Anurag1734/cuda-error-resolution-analysis
07
1[2 {3 "post_stream": {4 "posts": [5 {6 "id": 460749,7 "name": "",8 "username": "weitaoliu",9 "avatar_template": "/letter_avatar_proxy/v4/letter/w/35a633/{size}.png",10 "created_at": "2024-12-04T10:23:51.660Z",11 "cooked": "<p>Hello all,</p>\n<p>I have an error when loading the network on GPU. I link the libTorch 2.5.1 with CUDA 12.4 to my codes. It works fine on my local machine (with GTX 1070) and a testing machine (with RTX 4070 Ti). However, when I move the codes to a computing node with A100, the solver does not work and throws an error when loading the network. The error looks like the following:</p>\n<pre><code class=\"lang-auto\">terminate called after throwing an instance of 'c10::Error'\n what(): _ivalue_ INTERNAL ASSERT FAILED at \"XXPath_To_Codes/ThirdParty/libtorchCUDA/include/torch/csrc/jit/api/object.h\":38, please report a bug to PyTorch. \nException raised from _ivalue at XXPath_to_Codes/ThirdParty/libtorchCUDA/include/torch/csrc/jit/api/object.h:38 (most recent call first):\n</code></pre>\n<p>I set up the same environment (cuda driver) on the computing node and have no idea how to address this issue. Do you have any suggestions?</p>\n<p>If I choose to load on the CPU, then there is no problem.</p>\n<p>Note that the A100 is split into 7 instances. I don’t know if that could be the issue.</p>\n<p>Thanks for your time.</p>\n<p>Best, Weitao.</p>",12 "post_number": 1,13 "post_type": 1,14 "posts_count": 1,15 "updated_at": "2024-12-04T10:23:51.660Z",16 "reply_count": 0,17 "reply_to_post_number": null,18 "quote_count": 0,19 "incoming_link_count": 59,20 "reads": 7,21 "readers_count": 6,22 "score": 296.4,23 "yours": false,24 "topic_id": 213794,25 "topic_slug": "error-when-load-network-on-gpu-libtorch-2-5-1",26 "display_username": "",27 "primary_group_name": null,28 "flair_name": null,29 "flair_url": null,30 "flair_bg_color": null,31 "flair_color": null,32 "flair_group_id": null,33 "badges_granted": [],34 "version": 1,35 "can_edit": false,36 "can_delete": false,37 "can_recover": false,38 "can_see_hidden_post": false,39 "can_wiki": false,40 "read": true,41 "user_title": null,42 "bookmarked": false,43 "actions_summary": [],44 "moderator": false,45 "admin": false,46 "staff": false,47 "user_id": 65591,48 "hidden": false,49 "trust_level": 1,50 "deleted_at": null,51 "user_deleted": false,52 "edit_reason": null,53 "can_view_edit_history": true,54 "wiki": false,55 "post_url": "/t/error-when-load-network-on-gpu-libtorch-2-5-1/213794/1",56 "can_accept_answer": false,57 "can_unaccept_answer": false,58 "accepted_answer": false,59 "topic_accepted_answer": null,60 "can_vote": false61 }62 ],63 "stream": [64 46074965 ]66 },67 "timeline_lookup": [68 [69 1,70 32571 ]72 ],73 "suggested_topics": [74 {75 "fancy_title": "Rewriting this without using libtorch",76 "id": 219857,77 "title": "Rewriting this without using libtorch",78 "slug": "rewriting-this-without-using-libtorch",79 "posts_count": 1,80 "reply_count": 0,81 "highest_post_number": 1,82 "image_url": null,83 "created_at": "2025-05-08T04:15:44.041Z",84 "last_posted_at": "2025-05-08T04:15:44.087Z",85 "bumped": true,86 "bumped_at": "2025-05-08T09:26:02.477Z",87 "archetype": "regular",88 "unseen": false,89 "pinned": false,90 "unpinned": null,91 "visible": true,92 "closed": false,93 "archived": false,94 "bookmarked": null,95 "liked": null,96 "tags_descriptions": {},97 "like_count": 0,98 "views": 50,99 "category_id": 11,100 "featured_link": null,101 "has_accepted_answer": false,102 "posters": [103 {104 "extras": "latest single",105 "description": "Original Poster, Most Recent Poster",106 "user": {107 "id": 56882,108 "username": "L_R",109 "name": "L R",110 "avatar_template": "/user_avatar/discuss.pytorch.org/l_r/{size}/30837_2.png",111 "trust_level": 1112 }113 }114 ]115 },116 {117 "fancy_title": "Access cuDNN convolution grad input and grad weight computation function",118 "id": 214064,119 "title": "Access cuDNN convolution grad input and grad weight computation function",120 "slug": "access-cudnn-convolution-grad-input-and-grad-weight-computation-function",121 "posts_count": 3,122 "reply_count": 0,123 "highest_post_number": 3,124 "image_url": null,125 "created_at": "2024-12-10T16:18:12.287Z",126 "last_posted_at": "2024-12-13T14:18:10.237Z",127 "bumped": true,128 "bumped_at": "2024-12-13T14:24:11.703Z",129 "archetype": "regular",130 "unseen": false,131 "pinned": false,132 "unpinned": null,133 "visible": true,134 "closed": false,135 "archived": false,136 "bookmarked": null,137 "liked": null,138 "tags_descriptions": {},139 "like_count": 0,140 "views": 133,141 "category_id": 11,142 "featured_link": null,143 "has_accepted_answer": true,144 "posters": [145 {146 "extras": "latest",147 "description": "Original Poster, Most Recent Poster",148 "user": {149 "id": 81429,150 "username": "flauron",151 "name": "",152 "avatar_template": "/letter_avatar_proxy/v4/letter/f/ba8739/{size}.png",153 "trust_level": 1154 }155 },156 {157 "extras": null,158 "description": "Frequent Poster, Accepted Answer",159 "user": {160 "id": 3534,161 "username": "ptrblck",162 "name": "",163 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",164 "admin": true,165 "moderator": true,166 "trust_level": 2167 }168 }169 ]170 },171 {172 "fancy_title": "Dynamic Function Selection Process in PyTorch Compilation for torch_cpu.so”",173 "id": 214921,174 "title": "Dynamic Function Selection Process in PyTorch Compilation for torch_cpu.so”",175 "slug": "dynamic-function-selection-process-in-pytorch-compilation-for-torch-cpu-so",176 "posts_count": 1,177 "reply_count": 0,178 "highest_post_number": 1,179 "image_url": "https://discuss.pytorch.org/uploads/default/optimized/3X/2/7/274979df051e2c77531d349b6e73876298dc0fd7_2_1024x656.jpeg",180 "created_at": "2025-01-03T09:23:14.477Z",181 "last_posted_at": "2025-01-03T09:23:14.520Z",182 "bumped": true,183 "bumped_at": "2025-01-03T09:23:14.520Z",184 "archetype": "regular",185 "unseen": false,186 "pinned": false,187 "unpinned": null,188 "visible": true,189 "closed": false,190 "archived": false,191 "bookmarked": null,192 "liked": null,193 "tags_descriptions": {},194 "like_count": 0,195 "views": 23,196 "category_id": 11,197 "featured_link": null,198 "has_accepted_answer": false,199 "posters": [200 {201 "extras": "latest single",202 "description": "Original Poster, Most Recent Poster",203 "user": {204 "id": 81828,205 "username": "risemeup1",206 "name": "Risemeup1",207 "avatar_template": "/user_avatar/discuss.pytorch.org/risemeup1/{size}/74854_2.png",208 "trust_level": 1209 }210 }211 ]212 },213 {214 "fancy_title": "Torch C++ program failed to compile with weird error",215 "id": 212131,216 "title": "Torch C++ program failed to compile with weird error",217 "slug": "torch-c-program-failed-to-compile-with-weird-error",218 "posts_count": 1,219 "reply_count": 0,220 "highest_post_number": 1,221 "image_url": null,222 "created_at": "2024-10-26T12:44:03.850Z",223 "last_posted_at": "2024-10-26T12:44:03.906Z",224 "bumped": true,225 "bumped_at": "2024-10-26T12:44:03.906Z",226 "archetype": "regular",227 "unseen": false,228 "pinned": false,229 "unpinned": null,230 "visible": true,231 "closed": false,232 "archived": false,233 "bookmarked": null,234 "liked": null,235 "tags_descriptions": {},236 "like_count": 0,237 "views": 59,238 "category_id": 11,239 "featured_link": null,240 "has_accepted_answer": false,241 "posters": [242 {243 "extras": "latest single",244 "description": "Original Poster, Most Recent Poster",245 "user": {246 "id": 80497,247 "username": "Thanh_Bao_Nguyen",248 "name": "Thanh Bao Nguyen",249 "avatar_template": "/user_avatar/discuss.pytorch.org/thanh_bao_nguyen/{size}/73581_2.png",250 "trust_level": 1251 }252 }253 ]254 },255 {256 "fancy_title": "Questions about onnx with aten::copy",257 "id": 220532,258 "title": "Questions about onnx with aten::copy",259 "slug": "questions-about-onnx-with-aten-copy",260 "posts_count": 3,261 "reply_count": 0,262 "highest_post_number": 3,263 "image_url": null,264 "created_at": "2025-06-03T08:38:16.945Z",265 "last_posted_at": "2025-09-23T20:56:19.099Z",266 "bumped": true,267 "bumped_at": "2025-09-23T20:56:19.099Z",268 "archetype": "regular",269 "unseen": false,270 "pinned": false,271 "unpinned": null,272 "visible": true,273 "closed": false,274 "archived": false,275 "bookmarked": null,276 "liked": null,277 "tags_descriptions": {},278 "like_count": 0,279 "views": 