e628798025d22bc6b53","8279de4f60a3484baae4c15e89345731","57b306baba344afca11aa5240f479b78","379903f6a51c4ad3ac85fbbf87c8cc55","28df7838b47a4c80828c220612c7e0f5","d0c89eb784944402a3544b2eb55f320b","3665d48b3177448e87b89390bc2b203f","eeb0139000354daca0941e61c9c643f0","325381ad60ec43cc9b5751962e9c2d4f","719a0527b42348978e6ce4ca198d9a2d","451543a8df6d4f009249624853d462ec","6dc9bee4fdb04e94944a1233209d0cbc","c9b0cc9725614591a713e667b6141b82","5997f651f25a45fcbcae8c83c04318d8","dfd9dee097ff448897403622b98197a2","99db021ef8ec43868d9ec00aa792cea3","781dfc174e1949f49038bfae99006393","5266d924e6f34db498e52d4d0619a40c","b1e7ac1904ed4f28a54c175cf0128259","dccc8bbb86b94f829847303aeba2cf87","7fbf0efcc06f415c81daab914ced1177","b9746207d1f14dff9355a6f540976143","014d1738b79d4e74a3cdebca3a346b39","b673f6d547ce4c26bc38ad0122501811","dcf24466e66c425b87cc6d1c505deccc","7e3980af11f74ccf9e0007d36255adab","2890d1edc1ff47e3bc2628d6e6ae2002","4afb99107f3c4a0a9e9d699a45cdf9c7","6677622b5750424aa95a557299f5a57d","516874da8f2f4aab8a760f1d5c12f47f","26d3ebf3f49743d0b9e76388861c2d1b","3c72c470222c4b1988dac81c0b7ce221","9452098cfb6b42b6b12eaedd2b3bc695","76d7366f16bb4b9e831fda22fcb115d0","9379699bc0394891b0123acb6c676573","aec28defc8d64c62823523c7b4f53cb2","74976e0346e7445e96af5eb8b1d7bdf3","9d2922d6a6c34301ae0f83643b32eb26","eb9ab50511e14a76a0f32b72c33f1788","2939d352038e45c5908b091adf763e4f","2ec494b3ac8f4656875efdb1d55950e5","9279de5d15ed42b6a612ae7b10425cf1","09b1487d037b499eb62776a4f75b87dd","884bc2bad1414739b1a111c875604bf7","0d724e40d0b54778927bb5e7cda34f7a","44f3881d4f3e4619b230df5523bbc0c9","b1c5358892454738a5f2e922a8f2ad48","752e7b570dd146749162eeee13785c38","471adb58375c4b769792835f43ce59b7","68341620497e45ad974dcea16db71968","5dd76602baf049ef8900395b5465b76f","3f6d150b5b664f69bfcadee012e40c5d","e574e8789e1d4d39901ee0291aa45583","61eb6966f88f4fc5a539e41881b01548","1a1c16a041ad4b8b909660d8543f3d19","672cffa2e6d54d8e9076845474e3601e","bcca69c2407e457da130d864cb4a2606","bfbb39a165e74248b0fdf4d12a245ecb","0dc6672bc0304e65a85495a7e655a756","c8c336f6d6e54036b390a82d5bbb342d","9e5be51dc927416d948fb053e3ad0c06","062c27293e6b43b1bb7d4788225f23ac","4d49ea92d59b4913b036e60db152c2c5","ab7c1a28f9a2464485c081140821fda8","da5810877aed41a4b863b49e9eddc33e","abc66d323fcf43669f6253af60199001","f0a32c67e78c45f0950a367ce1c36fa3","84334da14fe746f99fa6c8ed6736dc1f","18d0b69fb388496e98a095c54ed8f396","71c049a6b87e49c88e39a3ae50cb8db4","1d85c3519e12402a8dab82a62a5054b0","e33110ab2a294bd7a5daba032bc9c005","48e5e07211414b418137898981ec464d","82d73c5e7a274957bdda9e073b41f364","f52ca7c881a44ac795fdb4f3a643cbf5","78ba42d49a694542a2ca4051e3036284","defd2965ccee448897678e03a2cb0f08","884b33a4e5ad4d098759cb9c1b44de75","ca5fe127bae74c61b34384ac0f817d1b","34892ebb1e9b44f89508d4d86c7f2c3c","659a3956fb6b47c094b91b0915a182cf","4ad3a0f3e4d449878ac78524349305be","ffa98713c3134ee98def68e49d17db55","c235e2c14acb4f6fb981698ef7b7d10f","54af40221e16408d922cb301b71f9135","d0cb4f189f9a4f29b12391de532b6c3a","d128d109a9ca41c3b8c118705d5dd19e","a5f41095d2b2444f82328edc5282270b","ab6fcebe7ca74226acb91d161c508f63","c84d64e9c66e45e19fed4c48e849a86b","c1dad8b695f14042a2bab52431030434","85cefa93669c484aa1a0cd2ad15cbdbf","7521f78c662941a58dd04e0e662092c6","3fe19361b7b945c39d1d1c08a72a8de4","096001b8bfa746e781dd058ff77392af","2e545288effe458597352b129414608d","a04f5b58f1834b79aefab5cc799ff05b","499320899a0646b29362d75e9c470d4f","829abcc9f4f74780a291fd1c288fe4b1","ff34c10b6e0f4c2c961f0b275ec91bb5","f4869e024876443e8afd530536776233","78a489f8f9be468bb19c987cd63a8872"]},"id":"90hzKWzZxwg8","executionInfo":{"status":"ok","timestamp":1764001992149,"user_tz":300,"elapsed":256007,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"6519ecef-ab69-43a9-bfba-194934ff9961"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Loading enriched JSONL from: /content/ufo_enriched_v2.jsonl\n","Total lines read: 242843\n","Header present: yes\n","Events for embed: 242842\n","\n","Loading embedding model: BAAI/bge-large-en-v1.5 on device: cuda\n"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.12/dist-packages/huggingface_hub/utils/_auth.py:94: UserWarning: \n","The secret `HF_TOKEN` does not exist in your Colab secrets.\n","To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.\n","You will be able to reuse this secret in all of your notebooks.\n","Please note that authentication is recommended but still optional to access public models or datasets.\n"," warnings.warn(\n"]},{"output_type":"display_data","data":{"text/plain":["modules.json: 0%| | 0.00/349 [00:00<?, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"3e5152e0342147ec8ef57fc25d051b98"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["config_sentence_transformers.json: 0%| | 0.00/124 [00:00<?, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"83e088c53e384a2eb052b857ba07e93d"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["README.md: 0.00B [00:00, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"9f0b3bb9fe8246d0a015b09f83a6f69f"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["sentence_bert_config.json: 0%| | 0.00/52.0 [00:00<?, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"379903f6a51c4ad3ac85fbbf87c8cc55"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["config.json: 0%| | 0.00/779 [00:00<?, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"dfd9dee097ff448897403622b98197a2"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["model.safetensors: 0%| | 0.00/1.34G [00:00<?, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"7e3980af11f74ccf9e0007d36255adab"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["tokenizer_config.json: 0%| | 0.00/366 [00:00<?, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"74976e0346e7445e96af5eb8b1d7bdf3"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["vocab.txt: 0.00B [00:00, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"752e7b570dd146749162eeee13785c38"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["tokenizer.json: 0.00B [00:00, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"0dc6672bc0304e65a85495a7e655a756"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["special_tokens_map.json: 0%| | 0.00/125 [00:00<?, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"71c049a6b87e49c88e39a3ae50cb8db4"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["config.json: 0%| | 0.00/191 [00:00<?, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"659a3956fb6b47c094b91b0915a182cf"}},"metadata":{}},{"output_type":"stream","name":"stdout","text":["\n","Encoding texts with BGE-Large…\n"]},{"output_type":"display_data","data":{"text/plain":["Batches: 0%| | 0/949 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"85cefa93669c484aa1a0cd2ad15cbdbf"}},"metadata":{}},{"output_type":"stream","name":"stdout","text":["\n","=== EMBEDDING SUMMARY ===\n","Num events embedded: 242,842\n","Embedding shape: (242842, 1024) (rows, dim)\n","Embedding dtype: float32\n","\n","Saved embeddings NPZ → /content/ufo_bge_large_embeddings_v1.npz\n","Saved meta JSON → /content/ufo_bge_large_embeddings_v1.meta.json\n","\n","Step 3 complete. Use this NPZ in Step 4 (semantic clustering/dedupe).\n"]}]},{"cell_type":"code","source":["#!/usr/bin/env python3\n","# -*- coding: utf-8 -*-\n","\"\"\"\n","Step 3 — BGE-Large embedding sanity check\n","\n","Validates:\n"," - NPZ file structure and shapes\n"," - Embedding norms (since we normalized)\n"," - Consistency between NPZ ufo_uuid list and enriched JSONL\n"," - Basic metadata consistency\n","\n","Inputs\n","------\n","- /content/ufo_enriched_v2.jsonl\n","- /content/ufo_bge_large_embeddings_v1.npz\n","- /content/ufo_bge_large_embeddings_v1.meta.json (optional but expected)\n","\n","Run\n","---\n","python ufo_step3_validate_embeddings.py\n","\"\"\"\n","\n","import os\n","import json\n","import numpy as np\n","\n","ENRICHED_PATH = \"/content/ufo_enriched_v2.jsonl\"\n","NPZ_PATH = \"/content/ufo_bge_large_embeddings_v1.npz\"\n","META_PATH = \"/content/ufo_bge_large_embeddings_v1.meta.json\"\n","\n","\n","# ------------------------------------------------------------\n","# Helpers\n","# ------------------------------------------------------------\n","def load_enriched_uuids(path):\n"," \"\"\"\n"," Stream the enriched JSONL and collect:\n"," - total event rows\n"," - set of ufo_uuid values\n"," Skips the header line that has 'dataset' / 'merge_version'.\n"," \"\"\"\n"," total_lines = 0\n"," event_rows = 0\n"," uuid_set = set()\n","\n"," header_seen = False\n","\n"," print(f\"Scanning enriched JSONL: {path}\")\n"," with open(path, \"r\", encoding=\"utf-8\") as f:\n"," for line in f:\n"," total_lines += 1\n"," line = line.strip()\n"," if not line:\n"," continue\n","\n"," row = json.loads(line)\n","\n"," # first header line\n"," if not header_seen and \"dataset\" in row and \"merge_version\" in row:\n"," header_seen = True\n"," continue\n","\n"," event_rows += 1\n"," ufo_id = row.get(\"ufo_uuid\")\n"," if isinstance(ufo_id, str):\n"," uuid_set.add(ufo_id)\n","\n"," print(f\" Total lines (incl. header): {total_lines:,}\")\n"," print(f\" Event rows (after header): {event_rows:,}\")\n"," print(f\" Unique event UUIDs: {len(uuid_set):,}\")\n"," return uuid_set, event_rows\n","\n","\n","def main():\n"," print(\"=== STEP 3 EMBEDDING VALIDATION ===\\n\")\n","\n"," # -----------------------------\n"," # 1) Check files exist\n"," # -----------------------------\n"," for p in (ENRICHED_PATH, NPZ_PATH):\n"," if not os.path.exists(p):\n"," raise FileNotFoundError(f\"Missing expected file: {p}\")\n","\n"," # META is optional but recommended\n"," meta = None\n"," if os.path.exists(META_PATH):\n"," with open(META_PATH, \"r\", encoding=\"utf-8\") as f:\n"," meta = json.load(f)\n"," print(\"Loaded meta JSON:\")\n"," print(json.dumps(meta, indent=2, ensure_ascii=False))\n"," else:\n"," print(f\"WARNING: meta JSON not found at {META_PATH}\")\n","\n"," # -----------------------------\n"," # 2) Load NPZ\n"," # -----------------------------\n"," print(\"\\nLoading NPZ embeddings:\", NPZ_PATH)\n"," npz = np.load(NPZ_PATH, allow_pickle=True)\n","\n"," if \"ufo_uuid\" not in npz or \"embeddings\" not in npz:\n"," raise KeyError(\"NPZ must contain 'ufo_uuid' and 'embeddings' arrays.\")\n","\n"," uuids_arr = npz[\"ufo_uuid\"]\n"," emb = npz[\"embeddings\"]\n","\n"," print(\"\\n=== NPZ STRUCTURE ===\")\n"," print(f\"ufo_uuid dtype: {uuids_arr.dtype}, shape: {uuids_arr.shape}\")\n"," print(f\"embeddings dtype: {emb.dtype}, shape: {emb.shape}\")\n","\n"," if emb.ndim != 2:\n"," raise ValueError(f\"Expected embeddings to be 2D; got shape {emb.shape}\")\n","\n"," n_rows, dim = emb.shape\n","\n"," if len(uuids_arr) != n_rows:\n"," raise ValueError(\n"," f\"Mismatch: len(ufo_uuid)={len(uuids_arr)} vs embeddings rows={n_rows}\"\n"," )\n","\n"," # If meta is present, verify embedding_dim\n"," if meta is not None and \"embedding_dim\" in meta:\n"," expected_dim = int(meta[\"embedding_dim\"])\n"," if dim != expected_dim:\n"," print(\n"," f\"WARNING: embedding_dim in meta={expected_dim}, \"\n"," f\"but embeddings shape dim={dim}\"\n"," )\n","\n"," # -----------------------------\n"," # 3) Norm statistics\n"," # -----------------------------\n"," print(\"\\n=== EMBEDDING NORM STATS ===\")\n"," # norms over axis=1 (row direction)\n"," norms = np.linalg.norm(emb, axis=1)\n"," print(f\"Min norm: {norms.min():.6f}\")\n"," print(f\"Max norm: {norms.max():.6f}\")\n"," print(f\"Mean norm: {norms.mean():.6f}\")\n"," print(f\"Std norm: {norms.std():.6f}\")\n","\n"," # If you normalized in encode(), these should all be ~1.0\n"," near_one_frac = float(np.mean((norms > 0.99) & (norms < 1.01)))\n"," print(f\"Frac norms in [0.99, 1.01]: {near_one_frac*100:.2f}%\")\n","\n"," # -----------------------------\n"," # 4) UUID integrity checks\n"," # -----------------------------\n"," print(\"\\n=== UUID INTEGRITY CHECK ===\")\n"," uuid_set_npz = set(map(str, uuids_arr.tolist()))\n"," print(f\"UUIDs in NPZ: {len(uuid_set_npz):,}\")\n","\n"," # Check if there are duplicates in NPZ\n"," if len(uuid_set_npz) != n_rows:\n"," print(\n"," f\"WARNING: NPZ has duplicate UUIDs: \"\n"," f\"rows={n_rows:,}, unique_uuid={len(uuid_set_npz):,}\"\n"," )\n"," else:\n"," print(\"No duplicate UUIDs in NPZ (good).\")\n","\n"," # -----------------------------\n"," # 5) Cross-check vs enriched JSONL\n"," # -----------------------------\n"," uuid_set_jsonl, event_rows = load_enriched_uuids(ENRICHED_PATH)\n","\n"," missing_in_jsonl = uuid_set_npz - uuid_set_jsonl\n"," missing_in_npz = uuid_set_jsonl - uuid_set_npz\n","\n"," print(\"\\n=== CROSS-CHECK (JSONL vs NPZ) ===\")\n"," print(f\"Events in JSONL (rows w/ UUID): {len(uuid_set_jsonl):,}\")\n"," print(f\"Events in NPZ (embedded UUIDs): {len(uuid_set_npz):,}\")\n","\n"," print(f\"UUIDs in NPZ but NOT in JSONL: {len(missing_in_jsonl):,}\")\n"," print(f\"UUIDs in JSONL but NOT in NPZ: {len(missing_in_npz):,}\")\n","\n"," if missing_in_jsonl:\n"," print(\" (This is unexpected – NPZ has UUIDs that JSONL doesn't.)\")\n"," if missing_in_npz:\n"," print(\n"," \" (Expected if you deliberately skipped some rows; \"\n"," \"otherwise, check Step 3 input filtering.)\"\n"," )\n","\n"," print(\"\\n=== DONE: STEP 3 VALIDATION ===\")\n"," print(\"If norms ~1 and UUID counts align, embeddings look good.