Anurag1734/cuda-error-resolution-analysis
07
1[2 {3 "post_stream": {4 "posts": [5 {6 "id": 255748,7 "name": "Kamer Ali Yüksel",8 "username": "Kamer_Ali_Yuksel",9 "avatar_template": "/user_avatar/discuss.pytorch.org/kamer_ali_yuksel/{size}/7641_2.png",10 "created_at": "2021-01-08T02:55:20.330Z",11 "cooked": "<p>The below statements doesn’t change the values in opt_weights[tr == kth].data</p>\n<p>opt_weights[tr == kth].data = dist.sample_n(opt_weights[tr == kth].shape[0])<br>\nopt_weights[tr == kth].data = dist.sample_n(opt_weights[tr == kth].shape[0]).data<br>\nopt_weights[tr == kth].data.copy_(dist.sample_n(opt_weights[tr == kth].shape[0]))</p>\n<p>I don’t get an error in above cases but as I said they’re not helpful in changing values.</p>\n<p>When I try to change it without .data part, I get the following.<br>\n*** RuntimeError: a leaf Variable that requires grad is being used in an in-place operation.</p>\n<p>What is the right way of overwriting certain values in a tensor with samples of a distribution?</p>",12 "post_number": 1,13 "post_type": 1,14 "posts_count": 3,15 "updated_at": "2021-01-08T02:56:05.565Z",16 "reply_count": 0,17 "reply_to_post_number": null,18 "quote_count": 0,19 "incoming_link_count": 37,20 "reads": 5,21 "readers_count": 4,22 "score": 186.0,23 "yours": false,24 "topic_id": 108230,25 "topic_slug": "setting-data-of-tensor-doesnt-change-it",26 "display_username": "Kamer Ali Yüksel",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": 13016,48 "hidden": false,49 "trust_level": 2,50 "deleted_at": null,51 "user_deleted": false,52 "edit_reason": null,53 "can_view_edit_history": true,54 "wiki": false,55 "post_url": "/t/setting-data-of-tensor-doesnt-change-it/108230/1",56 "can_accept_answer": false,57 "can_unaccept_answer": false,58 "accepted_answer": false,59 "topic_accepted_answer": true,60 "can_vote": false61 },62 {63 "id": 255750,64 "name": "Kamer Ali Yüksel",65 "username": "Kamer_Ali_Yuksel",66 "avatar_template": "/user_avatar/discuss.pytorch.org/kamer_ali_yuksel/{size}/7641_2.png",67 "created_at": "2021-01-08T03:09:02.696Z",68 "cooked": "<p>I have managed to do it via index_copy_ , is this the right solution?</p>\n<p>opt_weights.data.index_copy_(0, (tr == kth).nonzero().flatten(), dist.sample_n(opt_weights[tr == kth].shape[0]))</p>",69 "post_number": 2,70 "post_type": 1,71 "posts_count": 3,72 "updated_at": "2021-01-09T12:28:49.996Z",73 "reply_count": 0,74 "reply_to_post_number": null,75 "quote_count": 0,76 "incoming_link_count": 1,77 "reads": 5,78 "readers_count": 4,79 "score": 6.0,80 "yours": false,81 "topic_id": 108230,82 "topic_slug": "setting-data-of-tensor-doesnt-change-it",83 "display_username": "Kamer Ali Yüksel",84 "primary_group_name": null,85 "flair_name": null,86 "flair_url": null,87 "flair_bg_color": null,88 "flair_color": null,89 "flair_group_id": null,90 "badges_granted": [],91 "version": 1,92 "can_edit": false,93 "can_delete": false,94 "can_recover": false,95 "can_see_hidden_post": false,96 "can_wiki": false,97 "read": true,98 "user_title": null,99 "bookmarked": false,100 "actions_summary": [],101 "moderator": false,102 "admin": false,103 "staff": false,104 "user_id": 13016,105 "hidden": false,106 "trust_level": 2,107 "deleted_at": null,108 "user_deleted": false,109 "edit_reason": null,110 "can_view_edit_history": true,111 "wiki": false,112 "post_url": "/t/setting-data-of-tensor-doesnt-change-it/108230/2",113 "can_accept_answer": false,114 "can_unaccept_answer": false,115 "accepted_answer": true,116 "topic_accepted_answer": true117 },118 {119 "id": 255885,120 "name": "Alban D",121 "username": "albanD",122 "avatar_template": "/user_avatar/discuss.pytorch.org/alband/{size}/215_2.png",123 "created_at": "2021-01-08T17:05:22.352Z",124 "cooked": "<p>Hi,</p>\n<p>In general, you should never use <code>.data</code> otherwise you might get unexpected behaviors like that.</p>\n<p>The other error comes from the fact that you modify a leaf Tensor inplace. You can add a <code>opt_weights = opt_weights.clone()</code> before this code to make sure you don’t modify the original Tensor.