Anurag1734/cuda-error-resolution-analysis
07
1[2 {3 "post_stream": {4 "posts": [5 {6 "id": 218325,7 "name": "Leigaoxiang",8 "username": "leigaoxiang1",9 "avatar_template": "/user_avatar/discuss.pytorch.org/leigaoxiang1/{size}/65312_2.png",10 "created_at": "2020-08-05T08:04:09.091Z",11 "cooked": "<p>TracerWarning: There are 2 live references to the data region being modified when tracing in-place operator copy_ (possibly due to an assignment). This might cause the trace to be incorrect, because all other views that also reference this data will not reflect this change in the trace! On the other hand, if all other views use the same memory chunk, but are disjoint (e.g. are outputs of torch.split), this might still be safe.<br>\npred_boxes[…, 0] = x.data + self.grid_x</p>\n<p>how to mdify the code?</p>",12 "post_number": 1,13 "post_type": 1,14 "posts_count": 1,15 "updated_at": "2020-08-05T08:04:09.091Z",16 "reply_count": 0,17 "reply_to_post_number": null,18 "quote_count": 0,19 "incoming_link_count": 8,20 "reads": 7,21 "readers_count": 6,22 "score": 41.4,23 "yours": false,24 "topic_id": 91693,25 "topic_slug": "onnx-error-then-convert-pth-to-onnx",26 "display_username": "Leigaoxiang",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": 20464,48 "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/onnx-error-then-convert-pth-to-onnx/91693/1",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 21832565 ]66 },67 "timeline_lookup": [68 [69 1,70 190871 ]72 ],73 "suggested_topics": [74 {75 "fancy_title": "From RAM to GRAM",76 "id": 212269,77 "title": "From RAM to GRAM",78 "slug": "from-ram-to-gram",79 "posts_count": 2,80 "reply_count": 0,81 "highest_post_number": 2,82 "image_url": null,83 "created_at": "2024-10-29T12:13:24.646Z",84 "last_posted_at": "2024-10-29T13:30:42.838Z",85 "bumped": true,86 "bumped_at": "2024-10-29T13:30:42.838Z",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": 1,98 "views": 151,99 "category_id": 1,100 "featured_link": null,101 "has_accepted_answer": false,102 "posters": [103 {104 "extras": null,105 "description": "Original Poster",106 "user": {107 "id": 71569,108 "username": "Manvel_Petrosyan",109 "name": "Manvel Petrosyan",110 "avatar_template": "/user_avatar/discuss.pytorch.org/manvel_petrosyan/{size}/66101_2.png",111 "trust_level": 1112 }113 },114 {115 "extras": "latest",116 "description": "Most Recent Poster",117 "user": {118 "id": 12117,119 "username": "Tony-Y",120 "name": "",121 "avatar_template": "/user_avatar/discuss.pytorch.org/tony-y/{size}/8373_2.png",122 "trust_level": 2123 }124 }125 ]126 },127 {128 "fancy_title": "Pt_main_thread occasionally gets stuck, leading to training time increased",129 "id": 213404,130 "title": "Pt_main_thread occasionally gets stuck, leading to training time increased",131 "slug": "pt-main-thread-occasionally-gets-stuck-leading-to-training-time-increased",132 "posts_count": 1,133 "reply_count": 0,134 "highest_post_number": 1,135 "image_url": "https://discuss.pytorch.org/uploads/default/optimized/3X/2/a/2afc851e8d5e13a3b7ec514e48189417e5182bb9_2_1024x236.png",136 "created_at": "2024-11-25T08:32:58.086Z",137 "last_posted_at": "2024-11-25T08:32:58.136Z",138 "bumped": true,139 "bumped_at": "2024-11-25T08:32:58.136Z",140 "archetype": "regular",141 "unseen": false,142 "pinned": false,143 "unpinned": null,144 "visible": true,145 "closed": false,146 "archived": false,147 "bookmarked": null,148 "liked": null,149 "tags_descriptions": {},150 "like_count": 0,151 "views": 312,152 "category_id": 1,153 "featured_link": null,154 "has_accepted_answer": false,155 "posters": [156 {157 "extras": "latest single",158 "description": "Original Poster, Most Recent Poster",159 "user": {160 "id": 72712,161 "username": "xgbj",162 "name": "Xgbj",163 "avatar_template": "/user_avatar/discuss.pytorch.org/xgbj/{size}/67179_2.png",164 "trust_level": 1165 }166 }167 ]168 },169 {170 "fancy_title": "How can I parallelize this module",171 "id": 213835,172 "title": "How can I parallelize this module",173 "slug": "how-can-i-parallelize-this-module",174 "posts_count": 3,175 "reply_count": 1,176 "highest_post_number": 3,177 "image_url": null,178 "created_at": "2024-12-05T06:12:00.024Z",179 "last_posted_at": "2024-12-05T16:01:14.675Z",180 "bumped": true,181 "bumped_at": "2024-12-05T16:01:14.675Z",182 "archetype": "regular",183 "unseen": false,184 "pinned": false,185 "unpinned": null,186 "visible": true,187 "closed": false,188 "archived": false,189 "bookmarked": null,190 "liked": null,191 "tags_descriptions": {},192 "like_count": 2,193 "views": 43,194 "category_id": 1,195 "featured_link": null,196 "has_accepted_answer": true,197 "posters": [198 {199 "extras": "latest",200 "description": "Original Poster, Most Recent Poster",201 "user": {202 "id": 60868,203 "username": "Fox11",204 "name": "Fox11",205 "avatar_template": "/user_avatar/discuss.pytorch.org/fox11/{size}/54756_2.png",206 "trust_level": 1207 }208 },209 {210 "extras": null,211 "description": "Frequent Poster, Accepted Answer",212 "user": {213 "id": 81100,214 "username": "Space4444",215 "name": "Space4444",216 "avatar_template": "/user_avatar/discuss.pytorch.org/space4444/{size}/74167_2.png",217 "trust_level": 1218 }219 }220 ]221 },222 {223 "fancy_title": "Pytorch Weighted Regression Loss Functions",224 "id": 212480,225 "title": "Pytorch Weighted Regression Loss Functions",226 "slug": "pytorch-weighted-regression-loss-functions",227 "posts_count": 4,228 "reply_count": 2,229 "highest_post_number": 4,230 "image_url": null,231 "created_at": "2024-11-03T17:06:32.230Z",232 "last_posted_at": "2024-11-04T23:07:20.590Z",233 "bumped": true,234 "bumped_at": "2024-11-04T23:07:20.590Z",235 "archetype": "regular",236 "unseen": false,237 "pinned": false,238 "unpinned": null,239 "visible": true,240 "closed": false,241 "archived": false,242 "bookmarked": null,243 "liked": null,244 "tags_descriptions": {},245 "like_count": 1,246 "views": 530,247 "category_id": 1,248 "featured_link": null,249 "has_accepted_answer": true,250 "posters": [251 {252 "extras": null,253 "description": "Original Poster",254 "user": {255 "id": 80345,256 "username": "Ajaikrish",257 "name": "Ajaikrishna Ramalingam",258 "avatar_template": "/user_avatar/discuss.pytorch.org/ajaikrish/{size}/72800_2.png",259 "trust_level": 1260 }261 },262 {263 "extras": "latest",264 "description": "Most Recent Poster, Accepted Answer",265 "user": {266 "id": 3534,267 "username": "ptrblck",268 "name": "",269 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",270 "admin": true,271 "moderator": true,272 "trust_level": 2273 }274 }275 ]276 },277 {278 "fancy_title": "Single core of TPU gives inference results different than the CPU results",279 "id": 215794,280 "title": "Single core of TPU gives inference results different than the CPU results",281 "slug": "single-core-of-tpu-gives-inference-results-different-than-the-cpu-results",282 "posts_count": 2,283 "reply_count": 0,284 "highest_post_number": 2,285 "image_url": null,286 "created_at": "2025-01-23T22:46:14.583Z",287 "last_posted_at": "2025-01-24T06:21:02.944Z",288 "bumped": true,289 "bumped_at": "2025-01-25T04:22:34.162Z",290 "archetype": "regular",291 "unseen": false,292 "pinned": false,293 "unpinned": null,294 "visible": true,295 "closed": false,296 "archived": false,297 "bookmarked": null,298 "liked": null,299 "tags_descriptions": {},300 "like_count": 1,301 "views": 44,302 "category_id": 1,303 "featured_link": null,304 "has_accepted_answer": false,305 "posters": [306 {307 "extras": null,308 "description": "Original Poster",309 "user": {310 "id": 82276,311 "username": "Mohamed_Amara",312 "name": "Mohamed Amara",313 "avatar_template": "/user_avatar/discuss.pytorch.org/mohamed_amara/{size}/75269_2.png",314 "trust_level": 1315 }316 },317 {318 "extras": "latest",319 "description": "Most Recent Poster",320 "user": {321 "id": 82281,322 "username": "frank58rich",323 "name": "",324 "avatar_template": "/letter_avatar_proxy/v4/letter/f/ce7236/{size}.png",325 "trust_level": 0326 }327 }328 ]329 }330 ],331 "tags_descriptions": {},332 "fancy_title": "Onnx error then convert pth to onnx",333 "id": 91693,334 "title": "Onnx error then convert pth to onnx",335 "posts_count": 1,336 "created_at": "2020-08-05T08:04:09.044Z",337 "views": 276,338 "reply_count": 0,339 "like_count": 0,340 "last_posted_at": "2020-08-05T08:04:09.091Z",341 "visible": true,342 "closed": false,343 "archived": false,344 "has_summary": false,345 "archetype": "regular",346 "slug": "onnx-error-then-convert-pth-to-onnx",347 "category_id": 1,348 "word_count": 