\")\n"," print(\"You can now safely move on to Step 4 (semantic clustering/dedupe).\")\n","\n","\n","if __name__ == \"__main__\":\n"," main()\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"l4CsJ3a12L1s","executionInfo":{"status":"ok","timestamp":1764002056078,"user_tz":300,"elapsed":34920,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"091f8de4-dcd0-470d-ed7a-571fb585c2d8"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["=== STEP 3 EMBEDDING VALIDATION ===\n","\n","Loaded meta JSON:\n","{\n"," \"dataset\": \"UFO Witness Corpus v2\",\n"," \"stage\": \"bge_large_embeddings_v1\",\n"," \"source_enriched_path\": \"/content/ufo_enriched_v2.jsonl\",\n"," \"output_npz\": \"/content/ufo_bge_large_embeddings_v1.npz\",\n"," \"embedding_model\": \"BAAI/bge-large-en-v1.5\",\n"," \"embedding_dim\": 1024,\n"," \"normalize_embeddings\": true,\n"," \"batch_size\": 256,\n"," \"num_events\": 242842\n","}\n","\n","Loading NPZ embeddings: /content/ufo_bge_large_embeddings_v1.npz\n","\n","=== NPZ STRUCTURE ===\n","ufo_uuid dtype: object, shape: (242842,)\n","embeddings dtype: float32, shape: (242842, 1024)\n","\n","=== EMBEDDING NORM STATS ===\n","Min norm: 1.000000\n","Max norm: 1.000000\n","Mean norm: 1.000000\n","Std norm: 0.000000\n","Frac norms in [0.99, 1.01]: 100.00%\n","\n","=== UUID INTEGRITY CHECK ===\n","UUIDs in NPZ: 242,839\n","WARNING: NPZ has duplicate UUIDs: rows=242,842, unique_uuid=242,839\n","Scanning enriched JSONL: /content/ufo_enriched_v2.jsonl\n"," Total lines (incl. header): 242,843\n"," Event rows (after header): 242,842\n"," Unique event UUIDs: 242,839\n","\n","=== CROSS-CHECK (JSONL vs NPZ) ===\n","Events in JSONL (rows w/ UUID): 242,839\n","Events in NPZ (embedded UUIDs): 242,839\n","UUIDs in NPZ but NOT in JSONL: 0\n","UUIDs in JSONL but NOT in NPZ: 0\n","\n","=== DONE: STEP 3 VALIDATION ===\n","If norms ~1 and UUID counts align, embeddings look good.\n","You can now safely move on to Step 4 (semantic clustering/dedupe).\n"]}]},{"cell_type":"code","source":["import zipfile\n","import os\n","\n","meta_path = \"/content/ufo_bge_large_embeddings_v1.meta.json\"\n","npz_path = \"/content/ufo_bge_large_embeddings_v1.npz\"\n","zip_path = \"/content/ufo_bge_large_embeddings_v1.zip\"\n","\n","if not os.path.exists(meta_path):\n"," raise FileNotFoundError(meta_path)\n","if not os.path.exists(npz_path):\n"," raise FileNotFoundError(npz_path)\n","\n","with zipfile.ZipFile(zip_path, \"w\", compression=zipfile.ZIP_DEFLATED) as zf:\n"," zf.write(meta_path, arcname=\"ufo_bge_large_embeddings_v1.meta.json\")\n"," zf.write(npz_path, arcname=\"ufo_bge_large_embeddings_v1.npz\")\n","\n","print(\"Created ZIP:\", zip_path)\n","\n","# If you want to download in Colab:\n","# from google.colab import files\n","# files.download(zip_path)\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"2oGi_uWA4JNg","executionInfo":{"status":"ok","timestamp":1763921133728,"user_tz":300,"elapsed":33563,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"6aca8da3-6c35-4740-c78f-1fc707ec8fab"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Created ZIP: /content/ufo_bge_large_embeddings_v1.zip\n"]}]},{"cell_type":"code","source":["pip install hnswlib"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"SV_ijh7D4Mjg","executionInfo":{"status":"ok","timestamp":1764002137124,"user_tz":300,"elapsed":36945,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"acbc708e-10ce-471d-bf41-2202e8c566e6"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Collecting hnswlib\n"," Downloading hnswlib-0.8.0.tar.gz (36 kB)\n"," Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n"," Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n"," Preparing metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n","Requirement already satisfied: numpy in /usr/local/lib/python3.12/dist-packages (from hnswlib) (2.0.2)\n","Building wheels for collected packages: hnswlib\n"," Building wheel for hnswlib (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n"," Created wheel for hnswlib: filename=hnswlib-0.8.0-cp312-cp312-linux_x86_64.whl size=2528300 sha256=f350c9be93cd94cdc69127bfd281dccdfa37f3d1db6a9d830b7c3e940fe45532\n"," Stored in directory: /root/.cache/pip/wheels/ac/39/b3/cbd7f9cbb76501d2d5fbc84956e70d0b94e788aac87bda465e\n","Successfully built hnswlib\n","Installing collected packages: hnswlib\n","Successfully installed hnswlib-0.8.0\n"]}]},{"cell_type":"code","source":["#!/usr/bin/env python3\n","# -*- coding: utf-8 -*-\n","\"\"\"\n","STEP 4: Semantic clustering / dedupe over UFO Witness Corpus v2\n","\n","Pipeline decisions:\n","- Operate at ROW level (each enriched row is a candidate event).\n","- Use BGE-large normalized embeddings (cosine via HNSW).\n","- Build clusters by thresholding cosine similarity and taking connected components.\n","- Preserve all original fields, including provenance.\n","- Output a new JSONL with one canonical row per semantic cluster.\n","\n","Canonical selection rule (per cluster):\n","- Prefer rows with a geotemporal anchor (has_semantic_anchor = True:\n"," coordinates OR usable year_context_available).\n","- Then prefer rows with valid coordinates (has_coordinates = True).\n","- Then prefer earliest valid date (parsed_date_fill ASC).\n","- Then prefer longest narrative (full_text_len DESC).\n","- Then tie-break by ufo_uuid ASC, row_idx ASC.\n","\n","NEW IN THIS VERSION:\n","- We DO NOT semantically cluster rows whose narrative text is non-informative\n"," (e.g. \"nan\", \"none\", \"null\", empty, etc.). Those remain singletons, which\n"," prevents one giant cluster of \"nan\" rows.\n","- We ONLY allow semantic clustering between rows whose full_text is long\n"," enough (>= MIN_TEXT_LEN_FOR_CLUSTERING characters). Short, generic one-line\n"," reports never chain into large clusters.\n","- Geotemporal anchors (coords/year) are used ONLY for picking the canonical\n"," representative inside each cluster; they do NOT gate clustering anymore.\n","\"\"\"\n","\n","import json\n","import os\n","from collections import defaultdict\n","\n","import numpy as np\n","import pandas as pd\n","\n","try:\n"," import hnswlib\n","except ImportError:\n"," raise SystemExit(\n"," \"hnswlib is required. Install it with:\\n\\n\"\n"," \" pip install hnswlib\\n\"\n"," )\n","\n","# ---------------------------------------------------------------------\n","# CONFIG\n","# ---------------------------------------------------------------------\n","META_PATH = \"/content/ufo_bge_large_embeddings_v1.meta.json\"\n","DEFAULT_NPZ_PATH = \"/content/ufo_bge_large_embeddings_v1.npz\"\n","DEFAULT_JSONL_PATH = \"/content/ufo_enriched_v2.jsonl\"\n","OUT_JSONL_PATH = \"/content/ufo_semantic_deduped_v1.jsonl\"\n","\n","# HNSW + clustering params (tunable but fixed here for reproducibility)\n","HNSW_SPACE = \"cosine\"\n","HNSW_M = 32\n","HNSW_EF_CONSTRUCT = 200\n","HNSW_EF_SEARCH = 100\n","HNSW_RANDOM_SEED = 42\n","\n","K_NEIGHBORS = 10 # top-K neighbors per point\n","SIM_THRESHOLD = 0.92 # cosine similarity threshold to consider \"same event\"\n","\n","# Text length gate for clustering: only long, detailed narratives are eligible\n","MIN_TEXT_LEN_FOR_CLUSTERING = 120\n","\n","# Text values that we treat as \"non-informative\" for semantic dedupe\n","# (these rows will not be merged via embeddings).\n","NULL_LIKE_TEXT = {\"\", \"nan\", \"none\", \"null\", \"n/a\", \"na\", \"<na>\"}\n","\n","\n","# ---------------------------------------------------------------------\n","# UNION-FIND (DISJOINT SET) FOR CLUSTERS\n","# ---------------------------------------------------------------------\n","class UnionFind:\n"," def __init__(self, n: int):\n"," self.parent = list(range(n))\n"," self.rank = [0] * n\n","\n"," def find(self, x: int) -> int:\n"," # Path compression\n"," if self.parent[x] != x:\n"," self.parent[x] = self.find(self.parent[x])\n"," return self.parent[x]\n","\n"," def union(self, a: int, b: int):\n"," ra = self.find(a)\n"," rb = self.find(b)\n"," if ra == rb:\n"," return\n"," # Union by rank\n"," if self.rank[ra] < self.rank[rb]:\n"," self.parent[ra] = rb\n"," elif self.rank[ra] > self.rank[rb]:\n"," self.parent[rb] = ra\n"," else:\n"," self.parent[rb] = ra\n"," self.rank[ra] += 1\n","\n","\n","# ---------------------------------------------------------------------\n","# SMALL HELPER: MAKE VALUES JSON-SERIALIZABLE\n","# ---------------------------------------------------------------------\n","def to_jsonable(v):\n"," \"\"\"\n"," Normalize Pandas/NumPy scalars and timestamps into JSON-safe types.\n"," - np.generic / np.bool_ -> native Python scalar\n"," - pd.Timestamp -> ISO string or None if NaT\n"," - list/tuple -> recurse element-wise\n"," - dict -> recurse value-wise\n"," - NaN/NaT -> None\n"," \"\"\"\n"," # NumPy scalar (int64, float64, bool_, etc.)\n"," if isinstance(v, (np.generic,)):\n"," return v.item()\n","\n"," # Pandas timestamp\n"," if isinstance(v, pd.Timestamp):\n"," if pd.isna(v):\n"," return None\n"," return v.isoformat()\n","\n"," # Lists / tuples: recurse\n"," if isinstance(v, (list, tuple)):\n"," return [to_jsonable(x) for x in v]\n","\n"," # Dicts: recurse into values\n"," if isinstance(v, dict):\n"," return {str(k): to_jsonable(val) for k, val in v.items()}\n","\n"," # Normalize NaN / NaT to None (scalar-like only)\n"," try:\n"," if pd.isna(v):\n"," return None\n"," except TypeError:\n"," # Non-numeric/non-datetime types -> ignore\n"," pass\n","\n"," return v\n","\n","\n","# ---------------------------------------------------------------------\n","# LOAD META + EMBEDDINGS\n","# ---------------------------------------------------------------------\n","print(\"=== STEP 4: Semantic clustering / dedupe ===\")\n","\n","if not os.path.exists(META_PATH):\n"," raise FileNotFoundError(f\"Meta JSON not found: {META_PATH}\")\n","\n","with open(META_PATH, \"r\", encoding=\"utf-8\") as f:\n"," meta = json.load(f)\n","\n","print(\"\\nLoaded embedding meta JSON:\")\n","print(json.dumps(meta, ensure_ascii=False, indent=2))\n","\n","npz_path = meta.get(\"output_npz\", DEFAULT_NPZ_PATH)\n","enriched_path = meta.get(\"source_enriched_path\", DEFAULT_JSONL_PATH)\n","\n","if not os.path.exists(npz_path):\n"," raise FileNotFoundError(f\"Embeddings NPZ not found: {npz_path}\")\n","if not os.path.exists(enriched_path):\n"," raise FileNotFoundError(f\"Enriched JSONL not found: {enriched_path}\")\n","\n","print(\"\\nLoading embeddings NPZ:\", npz_path)\n","npz = np.load(npz_path, allow_pickle=True)\n","\n","# Expect at least: embeddings, ufo_uuid\n","if \"embeddings\" not in npz or \"ufo_uuid\" not in npz:\n"," raise ValueError(\"NPZ must contain 'embeddings' and 'ufo_uuid' arrays.\")\n","\n","embeddings = npz[\"embeddings\"]\n","ufo_uuid_arr = npz[\"ufo_uuid\"]\n","\n","n_vecs, dim = embeddings.shape\n","print(f\"Embeddings shape: {embeddings.shape}\")\n","print(\"First few UUIDs:\", ufo_uuid_arr[:5])\n","\n","# Sanity: norms ~1\n","norms = np.linalg.norm(embeddings, axis=1)\n","print(\"\\nEmbedding norm stats:\")\n","print(\" min:\", float(norms.min()))\n","print(\" max:\", float(norms.max()))\n","print(\" mean:\", float(norms.mean()))\n","\n","# ---------------------------------------------------------------------\n","# LOAD ENRICHED JSONL (SKIP HEADER) & ALIGN\n","# ---------------------------------------------------------------------\n","print(\"\\nLoading enriched JSONL rows from:\", enriched_path)\n","\n","rows = []\n","header_meta = None\n","\n","with open(enriched_path, \"r\", encoding=\"utf-8\") as f:\n"," # First line is dataset header\n"," first_line = f.readline()\n"," header_meta = json.loads(first_line)\n","\n"," for idx, line in enumerate(f):\n"," line = line.strip()\n"," if not line:\n"," continue\n"," row = json.loads(line)\n"," row[\"__row_idx\"] = idx # 0..N-1\n"," rows.append(row)\n","\n","print(f\"Loaded rows (excluding header): {len(rows)}\")\n","\n","if len(rows) != n_vecs:\n"," raise ValueError(\n"," f\"Mismatch: JSONL rows ({len(rows)}) vs embeddings ({n_vecs}). \"\n"," \"These must match 1:1 in order.\"\n"," )\n","\n","# Build DataFrame for canonical selection later\n","df = pd.DataFrame(rows)\n","df[\"row_idx\"] = df[\"__row_idx\"].astype(int)\n","df.drop(columns=[\"__row_idx\"], inplace=True)\n","\n","# Sanity: ufo_uuid alignment\n","if \"ufo_uuid\" not in df.columns:\n"," raise ValueError(\"Expected 'ufo_uuid' column in enriched JSONL rows.\")\n","\n","jsonl_uuid = df[\"ufo_uuid\"].astype(str).to_numpy()\n","npz_uuid = ufo_uuid_arr.astype(str)\n","\n","if len(jsonl_uuid) != len(npz_uuid):\n"," raise ValueError(\"UUID length mismatch between JSONL rows and NPZ.\")\n","\n","uuid_mismatch = np.sum(jsonl_uuid != npz_uuid)\n","print(f\"UUID positional mismatches between JSONL and NPZ: {uuid_mismatch}\")\n","\n","if uuid_mismatch > 0:\n"," print(\"WARNING: There are positional UUID mismatches. \"\n"," \"We will still trust row-order alignment, but investigate if needed.\")\n","\n","# ---------------------------------------------------------------------\n","# DERIVE GEOTEMPORAL ANCHOR FLAG (for canonical selection ONLY)\n","# ---------------------------------------------------------------------\n","print(\"\\nComputing geotemporal anchor mask (coords/year)...