94,280 "category_id": 11,281 "featured_link": null,282 "has_accepted_answer": false,283 "posters": [284 {285 "extras": null,286 "description": "Original Poster",287 "user": {288 "id": 84559,289 "username": "ZongYingLi",290 "name": "Zoey",291 "avatar_template": "/user_avatar/discuss.pytorch.org/zongyingli/{size}/77250_2.png",292 "trust_level": 0293 }294 },295 {296 "extras": null,297 "description": "Frequent Poster",298 "user": {299 "id": 78425,300 "username": "Vitali-Matteo",301 "name": "Matteo",302 "avatar_template": "/user_avatar/discuss.pytorch.org/vitali-matteo/{size}/72295_2.png",303 "trust_level": 1304 }305 },306 {307 "extras": "latest",308 "description": "Most Recent Poster",309 "user": {310 "id": 56628,311 "username": "justinchuby",312 "name": "Justin",313 "avatar_template": "/user_avatar/discuss.pytorch.org/justinchuby/{size}/50395_2.png",314 "trust_level": 1315 }316 }317 ]318 }319 ],320 "tags_descriptions": {},321 "fancy_title": "Error when load network on GPU libTorch 2.5.1",322 "id": 213794,323 "title": "Error when load network on GPU libTorch 2.5.1",324 "posts_count": 1,325 "created_at": "2024-12-04T10:23:51.600Z",326 "views": 165,327 "reply_count": 0,328 "like_count": 0,329 "last_posted_at": "2024-12-04T10:23:51.660Z",330 "visible": true,331 "closed": false,332 "archived": false,333 "has_summary": false,334 "archetype": "regular",335 "slug": "error-when-load-network-on-gpu-libtorch-2-5-1",336 "category_id": 11,337 "word_count": 191,338 "deleted_at": null,339 "user_id": 65591,340 "featured_link": null,341 "pinned_globally": false,342 "pinned_at": null,343 "pinned_until": null,344 "image_url": null,345 "slow_mode_seconds": 0,346 "draft": null,347 "draft_key": "topic_213794",348 "draft_sequence": null,349 "unpinned": null,350 "pinned": false,351 "current_post_number": 1,352 "highest_post_number": 1,353 "deleted_by": null,354 "actions_summary": [355 {356 "id": 4,357 "count": 0,358 "hidden": false,359 "can_act": false360 },361 {362 "id": 8,363 "count": 0,364 "hidden": false,365 "can_act": false366 },367 {368 "id": 10,369 "count": 0,370 "hidden": false,371 "can_act": false372 },373 {374 "id": 7,375 "count": 0,376 "hidden": false,377 "can_act": false378 }379 ],380 "chunk_size": 20,381 "bookmarked": false,382 "topic_timer": null,383 "message_bus_last_id": 0,384 "participant_count": 1,385 "show_read_indicator": false,386 "thumbnails": null,387 "slow_mode_enabled_until": null,388 "can_vote": false,389 "vote_count": 0,390 "user_voted": false,391 "discourse_zendesk_plugin_zendesk_id": null,392 "discourse_zendesk_plugin_zendesk_url": "https://your-url.zendesk.com/agent/tickets/",393 "details": {394 "can_edit": false,395 "notification_level": 1,396 "participants": [397 {398 "id": 65591,399 "username": "weitaoliu",400 "name": "",401 "avatar_template": "/letter_avatar_proxy/v4/letter/w/35a633/{size}.png",402 "post_count": 1,403 "primary_group_name": null,404 "flair_name": null,405 "flair_url": null,406 "flair_color": null,407 "flair_bg_color": null,408 "flair_group_id": null,409 "trust_level": 1410 }411 ],412 "created_by": {413 "id": 65591,414 "username": "weitaoliu",415 "name": "",416 "avatar_template": "/letter_avatar_proxy/v4/letter/w/35a633/{size}.png"417 },418 "last_poster": {419 "id": 65591,420 "username": "weitaoliu",421 "name": "",422 "avatar_template": "/letter_avatar_proxy/v4/letter/w/35a633/{size}.png"423 }424 },425 "bookmarks": []426 },427 {428 "post_stream": {429 "posts": [430 {431 "id": 460744,432 "name": "Find Definition",433 "username": "FindDefinition",434 "avatar_template": "/user_avatar/discuss.pytorch.org/finddefinition/{size}/41948_2.png",435 "created_at": "2024-12-04T09:17:46.961Z",436 "cooked": "<p>I’m working on a large model TP training. the prep part of this model runs fast and contains small number of parameters, so I don’t want to design plans for them to decrease maintain cost.</p>\n<ol>\n<li>If a linear layer have no parallel plan (not in transformer blocks) and have replicate input, each device inside same TP group will have same weight gradient, do they get all-reduced across TP group?</li>\n<li>If a linear layer have no parallel plan and have <strong>Shard</strong> input (local tensor), each device inside same TP group will have <strong>different</strong> weight gradient, do they get all-reduced across TP group?