</p>",125 "post_number": 3,126 "post_type": 1,127 "posts_count": 3,128 "updated_at": "2021-01-08T17:05:22.352Z",129 "reply_count": 0,130 "reply_to_post_number": null,131 "quote_count": 0,132 "incoming_link_count": 3,133 "reads": 3,134 "readers_count": 2,135 "score": 30.6,136 "yours": false,137 "topic_id": 108230,138 "topic_slug": "setting-data-of-tensor-doesnt-change-it",139 "display_username": "Alban D",140 "primary_group_name": null,141 "flair_name": null,142 "flair_url": null,143 "flair_bg_color": null,144 "flair_color": null,145 "flair_group_id": null,146 "badges_granted": [],147 "version": 1,148 "can_edit": false,149 "can_delete": false,150 "can_recover": false,151 "can_see_hidden_post": false,152 "can_wiki": false,153 "read": true,154 "user_title": "",155 "bookmarked": false,156 "actions_summary": [157 {158 "id": 2,159 "count": 1160 }161 ],162 "moderator": true,163 "admin": true,164 "staff": true,165 "user_id": 211,166 "hidden": false,167 "trust_level": 4,168 "deleted_at": null,169 "user_deleted": false,170 "edit_reason": null,171 "can_view_edit_history": true,172 "wiki": false,173 "post_url": "/t/setting-data-of-tensor-doesnt-change-it/108230/3",174 "can_accept_answer": false,175 "can_unaccept_answer": false,176 "accepted_answer": false,177 "topic_accepted_answer": true178 }179 ],180 "stream": [181 255748,182 255750,183 255885184 ]185 },186 "timeline_lookup": [187 [188 1,189 1752190 ],191 [192 3,193 1751194 ]195 ],196 "suggested_topics": [197 {198 "fancy_title": "DDP leads to Out of Memory error",199 "id": 213422,200 "title": "DDP leads to Out of Memory error",201 "slug": "ddp-leads-to-out-of-memory-error",202 "posts_count": 1,203 "reply_count": 0,204 "highest_post_number": 1,205 "image_url": null,206 "created_at": "2024-11-25T19:51:07.570Z",207 "last_posted_at": "2024-11-25T19:51:07.613Z",208 "bumped": true,209 "bumped_at": "2024-11-25T19:51:07.613Z",210 "archetype": "regular",211 "unseen": false,212 "pinned": false,213 "unpinned": null,214 "visible": true,215 "closed": false,216 "archived": false,217 "bookmarked": null,218 "liked": null,219 "tags_descriptions": {},220 "like_count": 0,221 "views": 58,222 "category_id": 1,223 "featured_link": null,224 "has_accepted_answer": false,225 "posters": [226 {227 "extras": "latest single",228 "description": "Original Poster, Most Recent Poster",229 "user": {230 "id": 37254,231 "username": "enterthevoidf22",232 "name": "",233 "avatar_template": "/user_avatar/discuss.pytorch.org/enterthevoidf22/{size}/29338_2.png",234 "trust_level": 2235 }236 }237 ]238 },239 {240 "fancy_title": "Modifying GPU operators",241 "id": 214319,242 "title": "Modifying GPU operators",243 "slug": "modifying-gpu-operators",244 "posts_count": 2,245 "reply_count": 0,246 "highest_post_number": 2,247 "image_url": null,248 "created_at": "2024-12-17T15:15:56.744Z",249 "last_posted_at": "2024-12-17T18:36:06.932Z",250 "bumped": true,251 "bumped_at": "2024-12-17T18:36:06.932Z",252 "archetype": "regular",253 "unseen": false,254 "pinned": false,255 "unpinned": null,256 "visible": true,257 "closed": false,258 "archived": false,259 "bookmarked": null,260 "liked": null,261 "tags_descriptions": {},262 "like_count": 2,263 "views": 233,264 "category_id": 1,265 "featured_link": null,266 "has_accepted_answer": false,267 "posters": [268 {269 "extras": null,270 "description": "Original Poster",271 "user": {272 "id": 74208,273 "username": "Izan_C_G",274 "name": "Izan C. G",275 "avatar_template": "/user_avatar/discuss.pytorch.org/izan_c_g/{size}/68535_2.png",276 "trust_level": 1277 }278 },279 {280 "extras": "latest",281 "description": "Most Recent Poster",282 "user": {283 "id": 3534,284 "username": "ptrblck",285 "name": "",286 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",287 "admin": true,288 "moderator": true,289 "trust_level": 2290 }291 }292 ]293 },294 {295 "fancy_title": "`num_features` parameter of `nn.InstanceNorm2d` does not change results",296 "id": 214610,297 "title": "`num_features` parameter of `nn.InstanceNorm2d` does not change results",298 "slug": "num-features-parameter-of-nn-instancenorm2d-does-not-change-results",299 "posts_count": 3,300 "reply_count": 1,301 "highest_post_number": 3,302 "image_url": null,303 "created_at": "2024-12-24T23:59:04.629Z",304 "last_posted_at": "2024-12-31T04:48:24.867Z",305 "bumped": true,306 "bumped_at": "2024-12-31T04:48:24.867Z",307 "archetype": "regular",308 "unseen": false,309 "pinned": false,310 "unpinned": null,311 "visible": true,312 "closed": false,313 "archived": false,314 "bookmarked": null,315 "liked": null,316 "tags_descriptions": {},317 "like_count": 0,318 "views": 46,319 "category_id": 1,320 "featured_link": null,321 "has_accepted_answer": true,322 "posters": [323 {324 "extras": "latest",325 "description": "Original Poster, Most Recent Poster",326 "user": {327 "id": 68684,328 "username": "aoot",329 "name": "",330 "avatar_template": "/letter_avatar_proxy/v4/letter/a/df788c/{size}.png",331 "trust_level": 2332 }333 },334 {335 "extras": null,336 "description": "Frequent Poster, Accepted Answer",337 "user": {338 "id": 3534,339 "username": "ptrblck",340 "name": "",341 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",342 "admin": true,343 "moderator": true,344 "trust_level": 2345 }346 }347 ]348 },349 {350 "fancy_title": "Initialization of a layer with different function than its forward function",351 "id": 213778,352 "title": "Initialization of a layer with different function than its forward function",353 "slug": "initialization-of-a-layer-with-different-function-than-its-forward-function",354 "posts_count": 3,355 "reply_count": 1,356 "highest_post_number": 4,357 "image_url": null,358 "created_at": "2024-12-04T04:21:29.888Z",359 "last_posted_at": "2024-12-05T00:42:20.599Z",360 "bumped": true,361 "bumped_at": "2024-12-05T01:18:39.381Z",362 "archetype": "regular",363 "unseen": false,364 "pinned": false,365 "unpinned": null,366 "visible": true,367 "closed": false,368 "archived": false,369 "bookmarked": null,370 "liked": null,371 "tags_descriptions": {},372 "like_count": 0,373 "views": 45,374 "category_id": 1,375 "featured_link": null,376 "has_accepted_answer": false,377 "posters": [378 {379 "extras": "latest",380 "description": "Original Poster, Most Recent Poster",381 "user": {382 "id": 71463,383 "username": "alireza378",384 "name": "",385 "avatar_template": "/user_avatar/discuss.pytorch.org/alireza378/{size}/61353_2.png",386 "trust_level": 1387 }388 },389 {390 "extras": null,391 "description": "Frequent Poster",392 "user": {393 "id": 3534,394 "username": "ptrblck",395 "name": "",396 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",397 "admin": true,398 "moderator": true,399 "trust_level": 2400 }401 }402 ]403 },404 {405 "fancy_title": "How do I remove torch_dtype=torch.float16 on Windows 11?",406 "id": 218446,407 "title": "How do I remove torch_dtype=torch.float16 on Windows 11?",408 "slug": "how-do-i-remove-torch-dtype-torch-float16-on-windows-11",409 "posts_count": 3,410 "reply_count": 1,411 "highest_post_number": 3,412 "image_url": null,413 "created_at": "2025-03-31T13:34:36.413Z",414 "last_posted_at": "2025-04-15T18:54:42.340Z",415 "bumped": true,416 "bumped_at": "2025-04-15T18:54:42.340Z",417 "archetype": "regular",418 "unseen": false,419 "pinned": false,420 "unpinned": null,421 "visible": true,422 "closed": false,423 "archived": false,424 "bookmarked": null,425 "liked": null,426 "tags_descriptions": {},427 "like_count": 0,428 "views": 181,429 "category_id": 1,430 "featured_link": null,431 "has_accepted_answer": false,432 "posters": [433 {434 "extras": null,435 "description": "Original Poster",436 "user": {437 "id": 83545,438 "username": "Taurus1983",439 "name": "",440 "avatar_template": "/letter_avatar_proxy/v4/letter/t/e9bcb4/{size}.png",441 "trust_level": 0442 }443 },444 {445 "extras": null,446 "description": "Frequent Poster",447 "user": {448 "id": 3534,449 "username": "ptrblck",450 "name": "",451 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",452 "admin": true,453 "moderator": true,454 "trust_level": 2455 }456 },457 {458 "extras": "latest",459 "description": "Most Recent Poster",460 "user": {461 "id": 83836,462 "username": "rxrue",463 "name": "",464 "avatar_template": "/user_avatar/discuss.pytorch.org/rxrue/{size}/76643_2.png",465 "trust_level": 0466 }467 }468 ]469 }470 ],471 "tags_descriptions": {},472 "fancy_title": "Setting .data of Tensor doesn’t change it",473 "id": 108230,474 "title": "Setting .data of Tensor doesn't change it",475 "posts_count": 3,476 "created_at": "2021-01-08T02:55:20.269Z",477 "views": 473,478 "reply_count": 0,479 "like_count": 1,480 "last_posted_at": "2021-01-08T17:05:22.352Z",481 "visible": true,482 "closed": false,483 "archived": false,484 "has_summary": false,485 "archetype": "regular",486 "slug": "setting-data-of-tensor-doesnt-change-it",487 "category_id": 1,488 "word_count": 191,489 "deleted_at": null,490 "user_id": 13016,491 "featured_link": null,492 "pinned_globally": false,493 "pinned_at": null,494 "pinned_until": null,495 "image_url": null,496 "slow_mode_seconds": 0,497 "draft": null,498 "draft_key": "topic_108230",499 "draft_sequence": null,500 "unpinned": null,501 "pinned": false,502 "current_post_number": 1,503 "highest_post_number": 3,504 "deleted_by": null,505 "actions_summary": [506 {507 "id": 4,508 "count": 0,509 "hidden": false,510 "can_act": false511 },512 {513 "id": 8,514 "count": 0,515 "hidden": false,516 "can_act": false517 },518 {519 "id": 10,520 "count": 0,521 "hidden": false,522 "can_act": false523 },524 {525 "id": 7,526 "count": 0,527 "hidden": false,528 "can_act": false529 }530 ],531 "chunk_size": 20,532 "bookmarked": false,533 "topic_timer": null,534 "message_bus_last_id": 0,535 "participant_count": 2,536 "show_read_indicator": false,537 "thumbnails": null,538 "slow_mode_enabled_until": null,539 "accepted_answer": {540 "post_number": 2,541 "username": "Kamer_Ali_Yuksel",542 "name": "Kamer Ali Yüksel",543 "excerpt": "I have managed to do it via index_copy_ , is this the right solution? \nopt_weights.data.index_copy_(0, (tr == kth).nonzero().flatten(), dist.sample_n(opt_weights[tr == kth].shape[0]))"544 },545 "can_vote": false,546 "vote_count": 0,547 "user_voted": false,548 "discourse_zendesk_plugin_zendesk_id": null,549 "discourse_zendesk_plugin_zendesk_url": "https://your-url.zendesk.com/agent/tickets/",550 "details": {551 "can_edit": false,552 "notification_level": 1,553 "participants": [554 {555 "id": 13016,556 "username": "Kamer_Ali_Yuksel",557 "name": "Kamer Ali Yüksel",558 "avatar_template": "/user_avatar/discuss.pytorch.org/kamer_ali_yuksel/{size}/7641_2.png",559 "post_count": 2,560 "primary_group_name": null,561 "flair_name": null,562 "flair_url": null,563 "flair_color": null,564 "flair_bg_color": null,565 "flair_group_id": null,566 "trust_level": 2567 },568 {569 "id": 211,570 "username": "albanD",571 "name": "Alban D",572 "avatar_template": "/user_avatar/discuss.pytorch.org/alband/{size}/215_2.png",573 "post_count": 1,574 "primary_group_name": null,575 "flair_name": null,576 "flair_url": null,577 "flair_color": null,578 "flair_bg_color": null,579 "flair_group_id": null,580 "admin": true,581 "moderator": true,582 "trust_level": 4583 }584 ],585 "created_by": {586 "id": 13016,587 "username": "Kamer_Ali_Yuksel",588 "name": "Kamer Ali Yüksel",589 "avatar_template": "/user_avatar/discuss.pytorch.org/kamer_ali_yuksel/{size}/7641_2.png"590 },591 "last_poster": {592 "id": 211,593 "username": "albanD",594 "name": "Alban D",595 "avatar_template": "/user_avatar/discuss.pytorch.org/alband/{size}/215_2.png"596 }597 },598 "bookmarks": []599 },600 {601 "post_stream": {602 "posts": [603 {604 "id": 255873,605 "name": "zuujhyt",606 "username": "111429",607 "avatar_template": "/letter_avatar_proxy/v4/letter/1/e0b2c6/{size}.png",608 "created_at": "2021-01-08T16:16:22.220Z",609 "cooked": "<p>Hi,<br>\nWhat’s the difference?<br>\nI create a dataset with torchvision.datasets.Kinetics400 class.<br>\nAnd get different length of dataloader with these two sampler.<br>\nAnd torch.utils.data.sampler.RandomSampler is correct to me.<br>\nThanks.