87,349 "deleted_at": null,350 "user_id": 20464,351 "featured_link": null,352 "pinned_globally": false,353 "pinned_at": null,354 "pinned_until": null,355 "image_url": null,356 "slow_mode_seconds": 0,357 "draft": null,358 "draft_key": "topic_91693",359 "draft_sequence": null,360 "unpinned": null,361 "pinned": false,362 "current_post_number": 1,363 "highest_post_number": 1,364 "deleted_by": null,365 "actions_summary": [366 {367 "id": 4,368 "count": 0,369 "hidden": false,370 "can_act": false371 },372 {373 "id": 8,374 "count": 0,375 "hidden": false,376 "can_act": false377 },378 {379 "id": 10,380 "count": 0,381 "hidden": false,382 "can_act": false383 },384 {385 "id": 7,386 "count": 0,387 "hidden": false,388 "can_act": false389 }390 ],391 "chunk_size": 20,392 "bookmarked": false,393 "topic_timer": null,394 "message_bus_last_id": 0,395 "participant_count": 1,396 "show_read_indicator": false,397 "thumbnails": null,398 "slow_mode_enabled_until": null,399 "can_vote": false,400 "vote_count": 0,401 "user_voted": false,402 "discourse_zendesk_plugin_zendesk_id": null,403 "discourse_zendesk_plugin_zendesk_url": "https://your-url.zendesk.com/agent/tickets/",404 "details": {405 "can_edit": false,406 "notification_level": 1,407 "participants": [408 {409 "id": 20464,410 "username": "leigaoxiang1",411 "name": "Leigaoxiang",412 "avatar_template": "/user_avatar/discuss.pytorch.org/leigaoxiang1/{size}/65312_2.png",413 "post_count": 1,414 "primary_group_name": null,415 "flair_name": null,416 "flair_url": null,417 "flair_color": null,418 "flair_bg_color": null,419 "flair_group_id": null,420 "trust_level": 1421 }422 ],423 "created_by": {424 "id": 20464,425 "username": "leigaoxiang1",426 "name": "Leigaoxiang",427 "avatar_template": "/user_avatar/discuss.pytorch.org/leigaoxiang1/{size}/65312_2.png"428 },429 "last_poster": {430 "id": 20464,431 "username": "leigaoxiang1",432 "name": "Leigaoxiang",433 "avatar_template": "/user_avatar/discuss.pytorch.org/leigaoxiang1/{size}/65312_2.png"434 }435 },436 "bookmarks": []437 },438 {439 "post_stream": {440 "posts": [441 {442 "id": 217217,443 "name": "Krister Viirsaar",444 "username": "kristerv",445 "avatar_template": "/letter_avatar_proxy/v4/letter/k/90db22/{size}.png",446 "created_at": "2020-07-31T10:59:36.001Z",447 "cooked": "<p>My task is to take an episode of a TV show and its subtitles. Then make the subtitle timings more accurate (from 200ms to 20ms). So I want to learn what is speech and what is not.</p>\n<p>I’ve now taken the audio, converted it into a spectrogram and separated each column of the spectogram to be a single data item. So now I have two arrays:</p>\n<pre><code class=\"lang-python\">print(train_speech.size()) # torch.Size([93482, 201])\nprint(train_silence.size()) # torch.Size([35038, 201])\n</code></pre>\n<p>All I want to do is a simple multi-linear NN to make a difference. <code>train_speech</code> is FFT’s of people talking and <code>train_silence</code> is no talking (used subtitles for the distinction).</p>\n<p>My question is what DataLoader can I use to take these into torch?</p>",448 "post_number": 1,449 "post_type": 1,450 "posts_count": 5,451 "updated_at": "2020-07-31T10:59:36.001Z",452 "reply_count": 0,453 "reply_to_post_number": null,454 "quote_count": 0,455 "incoming_link_count": 10,456 "reads": 6,457 "readers_count": 5,458 "score": 51.2,459 "yours": false,460 "topic_id": 91216,461 "topic_slug": "load-data-from-variables-into-classes",462 "display_username": "Krister Viirsaar",463 "primary_group_name": null,464 "flair_name": null,465 "flair_url": null,466 "flair_bg_color": null,467 "flair_color": null,468 "flair_group_id": null,469 "badges_granted": [],470 "version": 1,471 "can_edit": false,472 "can_delete": false,473 "can_recover": false,474 "can_see_hidden_post": false,475 "can_wiki": false,476 "read": true,477 "user_title": null,478 "bookmarked": false,479 "actions_summary": [],480 "moderator": false,481 "admin": false,482 "staff": false,483 "user_id": 35002,484 "hidden": false,485 "trust_level": 1,486 "deleted_at": null,487 "user_deleted": false,488 "edit_reason": null,489 "can_view_edit_history": true,490 "wiki": false,491 "post_url": "/t/load-data-from-variables-into-classes/91216/1",492 "can_accept_answer": false,493 "can_unaccept_answer": false,494 "accepted_answer": false,495 "topic_accepted_answer": null,496 "can_vote": false497 },498 {499 "id": 217417,500 "name": "",501 "username": "ptrblck",502 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",503 "created_at": "2020-08-01T07:24:13.331Z",504 "cooked": "<p>There is one <a href=\"https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader\"><code>DataLoader</code></a>, which accepts a <code>Dataset</code> and provides different functionalities such as shuffling, creating batches using multiple workers, etc.