\")\n","\n","if \"latitude\" in df.columns and \"longitude\" in df.columns:\n"," coord_mask = df[\"latitude\"].notna() & df[\"longitude\"].notna()\n","else:\n"," coord_mask = pd.Series(False, index=df.index)\n","\n","if \"year_context_available\" in df.columns:\n"," year_mask = df[\"year_context_available\"].astype(bool)\n","elif \"year\" in df.columns:\n"," # Fallback: treat year >= 0 as usable\n"," year_mask = df[\"year\"].astype(int) >= 0\n","else:\n"," year_mask = pd.Series(False, index=df.index)\n","\n","df[\"has_semantic_anchor\"] = coord_mask | year_mask\n","anchor_mask = df[\"has_semantic_anchor\"].to_numpy()\n","\n","# ---------------------------------------------------------------------\n","# TEXT INFORMATIVENESS + LENGTH MASKS\n","# ---------------------------------------------------------------------\n","print(\"\\nComputing text informativeness mask for semantic clustering...\")\n","\n","df[\"full_text\"] = df.get(\"full_text\", df.get(\"summary\", \"\")).astype(str)\n","full_text_norm = df[\"full_text\"].str.strip().str.lower()\n","df[\"is_informative_text\"] = ~full_text_norm.isin(NULL_LIKE_TEXT)\n","is_informative = df[\"is_informative_text\"].to_numpy()\n","\n","num_non_informative = int((~df[\"is_informative_text\"]).sum())\n","num_informative = int(df[\"is_informative_text\"].sum())\n","\n","print(f\"Non-informative text rows: {num_non_informative}\")\n","print(f\"Informative text rows: {num_informative}\")\n","\n","# Length-based gate: only long, detailed narratives can form semantic edges\n","df[\"full_text_len\"] = df[\"full_text\"].str.len()\n","df[\"is_long_enough_for_cluster\"] = df[\"full_text_len\"] >= MIN_TEXT_LEN_FOR_CLUSTERING\n","long_enough_mask = df[\"is_long_enough_for_cluster\"].to_numpy()\n","\n","# ---------------------------------------------------------------------\n","# BUILD HNSW INDEX\n","# ---------------------------------------------------------------------\n","print(\"\\nBuilding HNSW index...\")\n","index = hnswlib.Index(space=HNSW_SPACE, dim=dim)\n","index.init_index(\n"," max_elements=n_vecs,\n"," ef_construction=HNSW_EF_CONSTRUCT,\n"," M=HNSW_M,\n"," random_seed=HNSW_RANDOM_SEED,\n",")\n","\n","# embeddings are already normalized; for space='cosine' that is ideal\n","ids = np.arange(n_vecs, dtype=np.int32)\n","index.add_items(embeddings, ids)\n","index.set_ef(HNSW_EF_SEARCH)\n","\n","print(f\"HNSW index built with {n_vecs} elements, dim={dim}.\")\n","\n","# ---------------------------------------------------------------------\n","# KNN QUERY + CLUSTERING VIA UNION-FIND\n","# ---------------------------------------------------------------------\n","print(\"\\nQuerying nearest neighbors and building clusters...\")\n","uf = UnionFind(n_vecs)\n","\n","BATCH = 10000\n","for start in range(0, n_vecs, BATCH):\n"," end = min(start + BATCH, n_vecs)\n"," batch_emb = embeddings[start:end]\n"," labels, distances = index.knn_query(batch_emb, k=K_NEIGHBORS)\n","\n"," for i_rel in range(end - start):\n"," i = start + i_rel\n"," neigh_ids = labels[i_rel]\n"," neigh_dists = distances[i_rel]\n","\n"," for j, dist in zip(neigh_ids, neigh_dists):\n"," if j == i:\n"," continue # skip self\n","\n"," # For space='cosine', distance = 1 - cos_sim\n"," cos_sim = 1.0 - float(dist)\n"," if cos_sim >= SIM_THRESHOLD:\n"," # Only cluster if BOTH i and j have:\n"," # - informative text (not \"nan\"/\"null\"/etc.)\n"," # - sufficiently long narrative (>= MIN_TEXT_LEN_FOR_CLUSTERING)\n"," if not (\n"," is_informative[i]\n"," and is_informative[j]\n"," and long_enough_mask[i]\n"," and long_enough_mask[j]\n"," ):\n"," continue\n","\n"," a, b = (i, j) if i < j else (j, i)\n"," uf.union(a, b)\n","\n","print(\"Finished neighbor queries and unions.\")\n","\n","# ---------------------------------------------------------------------\n","# EXTRACT CLUSTERS\n","# ---------------------------------------------------------------------\n","print(\"\\nExtracting connected components (semantic clusters)...\")\n","\n","root_to_indices = defaultdict(list)\n","for idx in range(n_vecs):\n"," root = uf.find(idx)\n"," root_to_indices[root].append(idx)\n","\n","num_clusters = len(root_to_indices)\n","cluster_sizes = [len(v) for v in root_to_indices.values()]\n","num_collapsed = n_vecs - num_clusters\n","\n","print(f\"Total events (rows): {n_vecs}\")\n","print(f\"Semantic clusters: {num_clusters}\")\n","print(f\"Collapsed rows (N - C): {num_collapsed}\")\n","print(\"Cluster size stats:\")\n","print(\" min:\", int(np.min(cluster_sizes)))\n","print(\" max:\", int(np.max(cluster_sizes)))\n","print(\" mean:\", float(np.mean(cluster_sizes)))\n","\n","# ---------------------------------------------------------------------\n","# CHOOSE CANONICAL ROW PER CLUSTER (DETERMINISTIC)\n","# ---------------------------------------------------------------------\n","print(\"\\nSelecting canonical row per cluster...\")\n","\n","# Parsed date (earlier is preferred). We treat unparseable as far-future.\n","df[\"parsed_date\"] = pd.to_datetime(df.get(\"date_time\", \"\"), errors=\"coerce\")\n","df[\"parsed_date_fill\"] = df[\"parsed_date\"].fillna(pd.Timestamp.max)\n","\n","# Geospatial integrity: has_coordinates (True if latitude & longitude present)\n","if \"latitude\" in df.columns and \"longitude\" in df.columns:\n"," df[\"has_coordinates\"] = df[\"latitude\"].notna() & df[\"longitude\"].notna()\n","else:\n"," df[\"has_coordinates\"] = False\n","\n","# Attach cluster root + size\n","cluster_root = np.zeros(n_vecs, dtype=np.int64)\n","cluster_size_arr = np.zeros(n_vecs, dtype=np.int64)\n","\n","for root, idxs in root_to_indices.items():\n"," size = len(idxs)\n"," for idx in idxs:\n"," cluster_root[idx] = root\n"," cluster_size_arr[idx] = size\n","\n","df[\"semantic_cluster_root\"] = cluster_root\n","df[\"semantic_cluster_size\"] = cluster_size_arr\n","\n","# Map each root to a consecutive cluster_id for nicer output\n","root_list = sorted(root_to_indices.keys())\n","root_to_cluster_id = {root: cid for cid, root in enumerate(root_list)}\n","df[\"semantic_cluster_id\"] = df[\"semantic_cluster_root\"].map(root_to_cluster_id).astype(int)\n","\n","# Choose canonical per cluster with prioritized rules:\n","# has_semantic_anchor DESC,\n","# has_coordinates DESC,\n","# parsed_date_fill ASC,\n","# full_text_len DESC,\n","# ufo_uuid ASC,\n","# row_idx ASC\n","canonical_indices = []\n","canonical_cluster_id = []\n","for root, idxs in root_to_indices.items():\n"," sub = df.loc[idxs]\n"," sub_sorted = sub.sort_values(\n"," by=[\n"," \"has_semantic_anchor\",\n"," \"has_coordinates\",\n"," \"parsed_date_fill\",\n"," \"full_text_len\",\n"," \"ufo_uuid\",\n"," \"row_idx\",\n"," ],\n"," ascending=[False, False, True, False, True, True],\n"," )\n"," best_row_idx = int(sub_sorted.iloc[0][\"row_idx\"])\n"," canonical_indices.append(best_row_idx)\n"," canonical_cluster_id.append(root_to_cluster_id[root])\n","\n","canonical_set = set(canonical_indices)\n","print(f\"Canonical rows selected: {len(canonical_set)} (should equal num_clusters={num_clusters})\")\n","\n","# ---------------------------------------------------------------------\n","# WRITE OUTPUT JSONL (HEADER + CANONICAL ROWS ONLY)\n","# ---------------------------------------------------------------------\n","print(f\"\\nWriting semantic-deduped JSONL to: {OUT_JSONL_PATH}\")\n","\n","header_out = {\n"," \"dataset\": header_meta.get(\"dataset\", \"UFO Witness Corpus v2\"),\n"," \"stage\": \"semantic_dedup_hnsw_cosine_v1\",\n"," \"source_enriched_path\": enriched_path,\n"," \"source_embeddings_npz\": npz_path,\n"," \"embedding_model\": meta.get(\"embedding_model\", \"BAAI/bge-large-en-v1.5\"),\n"," \"embedding_dim\": int(meta.get(\"embedding_dim\", embeddings.shape[1])),\n"," \"normalize_embeddings\": bool(meta.get(\"normalize_embeddings\", True)),\n"," \"hnswlib_params\": {\n"," \"space\": HNSW_SPACE,\n"," \"M\": HNSW_M,\n"," \"ef_construction\": HNSW_EF_CONSTRUCT,\n"," \"ef_search\": HNSW_EF_SEARCH,\n"," \"random_seed\": HNSW_RANDOM_SEED,\n"," },\n"," \"semantic_dedup_params\": {\n"," \"k_neighbors\": K_NEIGHBORS,\n"," \"sim_threshold\": SIM_THRESHOLD,\n"," \"min_text_len_for_clustering\": MIN_TEXT_LEN_FOR_CLUSTERING,\n"," \"canonical_priority\": [\n"," \"has_semantic_anchor DESC\",\n"," \"has_coordinates DESC\",\n"," \"parsed_date_fill ASC\",\n"," \"full_text_len DESC\",\n"," \"ufo_uuid ASC\",\n"," \"row_idx ASC\",\n"," ],\n"," \"text_informative_filter\": {\n"," \"null_like\": sorted(list(NULL_LIKE_TEXT)),\n"," },\n"," \"anchor_filter\": \"none (anchors affect canonical selection only, not clustering)\",\n"," },\n"," \"num_input_rows\": int(n_vecs),\n"," \"num_clusters\": int(num_clusters),\n"," \"num_collapsed_rows\": int(num_collapsed),\n"," \"num_informative_text_rows\": int(num_informative),\n"," \"num_non_informative_text_rows\": int(num_non_informative),\n","}\n","\n","with open(OUT_JSONL_PATH, \"w\", encoding=\"utf-8\") as fout:\n"," # Header line\n"," fout.write(json.dumps(header_out, ensure_ascii=False) + \"\\n\")\n","\n"," for root, idxs in root_to_indices.items():\n"," cid = root_to_cluster_id[root]\n","\n"," sub = df.loc[idxs]\n"," sub_sorted = sub.sort_values(\n"," by=[\n"," \"has_semantic_anchor\",\n"," \"has_coordinates\",\n"," \"parsed_date_fill\",\n"," \"full_text_len\",\n"," \"ufo_uuid\",\n"," \"row_idx\",\n"," ],\n"," ascending=[False, False, True, False, True, True],\n"," )\n"," best = sub_sorted.iloc[0]\n","\n"," # build output record\n"," rec = {}\n","\n"," # copy all original columns except helper/cluster internals\n"," for col in df.columns:\n"," if col in {\n"," \"row_idx\",\n"," \"full_text_len\",\n"," \"parsed_date\",\n"," \"parsed_date_fill\",\n"," \"has_coordinates\",\n"," \"is_informative_text\",\n"," \"is_long_enough_for_cluster\",\n"," \"has_semantic_anchor\",\n"," \"semantic_cluster_root\",\n"," \"semantic_cluster_id\",\n"," \"semantic_cluster_size\",\n"," }:\n"," continue\n"," rec[col] = best[col]\n","\n"," # attach semantic cluster info\n"," rec[\"semantic_cluster_id\"] = int(cid)\n"," rec[\"semantic_cluster_size\"] = int(len(idxs))\n"," rec[\"semantic_cluster_members\"] = [\n"," str(df.loc[i, \"ufo_uuid\"]) for i in idxs\n"," ]\n","\n"," # normalize types for JSON\n"," clean_rec = {k: to_jsonable(v) for k, v in rec.items()}\n","\n"," fout.write(json.dumps(clean_rec, ensure_ascii=False) + \"\\n\")\n","\n","print(\"=== DONE: Semantic dedupe written. ===\")\n","print(f\"Input rows: {n_vecs}\")\n","print(f\"Output rows (canonical): {num_clusters}\")\n","print(f\"Collapsed rows: {num_collapsed}\")\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"K7y39uC34qKP","executionInfo":{"status":"ok","timestamp":1764004306213,"user_tz":300,"elapsed":2097858,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"a1b70914-b15e-4715-fcdc-ebd502a98396"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["=== STEP 4: Semantic clustering / dedupe ===\n","\n","Loaded embedding meta JSON:\n","{\n"," \"dataset\": \"UFO Witness Corpus v2\",\n"," \"stage\": \"bge_large_embeddings_v1\",\n"," \"source_enriched_path\": \"/content/ufo_enriched_v2.jsonl\",\n"," \"output_npz\": \"/content/ufo_bge_large_embeddings_v1.npz\",\n"," \"embedding_model\": \"BAAI/bge-large-en-v1.5\",\n"," \"embedding_dim\": 1024,\n"," \"normalize_embeddings\": true,\n"," \"batch_size\": 256,\n"," \"num_events\": 242842\n","}\n","\n","Loading embeddings NPZ: /content/ufo_bge_large_embeddings_v1.npz\n","Embeddings shape: (242842, 1024)\n","First few UUIDs: ['cc3690df331d32c1abd8954e41ce94de' '4f47228de56dbaa8215df1909f4292df'\n"," 'a735c2455e6c6a14ba91362f2680edf4' '12961556ef8409be970e839f0b8bc5a8'\n"," '962866b542a3c0c34c98ca3d281307ac']\n","\n","Embedding norm stats:\n"," min: 0.9999998807907104\n"," max: 1.0000001192092896\n"," mean: 1.0\n","\n","Loading enriched JSONL rows from: /content/ufo_enriched_v2.jsonl\n","Loaded rows (excluding header): 242842\n","UUID positional mismatches between JSONL and NPZ: 0\n","\n","Computing geotemporal anchor mask (coords/year)...\n","\n","Computing text informativeness mask for semantic clustering...\n","Non-informative text rows: 107944\n","Informative text rows: 134898\n","\n","Building HNSW index...\n","HNSW index built with 242842 elements, dim=1024.\n","\n","Querying nearest neighbors and building clusters...\n","Finished neighbor queries and unions.\n","\n","Extracting connected components (semantic clusters)...\n","Total events (rows): 242842\n","Semantic clusters: 242633\n","Collapsed rows (N - C): 209\n","Cluster size stats:\n"," min: 1\n"," max: 46\n"," mean: 1.0008613832413562\n","\n","Selecting canonical row per cluster...\n","Canonical rows selected: 242633 (should equal num_clusters=242633)\n","\n","Writing semantic-deduped JSONL to: /content/ufo_semantic_deduped_v1.jsonl\n","=== DONE: Semantic dedupe written. ===\n","Input rows: 242842\n","Output rows (canonical): 242633\n","Collapsed rows: 209\n"]}]},{"cell_type":"code","source":["#!/usr/bin/env python3\n","# -*- coding: utf-8 -*-\n","\"\"\"\n","STEP 4B: Semantic cluster sanity check\n","\n","Purpose:\n","- Inspect the semantic cluster size distribution from STEP 4.