</p>",610 "post_number": 1,611 "post_type": 1,612 "posts_count": 1,613 "updated_at": "2021-01-08T16:16:22.220Z",614 "reply_count": 0,615 "reply_to_post_number": null,616 "quote_count": 0,617 "incoming_link_count": 106,618 "reads": 5,619 "readers_count": 4,620 "score": 546.0,621 "yours": false,622 "topic_id": 108280,623 "topic_slug": "whats-the-difference-between-torchvision-datasets-samplers-randomclipsampler-and-torch-utils-data-sampler-randomsampler",624 "display_username": "zuujhyt",625 "primary_group_name": null,626 "flair_name": null,627 "flair_url": null,628 "flair_bg_color": null,629 "flair_color": null,630 "flair_group_id": null,631 "badges_granted": [],632 "version": 1,633 "can_edit": false,634 "can_delete": false,635 "can_recover": false,636 "can_see_hidden_post": false,637 "can_wiki": false,638 "read": true,639 "user_title": null,640 "bookmarked": false,641 "actions_summary": [642 {643 "id": 2,644 "count": 1645 }646 ],647 "moderator": false,648 "admin": false,649 "staff": false,650 "user_id": 39605,651 "hidden": false,652 "trust_level": 1,653 "deleted_at": null,654 "user_deleted": false,655 "edit_reason": null,656 "can_view_edit_history": true,657 "wiki": false,658 "post_url": "/t/whats-the-difference-between-torchvision-datasets-samplers-randomclipsampler-and-torch-utils-data-sampler-randomsampler/108280/1",659 "can_accept_answer": false,660 "can_unaccept_answer": false,661 "accepted_answer": false,662 "topic_accepted_answer": null,663 "can_vote": false664 }665 ],666 "stream": [667 255873668 ]669 },670 "timeline_lookup": [671 [672 1,673 1751674 ]675 ],676 "suggested_topics": [677 {678 "fancy_title": "Can’t find package of pytorch1.8.1 and cudatoolkit 11.3",679 "id": 212679,680 "title": "Can't find package of pytorch1.8.1 and cudatoolkit 11.3",681 "slug": "cant-find-package-of-pytorch1-8-1-and-cudatoolkit-11-3",682 "posts_count": 20,683 "reply_count": 17,684 "highest_post_number": 20,685 "image_url": "https://discuss.pytorch.org/uploads/default/original/3X/a/5/a586c6e485f1fb6ae6938f37d1acc2a2efc75811.png",686 "created_at": "2024-11-08T03:25:36.390Z",687 "last_posted_at": "2024-12-17T06:44:51.637Z",688 "bumped": true,689 "bumped_at": "2024-12-24T05:24:21.059Z",690 "archetype": "regular",691 "unseen": false,692 "pinned": false,693 "unpinned": null,694 "visible": true,695 "closed": false,696 "archived": false,697 "bookmarked": null,698 "liked": null,699 "tags_descriptions": {},700 "like_count": 0,701 "views": 1063,702 "category_id": 1,703 "featured_link": null,704 "has_accepted_answer": false,705 "posters": [706 {707 "extras": null,708 "description": "Original Poster",709 "user": {710 "id": 80759,711 "username": "yanzi_shi",712 "name": "yanzi shi",713 "avatar_template": "/user_avatar/discuss.pytorch.org/yanzi_shi/{size}/73842_2.png",714 "trust_level": 1715 }716 },717 {718 "extras": null,719 "description": "Frequent Poster",720 "user": {721 "id": 3534,722 "username": "ptrblck",723 "name": "",724 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",725 "admin": true,726 "moderator": true,727 "trust_level": 2728 }729 },730 {731 "extras": "latest",732 "description": "Most Recent Poster",733 "user": {734 "id": 12117,735 "username": "Tony-Y",736 "name": "",737 "avatar_template": "/user_avatar/discuss.pytorch.org/tony-y/{size}/8373_2.png",738 "trust_level": 2739 }740 }741 ]742 },743 {744 "fancy_title": "Fitting parameters of custom model with pytorch",745 "id": 217919,746 "title": "Fitting parameters of custom model with pytorch",747 "slug": "fitting-parameters-of-custom-model-with-pytorch",748 "posts_count": 2,749 "reply_count": 0,750 "highest_post_number": 2,751 "image_url": null,752 "created_at": "2025-03-16T19:18:26.757Z",753 "last_posted_at": "2025-03-17T03:08:29.653Z",754 "bumped": true,755 "bumped_at": "2025-03-17T03:08:29.653Z",756 "archetype": "regular",757 "unseen": false,758 "pinned": false,759 "unpinned": null,760 "visible": true,761 "closed": false,762 "archived": false,763 "bookmarked": null,764 "liked": null,765 "tags_descriptions": {},766 "like_count": 0,767 "views": 29,768 "category_id": 1,769 "featured_link": null,770 "has_accepted_answer": false,771 "posters": [772 {773 "extras": null,774 "description": "Original Poster",775 "user": {776 "id": 83311,777 "username": "fracton",778 "name": "Fracton",779 "avatar_template": "/letter_avatar_proxy/v4/letter/f/dbc845/{size}.png",780 "trust_level": 0781 }782 },783 {784 "extras": "latest",785 "description": "Most Recent Poster",786 "user": {787 "id": 18088,788 "username": "KFrank",789 "name": "K. Frank",790 "avatar_template": "/letter_avatar_proxy/v4/letter/k/ecb155/{size}.png",791 "trust_level": 2792 }793 }794 ]795 },796 {797 "fancy_title": "Issues with Conv2D Formula for Padding in Integer-Arithmetic-Only Inference",798 "id": 218392,799 "title": "Issues with Conv2D Formula for Padding in Integer-Arithmetic-Only Inference",800 "slug": "issues-with-conv2d-formula-for-padding-in-integer-arithmetic-only-inference",801 "posts_count": 1,802 "reply_count": 0,803 "highest_post_number": 1,804 "image_url": "https://discuss.pytorch.org/uploads/default/original/3X/e/b/eb09db80a2f24b5a79f58c446a0135d75d5b4212.png",805 "created_at": "2025-03-29T09:03:04.763Z",806 "last_posted_at": "2025-03-29T09:03:04.809Z",807 "bumped": true,808 "bumped_at": "2025-03-29T09:03:04.809Z",809 "archetype": "regular",810 "unseen": false,811 "pinned": false,812 "unpinned": null,813 "visible": true,814 "closed": false,815 "archived": false,816 "bookmarked": null,817 "liked": null,818 "tags_descriptions": {},819 "like_count": 0,820 "views": 23,821 "category_id": 1,822 "featured_link": null,823 "has_accepted_answer": false,824 "posters": [825 {826 "extras": "latest single",827 "description": "Original Poster, Most Recent Poster",828 "user": {829 "id": 82558,830 "username": "xiechengyan",831 "name": "承諺 謝",832 "avatar_template": "/letter_avatar_proxy/v4/letter/x/a8b319/{size}.png",833 "trust_level": 1834 }835 }836 ]837 },838 {839 "fancy_title": "`torch.multinomial`: sampling once vs in batches gives different results",840 "id": 212817,841 "title": "`torch.multinomial`: sampling once vs in batches gives different results",842 "slug": "torch-multinomial-sampling-once-vs-in-batches-gives-different-results",843 "posts_count": 4,844 "reply_count": 2,845 "highest_post_number": 4,846 "image_url": null,847 "created_at": "2024-11-11T19:57:26.792Z",848 "last_posted_at": "2024-11-12T13:29:06.085Z",849 "bumped": true,850 "bumped_at": "2024-11-12T13:29:06.085Z",851 "archetype": "regular",852 "unseen": false,853 "pinned": false,854 "unpinned": null,855 "visible": true,856 "closed": false,857 "archived": false,858 "bookmarked": null,859 "liked": null,860 "tags_descriptions": {},861 "like_count": 0,862 "views": 138,863 "category_id": 1,864 "featured_link": null,865 "has_accepted_answer": false,866 "posters": [867 {868 "extras": null,869 "description": "Original Poster",870 "user": {871 "id": 80827,872 "username": "marinegor",873 "name": "Egor Marin",874 "avatar_template": "/user_avatar/discuss.pytorch.org/marinegor/{size}/73923_2.png",875 "trust_level": 1876 }877 },878 {879 "extras": "latest",880 "description": "Most Recent Poster",881 "user": {882 "id": 3534,883 "username": "ptrblck",884 "name": "",885 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",886 "admin": true,887 "moderator": true,888 "trust_level": 2889 }890 }891 ]892 },893 {894 "fancy_title": "Apply linalg.solve if matrix is not singular, return arbitrary vector otherwise",895 "id": 218914,896 "title": "Apply linalg.solve if matrix is not singular, return arbitrary vector otherwise",897 "slug": "apply-linalg-solve-if-matrix-is-not-singular-return-arbitrary-vector-otherwise",898 "posts_count": 2,899 "reply_count": 0,900 "highest_post_number": 2,901 "image_url": null,902 "created_at": "2025-04-09T17:23:08.540Z",903 "last_posted_at": "2025-04-10T00:59:42.916Z",904 "bumped": true,905 "bumped_at": "2025-04-10T00:59:42.916Z",906 "archetype": "regular",907 "unseen": false,908 "pinned": false,909 "unpinned": null,910 "visible": true,911 "closed": false,912 "archived": false,913 "bookmarked": null,914 "liked": null,915 "tags_descriptions": {},916 "like_count": 0,917 "views": 68,918 "category_id": 1,919 "featured_link": null,920 "has_accepted_answer": false,921 "posters": [922 {923 "extras": null,924 "description": "Original Poster",925 "user": {926 "id": 70117,927 "username": "Stingpie_52",928 "name": "Stingpie 52",929 "avatar_template": "/user_avatar/discuss.pytorch.org/stingpie_52/{size}/62527_2.png",930 "trust_level": 1931 }932 },933 {934 "extras": "latest",935 "description": "Most Recent Poster",936 "user": {937 "id": 18088,938 "username": "KFrank",939 "name": "K. Frank",940 "avatar_template": "/letter_avatar_proxy/v4/letter/k/ecb155/{size}.png",941 "trust_level": 2942 }943 }944 ]945 }946 ],947 "tags_descriptions": {},948 "fancy_title": "What’s the difference between `torchvision.datasets.samplers.RandomClipSampler` and `torch.utils.data.sampler.RandomSampler`",949 "id": 108280,950 "title": "What's the difference between `torchvision.datasets.samplers.RandomClipSampler` and `torch.utils.data.sampler.RandomSampler`",951 "posts_count": 1,952 "created_at": "2021-01-08T16:16:22.160Z",953 "views": 459,954 "reply_count": 0,955 "like_count": 1,956 "last_posted_at": "2021-01-08T16:16:22.220Z",957 "visible": true,958 "closed": false,959 "archived": false,960 "has_summary": false,961 "archetype": "regular",962 "slug": "whats-the-difference-between-torchvision-datasets-samplers-randomclipsampler-and-torch-utils-data-sampler-randomsampler",963 "category_id": 1,964 "word_count": 35,965 "deleted_at": null,966 "user_id": 39605,967 "featured_link": null,968 "pinned_globally": false,969 "pinned_at": null,970 "pinned_until": null,971 "image_url": null,972 "slow_mode_seconds": 0,973 "draft": null,974 "draft_key": "topic_108280",975 "draft_sequence": null,976 "unpinned": null,977 "pinned": false,978 "current_post_number": 1,979 "highest_post_number": 1,980 "deleted_by": null,981 "actions_summary": [982 {983 "id": 4,984 "count": 0,985 "hidden": false,986 "can_act": false987 },988 {989 "id": 8,990 "count": 0,991 "hidden": false,992 "can_act": false993 },994 {995 "id": 10,996 "count": 0,997 "hidden": false,998 "can_act": false999 },1000 {1001 "id": 7,1002 "count": 0,1003 "hidden": false,1004 "can_act": false1005 }1006 ],1007 "chunk_size": 20,1008 "bookmarked": false,1009 "topic_timer": null,1010 "message_bus_last_id": 0,1011 "participant_count": 1,1012 "show_read_indicator": false,1013 "thumbnails": null,1014 "slow_mode_enabled_until": null,1015 "can_vote": false,1016 "vote_count": 0,1017 "user_voted": false,1018 "discourse_zendesk_plugin_zendesk_id": null,1019 "discourse_zendesk_plugin_zendesk_url": "https://your-url.zendesk.com/agent/tickets/",1020 "details": {1021 "can_edit": false,1022 "notification_level": 1,1023 "participants": [1024 {1025 "id": 39605,1026 "username": "111429",1027 "name": "zuujhyt",1028 "avatar_template": "/letter_avatar_proxy/v4/letter/1/e0b2c6/{size}.png",1029 "post_count": 1,1030 "primary_group_name": null,1031 "flair_name": null,1032 "flair_url": null,1033 "flair_color": null,1034 "flair_bg_color": null,1035 "flair_group_id": null,1036 "trust_level": 11037 }1038 ],1039 "created_by": {1040 "id": 39605,1041 "username": "111429",1042 "name": "zuujhyt",1043 "avatar_template": "/letter_avatar_proxy/v4/letter/1/e0b2c6/{size}.png"1044 },1045 "last_poster": {1046 "id": 39605,1047 "username": "111429",1048 "name": "zuujhyt",1049 "avatar_template": "/letter_avatar_proxy/v4/letter/1/e0b2c6/{size}.png"1050 }1051 },1052 "bookmarks": []1053 },1054 {1055 "post_stream": {1056 "posts": [1057 {1058 "id": 255494,1059 "name": "Bruce",1060 "username": "Bruce_zhuang",1061 "avatar_template": "/user_avatar/discuss.pytorch.org/bruce_zhuang/{size}/10713_2.png",1062 "created_at": "2021-01-07T07:08:46.672Z",1063 "cooked": "<p>I need to somewhat move the activations to cpu after the computation graph is generated. For instance, some computation graphs are generated in the GPU as follows:</p>\n<blockquote>\n<p>input1, input2 = torch.randn(128,3,32,32)<br>\noutput1 = model(input)<br>\noutput2 = model(input)</p>\n</blockquote>\n<p>which would need the memory of about <code>2M</code> with 2 computational graphs generated. Now I need to place the activations associated with <code>output1</code> to <code>cpu</code>, such that the memory in GPU would fall down to about <code>M</code>. This should make sense, but I don’t know how to implement it. Thanks!