</p>\n<p>To create a custom <code>Dataset</code> you could have a look at <a href=\"https://pytorch.org/tutorials/beginner/data_loading_tutorial.html\">this tutorial</a>. <img src=\"https://discuss.pytorch.org/images/emoji/apple/slight_smile.png?v=9\" title=\":slight_smile:\" class=\"emoji\" alt=\":slight_smile:\"></p>",505 "post_number": 2,506 "post_type": 1,507 "posts_count": 5,508 "updated_at": "2020-08-01T07:24:13.331Z",509 "reply_count": 0,510 "reply_to_post_number": null,511 "quote_count": 0,512 "incoming_link_count": 0,513 "reads": 5,514 "readers_count": 4,515 "score": 1.0,516 "yours": false,517 "topic_id": 91216,518 "topic_slug": "load-data-from-variables-into-classes",519 "display_username": "",520 "primary_group_name": null,521 "flair_name": null,522 "flair_url": null,523 "flair_bg_color": null,524 "flair_color": null,525 "flair_group_id": null,526 "badges_granted": [],527 "version": 1,528 "can_edit": false,529 "can_delete": false,530 "can_recover": false,531 "can_see_hidden_post": false,532 "can_wiki": false,533 "link_counts": [534 {535 "url": "https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader",536 "internal": false,537 "reflection": false,538 "title": "torch.utils.data — PyTorch 1.6.0 documentation",539 "clicks": 2540 },541 {542 "url": "https://pytorch.org/tutorials/beginner/data_loading_tutorial.html",543 "internal": false,544 "reflection": false,545 "title": "Writing Custom Datasets, DataLoaders and Transforms — PyTorch Tutorials 1.6.0 documentation",546 "clicks": 1547 }548 ],549 "read": true,550 "user_title": "",551 "bookmarked": false,552 "actions_summary": [],553 "moderator": true,554 "admin": true,555 "staff": true,556 "user_id": 3534,557 "hidden": false,558 "trust_level": 2,559 "deleted_at": null,560 "user_deleted": false,561 "edit_reason": null,562 "can_view_edit_history": true,563 "wiki": false,564 "post_url": "/t/load-data-from-variables-into-classes/91216/2",565 "can_accept_answer": false,566 "can_unaccept_answer": false,567 "accepted_answer": false,568 "topic_accepted_answer": null569 },570 {571 "id": 217809,572 "name": "Krister Viirsaar",573 "username": "kristerv",574 "avatar_template": "/letter_avatar_proxy/v4/letter/k/90db22/{size}.png",575 "created_at": "2020-08-03T11:49:48.943Z",576 "cooked": "<p>What I don’t get is that my data is already a simple tensor… It doesn’t make sense to me that I need to create a separate abstraction just to fetch numbers from a few arrays…</p>",577 "post_number": 3,578 "post_type": 1,579 "posts_count": 5,580 "updated_at": "2020-08-03T11:49:48.943Z",581 "reply_count": 1,582 "reply_to_post_number": null,583 "quote_count": 0,584 "incoming_link_count": 2,585 "reads": 4,586 "readers_count": 3,587 "score": 15.8,588 "yours": false,589 "topic_id": 91216,590 "topic_slug": "load-data-from-variables-into-classes",591 "display_username": "Krister Viirsaar",592 "primary_group_name": null,593 "flair_name": null,594 "flair_url": null,595 "flair_bg_color": null,596 "flair_color": null,597 "flair_group_id": null,598 "badges_granted": [],599 "version": 1,600 "can_edit": false,601 "can_delete": false,602 "can_recover": false,603 "can_see_hidden_post": false,604 "can_wiki": false,605 "read": true,606 "user_title": null,607 "bookmarked": false,608 "actions_summary": [],609 "moderator": false,610 "admin": false,611 "staff": false,612 "user_id": 35002,613 "hidden": false,614 "trust_level": 1,615 "deleted_at": null,616 "user_deleted": false,617 "edit_reason": null,618 "can_view_edit_history": true,619 "wiki": false,620 "post_url": "/t/load-data-from-variables-into-classes/91216/3",621 "can_accept_answer": false,622 "can_unaccept_answer": false,623 "accepted_answer": false,624 "topic_accepted_answer": null625 },626 {627 "id": 218061,628 "name": "",629 "username": "ptrblck",630 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",631 "created_at": "2020-08-04T08:51:42.073Z",632 "cooked": "<p>If your data is already stored as tensors, you can just use <a href=\"https://pytorch.org/docs/stable/data.html#torch.utils.data.TensorDataset\"><code>TensorDataset</code></a> or completely skip the abstraction and just feed to data to your model.