\n","- Identify the largest cluster (e.g., 100k+ members) and analyze it:\n"," * shapes / countries distribution\n"," * coordinate coverage\n"," * text length stats\n"," * a few random example reports\n","\n","Inputs:\n"," - /content/ufo_semantic_deduped_v1.jsonl\n"," (output of STEP 4 semantic dedupe; 1 line header + canonical rows)\n"," - /content/ufo_enriched_v2.jsonl\n"," (full enriched dataset with all original rows)\n","\n","This script does NOT modify any files; it only prints diagnostics.\n","\"\"\"\n","\n","import json\n","import os\n","from collections import Counter\n","\n","import numpy as np\n","import pandas as pd\n","import random\n","\n","SEM_DEDUP_PATH = \"/content/ufo_semantic_deduped_v1.jsonl\"\n","ENRICHED_PATH = \"/content/ufo_enriched_v2.jsonl\"\n","\n","# Number of random samples to show from the mega cluster\n","N_SAMPLES = 10\n","\n","print(\"=== STEP 4B: Semantic cluster sanity check ===\\n\")\n","\n","if not os.path.exists(SEM_DEDUP_PATH):\n"," raise FileNotFoundError(f\"Semantic-deduped JSONL not found: {SEM_DEDUP_PATH}\")\n","\n","if not os.path.exists(ENRICHED_PATH):\n"," raise FileNotFoundError(f\"Enriched JSONL not found: {ENRICHED_PATH}\")\n","\n","# ---------------------------------------------------------------------\n","# 1) LOAD SEMANTIC-DEDUPED FILE & BUILD CLUSTER SIZE DISTRIBUTION\n","# ---------------------------------------------------------------------\n","print(f\"Loading semantic-deduped file: {SEM_DEDUP_PATH}\")\n","\n","cluster_sizes = []\n","cluster_id_to_row = {} # cluster_id -> canonical row (dict)\n","max_cluster_row = None\n","max_size = 0\n","\n","with open(SEM_DEDUP_PATH, \"r\", encoding=\"utf-8\") as f:\n"," header_line = f.readline().strip()\n"," header = json.loads(header_line)\n"," print(\"Header (summary):\")\n"," print(json.dumps(header, indent=2, ensure_ascii=False))\n"," print()\n","\n"," for line in f:\n"," line = line.strip()\n"," if not line:\n"," continue\n"," row = json.loads(line)\n","\n"," cid = int(row.get(\"semantic_cluster_id\", -1))\n"," csize = int(row.get(\"semantic_cluster_size\", 1))\n","\n"," cluster_sizes.append(csize)\n"," cluster_id_to_row[cid] = row\n","\n"," if csize > max_size:\n"," max_size = csize\n"," max_cluster_row = row\n","\n","num_clusters = len(cluster_sizes)\n","print(f\"Canonical rows (clusters): {num_clusters}\")\n","print(f\"Min cluster size: {int(np.min(cluster_sizes))}\")\n","print(f\"Max cluster size: {int(np.max(cluster_sizes))}\")\n","print(f\"Mean cluster size: {float(np.mean(cluster_sizes)):.3f}\")\n","for q in [50, 75, 90, 95, 99]:\n"," print(f\"{q}th percentile size: {int(np.quantile(cluster_sizes, q/100.0))}\")\n","print()\n","\n","if max_cluster_row is None:\n"," raise RuntimeError(\"No clusters found in semantic-deduped file.\")\n","\n","mega_cid = int(max_cluster_row[\"semantic_cluster_id\"])\n","mega_size = int(max_cluster_row[\"semantic_cluster_size\"])\n","mega_members = max_cluster_row.get(\"semantic_cluster_members\", [])\n","\n","print(\"=== LARGEST CLUSTER ===\")\n","print(f\"Cluster ID: {mega_cid}\")\n","print(f\"Cluster size: {mega_size}\")\n","print(f\"Members listed: {len(mega_members)}\")\n","print()\n","\n","# ---------------------------------------------------------------------\n","# 2) LOAD ENRICHED JSONL AND BUILD ufo_uuid -> ROW MAP\n","# ---------------------------------------------------------------------\n","print(f\"Loading enriched JSONL for cluster inspection: {ENRICHED_PATH}\")\n","\n","uuid_to_row = {}\n","\n","with open(ENRICHED_PATH, \"r\", encoding=\"utf-8\") as f:\n"," # first line is header\n"," enriched_header_line = f.readline().strip()\n"," _ = json.loads(enriched_header_line) # we don't actually need the contents here\n","\n"," for line in f:\n"," line = line.strip()\n"," if not line:\n"," continue\n"," row = json.loads(line)\n"," uid = str(row.get(\"ufo_uuid\"))\n"," if uid:\n"," uuid_to_row[uid] = row\n","\n","print(f\"Total enriched rows with UUID: {len(uuid_to_row):,}\")\n","print()\n","\n","# ---------------------------------------------------------------------\n","# 3) EXTRACT MEGA CLUSTER MEMBERS FROM ENRICHED DATA\n","# ---------------------------------------------------------------------\n","print(\"Extracting mega cluster members from enriched data...\")\n","\n","mega_rows = []\n","missing_count = 0\n","\n","for uid in mega_members:\n"," uids = str(uid)\n"," r = uuid_to_row.get(uids)\n"," if r is not None:\n"," mega_rows.append(r)\n"," else:\n"," missing_count += 1\n","\n","print(f\"Mega cluster rows found: {len(mega_rows)}\")\n","print(f\"Mega cluster rows missing: {missing_count}\")\n","print()\n","\n","if not mega_rows:\n"," raise RuntimeError(\"No rows for mega cluster were found in the enriched data.\")\n","\n","mega_df = pd.DataFrame(mega_rows)\n","\n","# ---------------------------------------------------------------------\n","# 4) STATS ABOUT THE MEGA CLUSTER\n","# ---------------------------------------------------------------------\n","print(\"=== Mega cluster stats ===\")\n","\n","# Shapes\n","shape_counts = Counter(mega_df.get(\"shape\", pd.Series(dtype=str)).astype(str))\n","print(\"\\nTop shapes in mega cluster:\")\n","for shape, cnt in shape_counts.most_common(10):\n"," print(f\" {shape!r:20s} -> {cnt}\")\n","\n","# Countries\n","country_counts = Counter(mega_df.get(\"country\", pd.Series(dtype=str)).astype(str))\n","print(\"\\nTop countries in mega cluster:\")\n","for country, cnt in country_counts.most_common(10):\n"," print(f\" {country!r:20s} -> {cnt}\")\n","\n","# Coordinate coverage\n","has_lat = mega_df.get(\"latitude\").notna() if \"latitude\" in mega_df.columns else pd.Series(False, index=mega_df.index)\n","has_lon = mega_df.get(\"longitude\").notna() if \"longitude\" in mega_df.columns else pd.Series(False, index=mega_df.index)\n","has_coords = has_lat & has_lon\n","num_coords = int(has_coords.sum())\n","print(f\"\\nRows with valid coordinates: {num_coords} / {len(mega_df)} \"\n"," f\"({100.0 * num_coords / len(mega_df):.2f}%)\")\n","\n","# Text length stats\n","text_series = mega_df.get(\"full_text\", mega_df.get(\"summary\", pd.Series(\"\", index=mega_df.index))).astype(str)\n","lengths = text_series.str.len().to_numpy()\n","\n","print(\"\\nText length stats (characters):\")\n","print(f\" min: {int(lengths.min())}\")\n","print(f\" max: {int(lengths.max())}\")\n","print(f\" mean: {float(lengths.mean()):.1f}\")\n","for q in [50, 75, 90, 95, 99]:\n"," print(f\" {q}th percentile: {int(np.quantile(lengths, q/100.0))}\")\n","\n","# Date distribution (if parseable)\n","parsed_dates = pd.to_datetime(mega_df.get(\"date_time\", \"\"), errors=\"coerce\")\n","valid_dates = parsed_dates.dropna()\n","if not valid_dates.empty:\n"," print(\"\\nDate range in mega cluster:\")\n"," print(f\" earliest: {valid_dates.min()}\")\n"," print(f\" latest: {valid_dates.max()}\")\n","else:\n"," print(\"\\nNo parseable dates in mega cluster.\")\n","\n","# ---------------------------------------------------------------------\n","# 5) RANDOM SAMPLE OF EXAMPLE REPORTS\n","# ---------------------------------------------------------------------\n","print(\"\\n=== Random sample of example reports from mega cluster ===\")\n","\n","if len(mega_rows) <= N_SAMPLES:\n"," sample_rows = mega_rows\n","else:\n"," sample_rows = random.sample(mega_rows, N_SAMPLES)\n","\n","for i, r in enumerate(sample_rows, 1):\n"," city = r.get(\"city\")\n"," state = r.get(\"state\")\n"," country = r.get(\"country\")\n"," dt = r.get(\"date_time\")\n"," shape = r.get(\"shape\")\n"," lat = r.get(\"latitude\")\n"," lon = r.get(\"longitude\")\n"," txt = r.get(\"full_text\") or r.get(\"summary\") or \"\"\n"," txt_snip = txt[:260].replace(\"\\n\", \" \")\n","\n"," print(\"-\" * 70)\n"," print(f\"Sample #{i}\")\n"," print(f\" ufo_uuid: {r.get('ufo_uuid')}\")\n"," print(f\" date_time: {dt}\")\n"," print(f\" location: {city}, {state}, {country}\")\n"," print(f\" coords: {lat}, {lon}\")\n"," print(f\" shape: {shape}\")\n"," print(f\" text[0:260]: {txt_snip!r}\")\n","\n","print(\"\\n=== DONE: Mega cluster sanity check complete. ===\")\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"GmUj7D1-4qGm","executionInfo":{"status":"ok","timestamp":1763934698614,"user_tz":300,"elapsed":132000,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"375420a5-29f3-4d5c-f001-90a4dbe7cf9d"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["=== STEP 4B: Semantic cluster sanity check ===\n","\n","Loading semantic-deduped file: /content/ufo_semantic_deduped_v1.jsonl\n","Header (summary):\n","{\n"," \"dataset\": \"UFO Witness Corpus v2\",\n"," \"stage\": \"semantic_dedup_hnsw_cosine_v1\",\n"," \"source_enriched_path\": \"/content/ufo_enriched_v2.jsonl\",\n"," \"source_embeddings_npz\": \"/content/ufo_bge_large_embeddings_v1.npz\",\n"," \"embedding_model\": \"BAAI/bge-large-en-v1.5\",\n"," \"embedding_dim\": 1024,\n"," \"normalize_embeddings\": true,\n"," \"hnswlib_params\": {\n"," \"space\": \"cosine\",\n"," \"M\": 32,\n"," \"ef_construction\": 200,\n"," \"ef_search\": 100,\n"," \"random_seed\": 42\n"," },\n"," \"semantic_dedup_params\": {\n"," \"k_neighbors\": 10,\n"," \"sim_threshold\": 0.92,\n"," \"min_text_len_for_clustering\": 120,\n"," \"canonical_priority\": [\n"," \"has_semantic_anchor DESC\",\n"," \"has_coordinates DESC\",\n"," \"parsed_date_fill ASC\",\n"," \"full_text_len DESC\",\n"," \"ufo_uuid ASC\",\n"," \"row_idx ASC\"\n"," ],\n"," \"text_informative_filter\": {\n"," \"null_like\": [\n"," \"\",\n"," \"<na>\",\n"," \"n/a\",\n"," \"na\",\n"," \"nan\",\n"," \"none\",\n"," \"null\"\n"," ]\n"," },\n"," \"anchor_filter\": \"none (anchors affect canonical selection only, not clustering)\"\n"," },\n"," \"num_input_rows\": 242842,\n"," \"num_clusters\": 242633,\n"," \"num_collapsed_rows\": 209,\n"," \"num_informative_text_rows\": 134898,\n"," \"num_non_informative_text_rows\": 107944\n","}\n","\n","Canonical rows (clusters): 242633\n","Min cluster size: 1\n","Max cluster size: 46\n","Mean cluster size: 1.001\n","50th percentile size: 1\n","75th percentile size: 1\n","90th percentile size: 1\n","95th percentile size: 1\n","99th percentile size: 1\n","\n","=== LARGEST CLUSTER ===\n","Cluster ID: 2008\n","Cluster size: 46\n","Members listed: 46\n","\n","Loading enriched JSONL for cluster inspection: /content/ufo_enriched_v2.jsonl\n","Total enriched rows with UUID: 242,839\n","\n","Extracting mega cluster members from enriched data...\n","Mega cluster rows found: 46\n","Mega cluster rows missing: 0\n","\n","=== Mega cluster stats ===\n","\n","Top shapes in mega cluster:\n"," 'light' -> 15\n"," 'unknown' -> 5\n"," 'other' -> 4\n"," 'sphere' -> 4\n"," 'diamond' -> 3\n"," 'nan' -> 3\n"," 'cylinder' -> 2\n"," 'flash' -> 2\n"," 'changing' -> 2\n"," 'cross' -> 1\n","\n","Top countries in mega cluster:\n"," 'nan' -> 46\n","\n","Rows with valid coordinates: 0 / 46 (0.00%)\n","\n","Text length stats (characters):\n"," min: 121\n"," max: 135\n"," mean: 129.5\n"," 50th percentile: 130\n"," 75th percentile: 133\n"," 90th percentile: 135\n"," 95th percentile: 135\n"," 99th percentile: 135\n","\n","No parseable dates in mega cluster.\n","\n","=== Random sample of example reports from mega cluster ===\n","----------------------------------------------------------------------\n","Sample #1\n"," ufo_uuid: 8d6c8766dd795058ba10bbfd4a5c9e91\n"," date_time: nan\n"," location: Kannapolis, NC, nan\n"," coords: None, None\n"," shape: diamond\n"," text[0:260]: 'Multiple sightings of star like object moving in rapid precise movements across the night sky. ((NUFORC Note: Sighting of Sirius? PD)'\n","----------------------------------------------------------------------\n","Sample #2\n"," ufo_uuid: 89c80a50588d8f94cec8c77b3af2492a\n"," date_time: nan\n"," location: Stony Brook, NY, nan\n"," coords: None, None\n"," shape: flash\n"," text[0:260]: 'Light in the sky, moving eratically, dimming and undimming, and staying still. ((NUFORC Note: Possible sighting of Sirius?? PD))'\n","----------------------------------------------------------------------\n","Sample #3\n"," ufo_uuid: 648b9934ba5d780e55bd8b5c4b6877d2\n"," date_time: nan\n"," location: New York, NY, nan\n"," coords: None, None\n"," shape: diamond\n"," text[0:260]: 'Blue, red and yellow lights flashing. It was moving back and forth very low. ((NUFORC Note: Possible sighting of Sirius? PD))'\n","----------------------------------------------------------------------\n","Sample #4\n"," ufo_uuid: dc9a00c2557a5c92c223b6d1f0c278c1\n"," date_time: nan\n"," location: Atlanta, GA, nan\n"," coords: None, None\n"," shape: light\n"," text[0:260]: 'Bright multy colored object..going in a zigzag motion,spinning around changing colors. ((NUFORC Note: Twinkling star?? PD))'\n","----------------------------------------------------------------------\n","Sample #5\n"," ufo_uuid: 92fdcf7b9c75c67c98d96557f5f6ff00\n"," date_time: nan\n"," location: Albuquerque, NM, nan\n"," coords: None, None\n"," shape: light\n"," text[0:260]: 'Like an average star but binoculars show it to be strong flashing lights of white, blue, red. ((NUFORC Note: \"Twinkling\" star? PD))'\n","----------------------------------------------------------------------\n","Sample #6\n"," ufo_uuid: 69a48c55cf9d0c07a177f5fe474b1819\n"," date_time: nan\n"," location: Harborton, VA, nan\n"," coords: None, None\n"," shape: light\n"," text[0:260]: 'bright object emitting