{"users":[{"id":3,"username":"rliaw","name":"Richard Liaw","avatar_template":"/user_avatar/discuss.ray.io/rliaw/{size}/24_2.png","primary_group_name":"Ray-Team","flair_name":"Ray-Team","flair_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/7/796fedd2b646b5e9ea776cb6287afced709a5676.png","flair_bg_color":"fff","flair_group_id":41,"admin":true,"moderator":true,"trust_level":2},{"id":7771,"username":"Saaketh","name":"Saaketh Surya Sodanapalli","avatar_template":"https://avatars.discourse-cdn.com/v4/letter/s/f04885/{size}.png","trust_level":0},{"id":7472,"username":"RunLLM","name":"RunLLM","avatar_template":"/user_avatar/discuss.ray.io/runllm/{size}/8249_2.png","primary_group_name":"Ray-Team","flair_name":"Ray-Team","flair_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/7/796fedd2b646b5e9ea776cb6287afced709a5676.png","flair_bg_color":"fff","flair_group_id":41,"trust_level":3},{"id":7761,"username":"wymli","name":"Wymli","avatar_template":"/user_avatar/discuss.ray.io/wymli/{size}/8517_2.png","trust_level":0},{"id":7638,"username":"markoov","name":"nlp.chen","avatar_template":"/user_avatar/discuss.ray.io/markoov/{size}/8417_2.png","trust_level":1},{"id":7749,"username":"WilliamOnVoyage","name":"Moliang","avatar_template":"/user_avatar/discuss.ray.io/williamonvoyage/{size}/8505_2.png","trust_level":0},{"id":2393,"username":"yzs","name":"","avatar_template":"/user_avatar/discuss.ray.io/yzs/{size}/3240_2.png","trust_level":0},{"id":1697,"username":"cade","name":"Cade Daniel","avatar_template":"/user_avatar/discuss.ray.io/cade/{size}/3837_2.png","primary_group_name":"Ray-Team","flair_name":"Ray-Team","flair_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/7/796fedd2b646b5e9ea776cb6287afced709a5676.png","flair_bg_color":"fff","flair_group_id":41,"trust_level":2},{"id":2651,"username":"HuangLianghong","name":"Huang Lianghong","avatar_template":"/user_avatar/discuss.ray.io/huanglianghong/{size}/3545_2.png","trust_level":1},{"id":1116,"username":"Chen_Shen","name":"Chen Shen","avatar_template":"/user_avatar/discuss.ray.io/chen_shen/{size}/1486_2.png","primary_group_name":"Ray-Team","flair_name":"Ray-Team","flair_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/7/796fedd2b646b5e9ea776cb6287afced709a5676.png","flair_bg_color":"fff","flair_group_id":41,"trust_level":2},{"id":7725,"username":"wweic","name":"Wei Chen","avatar_template":"/user_avatar/discuss.ray.io/wweic/{size}/8492_2.png","trust_level":0},{"id":999,"username":"heng2j","name":"Heng2j","avatar_template":"/user_avatar/discuss.ray.io/heng2j/{size}/1342_2.png","trust_level":1},{"id":3923,"username":"alikleit","name":"Ali Kleit","avatar_template":"/user_avatar/discuss.ray.io/alikleit/{size}/5206_2.png","trust_level":1},{"id":3285,"username":"robcaulk","name":"Robert Caulk","avatar_template":"/user_avatar/discuss.ray.io/robcaulk/{size}/4455_2.png","trust_level":1},{"id":2823,"username":"Zac","name":"Zac","avatar_template":"/user_avatar/discuss.ray.io/zac/{size}/3822_2.png","trust_level":1},{"id":7691,"username":"BitCalSaul","name":"Bit Cal Saul","avatar_template":"/user_avatar/discuss.ray.io/bitcalsaul/{size}/8468_2.png","trust_level":0},{"id":7646,"username":"psnilesh","name":"Nilesh PS","avatar_template":"/user_avatar/discuss.ray.io/psnilesh/{size}/8424_2.png","trust_level":1},{"id":1357,"username":"Vipul_Gupta","name":"Vipul Gupta .","avatar_template":"/user_avatar/discuss.ray.io/vipul_gupta/{size}/1806_2.png","trust_level":0},{"id":2419,"username":"justinvyu","name":"Justin Yu","avatar_template":"/user_avatar/discuss.ray.io/justinvyu/{size}/4309_2.png","primary_group_name":"Ray-Team","flair_name":"Ray-Team","flair_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/7/796fedd2b646b5e9ea776cb6287afced709a5676.png","flair_bg_color":"fff","flair_group_id":41,"trust_level":4},{"id":4069,"username":"Magic_Liu_Liu","name":"Magic Liu Liu","avatar_template":"/user_avatar/discuss.ray.io/magic_liu_liu/{size}/5358_2.png","trust_level":0},{"id":7600,"username":"0x6A4B","name":null,"avatar_template":"https://avatars.discourse-cdn.com/v4/letter/0/34f0e0/{size}.png","trust_level":0},{"id":6553,"username":"Tairui_Wang","name":"","avatar_template":"https://avatars.discourse-cdn.com/v4/letter/t/c68b51/{size}.png","trust_level":1},{"id":4800,"username":"hpguo","name":"Hongpeng Guo","avatar_template":"/user_avatar/discuss.ray.io/hpguo/{size}/7247_2.png","primary_group_name":"Ray-Team","flair_name":"Ray-Team","flair_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/7/796fedd2b646b5e9ea776cb6287afced709a5676.png","flair_bg_color":"fff","flair_group_id":41,"trust_level":4},{"id":7466,"username":"JinLiang","name":"KingBright","avatar_template":"/user_avatar/discuss.ray.io/jinliang/{size}/8239_2.png","trust_level":0},{"id":6812,"username":"christina","name":"Christina