</p>",1064 "post_number": 1,1065 "post_type": 1,1066 "posts_count": 4,1067 "updated_at": "2021-01-07T07:08:46.672Z",1068 "reply_count": 0,1069 "reply_to_post_number": null,1070 "quote_count": 0,1071 "incoming_link_count": 142,1072 "reads": 20,1073 "readers_count": 19,1074 "score": 714.0,1075 "yours": false,1076 "topic_id": 108146,1077 "topic_slug": "move-activations-to-device",1078 "display_username": "Bruce",1079 "primary_group_name": null,1080 "flair_name": null,1081 "flair_url": null,1082 "flair_bg_color": null,1083 "flair_color": null,1084 "flair_group_id": null,1085 "badges_granted": [],1086 "version": 1,1087 "can_edit": false,1088 "can_delete": false,1089 "can_recover": false,1090 "can_see_hidden_post": false,1091 "can_wiki": false,1092 "read": true,1093 "user_title": "",1094 "bookmarked": false,1095 "actions_summary": [],1096 "moderator": false,1097 "admin": false,1098 "staff": false,1099 "user_id": 17345,1100 "hidden": false,1101 "trust_level": 1,1102 "deleted_at": null,1103 "user_deleted": false,1104 "edit_reason": null,1105 "can_view_edit_history": true,1106 "wiki": false,1107 "post_url": "/t/move-activations-to-device/108146/1",1108 "can_accept_answer": false,1109 "can_unaccept_answer": false,1110 "accepted_answer": false,1111 "topic_accepted_answer": null,1112 "can_vote": false1113 },1114 {1115 "id": 255650,1116 "name": "Richard Zou",1117 "username": "richard",1118 "avatar_template": "/user_avatar/discuss.pytorch.org/richard/{size}/1861_2.png",1119 "created_at": "2021-01-07T16:22:52.155Z",1120 "cooked": "<p>You can move <code>output1</code> to cpu by doing <code>output1.cpu()</code>. Or are you asking how to move activations that were saved by the PyTorch autograd engine to CPU to avoid OOM-ing?</p>",1121 "post_number": 2,1122 "post_type": 1,1123 "posts_count": 4,1124 "updated_at": "2021-01-07T16:22:52.155Z",1125 "reply_count": 1,1126 "reply_to_post_number": null,1127 "quote_count": 0,1128 "incoming_link_count": 6,1129 "reads": 18,1130 "readers_count": 17,1131 "score": 38.6,1132 "yours": false,1133 "topic_id": 108146,1134 "topic_slug": "move-activations-to-device",1135 "display_username": "Richard Zou",1136 "primary_group_name": null,1137 "flair_name": null,1138 "flair_url": null,1139 "flair_bg_color": null,1140 "flair_color": null,1141 "flair_group_id": null,1142 "badges_granted": [],1143 "version": 1,1144 "can_edit": false,1145 "can_delete": false,1146 "can_recover": false,1147 "can_see_hidden_post": false,1148 "can_wiki": false,1149 "read": true,1150 "user_title": "",1151 "bookmarked": false,1152 "actions_summary": [],1153 "moderator": true,1154 "admin": false,1155 "staff": true,1156 "user_id": 3718,1157 "hidden": false,1158 "trust_level": 2,1159 "deleted_at": null,1160 "user_deleted": false,1161 "edit_reason": null,1162 "can_view_edit_history": true,1163 "wiki": false,1164 "post_url": "/t/move-activations-to-device/108146/2",1165 "can_accept_answer": false,1166 "can_unaccept_answer": false,1167 "accepted_answer": false,1168 "topic_accepted_answer": null1169 },1170 {1171 "id": 255770,1172 "name": "Bruce",1173 "username": "Bruce_zhuang",1174 "avatar_template": "/user_avatar/discuss.pytorch.org/bruce_zhuang/{size}/10713_2.png",1175 "created_at": "2021-01-08T07:01:54.623Z",1176 "cooked": "<p>Yes. I am trying to move all the activations linked with <code>output1</code>, not just the tensor itself.</p>",1177 "post_number": 3,1178 "post_type": 1,1179 "posts_count": 4,1180 "updated_at": "2021-01-08T07:01:54.623Z",1181 "reply_count": 0,1182 "reply_to_post_number": 2,1183 "quote_count": 0,1184 "incoming_link_count": 0,1185 "reads": 14,1186 "readers_count": 13,1187 "score": 2.8,1188 "yours": false,1189 "topic_id": 108146,1190 "topic_slug": "move-activations-to-device",1191 "display_username": "Bruce",1192 "primary_group_name": null,1193 "flair_name": null,1194 "flair_url": null,1195 "flair_bg_color": null,1196 "flair_color": null,1197 "flair_group_id": null,1198 "badges_granted": [],1199 "version": 1,1200 "can_edit": false,