colored lights moving slowly up and south from 35 to 60 degress elevation east. ((NUFORC Note: Sirius?? PD))'\n","----------------------------------------------------------------------\n","Sample #7\n"," ufo_uuid: db65d113d7f69dfdda32518cd5c5c484\n"," date_time: nan\n"," location: Washoe, NV, nan\n"," coords: None, None\n"," shape: formation\n"," text[0:260]: 'Bright multicolored obj. in the sky hovering, then moving quickly in multiple dir.. ((NUFORC Note: Possible twinkling star? PD))'\n","----------------------------------------------------------------------\n","Sample #8\n"," ufo_uuid: b21467d3d16eaaa94ae18ef0037257bb\n"," date_time: nan\n"," location: New Albany, PA, nan\n"," coords: None, None\n"," shape: light\n"," text[0:260]: 'Small flashing star-like object, primrily white light with smaller red & green lights. ((NUFORC Note: Sirius would be in SW sky. PD)'\n","----------------------------------------------------------------------\n","Sample #9\n"," ufo_uuid: 41434181bd46dcb4347d2b462f752181\n"," date_time: nan\n"," location: Sumrall, MS, nan\n"," coords: None, None\n"," shape: other\n"," text[0:260]: 'star like object making circular and zigzag pattrens almost like its searching for something. ((NUFORC Note: Twinkling star?? PD))'\n","----------------------------------------------------------------------\n","Sample #10\n"," ufo_uuid: 325317b0e5041fc3114cdc4b9c98a330\n"," date_time: nan\n"," location: Cardiff, CA, nan\n"," coords: None, None\n"," shape: changing\n"," text[0:260]: 'Very bright, stationary bright object, cycling thru reds, blues, greens and white. ((NUFORC Note: Twinkling star?? PD))'\n","\n","=== DONE: Mega cluster sanity check complete. ===\n"]}]},{"cell_type":"code","source":["!gzip -9 /content/ufo_semantic_deduped_v1.jsonl\n"],"metadata":{"id":"pf-xNDZZIBNi"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["!zip -j /content/ufo_v2_clustered_bm25_v1.zip /content/ufo_v2_clustered_bm25_v1.jsonl"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"DCGC_2zRIBDM","executionInfo":{"status":"ok","timestamp":1764014681185,"user_tz":300,"elapsed":132332,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"70651e20-20eb-4bc9-ebda-06982c126ee6"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":[" adding: ufo_v2_clustered_bm25_v1.jsonl (deflated 89%)\n"]}]},{"cell_type":"code","source":["#!/usr/bin/env python3\n","# -*- coding: utf-8 -*-\n","\"\"\"\n","STEP 5: Topic clustering (HDBSCAN + PCA) + BM25-style cluster keywords\n","over the semantic-deduped UFO Witness Corpus v2.\n","\n","Inputs:\n"," - /content/ufo_semantic_deduped_v1.jsonl (canonical, semantically deduped)\n"," - /content/ufo_bge_large_embeddings_v1.npz (BGE-large embeddings for ALL events)\n","\n","Outputs:\n"," - /content/ufo_v2_clustered_bm25_v1.jsonl (single merged artifact)\n","\n","Pipeline:\n"," 1. Load semantic-deduped canonical rows.\n"," 2. Map canonical ufo_uuid -> original BGE embeddings (1024D).\n"," 3. PCA(1024 -> 64) for stability + speed.\n"," 4. HDBSCAN on 64D for unsupervised topic clusters.\n"," - Only cluster rows with sufficiently long, informative text.\n"," - Others get cluster_id = -1 (noise/unclustered).\n"," 5. BM25-style cluster keywords:\n"," - Tokenize all texts.\n"," - Compute df / idf across corpus.\n"," - For each cluster, compute tf_cluster * idf to get top tokens.\n"," 6. Write one JSONL with:\n"," - Original canonical fields (including semantic_cluster_* from Step 4).\n"," - hdbscan_cluster_id, hdbscan_cluster_size\n"," - hdbscan_cluster_top_terms (list of strings)\n"," - All configs in header for full reproducibility.\n","\"\"\"\n","\n","import json\n","import os\n","import re\n","from collections import Counter, defaultdict\n","\n","import numpy as np\n","import pandas as pd\n","\n","try:\n"," import hdbscan\n","except ImportError:\n"," raise SystemExit(\n"," \"hdbscan is required. Install it with:\\n\\n\"\n"," \" pip install hdbscan\\n\"\n"," )\n","\n","from sklearn.decomposition import PCA\n","\n","# ---------------------------------------------------------------------\n","# CONFIG\n","# ---------------------------------------------------------------------\n","SEMANTIC_DEDUP_JSONL = \"/content/ufo_semantic_deduped_v1.jsonl\"\n","EMBEDDINGS_NPZ_PATH = \"/content/ufo_bge_large_embeddings_v1.npz\"\n","OUT_JSONL_PATH = \"/content/ufo_v2_clustered_bm25_v1.jsonl\"\n","\n","# PCA config\n","PCA_DIM = 64\n","PCA_RANDOM_SEED = 42\n","\n","# HDBSCAN config\n","# Tuned for more granular topics:\n","# - min_cluster_size = 15 (smaller, more clusters)\n","# - cluster_selection_method='leaf' to keep leaf clusters.\n","HDBSCAN_MIN_CLUSTER_SIZE = 15\n","HDBSCAN_MIN_SAMPLES = 5\n","HDBSCAN_METRIC = \"euclidean\"\n","HDBSCAN_CLUSTER_SELECTION_METHOD = \"leaf\"\n","HDBSCAN_ALPHA = 1.0 # explicit, used in leaf selection\n","\n","# Text filters for clustering\n","NULL_LIKE_TEXT = {\"\", \"nan\", \"none\", \"null\", \"n/a\", \"na\", \"<na>\"}\n","MIN_TEXT_LEN_FOR_CLUSTERING = 120 # only cluster reasonably long narratives\n","\n","# BM25-style keyword scoring\n","BM25_TOP_TERMS_PER_CLUSTER = 15\n","BM25_MIN_DF = 5 # ignore tokens that appear in fewer than this many docs\n","\n","# INTENTIONALLY MINIMAL STOPWORD LIST.\n","# We only strip pure grammatical glue. Domain words like \"light\", \"object\",\n","# \"sky\", \"triangle\", \"bright\", etc. are *the signal* in UFO narratives and\n","# must NOT be removed.\n","STOPWORDS = {\n"," \"the\", \"and\", \"a\", \"an\", \"of\", \"to\", \"in\", \"for\", \"on\", \"at\", \"with\",\n"," \"it\", \"is\", \"was\", \"were\", \"be\", \"been\", \"are\",\n"," \"this\", \"that\", \"these\", \"those\",\n"," \"from\", \"by\", \"or\", \"as\", \"but\",\n"," \"have\", \"has\", \"had\", \"not\", \"no\",\n"," \"i\", \"we\", \"you\", \"they\", \"he\", \"she\", \"them\", \"his\", \"her\",\n","}\n","\n","\n","# ---------------------------------------------------------------------\n","# HELPER: JSON-SERIALIZABLE\n","# ---------------------------------------------------------------------\n","def to_jsonable(v):\n"," \"\"\"\n"," Normalize Pandas/NumPy scalars and timestamps into JSON-safe types.\n"," - np.generic / np.bool_ -> native Python scalar\n"," - pd.Timestamp -> ISO string or None if NaT\n"," - list/tuple -> recurse element-wise\n"," - dict -> recurse value-wise\n"," - NaN/NaT -> None\n"," \"\"\"\n"," if isinstance(v, (np.generic,)):\n"," return v.item()\n","\n"," if isinstance(v, pd.Timestamp):\n"," if pd.isna(v):\n"," return None\n"," return v.isoformat()\n","\n"," if isinstance(v, (list, tuple)):\n"," return [to_jsonable(x) for x in v]\n","\n"," if isinstance(v, dict):\n"," # avoid NameError by using local names\n"," return {str(kk): to_jsonable(vv) for kk, vv in v.items()}\n","\n"," try:\n"," if pd.isna(v):\n"," return None\n"," except TypeError:\n"," pass\n","\n"," return v\n","\n","\n","# ---------------------------------------------------------------------\n","# HELPER: TOKENIZE\n","# ---------------------------------------------------------------------\n","WORD_RE = re.compile(r\"[a-z0-9]+\")\n","\n","def tokenize(text: str):\n"," text = text.lower()\n"," return WORD_RE.findall(text)\n","\n","\n","# ---------------------------------------------------------------------\n","# LOAD SEMANTIC-DEDUPED CANONICAL FILE\n","# ---------------------------------------------------------------------\n","print(\"=== STEP 5: HDBSCAN + BM25 cluster labeling ===\")\n","\n","if not os.path.exists(SEMANTIC_DEDUP_JSONL):\n"," raise FileNotFoundError(f\"Semantic-deduped JSONL not found: {SEMANTIC_DEDUP_JSONL}\")\n","\n","print(\"\\nLoading semantic-deduped canonical JSONL:\", SEMANTIC_DEDUP_JSONL)\n","\n","rows = []\n","header_meta = None\n","with open(SEMANTIC_DEDUP_JSONL, \"r\", encoding=\"utf-8\") as f:\n"," first_line = f.readline()\n"," header_meta = json.loads(first_line)\n","\n"," for line in f:\n"," line = line.strip()\n"," if not line:\n"," continue\n"," rows.append(json.loads(line))\n","\n","df = pd.DataFrame(rows)\n","n_docs = len(df)\n","print(f\"Canonical rows loaded: {n_docs}\")\n","\n","if \"ufo_uuid\" not in df.columns:\n"," raise ValueError(\"Expected 'ufo_uuid' column in semantic-deduped file.\")\n","\n","# Choose text column\n","TEXT_COL_CANDIDATES = [\"full_text\", \"text\", \"summary\"]\n","text_col = None\n","for c in TEXT_COL_CANDIDATES:\n"," if c in df.columns:\n"," text_col = c\n"," break\n","\n","if text_col is None:\n"," raise ValueError(\n"," \"No text column found. Expected one of: \"\n"," f\"{TEXT_COL_CANDIDATES}, but got columns: {list(df.columns)}\"\n"," )\n","\n","df[text_col] = df[text_col].fillna(\"\").astype(str)\n","texts = df[text_col].to_numpy()\n","\n","# ---------------------------------------------------------------------\n","# LOAD EMBEDDINGS NPZ AND MAP CANONICAL ROWS\n","# ---------------------------------------------------------------------\n","if not os.path.exists(EMBEDDINGS_NPZ_PATH):\n"," raise FileNotFoundError(f\"Embeddings NPZ not found: {EMBEDDINGS_NPZ_PATH}\")\n","\n","print(\"\\nLoading embeddings NPZ:\", EMBEDDINGS_NPZ_PATH)\n","npz = np.load(EMBEDDINGS_NPZ_PATH, allow_pickle=True)\n","\n","if \"embeddings\" not in npz or \"ufo_uuid\" not in npz:\n"," raise ValueError(\"NPZ must contain 'embeddings' and 'ufo_uuid' arrays.\")\n","\n","emb_all = npz[\"embeddings\"]\n","uuids_all = npz[\"ufo_uuid\"].astype(str)\n","n_all, emb_dim = emb_all.shape\n","\n","print(f\"Embeddings shape (full): {emb_all.shape}\")\n","\n","uuid_to_idx = {u: i for i, u in enumerate(uuids_all)}\n","\n","canonical_uuids = df[\"ufo_uuid\"].astype(str).to_numpy()\n","canon_indices = np.empty(n_docs, dtype=np.int64)\n","\n","missing = 0\n","for i, u in enumerate(canonical_uuids):\n"," idx = uuid_to_idx.get(u, -1)\n"," if idx < 0:\n"," missing += 1\n"," canon_indices[i] = -1\n"," else:\n"," canon_indices[i] = idx\n","\n","if missing > 0:\n"," raise ValueError(\n"," f\"{missing} canonical rows have ufo_uuid not found in embeddings NPZ.\"\n"," )\n","\n","emb_canon = emb_all[canon_indices]\n","print(f\"Canonical embeddings shape: {emb_canon.shape}\")\n","\n","# ---------------------------------------------------------------------\n","# BUILD TEXT FILTER MASK FOR CLUSTERING\n","# ---------------------------------------------------------------------\n","print(\"\\nComputing clustering text mask...\")\n","\n","full_text_norm = pd.Series(texts).str.strip().str.lower()\n","is_non_informative = full_text_norm.isin(NULL_LIKE_TEXT)\n","is_informative = ~is_non_informative\n","\n","text_len = full_text_norm.str.len()\n","long_enough = text_len >= MIN_TEXT_LEN_FOR_CLUSTERING\n","\n","cluster_mask = (is_informative & long_enough).to_numpy()\n","n_clusterable = int(cluster_mask.sum())\n","\n","print(f\"Clusterable docs (informative & len >= {MIN_TEXT_LEN_FOR_CLUSTERING}): {n_clusterable}\")\n","print(f\"Non-clusterable docs: {n_docs - n_clusterable}\")\n","\n","# ---------------------------------------------------------------------\n","# PCA REDUCTION\n","# ---------------------------------------------------------------------\n","print(\"\\nRunning PCA reduction (1024 ->\", PCA_DIM, \")...\")\n","\n","pca = PCA(\n"," n_components=PCA_DIM,\n"," random_state=PCA_RANDOM_SEED,\n",")\n","X_pca = pca.fit_transform(emb_canon)\n","print(\"PCA embedding shape:\", X_pca.shape)\n","\n","# ---------------------------------------------------------------------\n","# HDBSCAN CLUSTERING\n","# ---------------------------------------------------------------------\n","print(\"\\nRunning HDBSCAN on clusterable subset...\")\n","\n","X_cluster = X_pca[cluster_mask]\n","\n","if n_clusterable == 0:\n"," print(\"WARNING: No docs passed clustering mask; all will be noise.\")\n"," labels_sub = np.full((0,), -1, dtype=int)\n","else:\n"," clusterer = hdbscan.HDBSCAN(\n"," min_cluster_size=HDBSCAN_MIN_CLUSTER_SIZE,\n"," min_samples=HDBSCAN_MIN_SAMPLES,\n"," metric=HDBSCAN_METRIC,\n"," cluster_selection_method=HDBSCAN_CLUSTER_SELECTION_METHOD,\n"," alpha=HDBSCAN_ALPHA,\n"," core_dist_n_jobs=-1, # use all cores\n"," )\n"," labels_sub = clusterer.fit_predict(X_cluster)\n","\n","# Map back to full doc index\n","hdbscan_labels = np.full(n_docs, -1, dtype=int)\n","clusterable_indices = np.where(cluster_mask)[0]\n","for i, doc_idx in enumerate(clusterable_indices):\n"," hdbscan_labels[doc_idx] = int(labels_sub[i])\n","\n","# Compute cluster sizes\n","cluster_size_counter = Counter(int(l) for l in hdbscan_labels if l >= 0)\n","hdbscan_cluster_size = np.array(\n"," [cluster_size_counter.get(int(l), 0) if l >= 0 else 0 for l in hdbscan_labels],\n"," dtype=int,\n",")\n","\n","num_clusters = len(cluster_size_counter)\n","num_noise = int((hdbscan_labels < 0).sum())\n","\n","print(\"\\nHDBSCAN clustering stats:\")\n","print(f\" HDBSCAN clusters (label >=0): {num_clusters}\")\n","print(f\" Noise / unclustered (label = -1): {num_noise}\")\n","\n","if num_clusters > 0:\n"," print(\" Max cluster size:\", max(cluster_size_counter.values()))\n"," print(\" Min cluster size:\", min(cluster_size_counter.values()))\n","else:\n"," print(\" No non-noise clusters found.\")\n","\n","# ---------------------------------------------------------------------\n","# BM25-STYLE CLUSTER KEYWORDS\n","# ---------------------------------------------------------------------\n","print(\"\\nComputing BM25-style cluster keywords...