Zhu","avatar_template":"/user_avatar/discuss.ray.io/christina/{size}/7542_2.png","primary_group_name":"Ray-Team","flair_name":"staff","flair_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/7/796fedd2b646b5e9ea776cb6287afced709a5676.png","flair_bg_color":"fff","flair_group_id":3,"moderator":true,"trust_level":4},{"id":7225,"username":"ypp","name":"Yue Peng","avatar_template":"/user_avatar/discuss.ray.io/ypp/{size}/8227_2.png","trust_level":1},{"id":7196,"username":"mitul93","name":"Mitul","avatar_template":"https://avatars.discourse-cdn.com/v4/letter/m/91b2a8/{size}.png","trust_level":1},{"id":1082,"username":"matthewdeng","name":"Matthew Deng","avatar_template":"/user_avatar/discuss.ray.io/matthewdeng/{size}/1446_2.png","primary_group_name":"Ray-Team","flair_name":"Ray-Team","flair_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/7/796fedd2b646b5e9ea776cb6287afced709a5676.png","flair_bg_color":"fff","flair_group_id":41,"trust_level":3},{"id":7414,"username":"gauravsuman","name":"Gaurav","avatar_template":"https://avatars.discourse-cdn.com/v4/letter/g/8c91f0/{size}.png","trust_level":0},{"id":7346,"username":"richardrex.xu","name":"淇 徐","avatar_template":"/user_avatar/discuss.ray.io/richardrex.xu/{size}/8104_2.png","trust_level":0},{"id":3246,"username":"H-Park","name":"Hunter Park","avatar_template":"/user_avatar/discuss.ray.io/h-park/{size}/4399_2.png","trust_level":1},{"id":6822,"username":"iykim","name":"인영 김","avatar_template":"/user_avatar/discuss.ray.io/iykim/{size}/7552_2.png","trust_level":1},{"id":7215,"username":"OmarAlmusa","name":"Omar Almusa","avatar_template":"/user_avatar/discuss.ray.io/omaralmusa/{size}/7963_2.png","trust_level":1},{"id":3507,"username":"milad_heidari","name":"milad heidari","avatar_template":"/user_avatar/discuss.ray.io/milad_heidari/{size}/4743_2.png","trust_level":1},{"id":6981,"username":"JmPearl","name":"","avatar_template":"https://avatars.discourse-cdn.com/v4/letter/j/f9ae1b/{size}.png","trust_level":0},{"id":7084,"username":"ishaan-mehta","name":"Ishaan Mehta","avatar_template":"/user_avatar/discuss.ray.io/ishaan-mehta/{size}/7822_2.png","trust_level":1},{"id":6881,"username":"navmarri1","name":null,"avatar_template":"https://avatars.discourse-cdn.com/v4/letter/n/9de0a6/{size}.png","trust_level":0},{"id":7101,"username":"Logan_Kinch","name":"Logan Kinch","avatar_template":"/user_avatar/discuss.ray.io/logan_kinch/{size}/7125_2.png","trust_level":1}],"primary_groups":[{"id":41,"name":"Ray-Team"}],"flair_groups":[{"id":41,"name":"Ray-Team","flair_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/7/796fedd2b646b5e9ea776cb6287afced709a5676.png","flair_bg_color":"fff","flair_color":""},{"id":3,"name":"staff","flair_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/7/796fedd2b646b5e9ea776cb6287afced709a5676.png","flair_bg_color":"fff","flair_color":""}],"topic_list":{"can_create_topic":false,"more_topics_url":"/c/ray-train/17?page=1","per_page":30,"top_tags":[],"topics":[{"fancy_title":"About the Ray Train category","id":3383,"title":"About the Ray Train category","slug":"about-the-ray-train-category","posts_count":1,"reply_count":0,"highest_post_number":1,"image_url":null,"created_at":"2021-08-29T04:56:19.367Z","last_posted_at":"2021-08-29T04:56:19.375Z","bumped":true,"bumped_at":"2025-03-05T19:49:49.677Z","archetype":"regular","unseen":false,"pinned":true,"unpinned":null,"excerpt":"For questions on distributed model training. Use Train to scale a model’s training process across CPUs, GPUs, or multiple nodes. \nRay Train is a scalable machine learning library for distributed training and fine-tuning. …","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":839,"like_count":0,"has_summary":false,"last_poster_username":"rliaw","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest single","description":"Original Poster, Most Recent Poster","user_id":3,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"Dynamic Resource Allocation","id":23571,"title":"Dynamic Resource Allocation","slug":"dynamic-resource-allocation","posts_count":2,"reply_count":1,"highest_post_number":2,"image_url":null,"created_at":"2026-06-25T17:39:36.192Z","last_posted_at":"2026-06-25T17:40:05.441Z","bumped":true,"bumped_at":"2026-06-25T17:40:05.441Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":338,"like_count":1,"has_summary":false,"last_poster_username":"RunLLM","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7771,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7472,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"Only 