</p>",633 "post_number": 4,634 "post_type": 1,635 "posts_count": 5,636 "updated_at": "2020-08-04T08:52:02.057Z",637 "reply_count": 0,638 "reply_to_post_number": 3,639 "quote_count": 0,640 "incoming_link_count": 0,641 "reads": 5,642 "readers_count": 4,643 "score": 1.0,644 "yours": false,645 "topic_id": 91216,646 "topic_slug": "load-data-from-variables-into-classes",647 "display_username": "",648 "primary_group_name": null,649 "flair_name": null,650 "flair_url": null,651 "flair_bg_color": null,652 "flair_color": null,653 "flair_group_id": null,654 "badges_granted": [],655 "version": 1,656 "can_edit": false,657 "can_delete": false,658 "can_recover": false,659 "can_see_hidden_post": false,660 "can_wiki": false,661 "link_counts": [662 {663 "url": "https://pytorch.org/docs/stable/data.html#torch.utils.data.TensorDataset",664 "internal": false,665 "reflection": false,666 "title": "torch.utils.data — PyTorch 1.6.0 documentation",667 "clicks": 0668 }669 ],670 "read": true,671 "user_title": "",672 "reply_to_user": {673 "id": 35002,674 "username": "kristerv",675 "name": "Krister Viirsaar",676 "avatar_template": "/letter_avatar_proxy/v4/letter/k/90db22/{size}.png"677 },678 "bookmarked": false,679 "actions_summary": [],680 "moderator": true,681 "admin": true,682 "staff": true,683 "user_id": 3534,684 "hidden": false,685 "trust_level": 2,686 "deleted_at": null,687 "user_deleted": false,688 "edit_reason": null,689 "can_view_edit_history": true,690 "wiki": false,691 "post_url": "/t/load-data-from-variables-into-classes/91216/4",692 "can_accept_answer": false,693 "can_unaccept_answer": false,694 "accepted_answer": false,695 "topic_accepted_answer": null696 },697 {698 "id": 218321,699 "name": "Krister Viirsaar",700 "username": "kristerv",701 "avatar_template": "/letter_avatar_proxy/v4/letter/k/90db22/{size}.png",702 "created_at": "2020-08-05T07:59:43.135Z",703 "cooked": "<p>Since I had two classes in separate variables I ended up making the custom class.</p>\n<pre><code class=\"lang-auto\">class MyDataset(Dataset):\n def __init__(self, speech, silence):\n self.data = list(map(lambda x: (x, 1), speech)) + list(map(lambda x: (x, 0), silence))\n \n def __getitem__(self, index):\n return self.data[index]\n \n def __len__(self):\n return len(self.data)\n\ntrain_ds = MyDataset(train_speech, train_silence)\ntrain_dl = DataLoader(train_ds, shuffle=True, batch_size=1024)\n</code></pre>\n<p>Thanks for helping me get through this.</p>",704 "post_number": 5,705 "post_type": 1,706 "posts_count": 5,707 "updated_at": "2020-08-05T07:59:43.135Z",708 "reply_count": 0,709 "reply_to_post_number": null,710 "quote_count": 0,711 "incoming_link_count": 0,712 "reads": 5,713 "readers_count": 4,714 "score": 1.0,715 "yours": false,716 "topic_id": 91216,717 "topic_slug": "load-data-from-variables-into-classes",718 "display_username": "Krister Viirsaar",719 "primary_group_name": null,720 "flair_name": null,721 "flair_url": null,722 "flair_bg_color": null,723 "flair_color": null,724 "flair_group_id": null,725 "badges_granted": [],726 "version": 1,727 "can_edit": false,728 "can_delete": false,729 "can_recover": false,730 "can_see_hidden_post": false,731 "can_wiki": false,732 "read": true,733 "user_title": null,734 "bookmarked": false,735 "actions_summary": [],736 "moderator": false,737 "admin": false,738 "staff": false,739 "user_id": 35002,740 "hidden": false,741 "trust_level": 1,742 "deleted_at": null,743 "user_deleted": false,744 "edit_reason": null,745 "can_view_edit_history": true,746 "wiki": false,747 "post_url": "/t/load-data-from-variables-into-classes/91216/5",748 "can_accept_answer": false,749 "can_unaccept_answer": false,750 "accepted_answer": false,751 "topic_accepted_answer": null752 }753 ],754 "stream": [755 217217,756 217417,757 217809,758 218061,759 218321760 ]761 },762 "timeline_lookup": [763 [764 