</li>\n<li>If a RMSNorm layer have <strong>Sequence Parallel</strong> plan and have <strong>Shard</strong> input, its weight becomes Replicate, when do they get all-reduced across TP group?</li>\n<li>optimizer will raise error mixed DTensor/Tensor if a model contains both local tensor (no parallel plan) and DTensor, we have two option for this, the first one is create two param groups and group them by type of tensor, the second one is create plans for all layers. which one is best in practice?</li>\n</ol>",437 "post_number": 1,438 "post_type": 1,439 "posts_count": 1,440 "updated_at": "2024-12-04T10:03:21.694Z",441 "reply_count": 0,442 "reply_to_post_number": null,443 "quote_count": 0,444 "incoming_link_count": 76,445 "reads": 7,446 "readers_count": 6,447 "score": 381.4,448 "yours": false,449 "topic_id": 213790,450 "topic_slug": "tensor-parallel-what-will-happen-if-some-modules-have-no-parallel-plan",451 "display_username": "Find Definition",452 "primary_group_name": null,453 "flair_name": null,454 "flair_url": null,455 "flair_bg_color": null,456 "flair_color": null,457 "flair_group_id": null,458 "badges_granted": [],459 "version": 4,460 "can_edit": false,461 "can_delete": false,462 "can_recover": false,463 "can_see_hidden_post": false,464 "can_wiki": false,465 "read": true,466 "user_title": null,467 "bookmarked": false,468 "actions_summary": [],469 "moderator": false,470 "admin": false,471 "staff": false,472 "user_id": 48769,473 "hidden": false,474 "trust_level": 1,475 "deleted_at": null,476 "user_deleted": false,477 "edit_reason": null,478 "can_view_edit_history": true,479 "wiki": false,480 "post_url": "/t/tensor-parallel-what-will-happen-if-some-modules-have-no-parallel-plan/213790/1",481 "can_accept_answer": false,482 "can_unaccept_answer": false,483 "accepted_answer": false,484 "topic_accepted_answer": null,485 "can_vote": false486 }487 ],488 "stream": [489 460744490 ]491 },492 "timeline_lookup": [493 [494 1,495 325496 ]497 ],498 "suggested_topics": [499 {500 "fancy_title": "Two models in distributed data parallel",501 "id": 216965,502 "title": "Two models in distributed data parallel",503 "slug": "two-models-in-distributed-data-parallel",504 "posts_count": 1,505 "reply_count": 0,506 "highest_post_number": 1,507 "image_url": null,508 "created_at": "2025-02-20T18:33:34.651Z",509 "last_posted_at": "2025-02-20T18:33:34.707Z",510 "bumped": true,511 "bumped_at": "2025-02-20T18:33:34.707Z",512 "archetype": "regular",513 "unseen": false,514 "pinned": false,515 "unpinned": null,516 "visible": true,517 "closed": false,518 "archived": false,519 "bookmarked": null,520 "liked": null,521 "tags_descriptions": {},522 "like_count": 0,523 "views": 57,524 "category_id": 12,525 "featured_link": null,526 "has_accepted_answer": false,527 "posters": [528 {529 "extras": "latest single",530 "description": "Original Poster, Most Recent Poster",531 "user": {532 "id": 82826,533 "username": "mf2055",534 "name": "mf2055",535 "avatar_template": "/user_avatar/discuss.pytorch.org/mf2055/{size}/75785_2.png",536 "trust_level": 0537 }538 }539 ]540 },541 {542 "fancy_title": "Supporting Autograd for Collectives",543 "id": 219430,544 "title": "Supporting Autograd for Collectives",545 "slug": "supporting-autograd-for-collectives",546 "posts_count": 7,547 "reply_count": 2,548 "highest_post_number": 7,549 "image_url": null,550 "created_at": "2025-04-24T18:27:01.695Z",551 "last_posted_at": "2025-05-06T13:02:37.140Z",552 "bumped": true,553 "bumped_at": "2025-05-06T13:14:10.582Z",554 "archetype": "regular",555 "unseen": false,556 "pinned": false,557 "unpinned": null,558 "visible": true,559 "closed": false,560 "archived": false,561 "bookmarked": null,562 "liked": null,563 "tags_descriptions": {},564 "like_count": 8,565 "views": 410,566 "category_id": 12,567 "featured_link": null,568 "has_accepted_answer": false,569 "posters": [570 {571 "extras": null,572 "description": "Original Poster",573 "user": {574 "id": 55883,575 "username": "wconstab",576 "name": "Will Constable",577 "avatar_template": "/user_avatar/discuss.pytorch.org/wconstab/{size}/49551_2.png",578 "trust_level": 2579 }580 },581 {582 "extras": null,583 "description": "Frequent Poster",584 "user": {585 "id": 42875,586 "username": "Chillee",587 "name": "Horace He",588 "avatar_template": "/user_avatar/discuss.pytorch.org/chillee/{size}/35574_2.png",589 "trust_level": 2590 }591 },592 {593 "extras": null,594 "description": "Frequent Poster",595 "user": {596 "id": 13180,597 "username": "wanchaol",598 "name": "Wanchao",599 "avatar_template": "/user_avatar/discuss.pytorch.org/wanchaol/{size}/50343_2.png",600 "trust_level": 2601 }602 },603 {604 "extras": "latest",605 "description": "Most Recent Poster",606 "user": {607 "id": 78505,608 "username": "tianyu",609 "name": "",610 "avatar_template": "/user_avatar/discuss.pytorch.org/tianyu/{size}/72368_2.png",611 "trust_level": 2612 }613 }614 ]615 },616 {617 "fancy_title": "DistributedSampler is NOT required for FSDP on a single cluster",618 "id": 214056,619 "title": "DistributedSampler is NOT required for FSDP on a single cluster",620 "slug": "distributedsampler-is-not-required-for-fsdp-on-a-single-cluster",621 "posts_count": 2,622 "reply_count": 0,623 "highest_post_number": 2,624 "image_url": null,625 "created_at": "2024-12-10T14:24:33.412Z",626 "last_posted_at": "2024-12-13T11:04:15.024Z",627 "bumped": true,628 "bumped_at": "2024-12-13T11:04:15.024Z",629 "archetype": "regular",630 "unseen": false,631 "pinned": false,632 "unpinned": null,633 "visible": true,634 "closed": false,635 "archived": false,636 "bookmarked": null,637 "liked": null,638 "tags_descriptions": {},639 "like_count": 1,640 "views": 64,641 "category_id": 12,642 "featured_link": null,643 "has_accepted_answer": true,644 "posters": [645 {646 "extras": "latest single",647 "description": "Original Poster, Most Recent Poster, Accepted Answer",648 "user": {649 "id": 34973,650 "username": "hH1sG0n3",651 "name": "",652 "avatar_template": "/letter_avatar_proxy/v4/letter/h/5fc32e/{size}.png",653 "trust_level": 1654 }655 }656 ]657 },658 {659 "fancy_title": "Torch DDP with AMP make deadlock",660 "id": 212192,661 "title": "Torch DDP with AMP make deadlock",662 "slug": "torch-ddp-with-amp-make-deadlock",663 "posts_count": 1,664 "reply_count": 0,665 "highest_post_number": 1,666 "image_url": "https://discuss.pytorch.org/uploads/default/optimized/3X/6/e/6ee6e88a1cff1d6781a1dd444ffc12ef5e45c275_2_1024x19.png",667 "created_at": "2024-10-28T07:16:15.894Z",668 "last_posted_at": "2024-10-28T07:16:15.968Z",669 "bumped": true,670 "bumped_at": "2024-10-28T07:22:03.162Z",671 "archetype": "regular",672 "unseen": false,673 "pinned": false,674 "unpinned": null,675 "visible": true,676 "closed": false,677 "archived": false,678 "bookmarked": null,679 "liked": null,680 "tags_descriptions": {},681 "like_count": 0,682 "views": 101,683 "category_id": 12,684 "featured_link": null,685 "has_accepted_answer": false,686 "posters": [687 {688 "extras": "latest single",689 "description": "Original Poster, Most Recent Poster",690 "user": {691 "id": 74787,692 "username": "taehyung",693 "name": "Taehyung Yu",694 "avatar_template": "/user_avatar/discuss.pytorch.org/taehyung/{size}/69057_2.png",695 "trust_level": 1696 }697 }698 ]699 },700 {701 "fancy_title": "FSDP2 and gradient w.r.t. inputs",702 "id": 221659,703 "title": "FSDP2 and gradient w.r.t. inputs",704 "slug": "fsdp2-and-gradient-w-r-t-inputs",705 "posts_count": 2,706 "reply_count": 0,707 "highest_post_number": 2,708 "image_url": null,709 "created_at": "2025-07-18T23:42:58.881Z",710 "last_posted_at": "2025-07-20T21:43:51.135Z",711 "bumped": true,712 "bumped_at": "2025-07-20T21:43:51.135Z",713 "archetype": "regular",714 "unseen": false,715 "pinned": false,716 "unpinned": null,717 "visible": true,718 "closed": false,719 "archived": false,720 "bookmarked": null,721 "liked": null,722 "tags_descriptions": {},723 "like_count": 0,724 "views": 49,725 "category_id": 12,726 "featured_link": null,727 "has_accepted_answer": false,728 "posters": [729 {730 "extras": null,731 "description": "Original Poster",732 "user": {733 "id": 85151,734 "username": "ccasert",735 "name": null,736 "avatar_template": "/letter_avatar_proxy/v4/letter/c/eada6e/{size}.png",737 "trust_level": 1738 }739 },740 {741 "extras": "latest",742 "description": "Most