10 ray data actor is working with ray train and the rest is just idle","id":23568,"title":"Only 10 ray data actor is working with ray train and the rest is just idle","slug":"only-10-ray-data-actor-is-working-with-ray-train-and-the-rest-is-just-idle","posts_count":19,"reply_count":18,"highest_post_number":20,"image_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/optimized/2X/9/9d895361c53a9c5daa3e02cb761094f787cc85ae_2_654x1024.png","created_at":"2026-06-12T09:52:24.500Z","last_posted_at":"2026-06-12T17:07:36.737Z","bumped":true,"bumped_at":"2026-06-12T17:07:36.737Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":527,"like_count":0,"has_summary":false,"last_poster_username":"RunLLM","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7761,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7472,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"Error with “column_names” when using Ray with TRL’s sft_trainer","id":23404,"title":"Error with \"column_names\" when using Ray with TRL's sft_trainer","slug":"error-with-column-names-when-using-ray-with-trls-sft-trainer","posts_count":7,"reply_count":3,"highest_post_number":7,"image_url":null,"created_at":"2025-12-22T08:09:06.288Z","last_posted_at":"2026-06-02T17:59:39.688Z","bumped":true,"bumped_at":"2026-06-02T17:59:39.688Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":545,"like_count":0,"has_summary":false,"last_poster_username":"WilliamOnVoyage","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7638,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster","user_id":7472,"primary_group_id":41,"flair_group_id":41},{"extras":"latest","description":"Most Recent Poster","user_id":7749,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"Are there any hacks to use nsys in Ray?","id":8097,"title":"Are there any hacks to use nsys in Ray?","slug":"are-there-any-hacks-to-use-nsys-in-ray","posts_count":13,"reply_count":6,"highest_post_number":13,"image_url":null,"created_at":"2022-11-01T13:54:33.534Z","last_posted_at":"2026-04-13T16:04:12.285Z","bumped":true,"bumped_at":"2026-04-13T16:04:12.285Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":2564,"like_count":1,"has_summary":false,"last_poster_username":"wweic","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":true,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":2393,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster, Accepted Answer","user_id":1697,"primary_group_id":41,"flair_group_id":41},{"extras":null,"description":"Frequent Poster","user_id":2651,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster","user_id":1116,"primary_group_id":41,"flair_group_id":41},{"extras":"latest","description":"Most Recent Poster","user_id":7725,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"VScode breakpoint will be bypassed even with local_mode=True","id":8493,"title":"VScode breakpoint will be bypassed even with local_mode=True","slug":"vscode-breakpoint-will-be-bypassed-even-with-local-mode-true","posts_count":8,"reply_count":4,"highest_post_number":8,"image_url":null,"created_at":"2022-11-30T22:43:46.244Z","last_posted_at":"2026-02-13T17:47:02.087Z","bumped":true,"bumped_at":"2026-02-13T17:47:02.087Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":2084,"like_count":1,"has_summary":false,"last_poster_username":"BitCalSaul","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":999,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":3923,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":3285,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster","user_id":2823,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7691,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"How to enable debug logs from Ray train’s internal checkpoint manager?","id":23467,"title":"How to enable debug logs from Ray train's internal checkpoint manager?","slug":"how-to-enable-debug-logs-from-ray-trains-internal-checkpoint-manager","posts_count":4,"reply_count":2,"highest_post_number":4,"image_url":null,"created_at":"2026-02-05T06:11:51.233Z","last_posted_at":"2026-02-05T06:16:56.354Z","bumped":true,"bumped_at":"2026-02-05T06:16:56.354Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":79,"like_count":0,"has_summary":false,"last_poster_username":"RunLLM","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7646,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7472,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"Model