\")\n","\n","# Tokenize all texts\n","tokens_all = []\n","df_counter = Counter() # document frequency\n","for i, txt in enumerate(texts):\n"," toks = tokenize(txt)\n"," toks = [t for t in toks if t not in STOPWORDS]\n"," tokens_all.append(toks)\n","\n"," seen = set()\n"," for t in toks:\n"," if t not in seen:\n"," df_counter[t] += 1\n"," seen.add(t)\n","\n","N = n_docs\n","idf = {}\n","for t, df_t in df_counter.items():\n"," if df_t < BM25_MIN_DF:\n"," continue\n"," # BM25-ish idf (no length norm, just weighting)\n"," idf_val = np.log(1.0 + (N - df_t + 0.5) / (df_t + 0.5))\n"," idf[t] = float(idf_val)\n","\n","print(f\"Vocabulary size after df >= {BM25_MIN_DF}: {len(idf)}\")\n","\n","# Build mapping from cluster_id -> doc indices\n","cluster_to_docs = defaultdict(list)\n","for idx, cid in enumerate(hdbscan_labels):\n"," if cid >= 0:\n"," cluster_to_docs[cid].append(idx)\n","\n","cluster_top_terms = {} # cid -> list[str]\n","for cid, doc_ids in cluster_to_docs.items():\n"," tf_cluster = Counter()\n"," for idx in doc_ids:\n"," for t in tokens_all[idx]:\n"," if t in idf: # skip tokens filtered out of idf\n"," tf_cluster[t] += 1\n","\n"," if not tf_cluster:\n"," cluster_top_terms[cid] = []\n"," continue\n","\n"," # score = tf_cluster * idf\n"," scores = []\n"," for t, tf_val in tf_cluster.items():\n"," scores.append((t, tf_val * idf.get(t, 0.0)))\n","\n"," scores.sort(key=lambda x: x[1], reverse=True)\n"," top_terms = [t for (t, _) in scores[:BM25_TOP_TERMS_PER_CLUSTER]]\n"," cluster_top_terms[cid] = top_terms\n","\n","print(\"Finished computing cluster keywords.\")\n","\n","# ---------------------------------------------------------------------\n","# WRITE MERGED OUTPUT JSONL\n","# ---------------------------------------------------------------------\n","print(f\"\\nWriting merged clustered+BM25 JSONL to: {OUT_JSONL_PATH}\")\n","\n","header_out = {\n"," \"dataset\": header_meta.get(\"dataset\", \"UFO Witness Corpus v2\"),\n"," \"stage\": \"semantic_dedup_hdbscan_bm25_v1\",\n"," \"source_semantic_dedup_jsonl\": SEMANTIC_DEDUP_JSONL,\n"," \"source_embeddings_npz\": EMBEDDINGS_NPZ_PATH,\n"," \"embedding_model\": header_meta.get(\"embedding_model\", \"BAAI/bge-large-en-v1.5\"),\n"," \"embedding_dim\": int(header_meta.get(\"embedding_dim\", emb_dim)),\n"," \"semantic_dedup_params\": header_meta.get(\"semantic_dedup_params\", {}),\n"," \"pca_params\": {\n"," \"n_components\": PCA_DIM,\n"," \"random_state\": PCA_RANDOM_SEED,\n"," \"fit_on\": \"canonical_embeddings\",\n"," },\n"," \"hdbscan_params\": {\n"," \"min_cluster_size\": HDBSCAN_MIN_CLUSTER_SIZE,\n"," \"min_samples\": HDBSCAN_MIN_SAMPLES,\n"," \"metric\": HDBSCAN_METRIC,\n"," \"cluster_selection_method\": HDBSCAN_CLUSTER_SELECTION_METHOD,\n"," \"alpha\": HDBSCAN_ALPHA,\n"," \"core_dist_n_jobs\": -1,\n"," },\n"," \"clustering_text_filter\": {\n"," \"min_text_len_for_clustering\": MIN_TEXT_LEN_FOR_CLUSTERING,\n"," \"null_like\": sorted(list(NULL_LIKE_TEXT)),\n"," },\n"," \"bm25_params\": {\n"," \"top_terms_per_cluster\": BM25_TOP_TERMS_PER_CLUSTER,\n"," \"min_df\": BM25_MIN_DF,\n"," \"stopwords_size\": len(STOPWORDS),\n"," \"tokenizer\": \"regex [a-z0-9]+, lowercased\",\n"," },\n"," \"num_docs_canonical\": int(n_docs),\n"," \"num_docs_clusterable\": int(n_clusterable),\n"," \"num_hdbscan_clusters\": int(num_clusters),\n"," \"num_hdbscan_noise_docs\": int(num_noise),\n","}\n","\n","with open(OUT_JSONL_PATH, \"w\", encoding=\"utf-8\") as fout:\n"," fout.write(json.dumps(header_out, ensure_ascii=False) + \"\\n\")\n","\n"," for i in range(n_docs):\n"," row = df.iloc[i].to_dict()\n","\n"," # Attach HDBSCAN info\n"," cid = int(hdbscan_labels[i])\n"," row[\"hdbscan_cluster_id\"] = cid\n"," row[\"hdbscan_cluster_size\"] = int(hdbscan_cluster_size[i])\n","\n"," if cid >= 0:\n"," row[\"hdbscan_cluster_top_terms\"] = cluster_top_terms.get(cid, [])\n"," else:\n"," row[\"hdbscan_cluster_top_terms\"] = []\n","\n"," clean_rec = {k: to_jsonable(v) for k, v in row.items()}\n"," fout.write(json.dumps(clean_rec, ensure_ascii=False) + \"\\n\")\n","\n","print(\"=== DONE: HDBSCAN + BM25 clustering written. ===\")\n","print(f\"Output rows: {n_docs}\")\n","print(f\"HDBSCAN clusters: {num_clusters}, noise docs: {num_noise}\")\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"gmnjTZKl8BV6","executionInfo":{"status":"ok","timestamp":1764014368148,"user_tz":300,"elapsed":645366,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"61b68703-f18e-4bc8-c918-5bc10dfcf84c"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["=== STEP 5: HDBSCAN + BM25 cluster labeling ===\n","\n","Loading semantic-deduped canonical JSONL: /content/ufo_semantic_deduped_v1.jsonl\n","Canonical rows loaded: 242633\n","\n","Loading embeddings NPZ: /content/ufo_bge_large_embeddings_v1.npz\n","Embeddings shape (full): (242842, 1024)\n","Canonical embeddings shape: (242633, 1024)\n","\n","Computing clustering text mask...\n","Clusterable docs (informative & len >= 120): 35027\n","Non-clusterable docs: 207606\n","\n","Running PCA reduction (1024 -> 64 )...\n","PCA embedding shape: (242633, 64)\n","\n","Running HDBSCAN on clusterable subset...\n"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.12/dist-packages/sklearn/utils/deprecation.py:151: FutureWarning: 'force_all_finite' was renamed to 'ensure_all_finite' in 1.6 and will be removed in 1.8.\n"," warnings.warn(\n","/usr/local/lib/python3.12/dist-packages/sklearn/utils/deprecation.py:151: FutureWarning: 'force_all_finite' was renamed to 'ensure_all_finite' in 1.6 and will be removed in 1.8.\n"," warnings.warn(\n"]},{"output_type":"stream","name":"stdout","text":["\n","HDBSCAN clustering stats:\n"," HDBSCAN clusters (label >=0): 7\n"," Noise / unclustered (label = -1): 242246\n"," Max cluster size: 111\n"," Min cluster size: 19\n","\n","Computing BM25-style cluster keywords...\n","Vocabulary size after df >= 5: 9082\n","Finished computing cluster keywords.\n","\n","Writing merged clustered+BM25 JSONL to: /content/ufo_v2_clustered_bm25_v1.jsonl\n","=== DONE: HDBSCAN + BM25 clustering written. ===\n","Output rows: 242633\n","HDBSCAN clusters: 7, noise docs: 242246\n"]}]},{"cell_type":"code","source":["import json, pandas as pd\n","\n","MASTER_PATH = \"/content/ufo_v2_clustered_bm25_v1.jsonl\"\n","\n","with open(MASTER_PATH, \"r\", encoding=\"utf-8\") as f:\n"," header = json.loads(f.readline())\n","header\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"HthUXlDSiS4C","executionInfo":{"status":"ok","timestamp":1764015960428,"user_tz":300,"elapsed":14,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"a4a8e871-f578-42a2-9208-ff6294edd442"},"execution_count":18,"outputs":[{"output_type":"execute_result","data":{"text/plain":["{'dataset': 'UFO Witness Corpus v2',\n"," 'stage': 'semantic_dedup_hdbscan_bm25_v1',\n"," 'source_semantic_dedup_jsonl': '/content/ufo_semantic_deduped_v1.jsonl',\n"," 'source_embeddings_npz': '/content/ufo_bge_large_embeddings_v1.npz',\n"," 'embedding_model': 'BAAI/bge-large-en-v1.5',\n"," 'embedding_dim': 1024,\n"," 'semantic_dedup_params': {'k_neighbors': 10,\n"," 'sim_threshold': 0.92,\n"," 'min_text_len_for_clustering': 120,\n"," 'canonical_priority': ['has_semantic_anchor DESC',\n"," 'has_coordinates DESC',\n"," 'parsed_date_fill ASC',\n"," 'full_text_len DESC',\n"," 'ufo_uuid ASC',\n"," 'row_idx ASC'],\n"," 'text_informative_filter': {'null_like': ['',\n"," '<na>',\n"," 'n/a',\n"," 'na',\n"," 'nan',\n"," 'none',\n"," 'null']},\n"," 'anchor_filter': 'none (anchors affect canonical selection only, not clustering)'},\n"," 'pca_params': {'n_components': 64,\n"," 'random_state': 42,\n"," 'fit_on': 'canonical_embeddings'},\n"," 'hdbscan_params': {'min_cluster_size': 15,\n"," 'min_samples': 5,\n"," 'metric': 'euclidean',\n"," 'cluster_selection_method': 'leaf',\n"," 'alpha': 1.0,\n"," 'core_dist_n_jobs': -1},\n"," 'clustering_text_filter': {'min_text_len_for_clustering': 120,\n"," 'null_like': ['', '<na>', 'n/a', 'na', 'nan', 'none', 'null']},\n"," 'bm25_params': {'top_terms_per_cluster': 15,\n"," 'min_df': 5,\n"," 'stopwords_size': 41,\n"," 'tokenizer': 'regex [a-z0-9]+, lowercased'},\n"," 'num_docs_canonical': 242633,\n"," 'num_docs_clusterable': 35027,\n"," 'num_hdbscan_clusters': 7,\n"," 'num_hdbscan_noise_docs': 242246}"]},"metadata":{},"execution_count":18}]},{"cell_type":"code","source":[],"metadata":{"id":"ET8VR4GHiS1f"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["required_cols = [\n"," \"ufo_uuid\",\n"," \"full_text\", # or your main narrative column\n"," \"semantic_cluster_id\",\n"," \"semantic_cluster_size\",\n"," \"semantic_cluster_members\",\n"," \"hdbscan_cluster_id\",\n"," \"hdbscan_cluster_size\",\n"," \"hdbscan_cluster_top_terms\",\n","]\n","\n","missing = [c for c in required_cols if c not in df.columns]\n","missing\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"woIfEU7CihKD","executionInfo":{"status":"ok","timestamp":1764015903348,"user_tz":300,"elapsed":5,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"896356bb-b3ce-4f78-d008-f2dab3491170"},"execution_count":17,"outputs":[{"output_type":"execute_result","data":{"text/plain":["['hdbscan_cluster_id', 'hdbscan_cluster_size', 'hdbscan_cluster_top_terms']"]},"metadata":{},"execution_count":17}]},{"cell_type":"code","source":["import json\n","import pandas as pd\n","\n","PATH = \"/content/ufo_v2_clustered_bm25_v1.jsonl\"\n","\n","rows = []\n","with open(PATH, \"r\", encoding=\"utf-8\") as f:\n"," header = json.loads(f.readline()) # keep if you want metadata\n"," for line in f:\n"," line = line.strip()\n"," if not line:\n"," continue\n"," rows.append(json.loads(line))\n","\n","df = pd.DataFrame(rows)\n","print(\"Rows loaded:\", len(df))\n","print(df.columns.tolist())\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"M_RckOOriSwB","executionInfo":{"status":"ok","timestamp":1764018122300,"user_tz":300,"elapsed":107035,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"ffbff170-8995-4825-ffc5-1e65e0ec5eab"},"execution_count":4,"outputs":[{"output_type":"stream","name":"stdout","text":["Rows loaded: 242633\n","['city', 'state', 'date_time', 'shape', 'duration_bucket', 'text_hash', 'raw_snapshot', 'raw_hash', 'summary', 'full_text', 'duration_seconds', 'latitude', 'longitude', 'country', 'year', 'year_context_available', 'ufo_uuid', 'lineage', 'merge_version', 'dedupe_method', 'nchs_nchs_top_five_leading_causes_of_death_united_states_1990_1950_2000_number_of_deaths', 'nchs_nchs_age_adjusted_death_rates_for_selected_major_causes_of_death_age_adjusted_death_rate', 'nchs_nchs_death_rates_and_life_expectancy_at_birth_average_life_expectancy_years', 'nchs_nchs_death_rates_and_life_expectancy_at_birth_age_adjusted_death_rate', 'nchs_nchs_leading_causes_of_death_united_states_deaths', 'nchs_nchs_leading_causes_of_death_united_states_age_adjusted_death_rate', 'nchs_nchs_potentially_excess_deaths_from_the_five_leading_causes_of_death_hhs_region', 'nchs_nchs_potentially_excess_deaths_from_the_five_leading_causes_of_death_observed_deaths', 'nchs_nchs_potentially_excess_deaths_from_the_five_leading_causes_of_death_population', 'nchs_nchs_potentially_excess_deaths_from_the_five_leading_causes_of_death_expected_deaths', 'nchs_nchs_potentially_excess_deaths_from_the_five_leading_causes_of_death_potentially_excess_deaths', 'nchs_nchs_potentially_excess_deaths_from_the_five_leading_causes_of_death_percent_potentially_excess_deaths', 'news_articles_total', 'news_unique_sources', 'news_source_entropy', 'news_total_words', 'news_mean_words', 'news_median_words', 'news_vocab_size', 'news_vocab_entropy', 'news_lexical_diversity', 'news_sentiment_mean', 'news_sentiment_std', 'news_ufo', 'news_uap', 'news_alien', 'news_saucer', 'news_disk', 'news_balloon', 'news_missile', 'news_nuclear', 'news_articles_total_z', 'news_unique_sources_z', 'news_source_entropy_z', 'news_total_words_z', 'news_mean_words_z', 'news_vocab_size_z', 'news_vocab_entropy_z', 'news_sentiment_mean_z', 'famous_birth_count', 'famous_death_count', 'famous_mean_age_of_death', 'famous_std_age_of_death', 'famous_manner_entropy', 'famous_country_entropy', 'famous_occupation_entropy', 'famous_gender_entropy', 'famous_homicide_rate', 'famous_birth_count_z', 'famous_death_count_z', 'famous_mean_age_of_death_z', 'famous_homicide_rate_z', 'famous_manner_entropy_z', 'famous_country_entropy_z', 'famous_occupation_entropy_z', 'famous_gender_entropy_z', 'gdp_gdp_growth_mean', 'gdp_gdp_growth_std', 'gdp_gdp_growth_min', 'gdp_gdp_growth_max', 'gdp_gdp_growth_median', 'disaster_disaster_deaths_total', 'disaster_disaster_deaths_mean', 'disaster_disaster_deaths_std', 'disaster_disaster_deaths_min', 'disaster_disaster_deaths_max', 'disaster_disaster_deaths_count', 'disaster_disaster_deaths_pct_change', 'patent_yearly_contextmerged_patent_count', 'patent_yearly_contextmerged_cpc_total', 'patent_yearly_contextmerged_gov_interest', 'patent_yearly_contextmerged_cit_back', 'arxiv_total_papers', 'arxiv_category_entropy', 'arxiv_math_co', 'arxiv_math_nt', 'arxiv_hep_th', 'arxiv_math_pr', 'arxiv_math_ag', 'arxiv_nlin_ps', 'arxiv_physics_chem_ph', 'arxiv_q_bio_mn', 'arxiv_math_na', 'arxiv_math_ra', 'arxiv_q_bio_pe', 