1,765 1913766 ],767 [768 2,769 1912770 ],771 [772 3,773 1909774 ],775 [776 5,777 1908778 ]779 ],780 "suggested_topics": [781 {782 "fancy_title": "CTC loss inputs and input lengths",783 "id": 219837,784 "title": "CTC loss inputs and input lengths",785 "slug": "ctc-loss-inputs-and-input-lengths",786 "posts_count": 1,787 "reply_count": 0,788 "highest_post_number": 1,789 "image_url": null,790 "created_at": "2025-05-07T13:49:43.750Z",791 "last_posted_at": "2025-05-07T13:49:43.799Z",792 "bumped": true,793 "bumped_at": "2025-05-07T13:49:43.799Z",794 "archetype": "regular",795 "unseen": false,796 "pinned": false,797 "unpinned": null,798 "visible": true,799 "closed": false,800 "archived": false,801 "bookmarked": null,802 "liked": null,803 "tags_descriptions": {},804 "like_count": 0,805 "views": 67,806 "category_id": 9,807 "featured_link": null,808 "has_accepted_answer": false,809 "posters": [810 {811 "extras": "latest single",812 "description": "Original Poster, Most Recent Poster",813 "user": {814 "id": 83384,815 "username": "alicemabille",816 "name": "Alice Mabille",817 "avatar_template": "/user_avatar/discuss.pytorch.org/alicemabille/{size}/76264_2.png",818 "trust_level": 1819 }820 }821 ]822 },823 {824 "fancy_title": "Issues with CTCLoss",825 "id": 216390,826 "title": "Issues with CTCLoss",827 "slug": "issues-with-ctcloss",828 "posts_count": 1,829 "reply_count": 0,830 "highest_post_number": 1,831 "image_url": null,832 "created_at": "2025-02-08T01:34:34.042Z",833 "last_posted_at": "2025-02-08T01:34:34.086Z",834 "bumped": true,835 "bumped_at": "2025-02-08T01:34:34.086Z",836 "archetype": "regular",837 "unseen": false,838 "pinned": false,839 "unpinned": null,840 "visible": true,841 "closed": false,842 "archived": false,843 "bookmarked": null,844 "liked": null,845 "tags_descriptions": {},846 "like_count": 0,847 "views": 63,848 "category_id": 9,849 "featured_link": null,850 "has_accepted_answer": false,851 "posters": [852 {853 "extras": "latest single",854 "description": "Original Poster, Most Recent Poster",855 "user": {856 "id": 82555,857 "username": "Butzti",858 "name": "Butzti",859 "avatar_template": "/user_avatar/discuss.pytorch.org/butzti/{size}/75538_2.png",860 "trust_level": 0861 }862 }863 ]864 },865 {866 "fancy_title": "[PR] Torchaudio incompatible with python flag -OO due to __doc__ being None",867 "id": 216897,868 "title": "[PR] Torchaudio incompatible with python flag -OO due to __doc__ being None",869 "slug": "pr-torchaudio-incompatible-with-python-flag-oo-due-to-doc-being-none",870 "posts_count": 3,871 "reply_count": 1,872 "highest_post_number": 3,873 "image_url": null,874 "created_at": "2025-02-19T14:32:46.063Z",875 "last_posted_at": "2025-02-19T15:16:06.680Z",876 "bumped": true,877 "bumped_at": "2025-02-19T15:16:06.680Z",878 "archetype": "regular",879 "unseen": false,880 "pinned": false,881 "unpinned": null,882 "visible": true,883 "closed": false,884 "archived": false,885 "bookmarked": null,886 "liked": null,887 "tags_descriptions": {},888 "like_count": 0,889 "views": 100,890 "category_id": 9,891 "featured_link": null,892 "has_accepted_answer": false,893 "posters": [894 {895 "extras": "latest",896 "description": "Original Poster, Most Recent Poster",897 "user": {898 "id": 82796,899 "username": "FremyCompany",900 "name": "François REMY",901 "avatar_template": "/user_avatar/discuss.pytorch.org/fremycompany/{size}/75760_2.png",902 "trust_level": 0903 }904 },905 {906 "extras": null,907 "description": "Frequent Poster",908 "user": {909 "id": 3534,910 "username": "ptrblck",911 "name": "",912 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",913 "admin": true,914 "moderator": true,915 "trust_level": 2916 }917 }918 ]919 },920 {921 "fancy_title": "Understanding channel dim in audio processing",922 "id": 212777,923 "title": "Understanding channel dim in audio processing",924 "slug": "understanding-channel-dim-in-audio-processing",925 "posts_count": 2,926 "reply_count": 0,927 "highest_post_number": 2,928 "image_url": null,929 "created_at": "2024-11-10T22:53:34.319Z",930 "last_posted_at": "2024-11-26T07:16:03.415Z",931 "bumped": true,932 "bumped_at": "2024-11-26T07:16:03.415Z",933 "archetype": "regular",934 "unseen": false,935 "pinned": false,936 "unpinned": null,937 "visible": true,938 "closed": false,939 "archived": false,940 "bookmarked": null,941 "liked": null,942 "tags_descriptions": {},943 "like_count": 1,944 "views": 73,945 "category_id": 9,946 "featured_link": null,947 "has_accepted_answer": false,948 "posters": [949 {950 "extras": null,951 "description": "Original Poster",952 "user": {953 "id": 80807,954 "username": "axion66",955 "name": "axion66",956 "avatar_template": "/letter_avatar_proxy/v4/letter/a/ec9cab/{size}.png",957 "trust_level": 0958 }959 },960 {961 "extras": "latest",962 "description": "Most Recent Poster",963 "user": {964 "id": 30366,965 "username": "shivammehta007",966 "name": "Shivam Mehta",967 "avatar_template": "/user_avatar/discuss.pytorch.org/shivammehta007/{size}/31230_2.png",968 "trust_level": 2969 }970 }971 ]972 },973 {974 "fancy_title": "Question about using pytorch-tcn vs. keras-tcn",975 "id": 220524,976 "title": "Question about using pytorch-tcn vs. keras-tcn",977 "slug": "question-about-using-pytorch-tcn-vs-keras-tcn",978 "posts_count": 1,979 "reply_count": 0,980 "highest_post_number": 1,981 "image_url": null,982 "created_at": "2025-06-02T18:54:46.961Z",983 "last_posted_at": "2025-06-02T18:54:47.014Z",984 "bumped": true,985 "bumped_at": "2025-06-03T12:16:59.851Z",986 "archetype": "regular",987 "unseen": false,988 "pinned": false,989 "unpinned": null,990 "visible": true,991 "closed": false,992 "archived": false,993 "bookmarked": null,994 "liked": null,995 "tags_descriptions": {},996 "like_count": 0,997 "views": 63,998 "category_id": 9,999 "featured_link": null,1000 "has_accepted_answer": false,1001 "posters": [1002 {1003 "extras": "latest single",1004 "description": "Original Poster, Most Recent Poster",1005 "user": {1006 "id": 84552,1007 "username": "User9393",1008 "name": "User911",1009 "avatar_template": "/user_avatar/discuss.pytorch.org/user9393/{size}/77243_2.png",1010 "trust_level": 01011 }1012 }1013 ]1014 }1015 ],1016 "tags_descriptions": {},1017 "fancy_title": "Load data from variables into classes",1018 "id": 91216,1019 "title": "Load data from variables into classes",1020 "posts_count": 5,1021 "created_at": "2020-07-31T10:59:35.945Z",1022 "views": 475,1023 "reply_count": 1,1024 "like_count": 0,1025 "last_posted_at": "2020-08-05T07:59:43.135Z",1026 "visible": true,1027 "closed": false,1028 "archived": false,1029 "has_summary": false,1030 "archetype": "regular",1031 "slug": "load-data-from-variables-into-classes",1032 "category_id": 9,1033 "word_count": 325,1034 "deleted_at": null,1035 "user_id": 35002,1036 "featured_link": null,1037 "pinned_globally": false,1038 "pinned_at": null,1039 "pinned_until": null,1040 "image_url": null,1041 "slow_mode_seconds": 0,1042 "draft": null,1043 "draft_key": "topic_91216",1044 "draft_sequence": null,1045 "unpinned": null,1046 "pinned": false,1047 "current_post_number": 1,1048 "highest_post_number": 5,1049 "deleted_by": null,1050 "actions_summary": [1051 {1052 "id": 4,1053 "count": 0,1054 "hidden": false,1055 "can_act": false1056 },1057 {1058 "id": 8,1059 "count": 0,1060 "hidden": false,1061 "can_act": false1062 },1063 {1064 "id": 10,1065 "count": 0,1066 "hidden": false,1067 "can_act": false1068 },1069 {1070 "id": 7,1071 "count": 0,1072 "hidden": false,1073 "can_act": false1074 }1075 ],1076 "chunk_size": 20,1077 "bookmarked": false,1078 "topic_timer": null,1079 "message_bus_last_id": 0,1080 "participant_count": 2,1081 "show_read_indicator": false,1082 "thumbnails": null,1083 "slow_mode_enabled_until": null,1084 "can_vote": false,1085 "vote_count": 0,1086 "user_voted": false,1087 "discourse_zendesk_plugin_zendesk_id": null,1088 "discourse_zendesk_plugin_zendesk_url": "https://your-url.zendesk.com/agent/tickets/",1089 "details": {1090 "can_edit": false,1091 "notification_level": 1,1092 "participants": [1093 {1094 "id": 35002,1095 "username": "kristerv",1096 "name": "Krister Viirsaar",1097 "avatar_template": "/letter_avatar_proxy/v4/letter/k/90db22/{size}.png",1098 "post_count": 3,1099 "primary_group_name": null,1100 "flair_name": null,1101 "flair_url": null,1102 "flair_color": null,1103 "flair_bg_color": null,1104 "flair_group_id": null,1105 "trust_level": 11106 },1107 {1108 "id": 3534,1109 "username": "ptrblck",1110 "name": "",1111 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",1112 "post_count": 