Parallelism in Ray","id":4617,"title":"Model Parallelism in Ray","slug":"model-parallelism-in-ray","posts_count":11,"reply_count":6,"highest_post_number":11,"image_url":null,"created_at":"2022-01-07T18:57:24.987Z","last_posted_at":"2026-01-09T04:07:10.445Z","bumped":true,"bumped_at":"2026-01-09T04:07:10.445Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":3407,"like_count":2,"has_summary":false,"last_poster_username":"psnilesh","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":1357,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":3,"primary_group_id":41,"flair_group_id":41},{"extras":null,"description":"Frequent Poster","user_id":2419,"primary_group_id":41,"flair_group_id":41},{"extras":null,"description":"Recent Poster","user_id":4069,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7646,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"How to integrate Megatron-Core with Ray Train v2 for large language model training?","id":23443,"title":"How to integrate Megatron-Core with Ray Train v2 for large language model training?","slug":"how-to-integrate-megatron-core-with-ray-train-v2-for-large-language-model-training","posts_count":4,"reply_count":2,"highest_post_number":4,"image_url":null,"created_at":"2026-01-09T03:52:31.865Z","last_posted_at":"2026-01-09T03:56:57.414Z","bumped":true,"bumped_at":"2026-01-09T03:56:57.414Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":254,"like_count":0,"has_summary":false,"last_poster_username":"RunLLM","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7646,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7472,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"What are the advantages of calling `ray.train.report` with upload_mode as `NO_UPLOAD`?","id":23429,"title":"What are the advantages of calling `ray.train.report` with upload_mode as `NO_UPLOAD`?","slug":"what-are-the-advantages-of-calling-ray-train-report-with-upload-mode-as-no-upload","posts_count":6,"reply_count":6,"highest_post_number":7,"image_url":null,"created_at":"2026-01-01T19:36:40.237Z","last_posted_at":"2026-01-01T19:46:08.364Z","bumped":true,"bumped_at":"2026-01-01T19:46:08.364Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":74,"like_count":0,"has_summary":false,"last_poster_username":"RunLLM","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7646,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7472,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"How to quickly check how many times a TorchTrainer job has restarted?","id":23424,"title":"How to quickly check how many times a TorchTrainer job has restarted?","slug":"how-to-quickly-check-how-many-times-a-torchtrainer-job-has-restarted","posts_count":4,"reply_count":2,"highest_post_number":4,"image_url":null,"created_at":"2025-12-29T21:24:50.985Z","last_posted_at":"2025-12-29T22:26:40.097Z","bumped":true,"bumped_at":"2025-12-29T22:26:40.097Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":80,"like_count":0,"has_summary":false,"last_poster_username":"RunLLM","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7646,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7472,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"High GPU Memory (DeepSpeed+HuggingFace+Ray)","id":23408,"title":"High GPU Memory (DeepSpeed+HuggingFace+Ray)","slug":"high-gpu-memory-deepspeed-huggingface-ray","posts_count":3,"reply_count":2,"highest_post_number":3,"image_url":null,"created_at":"2025-12-23T10:08:07.172Z","last_posted_at":"2025-12-23T10:19:06.242Z","bumped":true,"bumped_at":"2025-12-23T10:19:06.242Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":60,"like_count":0,"has_summary":false,"last_poster_username":"markoov","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":7638,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster","user_id":7472,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"Training issues with MultiworkerMirroredStrategy","id":23335,"title":"Training issues with MultiworkerMirroredStrategy","slug":"training-issues-with-multiworkermirroredstrategy","posts_count":6,"reply_count":4,"highest_post_number":6,"image_url":null,"created_at":"2025-11-25T16:18:40.404Z","last_posted_at":"2025-11-25T18:00:46.016Z","bumped":true,"bumped_at":"2025-11-25T18:00:46.016Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":118,"like_count":0,"has_summary":false,"last_poster_username":"RunLLM","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7600,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7472,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"Long initialization time