Recent Poster",743 "user": {744 "id": 6225,745 "username": "yf225",746 "name": "PyTorch Developer, Meta",747 "avatar_template": "/user_avatar/discuss.pytorch.org/yf225/{size}/3418_2.png",748 "trust_level": 2749 }750 }751 ]752 }753 ],754 "tags_descriptions": {},755 "fancy_title": "[Tensor Parallel] what will happen if some modules have no parallel plan?",756 "id": 213790,757 "title": "[Tensor Parallel] what will happen if some modules have no parallel plan?",758 "posts_count": 1,759 "created_at": "2024-12-04T09:17:46.892Z",760 "views": 86,761 "reply_count": 0,762 "like_count": 0,763 "last_posted_at": "2024-12-04T09:17:46.961Z",764 "visible": true,765 "closed": false,766 "archived": false,767 "has_summary": false,768 "archetype": "regular",769 "slug": "tensor-parallel-what-will-happen-if-some-modules-have-no-parallel-plan",770 "category_id": 12,771 "word_count": 189,772 "deleted_at": null,773 "user_id": 48769,774 "featured_link": null,775 "pinned_globally": false,776 "pinned_at": null,777 "pinned_until": null,778 "image_url": null,779 "slow_mode_seconds": 0,780 "draft": null,781 "draft_key": "topic_213790",782 "draft_sequence": null,783 "unpinned": null,784 "pinned": false,785 "current_post_number": 1,786 "highest_post_number": 1,787 "deleted_by": null,788 "actions_summary": [789 {790 "id": 4,791 "count": 0,792 "hidden": false,793 "can_act": false794 },795 {796 "id": 8,797 "count": 0,798 "hidden": false,799 "can_act": false800 },801 {802 "id": 10,803 "count": 0,804 "hidden": false,805 "can_act": false806 },807 {808 "id": 7,809 "count": 0,810 "hidden": false,811 "can_act": false812 }813 ],814 "chunk_size": 20,815 "bookmarked": false,816 "topic_timer": null,817 "message_bus_last_id": 0,818 "participant_count": 1,819 "show_read_indicator": false,820 "thumbnails": null,821 "slow_mode_enabled_until": null,822 "can_vote": false,823 "vote_count": 0,824 "user_voted": false,825 "discourse_zendesk_plugin_zendesk_id": null,826 "discourse_zendesk_plugin_zendesk_url": "https://your-url.zendesk.com/agent/tickets/",827 "details": {828 "can_edit": false,829 "notification_level": 1,830 "participants": [831 {832 "id": 48769,833 "username": "FindDefinition",834 "name": "Find Definition",835 "avatar_template": "/user_avatar/discuss.pytorch.org/finddefinition/{size}/41948_2.png",836 "post_count": 1,837 "primary_group_name": null,838 "flair_name": null,839 "flair_url": null,840 "flair_color": null,841 "flair_bg_color": null,842 "flair_group_id": null,843 "trust_level": 1844 }845 ],846 "created_by": {847 "id": 48769,848 "username": "FindDefinition",849 "name": "Find Definition",850 "avatar_template": "/user_avatar/discuss.pytorch.org/finddefinition/{size}/41948_2.png"851 },852 "last_poster": {853 "id": 48769,854 "username": "FindDefinition",855 "name": "Find Definition",856 "avatar_template": "/user_avatar/discuss.pytorch.org/finddefinition/{size}/41948_2.png"857 }858 },859 "bookmarks": []860 },861 {862 "post_stream": {863 "posts": [864 {865 "id": 460715,866 "name": "Silver",867 "username": "silveroxides",868 "avatar_template": "/user_avatar/discuss.pytorch.org/silveroxides/{size}/74338_2.png",869 "created_at": "2024-12-03T23:04:47.118Z",870 "cooked": "<p>I suggest adding the install commands for installing torch on intel xpu on the front page considering the upcoming launch. Taking a look at their own site it is not much different from the Pytorch Get Started page and would be useful if such an instruction was in place on the official page before the launch of the new SKUs in 10 days. <a href=\"https://intel.github.io/intel-extension-for-pytorch/index.html#installation?platform=gpu&version=v2.3.110%2bxpu&os=windows&package=pip\" rel=\"noopener nofollow ugc\">https://intel.github.io/intel-extension-for-pytorch/index.html#installation?platform=gpu&version=v2.3.110%2bxpu&os=windows&package=pip</a></p>",871 "post_number": 1,872 "post_type": 1,873 "posts_count": 2,874 "updated_at": "2024-12-03T23:04:47.118Z",875 "reply_count": 0,876 "reply_to_post_number": null,877 "quote_count": 0,878 "incoming_link_count": 79,879 "reads": 6,880 "readers_count": 5,881 "score": 411.2,882 "yours": false,883 "topic_id": 213768,884 "topic_slug": "add-install-instruction-for-intel-gpu-xpu",885 "display_username": "Silver",886 "primary_group_name": null,887 "flair_name": null,888 "flair_url": null,889 "flair_bg_color": null,890 "flair_color": null,891 "flair_group_id": null,892 "badges_granted": [],893 "version": 1,894 "can_edit": false,895 "can_delete": false,896 "can_recover": false,897 "can_see_hidden_post": false,898 "can_wiki": false,899 "link_counts": [900 {901 "url": "https://intel.github.io/intel-extension-for-pytorch/index.html#installation?platform=gpu&version=v2.3.110%2bxpu&os=windows&package=pip",902 "internal": false,903 "reflection": false,904 "title": "Welcome to Intel® Extension for PyTorch* Documentation!",905 "clicks": 19906 }907 ],908 "read": true,909 "user_title": null,910 "bookmarked": false,911 "actions_summary": [912 {913 "id": 2,914 "count": 1915 }916 ],917 "moderator": false,918 "admin": false,919 "staff": false,920 "user_id": 81288,921 "hidden": false,922 "trust_level": 1,923 "deleted_at": null,924 "user_deleted": false,925 "edit_reason": null,926 "can_view_edit_history": true,927 "wiki": false,928 "post_url": "/t/add-install-instruction-for-intel-gpu-xpu/213768/1",929 "can_accept_answer": false,930 "can_unaccept_answer": false,931 "accepted_answer": false,932 "topic_accepted_answer": null,933 "can_vote": false934 },935 {936 "id": 460736,937 "name": "Aknw Fen",938 "username": "Aknw_Fen",939 "avatar_template": "/user_avatar/discuss.pytorch.org/aknw_fen/{size}/74156_2.png",940 "created_at": "2024-12-04T06:35:39.319Z",941 "cooked": "<p>that’s so cool, i hope they add it, many users have intel core processors aside from xpu devices, and will likely make a diff with avx512 instructions used, isn’t it?</p>",942 "post_number": 2,943 "post_type": 1,944 "posts_count": 2,945 "updated_at": "2024-12-04T06:38:57.798Z",946 "reply_count": 0,947 "reply_to_post_number": null,948 "quote_count": 0,949 "incoming_link_count": 1,950 "reads": 5,951 "readers_count": 4,952 "score": 6.0,953 "yours": false,954 "topic_id": 213768,955 "topic_slug": "add-install-instruction-for-intel-gpu-xpu",956 "display_username": "Aknw Fen",957 "primary_group_name": null,958 "flair_name": null,959 "flair_url": null,960 "flair_bg_color": null,961 "flair_color": null,962 "flair_group_id": null,963 "badges_granted": [],964 "version": 1,965 "can_edit": false,966 "can_delete": false,967 "can_recover": false,968 "can_see_hidden_post": false,969 "can_wiki": false,970 "read": true,971 "user_title": null,972 "bookmarked": false,973 "actions_summary": [],974 "moderator": false,975 "admin": false,976 "staff": false,977 "user_id": 81089,978 "hidden": false,979 "trust_level": 2,980 "deleted_at": null,981 "user_deleted": false,982 "edit_reason": null,983 "can_view_edit_history": true,984 "wiki": false,985 "post_url": "/t/add-install-instruction-for-intel-gpu-xpu/213768/2",986 "can_accept_answer": false,987 "can_unaccept_answer": false,988 "accepted_answer": false,989 "topic_accepted_answer": null990 }991 ],992 "stream": [993 460715,994 460736995 ]996 },997 "timeline_lookup": [998 [999 1,1000 3261001 ],1002 [1003 2,1004 3251005 ]1006 ],1007 "suggested_topics": [1008 {1009 "fancy_title": "New PyTorch Sphinx Theme - Feedback Requested",1010 "id": 219351,1011 "title": "New PyTorch Sphinx Theme - Feedback Requested",1012 "slug": "new-pytorch-sphinx-theme-feedback-requested",1013 "posts_count": 1,1014 "reply_count": 0,1015 "highest_post_number": 1,1016 "image_url": null,1017 "created_at": "2025-04-22T19:39:57.390Z",1018 "last_posted_at": "2025-04-22T19:39:57.442Z",1019 "bumped": true,1020 "bumped_at": "2025-04-22T19:53:54.534Z",1021 "archetype": "regular",1022 "unseen": false,1023 "pinned": false,1024 "unpinned": null,1025 "visible": true,1026 "closed": false,1027 "archived": false,1028 "bookmarked": null,1029 "liked": null,1030 "tags_descriptions": {},1031 "like_count": 0,1032 "views": 80,1033 "category_id": 3,1034 "featured_link": null,1035 "has_accepted_answer": false,1036 "posters": [1037 {1038 "extras": "latest single",1039 "description": "Original Poster, Most Recent Poster",1040 "user": {1041 "id": 62819,1042 "username": "svekars",1043 "name": "Svetlana",1044 "avatar_template": "/user_avatar/discuss.pytorch.org/svekars/{size}/56746_2.png",1045 "trust_level": 11046 }1047 }1048 ]1049 },1050 {1051 "fancy_title": "Announcement: Share Your Feedback on PyTorch Docs and Tutorial",1052 "id": 217813,1053 "title": "Announcement: Share Your Feedback on PyTorch Docs and Tutorial",1054 "slug": "announcement-share-your-feedback-on-pytorch-docs-and-tutorial",1055 "posts_count": 1,1056 "reply_count": 0,1057 "highest_post_number": 1,1058 "image_url": null,1059 "created_at": "2025-03-13T14:44:18.138Z",1060 "last_posted_at": "2025-03-13T14:44:18.214Z",1061 "bumped": true,1062 "bumped_at": "2025-03-13T14:44:18.214Z",1063 "archetype": "regular",1064 "unseen": false,1065 "pinned": false,1066 "unpinned": null,1067 "visible": true,1068 "closed": false,1069 "archived": false,1070 "bookmarked": null,1071 "liked": null,1072 "tags_descriptions": {},1073 "like_count": 0,1074 "views": 32,1075 "category_id": 3,1076 "featured_link": null,1077 "has_accepted_answer": false,1078 "posters": [1079 {1080 "extras": "latest single",1081 "description": "Original Poster, Most Recent Poster",1082 "user": {1083 "id": 83252,1084 "username": "sekyondaMeta",1085 "name": "Sekyonda ",1086 "avatar_template": "/user_avatar/discuss.pytorch.org/sekyondameta/{size}/76147_2.png",1087 "trust_level": 11088 }1089 }1090 ]1091 },1092 {1093 "fancy_title": "Industry-Related Groups for Pytorch Forums",1094 "id": 222518,1095 "title": "Industry-Related Groups for Pytorch Forums",1096 "slug": "industry-related-groups-for-pytorch-forums",1097 "posts_count": 1,1098 "reply_count": 0,1099 "highest_post_number": 1,1100 "image_url": null,1101 "created_at": "2025-08-20T16:39:35.373Z",1102 "last_posted_at": "2025-08-20T16:39:35.437Z",1103 "bumped": true,1104 "bumped_at": "2025-08-20T16:39:35.437Z",1105 "archetype": "regular",1106 "unseen": false,1107 "pinned": false,1108 "unpinned": null,1109 "visible": true,1110 "closed": false,1111 "archived": false,1112 "bookmarked": null,1113 "liked": null,1114 "tags_descriptions": {},1115 "like_count": 0,1116 "views": 37,1117 "category_id": 3,1118 "featured_link": null,1119 "has_accepted_answer": false,1120 "posters": [1121 {1122 "extras": "latest single",1123 "description": "Original Poster, Most Recent Poster",1124 "user": {1125 "id": 77908,1126 "username": "mycul",1127 "name": "",1128 "avatar_template": "/user_avatar/discuss.pytorch.org/mycul/{size}/72394_2.png",1129 "trust_level": 21130 }1131 }1132 ]1133 },1134 {1135 "fancy_title": "Executorch - Encoder/Decoder models?",1136 "id": 213305,1137 "title": "Executorch - Encoder/Decoder models?",1138 "slug": "executorch-encoder-decoder-models",1139 "posts_count": 2,1140 "reply_count": 0,1141 "highest_post_number": 2,1142 "image_url": null,1143 "created_at": "2024-11-22T11:38:56.139Z",1144 "last_posted_at": "2024-11-22T23:11:43.648Z",1145 "bumped": true,1146 "bumped_at": "2024-11-22T23:11:43.648Z",1147 "archetype": "regular",1148 "unseen": false,1149 "pinned": false,1150 "unpinned": null,1151 "visible": true,1152 "closed": false,1153 "archived": false,1154 "bookmarked": null,1155 "liked": null,1156 "tags_descriptions": {},1157 "like_count": 0,1158 "views": 262,1159 "category_id": 42,1160 "featured_link": null,1161 "has_accepted_answer": false,1162 "posters": [1163 {1164 "extras": null,1165 "description": "Original Poster",1166 "user": {1167 "id": 81023,1168 "username": "cptspacemanspiff",1169 "name": "",1170 "avatar_template": "/user_avatar/discuss.pytorch.org/cptspacemanspiff/{size}/74098_2.png",1171 "trust_level": 11172 }1173 },1174 {1175 "extras": "latest",1176 "description": "Most Recent Poster",1177 "user": {1178 "id": 3534,1179 "username": "ptrblck",1180 "name": "",1181 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",1182 "admin": true,1183 "moderator": true,1184 "trust_level": 21185 }1186 }1187 ]1188 },1189 {1190 "fancy_title": "Reshaping tensors while using model parallelism",1191 "id": 218591,1192 "title": "Reshaping tensors while using model parallelism",1193 "slug": "reshaping-tensors-while-using-model-parallelism",1194 "posts_count": 1,1195 "reply_count": 0,1196 "highest_post_number": 1,1197 "image_url": null,1198 "created_at": "2025-04-03T18:00:02.837Z",1199 "last_posted_at": "2025-04-03T18:00:02.891Z",1200 "bumped": true,