'arxiv_q_bio_cb', 'arxiv_quant_ph', 'arxiv_physics_optics', 'arxiv_physics_comp_ph', 'arxiv_cond_mat_stat_mech', 'arxiv_cond_mat_mtrl_sci', 'arxiv_nlin_si', 'arxiv_cs_ne', 'arxiv_cs_ai', 'arxiv_math_fa', 'arxiv_math_cv', 'arxiv_physics_soc_ph', 'arxiv_math_ph', 'arxiv_math_mp', 'arxiv_physics_gen_ph', 'arxiv_math_oa', 'arxiv_math_qa', 'arxiv_math_rt', 'arxiv_math_dg', 'arxiv_physics_data_an', 'arxiv_cs_ce', 'arxiv_cs_ms', 'arxiv_cs_na', 'arxiv_astro_ph', 'arxiv_math_ac', 'arxiv_cs_cc', 'arxiv_cond_mat_other', 'arxiv_math_kt', 'arxiv_math_gt', 'arxiv_math_ds', 'arxiv_math_ap', 'arxiv_cond_mat_mes_hall', 'arxiv_nucl_th', 'arxiv_gr_qc', 'arxiv_physics_class_ph', 'arxiv_hep_ph', 'arxiv_hep_ex', 'arxiv_physics_bio_ph', 'arxiv_q_bio_bm', 'arxiv_nlin_cg', 'arxiv_cs_dm', 'arxiv_cs_lo', 'arxiv_nucl_ex', 'arxiv_math_sg', 'arxiv_cs_it', 'arxiv_math_it', 'arxiv_math_st', 'arxiv_stat_th', 'arxiv_cond_mat_str_el', 'arxiv_q_bio_ot', 'arxiv_physics_flu_dyn', 'arxiv_cond_mat_soft', 'arxiv_cond_mat_dis_nn', 'arxiv_math_ca', 'arxiv_cond_mat_supr_con', 'arxiv_physics_geo_ph', 'arxiv_nlin_cd', 'arxiv_physics_atom_ph', 'arxiv_physics_ed_ph', 'arxiv_math_oc', 'arxiv_math_gr', 'arxiv_math_at', 'arxiv_math_ct', 'arxiv_math_lo', 'arxiv_cs_ni', 'arxiv_math_sp', 'arxiv_q_bio_qm', 'arxiv_q_bio_nc', 'arxiv_cs_pf', 'arxiv_stat_me', 'arxiv_stat_ap', 'arxiv_math_gm', 'arxiv_cs_ds', 'arxiv_cs_cr', 'arxiv_cs_se', 'arxiv_math_gn', 'arxiv_stat_co', 'arxiv_cs_ar', 'arxiv_cs_lg', 'arxiv_cs_sc', 'arxiv_math_mg', 'arxiv_q_bio_gn', 'arxiv_cs_cg', 'arxiv_cs_cv', 'arxiv_hep_lat', 'arxiv_physics_ins_det', 'arxiv_cs_oh', 'arxiv_cs_db', 'arxiv_cs_dl', 'arxiv_cs_hc', 'arxiv_cs_pl', 'arxiv_physics_space_ph', 'arxiv_physics_plasm_ph', 'arxiv_cs_ir', 'arxiv_cs_cy', 'arxiv_cs_gt', 'arxiv_cs_dc', 'arxiv_q_bio_sc', 'arxiv_cs_ma', 'arxiv_cs_cl', 'arxiv_physics_med_ph', 'arxiv_physics_atm_clus', 'arxiv_physics_hist_ph', 'arxiv_math_ho', 'arxiv_nlin_ao', 'arxiv_physics_ao_ph', 'arxiv_physics_acc_ph', 'arxiv_cs_mm', 'arxiv_cs_ro', 'arxiv_stat_ml', 'arxiv_q_bio_to', 'arxiv_physics_pop_ph', 'arxiv_cs_gl', 'arxiv_cs_os', 'arxiv_cs_gr', 'arxiv_cs_sd', 'arxiv_adap_org', 'arxiv_q_bio', 'arxiv_cond_mat', 'arxiv_alg_geom', 'arxiv_dg_ga', 'arxiv_q_alg', 'arxiv_chem_ph', 'arxiv_chao_dyn', 'arxiv_patt_sol', 'arxiv_cmp_lg', 'arxiv_comp_gas', 'arxiv_mtrl_th', 'arxiv_supr_con', 'arxiv_solv_int', 'arxiv_funct_an', 'arxiv_atom_ph', 'arxiv_acc_phys', 'arxiv_plasm_ph', 'arxiv_q_fin_pr', 'arxiv_q_fin_gn', 'arxiv_q_fin_st', 'arxiv_q_fin_cp', 'arxiv_q_fin_pm', 'arxiv_q_fin_rm', 'arxiv_q_fin_tr', 'arxiv_bayes_an', 'arxiv_ao_sci', 'arxiv_astro_ph_ep', 'arxiv_astro_ph_sr', 'arxiv_astro_ph_co', 'arxiv_cond_mat_quant_gas', 'arxiv_astro_ph_he', 'arxiv_cs_fl', 'arxiv_astro_ph_ga', 'arxiv_astro_ph_im', 'arxiv_cs_sy', 'arxiv_cs_si', 'arxiv_stat_ot', 'arxiv_cs_et', 'arxiv_q_fin_ec', 'arxiv_q_fin_mf', 'arxiv_econ_em', 'arxiv_eess_iv', 'arxiv_physics_app_ph', 'arxiv_eess_sp', 'arxiv_eess_as', 'arxiv_econ_th', 'arxiv_econ_gn', 'arxiv_eess_sy', 'space_launch_total', 'space_launch_mean', 'space_launch_std', 'space_countries_reporting', 'space_max_country_launch', 'space_min_country_launch', 'space_launch_total_z', 'space_launch_mean_z', 'space_countries_reporting_z', 'nuclear_yearly_context_number_of_nuclear_weapons_tests_year', 'nuclear_yearly_context_number_of_nuclear_weapons_tests_number_of_nuclear_weapons_tests', 'nuclear_yearly_context_nuclear_warhead_inventories_year', 'nuclear_yearly_context_nuclear_warhead_inventories_number_of_deployed_strategic_nuclear_warheads', 'nuclear_yearly_context_nuclear_warhead_inventories_number_of_deployed_nonstrategic_nuclear_warheads', 'nuclear_yearly_context_nuclear_warhead_inventories_number_of_nondeployed_nuclear_warheads_in_reserve', 'nuclear_yearly_context_nuclear_warhead_inventories_number_of_retired_nuclear_warheads', 'energy_country_reporting', 'energy_country_entropy', 'energy_biodiesel_cons_kboed_sum', 'energy_biodiesel_cons_kboed_mean', 'energy_biodiesel_cons_kboed_count', 'energy_biodiesel_cons_pj_sum', 'energy_biodiesel_cons_pj_mean', 'energy_biodiesel_cons_pj_count', 'energy_biofuels_cons_ej_sum', 'energy_biofuels_cons_ej_mean', 'energy_biofuels_cons_ej_count', 'energy_biofuels_cons_kbd_sum', 'energy_biofuels_cons_kbd_mean', 'energy_biofuels_cons_kbd_count', 'energy_biofuels_cons_kboed_sum', 'energy_biofuels_cons_kboed_mean', 'energy_biofuels_cons_kboed_count', 'energy_biofuels_cons_pj_sum', 'energy_biofuels_cons_pj_mean', 'energy_biofuels_cons_pj_count', 'energy_biogeo_ej_sum', 'energy_biogeo_ej_mean', 'energy_biogeo_ej_count', 'energy_biogeo_twh_sum', 'energy_biogeo_twh_mean', 'energy_biogeo_twh_count', 'energy_biogeo_twh_net_sum', 'energy_biogeo_twh_net_mean', 'energy_biogeo_twh_net_count', 'energy_co2_combust_mtco2_sum', 'energy_co2_combust_mtco2_mean', 'energy_co2_combust_mtco2_count', 'energy_co2_combust_pc_sum', 'energy_co2_combust_pc_mean', 'energy_co2_combust_pc_count', 'energy_co2_combust_per_ej_sum', 'energy_co2_combust_per_ej_mean', 'energy_co2_combust_per_ej_count', 'energy_coalcons_ej_sum', 'energy_coalcons_ej_mean', 'energy_coalcons_ej_count', 'energy_electbyfuel_hydro_sum', 'energy_electbyfuel_hydro_mean', 'energy_electbyfuel_hydro_count', 'energy_electbyfuel_nuclear_sum', 'energy_electbyfuel_nuclear_mean', 'energy_electbyfuel_nuclear_count', 'energy_electbyfuel_ren_power_sum', 'energy_electbyfuel_ren_power_mean', 'energy_electbyfuel_ren_power_count', 'energy_ethanol_cons_kboed_sum', 'energy_ethanol_cons_kboed_mean', 'energy_ethanol_cons_kboed_count', 'energy_ethanol_cons_pj_sum', 'energy_ethanol_cons_pj_mean', 'energy_ethanol_cons_pj_count', 'energy_fuel_oil_cons_kbd_sum', 'energy_fuel_oil_cons_kbd_mean', 'energy_fuel_oil_cons_kbd_count', 'energy_gascons_bcfd_sum', 'energy_gascons_bcfd_mean', 'energy_gascons_bcfd_count', 'energy_gascons_bcm_sum', 'energy_gascons_bcm_mean', 'energy_gascons_bcm_count', 'energy_gascons_ej_sum', 'energy_gascons_ej_mean', 'energy_gascons_ej_count', 'energy_hydro_ej_sum', 'energy_hydro_ej_mean', 'energy_hydro_ej_count', 'energy_hydro_twh_sum', 'energy_hydro_twh_mean', 'energy_hydro_twh_count', 'energy_hydro_twh_net_sum', 'energy_hydro_twh_net_mean', 'energy_hydro_twh_net_count', 'energy_light_dist_cons_kbd_sum', 'energy_light_dist_cons_kbd_mean', 'energy_light_dist_cons_kbd_count', 'energy_liqcons_kbd_sum', 'energy_liqcons_kbd_mean', 'energy_liqcons_kbd_count', 'energy_middle_dist_cons_kbd_sum', 'energy_middle_dist_cons_kbd_mean', 'energy_middle_dist_cons_kbd_count', 'energy_nuclear_ej_sum', 'energy_nuclear_ej_mean', 'energy_nuclear_ej_count', 'energy_nuclear_twh_sum', 'energy_nuclear_twh_mean', 'energy_nuclear_twh_count', 'energy_nuclear_twh_net_sum', 'energy_nuclear_twh_net_mean', 'energy_nuclear_twh_net_count', 'energy_oilcons_ej_sum', 'energy_oilcons_ej_mean', 'energy_oilcons_ej_count', 'energy_oilcons_kbd_sum', 'energy_oilcons_kbd_mean', 'energy_oilcons_kbd_count', 'energy_oilcons_mt_sum', 'energy_oilcons_mt_mean', 'energy_oilcons_mt_count', 'energy_oilprod_crudecond_kbd_sum', 'energy_oilprod_crudecond_kbd_mean', 'energy_oilprod_crudecond_kbd_count', 'energy_oilprod_kbd_sum', 'energy_oilprod_kbd_mean', 'energy_oilprod_kbd_count', 'energy_oilprod_mt_sum', 'energy_oilprod_mt_mean', 'energy_oilprod_mt_count', 'energy_oilprod_ngl_kbd_sum', 'energy_oilprod_ngl_kbd_mean', 'energy_oilprod_ngl_kbd_count', 'energy_other_oil_cons_kbd_sum', 'energy_other_oil_cons_kbd_mean', 'energy_other_oil_cons_kbd_count', 'energy_pop_sum', 'energy_pop_mean', 'energy_pop_count', 'energy_refcap_kbd_sum', 'energy_refcap_kbd_mean', 'energy_refcap_kbd_count', 'energy_ren_power_ej_sum', 'energy_ren_power_ej_mean', 'energy_ren_power_ej_count', 'energy_ren_power_twh_sum', 'energy_ren_power_twh_mean', 'energy_ren_power_twh_count', 'energy_ren_power_twh_net_sum', 'energy_ren_power_twh_net_mean', 'energy_ren_power_twh_net_count', 'energy_renewables_ej_sum', 'energy_renewables_ej_mean', 'energy_renewables_ej_count', 'energy_solar_ej_sum', 'energy_solar_ej_mean', 'energy_solar_ej_count', 'energy_solar_twh_sum', 'energy_solar_twh_mean', 'energy_solar_twh_count', 'energy_solar_twh_net_sum', 'energy_solar_twh_net_mean', 'energy_solar_twh_net_count', 'energy_tes_ej_sum', 'energy_tes_ej_mean', 'energy_tes_ej_count', 'energy_tes_gj_pc_sum', 'energy_tes_gj_pc_mean', 'energy_tes_gj_pc_count', 'energy_wind_ej_sum', 'energy_wind_ej_mean', 'energy_wind_ej_count', 'energy_wind_twh_sum', 'energy_wind_twh_mean', 'energy_wind_twh_count', 'energy_wind_twh_net_sum', 'energy_wind_twh_net_mean', 'energy_wind_twh_net_count', 'energy_gasprod_bcfd_sum', 'energy_gasprod_bcfd_mean', 'energy_gasprod_bcfd_count', 'energy_gasprod_bcm_sum', 'energy_gasprod_bcm_mean', 'energy_gasprod_bcm_count', 'energy_gasprod_ej_sum', 'energy_gasprod_ej_mean', 'energy_gasprod_ej_count', 'energy_gasflared_bcm_sum', 'energy_gasflared_bcm_mean', 'energy_gasflared_bcm_count', 'energy_gasflared_mtco2_sum', 'energy_gasflared_mtco2_mean', 'energy_gasflared_mtco2_count', 'energy_diesel_gasoil_cons_kbd_sum', 'energy_diesel_gasoil_cons_kbd_mean', 'energy_diesel_gasoil_cons_kbd_count', 'energy_gasoline_cons_kbd_sum', 'energy_gasoline_cons_kbd_mean', 'energy_gasoline_cons_kbd_count', 'energy_kerosene_cons_kbd_sum', 'energy_kerosene_cons_kbd_mean', 'energy_kerosene_cons_kbd_count', 'energy_lpg_cons_kbd_sum', 'energy_lpg_cons_kbd_mean', 'energy_lpg_cons_kbd_count', 'energy_naphtha_cons_kbd_sum', 'energy_naphtha_cons_kbd_mean', 'energy_naphtha_cons_kbd_count', 'energy_refcaputil_pct_sum', 'energy_refcaputil_pct_mean', 'energy_refcaputil_pct_count', 'energy_refthru_kbd_sum', 'energy_refthru_kbd_mean', 'energy_refthru_kbd_count', 'energy_coalprod_ej_sum', 'energy_coalprod_ej_mean', 'energy_coalprod_ej_count', 'energy_coalprod_mt_sum', 'energy_coalprod_mt_mean', 'energy_coalprod_mt_count', 'energy_elect_twh_sum', 'energy_elect_twh_mean', 'energy_elect_twh_count', 'energy_electbyfuel_coal_sum', 'energy_electbyfuel_coal_mean', 'energy_electbyfuel_coal_count', 'energy_electbyfuel_gas_sum', 'energy_electbyfuel_gas_mean', 'energy_electbyfuel_gas_count', 'energy_electbyfuel_oil_sum', 'energy_electbyfuel_oil_mean', 'energy_electbyfuel_oil_count', 'energy_electbyfuel_other_sum', 'energy_electbyfuel_other_mean', 'energy_electbyfuel_other_count', 'energy_electbyfuel_total_sum', 'energy_electbyfuel_total_mean', 'energy_electbyfuel_total_count', 'energy_biodiesel_prod_kboed_sum', 'energy_biodiesel_prod_kboed_mean', 'energy_biodiesel_prod_kboed_count', 'energy_biodiesel_prod_pj_sum', 'energy_biodiesel_prod_pj_mean', 'energy_biodiesel_prod_pj_count', 'energy_biofuels_prod_kbd_sum', 'energy_biofuels_prod_kbd_mean', 'energy_biofuels_prod_kbd_count', 'energy_biofuels_prod_kboed_sum', 'energy_biofuels_prod_kboed_mean', 'energy_biofuels_prod_kboed_count', 'energy_biofuels_prod_pj_sum', 'energy_biofuels_prod_pj_mean', 'energy_biofuels_prod_pj_count', 'energy_co2_mtco2_sum', 'energy_co2_mtco2_mean', 'energy_co2_mtco2_count', 'energy_ethanol_prod_kboed_sum', 'energy_ethanol_prod_kboed_mean', 'energy_ethanol_prod_kboed_count', 'energy_ethanol_prod_pj_sum', 'energy_ethanol_prod_pj_mean', 'energy_ethanol_prod_pj_count', 'energy_methane_process_mtco2_sum', 'energy_methane_process_mtco2_mean', 'energy_methane_process_mtco2_count', 'energy_cobalt_kt_sum', 'energy_cobalt_kt_mean', 'energy_cobalt_kt_count', 'energy_graphite_kt_sum', 'energy_graphite_kt_mean', 'energy_graphite_kt_count', 'energy_lithium_kt_sum', 'energy_lithium_kt_mean', 'energy_lithium_kt_count', 'energy_rareearths_kt_sum', 'energy_rareearths_kt_mean', 'energy_rareearths_kt_count', 'energy_cobaltres_kt_sum', 'energy_cobaltres_kt_mean', 'energy_cobaltres_kt_count', 'energy_graphiteres_kt_sum', 'energy_graphiteres_kt_mean', 'energy_graphiteres_kt_count', 'energy_lithiumres_kt_sum', 'energy_lithiumres_kt_mean', 'energy_lithiumres_kt_count', 'energy_rareearthsres_kt_sum', 'energy_rareearthsres_kt_mean', 'energy_rareearthsres_kt_count', 'global_temp_temp_anomaly_mean', 'global_temp_temp_anomaly_std', 'global_temp_temp_anomaly_min', 'global_temp_temp_anomaly_max', 'global_temp_temp_anomaly_count', 'global_temp_temp_anomaly_pct_change', 'market_spx_mean', 'market_spx_std', 'market_spx_min', 'market_spx_max', 'market_spx_pct_change', 'market_spx_vol_mean', 'market_dji_mean', 'market_dji_std', 'market_dji_min', 'market_dji_max', 'market_dji_pct_change', 'market_dji_vol_mean', 'market_dax_mean', 'market_dax_std', 'market_dax_min', 'market_dax_max', 'market_dax_pct_change', 'market_dax_vol_mean', 'gdelt_yearly_context_records', 'gdelt_yearly_context_events_total', 'gdelt_yearly_context_articles_total', 'gdelt_yearly_context_goldstein_mean', 'gdelt_yearly_context_goldstein_std', 'gdelt_yearly_context_quad_entropy', 'gdelt_yearly_context_cameo_entropy', 'gdelt_yearly_context_coop_total', 'gdelt_yearly_context_conflict_total', 'gdelt_yearly_context_conflict_ratio', 'gdelt_yearly_context_source_lat_mean', 'gdelt_yearly_context_source_long_mean', 'gdelt_yearly_context_target_lat_mean', 'gdelt_yearly_context_target_long_mean', 'gdelt_yearly_context_geo_spread_mean_km', 'gdelt_yearly_context_cameo_distribution', 'gdelt_yearly_context_events_total_zscore', 'gdelt_yearly_context_articles_total_zscore', 'gdelt_yearly_context_goldstein_mean_zscore', 