to initialize_session with large scale dataset","id":21545,"title":"Long initialization time to initialize_session with large scale dataset","slug":"long-initialization-time-to-initialize-session-with-large-scale-dataset","posts_count":6,"reply_count":2,"highest_post_number":6,"image_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/optimized/2X/d/d8608d5837fd446771b2ea45b65e750c4e877834_2_1024x443.png","created_at":"2025-01-21T23:07:17.901Z","last_posted_at":"2025-10-07T19:02:14.691Z","bumped":true,"bumped_at":"2025-10-07T19:02:14.691Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":168,"like_count":0,"has_summary":false,"last_poster_username":"RunLLM","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":true,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster, Accepted Answer","user_id":6553,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":4800,"primary_group_id":41,"flair_group_id":41},{"extras":null,"description":"Recent Poster","user_id":2419,"primary_group_id":41,"flair_group_id":41},{"extras":"latest","description":"Most Recent Poster","user_id":7472,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"Help needed with a simple demo","id":23109,"title":"Help needed with a simple demo","slug":"help-needed-with-a-simple-demo","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2025-09-16T10:41:06.060Z","last_posted_at":"2025-09-22T23:50:35.082Z","bumped":true,"bumped_at":"2025-09-22T23:50:35.082Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":62,"like_count":0,"has_summary":false,"last_poster_username":"christina","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7466,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":6812,"primary_group_id":41,"flair_group_id":3}]},{"fancy_title":"How to get dataset shard size in each train worker","id":23094,"title":"How to get dataset shard size in each train worker","slug":"how-to-get-dataset-shard-size-in-each-train-worker","posts_count":1,"reply_count":0,"highest_post_number":1,"image_url":null,"created_at":"2025-09-11T08:16:41.501Z","last_posted_at":"2025-09-11T08:16:41.572Z","bumped":true,"bumped_at":"2025-09-11T08:16:41.572Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":46,"like_count":0,"has_summary":false,"last_poster_username":"ypp","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest single","description":"Original Poster, Most Recent Poster","user_id":7225,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"What is the correct way of using get_dataset_shard?","id":23093,"title":"What is the correct way of using get_dataset_shard?","slug":"what-is-the-correct-way-of-using-get-dataset-shard","posts_count":1,"reply_count":0,"highest_post_number":1,"image_url":null,"created_at":"2025-09-11T07:59:42.786Z","last_posted_at":"2025-09-11T07:59:42.848Z","bumped":true,"bumped_at":"2025-09-11T07:59:45.991Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":49,"like_count":0,"has_summary":false,"last_poster_username":"mitul93","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest single","description":"Original Poster, Most Recent Poster","user_id":7196,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"Training loop stuck at StreamSplitDataIterator","id":23069,"title":"Training loop stuck at StreamSplitDataIterator","slug":"training-loop-stuck-at-streamsplitdataiterator","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2025-09-04T14:50:48.750Z","last_posted_at":"2025-09-05T13:52:53.518Z","bumped":true,"bumped_at":"2025-09-05T13:52:53.518Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":111,"like_count":0,"has_summary":false,"last_poster_username":"mitul93","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":true,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest single","description":"Original Poster, Most Recent Poster, Accepted Answer","user_id":7196,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"_collate_fn argument removed from ray.data.DataIterator.iter_batches","id":23052,"title":"_collate_fn argument removed from ray.data.DataIterator.iter_batches","slug":"collate-fn-argument-removed-from-ray-data-dataiterator-iter-batches","posts_count":3,"reply_count":0,"highest_post_number":3,"image_url":null,"created_at":"2025-08-27T14:47:07.361Z","last_posted_at":"2025-08-28T08:52:28.684Z","bumped":true,"bumped_at":"2025-08-28T08:52:28.684Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":120,"like_count":0,"has_summary":false,"last_poster_username":"mitul93","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":7196,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster","user_id":1082,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"Ray