2,1113 "primary_group_name": null,1114 "flair_name": null,1115 "flair_url": null,1116 "flair_color": null,1117 "flair_bg_color": null,1118 "flair_group_id": null,1119 "admin": true,1120 "moderator": true,1121 "trust_level": 21122 }1123 ],1124 "created_by": {1125 "id": 35002,1126 "username": "kristerv",1127 "name": "Krister Viirsaar",1128 "avatar_template": "/letter_avatar_proxy/v4/letter/k/90db22/{size}.png"1129 },1130 "last_poster": {1131 "id": 35002,1132 "username": "kristerv",1133 "name": "Krister Viirsaar",1134 "avatar_template": "/letter_avatar_proxy/v4/letter/k/90db22/{size}.png"1135 },1136 "links": [1137 {1138 "url": "https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader",1139 "title": "torch.utils.data — PyTorch 1.6.0 documentation",1140 "internal": false,1141 "attachment": false,1142 "reflection": false,1143 "clicks": 2,1144 "user_id": 3534,1145 "domain": "pytorch.org",1146 "root_domain": "pytorch.org"1147 },1148 {1149 "url": "https://pytorch.org/tutorials/beginner/data_loading_tutorial.html",1150 "title": "Writing Custom Datasets, DataLoaders and Transforms — PyTorch Tutorials 1.6.0 documentation",1151 "internal": false,1152 "attachment": false,1153 "reflection": false,1154 "clicks": 1,1155 "user_id": 3534,1156 "domain": "pytorch.org",1157 "root_domain": "pytorch.org"1158 }1159 ]1160 },1161 "bookmarks": []1162 },1163 {1164 "post_stream": {1165 "posts": [1166 {1167 "id": 218268,1168 "name": "",1169 "username": "legoh",1170 "avatar_template": "/letter_avatar_proxy/v4/letter/l/82dd89/{size}.png",1171 "created_at": "2020-08-05T05:46:32.210Z",1172 "cooked": "<p>Hi, I’m facing a bit of an issue in terms of running an algorithm on top of a computational graph. Some context, <code>m_alpha_beta</code> and <code>m_beta_alpha</code> are parameters that are a result from some neural network computation. I’m trying to run an algorithm on top of these parameters, followed by taking the loss of the output and backpropagating it to learn the network. At this stage, training is a problem because I’m getting an error of “RuntimeError: leaf variable has been moved into the graph interior”, which I suspect is due to in-place operations? My model will return <code>m_alpha_beta_k</code> and <code>m_beta_alpha_k</code> and I will calculate some cross-entropy loss based on it.</p>\n<p>Is there any way to prevent this issue or to code it better?</p>\n<pre><code class=\"lang-auto\">def simp_min_sum_batch(self, m_alpha_beta, m_beta_alpha):\n n = m_alpha_beta.size(1)\n # Define message vectors\n # Each row belongs to one of the nodes, we have n alpha nodes and n beta nodes\n # For row i and column j, each entry will be the message vector of alpha_i to beta_j, M_{alpha_i -> beta_j} of length n,\n # where each entry of this vector is m_{alpha_i -> beta_j} (q)\n m_alpha_beta_k = torch.zeros((m_alpha_beta.size(0), m_alpha_beta.size(1), m_alpha_beta.size(2)),\n device=m_alpha_beta.device, requires_grad=True)\n m_beta_alpha_k = torch.zeros((m_beta_alpha.size(0), m_beta_alpha.size(1), m_beta_alpha.size(2)),\n device=m_beta_alpha.device, requires_grad=True)\n\n # Message passing\n for i in range(n):\n m_beta_alpha_k[:, :, i] = m_alpha_beta[:, i, :] - torch.max(\n torch.cat((m_alpha_beta[:, :i, :], m_alpha_beta[:, (i + 1):, :]), dim=1), dim=1)[0]\n m_alpha_beta_k[:, :, i] = m_alpha_beta[:, :, i] - torch.max(\n torch.cat((m_beta_alpha[:, :i, :], m_beta_alpha[:, (i + 1):, :]), dim=1), dim=1)[0]\n\n return m_alpha_beta_k, m_beta_alpha_k\n</code></pre>",1173 "post_number": 1,1174 "post_type": 1,1175 "posts_count": 4,1176 "updated_at": "2020-08-05T05:46:32.210Z",1177 "reply_count": 0,1178 "reply_to_post_number": null,1179 "quote_count": 0,1180 "incoming_link_count": 24,1181 "reads": 8,1182 "readers_count": 7,1183 "score": 121.6,1184 "yours": false,1185 "topic_id": 91674,1186 "topic_slug": "how-to-prevent-leaf-variable-from-moving-into-graph-interior",1187 "display_username": "",1188 "primary_group_name": null,1189 "flair_name": null,1190 "flair_url": null,1191 "flair_bg_color": null,1192 "flair_color": null,1193 "flair_group_id": null,1194 "badges_granted": [],1195 "version": 1,1196 "can_edit": false,1197 "can_delete": false,1198 "can_recover": false,1199 "can_see_hidden_post": false,1200 "can_wiki": false,