'gdelt_yearly_context_geo_spread_mean_km_zscore', 'gdelt_yearly_context_conflict_ratio_zscore', 'ucdp_events_total', 'ucdp_unique_conflicts', 'ucdp_unique_dyads', 'ucdp_state_violence_events', 'ucdp_non_state_violence_events', 'ucdp_one_sided_violence_events', 'ucdp_deaths_a_sum', 'ucdp_deaths_b_sum', 'ucdp_deaths_civilians_sum', 'ucdp_deaths_unknown_sum', 'ucdp_fatalities_best_sum', 'ucdp_fatalities_high_sum', 'ucdp_fatalities_low_sum', 'ucdp_fatalities_best_avg', 'ucdp_fatalities_high_avg', 'ucdp_fatalities_low_avg', 'ucdp_countries_covered', 'ucdp_clear_events', 'ucdp_unclear_events', 'air_air_passengers_total', 'air_air_passengers_mean', 'air_air_passengers_std', 'air_air_passengers_min', 'air_air_passengers_max', 'air_air_passengers_count', 'air_air_passengers_pct_change', 'moon_altitude_deg', 'moon_azimuth_deg', 'moon_distance_km', 'moon_illumination', 'moon_phase_age_days', 'geomagnetic_latitude', 'nearest_airport_code', 'nearest_airport_km', 'nearest_airport_source', 'semantic_cluster_id', 'semantic_cluster_size', 'semantic_cluster_members', 'hdbscan_cluster_id', 'hdbscan_cluster_size', 'hdbscan_cluster_top_terms']\n"]}]},{"cell_type":"code","source":["# 1) How many docs per cluster (including noise)\n","df[\"hdbscan_cluster_id\"].value_counts().sort_index()\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":366},"id":"auOtsBqsqrBK","executionInfo":{"status":"ok","timestamp":1764018298887,"user_tz":300,"elapsed":54,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"4238d0fd-f212-4113-d9fa-fd6fc1df9e77"},"execution_count":5,"outputs":[{"output_type":"execute_result","data":{"text/plain":["hdbscan_cluster_id\n","-1 242246\n"," 0 25\n"," 1 19\n"," 2 52\n"," 3 111\n"," 4 85\n"," 5 26\n"," 6 69\n","Name: count, dtype: int64"],"text/html":["<div>\n","<style scoped>\n"," .dataframe tbody tr th:only-of-type {\n"," vertical-align: middle;\n"," }\n","\n"," .dataframe tbody tr th {\n"," vertical-align: top;\n"," }\n","\n"," .dataframe thead th {\n"," text-align: right;\n"," }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n"," <thead>\n"," <tr style=\"text-align: right;\">\n"," <th></th>\n"," <th>count</th>\n"," </tr>\n"," <tr>\n"," <th>hdbscan_cluster_id</th>\n"," <th></th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>-1</th>\n"," <td>242246</td>\n"," </tr>\n"," <tr>\n"," <th>0</th>\n"," <td>25</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>19</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>52</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>111</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>85</td>\n"," </tr>\n"," <tr>\n"," <th>5</th>\n"," <td>26</td>\n"," </tr>\n"," <tr>\n"," <th>6</th>\n"," <td>69</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div><br><label><b>dtype:</b> int64</label>"]},"metadata":{},"execution_count":5}]},{"cell_type":"code","source":["# 2) Fraction that actually landed in a cluster\n","total = len(df)\n","noise = (df[\"hdbscan_cluster_id\"] < 0).sum()\n","clustered = total - noise\n","total, clustered, noise, clustered / total\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"J0Xr5nsRqq6C","executionInfo":{"status":"ok","timestamp":1764018342999,"user_tz":300,"elapsed":15,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"eae28179-5c11-40d4-c255-6a77dab1eacd"},"execution_count":6,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(242633, np.int64(387), np.int64(242246), np.float64(0.0015950015043295841))"]},"metadata":{},"execution_count":6}]},{"cell_type":"code","source":["# 3) Inspect clusters + top terms + sample rows\n","text_col = \"full_text\" # or \"summary\" if that's what you used\n","\n","for cid in sorted(df[\"hdbscan_cluster_id\"].unique()):\n"," if cid < 0:\n"," continue\n"," sub = df[df[\"hdbscan_cluster_id\"] == cid]\n"," print(\"=\"*80)\n"," print(f\"Cluster {cid} | size={len(sub)}\")\n"," print(\"Top terms:\", sub[\"hdbscan_cluster_top_terms\"].iloc[0])\n"," print(\"Sample rows:\")\n"," for _, row in sub.sample(min(5, len(sub)), random_state=0)[\n"," [\"date_time\", \"city\", \"state\", \"shape\", text_col]\n"," ].iterrows():\n"," print(\"---\", row[\"date_time\"], \"|\", row[\"city\"], row[\"state\"], \"|\", row[\"shape\"])\n"," print(row[text_col][:300].replace(\"\\n\", \" \"))\n"," print()\n"],"metadata":{"id":"FVUAt0Qfr6Sg","executionInfo":{"status":"ok","timestamp":1764018355521,"user_tz":300,"elapsed":24,"user":{"displayName":"bewithyourbreath","userId":"03865340285477226404"}},"outputId":"36868150-19bc-48c3-f1eb-353d63e1a09b","colab":{"base_uri":"https://localhost:8080/"}},"execution_count":7,"outputs":[{"output_type":"stream","name":"stdout","text":["================================================================================\n","Cluster 0 | size=25\n","Top terms: ['2003', 'occurred', 'reported', 'entered', '10', '00', 'posted', 'am', 'report', 'ago', 'location', '2004', '23', 'happened', '2005']\n","Sample rows:\n","--- nan | Gulfport MS | triangle\n","NOTE: I received a letter from the witness on 07/16/00. I sent her a letter asking if she would like to file an offfical report with \n","\n","--- nan | Cape Coral FL | triangle\n","I never reported this, but since I found this website, thought I would in case anyone else may have reported something around the same\n","\n","--- nan | Seattle WA | circle\n","This event occurred many years ago when I was eight. I have never spoken publicly of it, but now, at the age of 45, I am interested in \n","\n","--- nan | Phoenix AZ | nan\n","Occurred : 5/19/2005 10:35 (Entered as : 05/19/1905 10:35) Reported: 5/20/2005 1:37:11 AM 01:37 Posted: 5/24/2005 Location: Phoenix (90\n","\n","--- nan | Montague NJ | circle\n","WELL,FIRST ID REALLY LIKE WHOMEVER IS GOING TO READ THIS TO KNOW THAT I AM REALLY HAVING A HARD TIME BELIEVING THIS MYSELF.I KEEP TRY\n","\n","================================================================================\n","Cluster 1 | size=19\n","Top terms: ['cigarette', 'my', 'outside', 'porch', 'went', 'smoke', 'smoking', 'out', 'stepped', 'before', 'back', 'pm', '2020', 'approximately', 'morning']\n","Sample rows:\n","--- nan | Lynnwood WA | other\n","I woke up at around 2:40- 3am one night cant remember the day. I came out to my balcony to smoke a cigarette I was just about to light\n","\n","--- nan | Colby KS | circle\n","My wife and I went outside at about 11:00pm to have a cigarette before going to bed. We were sitting on our front porch smoking when my\n","\n","--- nan | Litchfield ME | light\n","November 15th 2009, I stepped out of my mother’s small home in Litchfield Maine to smoke a cigarette. I usually stand on the porch its\n","\n","--- nan | San Francisco CA | other\n","At roughly 10:10pm, on Tuesday, October 17, 2006, I went outside to have a cigarette. The observations that follow took place while I\n","\n","--- nan | Olathe KS | triangle\n","My fiance and cousin were outside on our front porch last night (March 25th 2012) at around 11:30PM smoking a cigarette before going to\n","\n","================================================================================\n","Cluster 2 | size=52\n","Top terms: ['my', 'driving', 'home', 'way', 'girlfriend', 'pm', 'around', 'bound', 'heading', 'work', 'south', 'traveling', 'approximately', 'about', 'back']\n","Sample rows:\n","--- nan | Melcroft PA | light\n","Hello, Tonight around 9:20pm I was traveling on route 711 in westmoreland county, just north of Normalville in the region of melcroft.\n","\n","--- nan | Plumsteadville PA | diamond\n","\t My girlfriend, her sister, my little sister, and myself were coming back from running errands on Friday 1/23/09. We were coming back\n","\n","--- nan | Dyersburg TN | cigar\n","I was in a vehicle on an exit ramp on I-155w waiting at a stop sign to turn left onto lake rd.(hwy 78) in dyersburg, tn. I noticed a l\n","\n","--- nan | North Highlands CA | triangle\n","Was driving home from work heading east bound on I-80 right near the I-80 split. While driving I happened to look over to my left ( whi\n","\n","--- nan | Benicia CA | light\n","My brother and I were driving from martinez to where we were staying in benicia. As we were heading northwest on 780 about to exit I lo\n","\n","================================================================================\n","Cluster 3 | size=111\n","Top terms: ['my', 'sky', 'outside', 'star', 'noticed', 'looked', 'saw', 'thought', 'looking', 'up', 'light', 'out', 'cigarette', 'what', 'night']\n","Sample rows:\n","--- nan | Surrey BC | light\n","at around 7:20pm i steped ouside for a after dinner smoke. looked up to in the sky looking at the stars , and saw an extremly bright li\n","\n","--- nan | Birmingham (UK/England) nan | unknown\n","I was out in my garden around 23:00pm looking at the stars, it was a clear night and a very mild temperature for November. But then i n\n","\n","--- nan | Oxnard CA | light\n","At aprx 10:30pm I was sitting in my patio looking at the night sky facing east it was clear out and I notice that what I thought was tw\n","\n","--- nan | Annapolis MD | other\n","My husband went out At 5 am and noticed what he called bright brush strokes in the sky that were lit up. Then we both saw it from outsi\n","\n","--- nan | Coos Bay OR | other\n","I was outside smoking a cigarette on my front porch and in front right above the baseball field it appeared very bright, like a star bu\n","\n","================================================================================\n","Cluster 4 | size=85\n","Top terms: ['satellites', 'starlink', 'line', 'lights', 'pd', 'note', 'nuforc', 'straight', 'spaced', 'moving', 'string', 'cluster', 'same', 'sky', 'evenly']\n","Sample rows:\n","--- nan | Ashland VA | light\n","5 lights equally spaced trailing each other. Appearance looked like satellite crossing the sky. ((Starlink satellites?))\n","\n","--- nan | Chesapeake VA | formation\n","Straight line of lights no flashing. Moved together from west to east. ((NUFORC Note: Cluster of \"Starlink\" satellites. PD))\n","\n","--- nan | Phoenix AZ | light\n","Lights in the sky moved to a similar spot in thr sky frome all directions. ((NUFORC Note: \"Starlink\" satellites? PD))\n","\n","--- nan | John Day OR | unknown\n","Witnessed 25 lights appear near south horizon, brighten and then fade away near straight up in sky. ((\"Starlink\" satellites??)\n","\n","--- nan | Llanasa (UK/Wales) nan | circle\n","Moving 30-40 lights in a linear movement in sincrenisation as being as stars. ((NUFORC Note: \"Starlink\" satellites?? PD))))\n","\n","================================================================================\n","Cluster 5 | size=26\n","Top terms: ['ufo', 'report', 'reported', 'nuforc', 'ufos', 'sighting', 'site', 'pd', 'note', 'www', 'national', 'net', 'submit', 'wish', 'ask']\n","Sample rows:\n","--- nan | Bucharest (Romania) nan | disk\n","Hello, I wish to submit an ufo report to you. I send this to you now because i did not know about your site until now(thank you Coast t\n","\n","--- nan | Sycamore IL | disk\n","Not that well versed in use of computer. I could send a very detailed description in an e-mail. A ufo was reported that same night in B\n","\n","--- nan | Mount Vernon IL | nan\n","UFOs reported to police in IL, (60 mi NW of Mt. Venon, IN) ((NUFORC Note: Report by Francis Ridge, experienced UFO investigator. PD))\n","\n","--- nan | Phuket (Thailand) nan | nan\n","Greetings from Thailand, coconut Island. I have not seen UFO with my eyes but since last night very strange occurrences. ((anonymous))\n","\n","--- nan | Evendale OH | unknown\n","This UFO event occurred in August or Semptember of 1961. I wish someone could shed some light on this matter. A lot of time has passe\n","\n","================================================================================\n","Cluster 6 | size=69\n","Top terms: ['note', 'nuforc', 'pd', 'star', 'sky', 'light', 'sirius', 'venus', 'sighting', 'bright', 'suspect', 'possible', 'stationary', 'like', 'twinkling']\n","Sample rows:\n","--- nan | Tessaloniki (Greece) nan | circle\n","Light changing shape and color on top of Salonika, l can see it from my town in Asomata. ((NUFORC Note: Sighting of Sirius? PD))\n","\n","--- nan | Lake Geneva WI | light\n","Bright light in Lake Genva sky-not a star or plane. ((NUFORC Note: If the witness was looking west, the object was Venus. PD))\n","\n","--- nan | Rocklin CA | triangle\n","I first saw the object at about 10:30 last night. It remained in one place. ((NUFORC Note: Possibly a twinkling star?? PD))\n","\n","--- nan | White Sulphur Springs MT | disk\n","S sky bright lights, white, green and red flashing with no movement for 20 min. +. ((NUFORC Note: Possible sighting of Sirius? PD))\n","\n","--- nan | Otego NY | sphere\n","Exceptionally bright obj. noted in the W sky at approx 8PM, which dropped below horizon after 1 hour. ((NUFORC Note: Venus. PD))\n","\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"BvA7lYoHr6Pe"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"w5LzlQ4rr6Mr"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"llzJ1LGur6J8"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"R9jj7XlOr6HD"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"o8w3IltHr6Ea"},"execution_count":null,"outputs":[]}]} |