Tensorflow/Pytorch trainer metrics","id":23009,"title":"Ray Tensorflow/Pytorch trainer metrics","slug":"ray-tensorflow-pytorch-trainer-metrics","posts_count":1,"reply_count":0,"highest_post_number":1,"image_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/optimized/2X/f/f35e8248b6e8e7f9392d2f9caf5fb7296e251933_2_1024x390.png","created_at":"2025-08-18T08:58:49.730Z","last_posted_at":"2025-08-18T08:58:49.812Z","bumped":true,"bumped_at":"2025-08-18T08:58:49.812Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":38,"like_count":0,"has_summary":false,"last_poster_username":"gauravsuman","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest single","description":"Original Poster, Most Recent Poster","user_id":7414,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"Save log for train v2 in specify dir","id":22858,"title":"Save log for train v2 in specify dir","slug":"save-log-for-train-v2-in-specify-dir","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2025-07-19T08:22:50.718Z","last_posted_at":"2025-07-21T17:16:06.478Z","bumped":true,"bumped_at":"2025-07-21T17:16:06.478Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":45,"like_count":0,"has_summary":false,"last_poster_username":"christina","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7346,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":6812,"primary_group_id":41,"flair_group_id":3}]},{"fancy_title":"TorchDiagGaussian from logits","id":22568,"title":"TorchDiagGaussian from logits","slug":"torchdiaggaussian-from-logits","posts_count":5,"reply_count":1,"highest_post_number":5,"image_url":null,"created_at":"2025-05-28T09:50:14.412Z","last_posted_at":"2025-06-05T18:41:50.639Z","bumped":true,"bumped_at":"2025-06-05T18:41:50.639Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":151,"like_count":0,"has_summary":false,"last_poster_username":"christina","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":true,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":3246,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster, Accepted Answer","user_id":6822,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":6812,"primary_group_id":41,"flair_group_id":3}]},{"fancy_title":"[Ray Train] XGBoostTrainer crashes with ActorDiedError when using num_workers > 1 and use_gpu=False","id":22562,"title":"[Ray Train] XGBoostTrainer crashes with ActorDiedError when using num_workers > 1 and use_gpu=False","slug":"ray-train-xgboosttrainer-crashes-with-actordiederror-when-using-num-workers-1-and-use-gpu-false","posts_count":1,"reply_count":0,"highest_post_number":1,"image_url":null,"created_at":"2025-05-26T20:38:05.328Z","last_posted_at":"2025-05-26T20:38:05.410Z","bumped":true,"bumped_at":"2025-05-26T20:38:05.410Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":55,"like_count":0,"has_summary":false,"last_poster_username":"OmarAlmusa","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest single","description":"Original Poster, Most Recent Poster","user_id":7215,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"How to report loss when using more than one worker?","id":22513,"title":"How to report loss when using more than one worker?","slug":"how-to-report-loss-when-using-more-than-one-worker","posts_count":3,"reply_count":1,"highest_post_number":3,"image_url":null,"created_at":"2025-05-19T12:29:28.534Z","last_posted_at":"2025-05-20T06:17:09.156Z","bumped":true,"bumped_at":"2025-05-20T06:17:09.156Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":79,"like_count":0,"has_summary":false,"last_poster_username":"mitul93","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":7196,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster","user_id":6812,"primary_group_id":41,"flair_group_id":3}]},{"fancy_title":"Ray train job gets killed with no errors!","id":11831,"title":"Ray train job gets killed with no errors!","slug":"ray-train-job-gets-killed-with-no-errors","posts_count":4,"reply_count":1,"highest_post_number":4,"image_url":null,"created_at":"2023-08-17T11:20:20.755Z","last_posted_at":"2025-05-19T05:53:36.427Z","bumped":true,"bumped_at":"2025-05-19T05:53:36.427Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":543,"like_count":0,"has_summary":false,"last_poster_username":"OmarAlmusa","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":true,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster, Accepted Answer","user_id":3507,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster","user_id":1082,"primary_group_id":41,"flair_group_id":41},{"extras":"latest","description":"Most Recent Poster","user_id":7215,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"XGBoostTrainer crashes with ActorDiedError when using num_workers > 1 and use_gpu=False","id":22507,"title":"XGBoostTrainer crashes with ActorDiedError when using num_workers > 1 and use_gpu=False","slug":"xgboosttrainer-crashes-with-actordiederror-when-using-num-workers-1-and-use-gpu-false","posts_count":1,"reply_count":0,"highest_post_number":1,"image_url":null,"created_at":"2025-05-18T19:10:43.665Z","last_posted_at":"2025-05-18T19:10:43.725Z","bumped":true,"bumped_at":"2025-05-18T19:10:43.725Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":41,"like_count":0,"has_summary":false,"last_poster_username":"OmarAlmusa","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest single","description":"Original Poster, Most Recent Poster","user_id":7215,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"WorkerCrashedError: The worker died unexpectedly while executing this task. Check python-core-worker-*.log files for more information","id":22506,"title":"WorkerCrashedError: The worker died unexpectedly while executing this task. Check python-core-worker-*.log files for more information","slug":"workercrashederror-the-worker-died-unexpectedly-while-executing-this-task-check-python-core-worker-log-files-for-more-information","posts_count":1,"reply_count":0,"highest_post_number":1,"image_url":null,"created_at":"2025-05-18T18:38:46.178Z","last_posted_at":"2025-05-18T18:38:46.240Z","bumped":true,"bumped_at":"2025-05-18T18:38:46.240Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":82,"like_count":0,"has_summary":false,"last_poster_username":"OmarAlmusa","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest single","description":"Original Poster, Most Recent Poster","user_id":7215,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"OSError when saving checkpoint with ray.train.lightning.RayTrainReportCallback","id":21905,"title":"OSError when saving checkpoint with ray.train.lightning.RayTrainReportCallback","slug":"oserror-when-saving-checkpoint-with-ray-train-lightning-raytrainreportcallback","posts_count":7,"reply_count":3,"highest_post_number":7,"image_url":null,"created_at":"2025-02-24T19:33:14.106Z","last_posted_at":"2025-05-07T22:06:49.346Z","bumped":true,"bumped_at":"2025-05-08T16:01:16.035Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":137,"like_count":0,"has_summary":false,"last_poster_username":"ishaan-mehta","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":null,"description":"Original Poster","user_id":6981,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster","user_id":6812,"primary_group_id":41,"flair_group_id":3},{"extras":"latest","description":"Most Recent Poster","user_id":7084,"primary_group_id":null,"flair_group_id":null}]},{"fancy_title":"Init device mesh in pytorch distributed","id":22371,"title":"Init device mesh in pytorch distributed","slug":"init-device-mesh-in-pytorch-distributed","posts_count":3,"reply_count":1,"highest_post_number":3,"image_url":null,"created_at":"2025-04-24T23:28:03.402Z","last_posted_at":"2025-04-26T18:08:48.960Z","bumped":true,"bumped_at":"2025-04-26T18:08:48.960Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":245,"like_count":0,"has_summary":false,"last_poster_username":"navmarri1","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":6881,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster","user_id":3,"primary_group_id":41,"flair_group_id":41}]},{"fancy_title":"Ray Train V2 with Ray Tune does not start another trial after a training run is TERMINATED","id":22330,"title":"Ray Train V2 with Ray Tune does not start another trial after a training run is TERMINATED","slug":"ray-train-v2-with-ray-tune-does-not-start-another-trial-after-a-training-run-is-terminated","posts_count":4,"reply_count":1,"highest_post_number":4,"image_url":"https://us1.discourse-cdn.com/flex020/uploads/ray/optimized/2X/b/bcf111c9f9979db7b4375085af4af8ecddfa59c8_2_1024x203.png","created_at":"2025-04-17T12:54:03.075Z","last_posted_at":"2025-04-17T19:51:41.373Z","bumped":true,"bumped_at":"2025-04-17T19:51:41.373Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":97,"like_count":0,"has_summary":false,"last_poster_username":"Logan_Kinch","category_id":17,"op_like_count":0,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"can_have_answer":true,"can_vote":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":7101,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Recent Poster","user_id":1082,"primary_group_id":41,"flair_group_id":41}]}]}}