Anurag1734/cuda-error-resolution-analysis
07
1[2 {3 "post_stream": {4 "posts": [5 {6 "id": 242500,7 "name": "Steven",8 "username": "Chen0729",9 "avatar_template": "/user_avatar/discuss.pytorch.org/chen0729/{size}/15587_2.png",10 "created_at": "2020-11-09T02:06:22.072Z",11 "cooked": "<p>Hi, All</p>\n<p>I have inquiry about how to convert tensorboard.summarywriter to C++<br>\nThe python code is<br>\nSelf.writer=tensorboardX.SummaryWriter()</p>\n<p>Any suggestions are welcome.</p>\n<p>Thanks!</p>",12 "post_number": 1,13 "post_type": 1,14 "posts_count": 2,15 "updated_at": "2020-11-09T02:06:22.072Z",16 "reply_count": 0,17 "reply_to_post_number": null,18 "quote_count": 0,19 "incoming_link_count": 59,20 "reads": 10,21 "readers_count": 9,22 "score": 297.0,23 "yours": false,24 "topic_id": 102108,25 "topic_slug": "about-conversion-of-summarywriter-to-c",26 "display_username": "Steven",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": "",42 "bookmarked": false,43 "actions_summary": [],44 "moderator": false,45 "admin": false,46 "staff": false,47 "user_id": 36581,48 "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/about-conversion-of-summarywriter-to-c/102108/1",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 "id": 242555,64 "name": "",65 "username": "ptrblck",66 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",67 "created_at": "2020-11-09T07:17:28.836Z",68 "cooked": "<p>I don’t think <code>tensorboard</code> has a built-in C++ API, but you might be able to find some 3rd party implementations.<br>\n<a href=\"https://github.com/pytorch/pytorch/issues/26097\">Here</a> is a tracker for a <a href=\"https://github.com/tensorflow/tensorboard/issues/2636\">feature request</a> to enable C++ logging, which apparently didn’t get much attention.</p>",69 "post_number": 2,70 "post_type": 1,71 "posts_count": 2,72 "updated_at": "2020-11-09T07:17:28.836Z",73 "reply_count": 0,74 "reply_to_post_number": null,75 "quote_count": 0,76 "incoming_link_count": 1,77 "reads": 9,78 "readers_count": 8,79 "score": 6.8,80 "yours": false,81 "topic_id": 102108,82 "topic_slug": "about-conversion-of-summarywriter-to-c",83 "display_username": "",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 "link_counts": [98 {99 "url": "https://github.com/pytorch/pytorch/issues/26097",100 "internal": false,101 "reflection": false,102 "title": "[RFC] TensorBoard extensions and improvements for PyTorch · Issue #26097 · pytorch/pytorch · GitHub",103 "clicks": 7104 },105 {106 "url": "https://github.com/tensorflow/tensorboard/issues/2636",107 "internal": false,108 "reflection": false,109 "title": "Feature Request: C++ logging from non-TensorFlow frameworks · Issue #2636 · tensorflow/tensorboard · GitHub",110 "clicks": 3111 }112 ],113 "read": true,114 "user_title": "",115 "bookmarked": false,116 "actions_summary": [],117 "moderator": true,118 "admin": true,119 "staff": true,120 "user_id": 3534,121 "hidden": false,122 "trust_level": 2,123 "deleted_at": null,124 "user_deleted": false,125 "edit_reason": null,126 "can_view_edit_history": true,127 "wiki": false,128 "post_url": "/t/about-conversion-of-summarywriter-to-c/102108/2",129 "can_accept_answer": false,130 "can_unaccept_answer": false,131 "accepted_answer": false,132 "topic_accepted_answer": null133 }134 ],135 "stream": [136 242500,137 242555138 ]139 },140 "timeline_lookup": [141 [142 1,143 1812144 ]145 ],146 "suggested_topics": [147 {148 "fancy_title": "Torch::cuda::is_available() always returns false",149 "id": 221281,150 "title": "Torch::cuda::is_available() always returns false",151 "slug": "torch-is-available-always-returns-false",152 "posts_count": 2,153 "reply_count": 0,154 "highest_post_number": 2,155 "image_url": null,156 "created_at": "2025-07-04T08:26:33.951Z",157 "last_posted_at": "2025-07-15T06:56:46.817Z",158 "bumped": true,159 "bumped_at": "2025-07-23T08:31:08.437Z",160 "archetype": "regular",161 "unseen": false,162 "pinned": false,163 "unpinned": null,164 "visible": true,165 "closed": false,166 "archived": false,167 "bookmarked": null,168 "liked": null,169 "unicode_title": "Torch::cuda::is_available() always returns false",170 "tags_descriptions": {},171 "like_count": 0,172 "views": 132,173 "category_id": 11,174 "featured_link": null,175 "has_accepted_answer": false,176 "posters": [177 {178 "extras": null,179 "description": "Original Poster",180 "user": {181 "id": 84940,182 "username": "marcleon",183 "name": "Marc",184 "avatar_template": "/letter_avatar_proxy/v4/letter/m/db5fbb/{size}.png",185 "trust_level": 1186 }187 },188 {189 "extras": "latest",190 "description": "Most Recent Poster",191 "user": {192 "id": 85084,193 "username": "qyq",194 "name": "",195 "avatar_template": "/letter_avatar_proxy/v4/letter/q/82dd89/{size}.png",196 "trust_level": 0197 }198 }199 ]200 },201 {202 "fancy_title": "Portable parallelization over batch dimension",203 "id": 217392,204 "title": "Portable parallelization over batch dimension",205 "slug": "portable-parallelization-over-batch-dimension",206 "posts_count": 1,207 "reply_count": 0,208 "highest_post_number": 1,209 "image_url": null,210 "created_at": "2025-03-03T18:07:43.894Z",211 "last_posted_at": "2025-03-03T18:07:44.027Z",212 "bumped": true,213 "bumped_at": "2025-03-03T18:07:44.027Z",214 "archetype": "regular",215 "unseen": false,216 "pinned": false,217 "unpinned": null,218 "visible": true,219 "closed": false,220 "archived": false,221 "bookmarked": null,222 "liked": null,223 "tags_descriptions": {},224 "like_count": 0,225 "views": 75,226 "category_id": 11,227 "featured_link": null,228 "has_accepted_answer": false,229 "posters": [230 {231 "extras": "latest single",232 "description": "Original Poster, Most Recent Poster",233 "user": {234 "id": 9258,235 "username": "BeachHut",236 "name": "",237 "avatar_template": "/letter_avatar_proxy/v4/letter/b/4da419/{size}.png",238 "trust_level": 1239 }240 }241 ]242 },243 {244 "fancy_title": "Rewriting this without using libtorch",245 "id": 219857,246 "title": "Rewriting this without using libtorch",247 "slug": "rewriting-this-without-using-libtorch",248 "posts_count": 1,249 "reply_count": 0,250 "highest_post_number": 1,251 "image_url": null,252 "created_at": "2025-05-08T04:15:44.041Z",253 "last_posted_at": "2025-05-08T04:15:44.087Z",254 "bumped": true,255 "bumped_at": "2025-05-08T09:26:02.477Z",256 "archetype": "regular",257 "unseen": false,258 "pinned": false,259 "unpinned": null,260 "visible": true,261 "closed": false,262 "archived": false,263 "bookmarked": null,264 "liked": null,265 "tags_descriptions": {},266 "like_count": 0,267 "views": 50,268 "category_id": 11,269 "featured_link": null,270 "has_accepted_answer": false,271 "posters": [272 {273 "extras": "latest single",274 "description": "Original Poster, Most Recent Poster",275 "user": {276 "id": 56882,277 "username": "L_R",278 "name": "L R",279 "avatar_template": "/user_avatar/discuss.pytorch.org/l_r/{size}/30837_2.png",280 "trust_level": 1281 }282 }283 ]284 },285 {286 "fancy_title": "Autograd about CUDAExtension",287 "id": 219960,288 "title": "Autograd about CUDAExtension",289 "slug": "autograd-about-cudaextension",290 "posts_count": 5,291 "reply_count": 3,292 "highest_post_number": 5,293 "image_url": null,294 "created_at": "2025-05-12T13:11:17.217Z",295 "last_posted_at": "2025-05-13T00:47:45.506Z",296 "bumped": true,297 "bumped_at": "2025-05-13T00:47:45.506Z",298 "archetype": "regular",299 "unseen": false,300 "pinned": false,301 "unpinned": null,302 "visible": true,303 "closed": false,304 "archived": false,305 "bookmarked": null,306 "liked": null,307 "tags_descriptions": {},308 "like_count": 0,309 "views": 124,310 "category_id": 11,311 "featured_link": null,312 "has_accepted_answer": false,313 "posters": [314 {315 "extras": "latest",316 "description": "Original Poster, Most Recent Poster",317 "user": {318 "id": 84244,319 "username": "XinyuFu",320 "name": "life",321 "avatar_template": "/letter_avatar_proxy/v4/letter/x/e95f7d/{size}.png",322 "trust_level": 0323 }324 },325 {326 "extras": null,327 "description": "Frequent Poster",328 "user": {329 "id": 3534,330 "username": "ptrblck",331 "name": "",332 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",333 "admin": true,334 "moderator": true,335 "trust_level": 2336 }337 }338 ]339 },340 {341 "fancy_title": "Pybind11v3 integration",342 "id": 222811,343 "title": "Pybind11v3 integration",344 "slug": "pybind11v3-integration",345 "posts_count": 1,346 "reply_count": 0,347 "highest_post_number": 1,348 "image_url": null,349 "created_at": "2025-09-01T18:10:29.371Z",350 "last_posted_at": "2025-09-01T18:10:29.418Z",351 "bumped": true,352 "bumped_at": "2025-09-01T18:10:29.418Z",353 "archetype": "regular",354 "unseen": false,355 "pinned": false,356 "unpinned": null,357 "visible": true,358 "closed": false,359 "archived": false,360 "bookmarked": null,361 "liked": null,362 "tags_descriptions": {},363 "like_count": 0,364 "views": 13,365 "category_id": 11,366 "featured_link": null,367 "has_accepted_answer": false,368 "posters": [369 {370 "extras": "latest single",371 "description": "Original Poster, Most Recent Poster",372 "user": {373 "id": 85405,374 "username": "pya",375 "name": null,376 "avatar_template": "/letter_avatar_proxy/v4/letter/p/919ad9/{size}.png",377 "trust_level": 1378 }379 }380 ]381 }382 ],383 "tags_descriptions": {},384 "fancy_title": "About conversion of summarywriter to C++",385 "id": 102108,386 "title": "About conversion of summarywriter to C++",387 "posts_count": 2,388 "created_at": "2020-11-09T02:06:22.010Z",389 "views": 566,390 "reply_count": 0,391 "like_count": 0,392 "last_posted_at": "2020-11-09T07:17:28.836Z",393 "visible": true,394 "closed": false,395 "archived": false,396 "has_summary": false,397 "archetype": "regular",398 "slug": "about-conversion-of-summarywriter-to-c",399 "category_id": 11,400 "word_count": 81,401 "deleted_at": null,402 "user_id": 36581,403 "featured_link": null,404 "pinned_globally": false,405 "pinned_at": null,406 "pinned_until": null,407 "image_url": null,408 "slow_mode_seconds": 0,409 "draft": null,410 "draft_key": "topic_102108",411 "draft_sequence": null,412 "unpinned": null,413 "pinned": false,414 "current_post_number": 1,415 "highest_post_number": 2,416 "deleted_by": null,417 "actions_summary": [418 {419 "id": 4,420 "count": 0,421 "hidden": false,422 "can_act": false423 },424 {425 "id": 8,426 "count": 0,427 "hidden": false,428 "can_act": false429 },430 {431 "id": 10,432 "count": 0,433 "hidden": false,434 "can_act": false435 },436 {437 "id": 7,438 "count": 0,439 "hidden": false,440 "can_act": false441 }442 ],443 "chunk_size": 20,444 "bookmarked": false,445 "topic_timer": null,446 "message_bus_last_id": 0,447 "participant_count": 2,448 "show_read_indicator": false,449 "thumbnails": null,450 "slow_mode_enabled_until": null,451 "can_vote": false,452 "vote_count": 0,453 "user_voted": false,454 "discourse_zendesk_plugin_zendesk_id": null,455 "discourse_zendesk_plugin_zendesk_url": "https://your-url.zendesk.com/agent/tickets/",456 "details": {457 "can_edit": false,458 "notification_level": 1,459 "participants": [460 {461 "id": 3534,462 "username": "ptrblck",463 "name": "",464 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",465 "post_count": 1,466 "primary_group_name": null,467 "flair_name": null,468 "flair_url": null,469 "flair_color": null,470 "flair_bg_color": null,471 "flair_group_id": null,472 "admin": true,473 "moderator": true,474 "trust_level": 2475 },476 {477 "id": 36581,478 "username": "Chen0729",479 "name": "Steven",480 "avatar_template": "/user_avatar/discuss.pytorch.org/chen0729/{size}/15587_2.png",481 "post_count": 1,482 "primary_group_name": null,483 "flair_name": null,484 "flair_url": null,485 "flair_color": null,486 "flair_bg_color": null,487 "flair_group_id": null,488 "trust_level": 1489 }490 ],491 "created_by": {492 "id": 36581,493 "username": "Chen0729",494 "name": "Steven",495 "avatar_template": "/user_avatar/discuss.pytorch.org/chen0729/{size}/15587_2.png"496 },497 "last_poster": {498 "id": 3534,499 "username": "ptrblck",500 "name": "",501 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png"502 },503 "links": [504 {505 "url": "https://github.com/pytorch/pytorch/issues/26097",506 "title": "[RFC] TensorBoard extensions and improvements for PyTorch · Issue #26097 · pytorch/pytorch · GitHub",507 "internal": false,508 "attachment": false,509 "reflection": false,510 "clicks": 7,511 "user_id": 3534,512 "domain": "github.com",513 "root_domain": "github.com"514 },515 {516 "url": "https://github.com/tensorflow/tensorboard/issues/2636",517 "title": "Feature Request: C++ logging from non-TensorFlow frameworks · Issue #2636 · tensorflow/tensorboard · GitHub",518 "internal": false,519 "attachment": false,520 "reflection": false,521 "clicks": 3,522 "user_id": 3534,523 "domain": "github.com",524 "root_domain": "github.com"525 }526 ]527 },528 "bookmarks": []529 },530 {531 "post_stream": {532 "posts": [533 {534 "id": 242462,535 "name": "Sandeep Pathania",536 "username": "Sandeep_Pathania",537 "avatar_template": "/user_avatar/discuss.pytorch.org/sandeep_pathania/{size}/29350_2.png",538 "created_at": "2020-11-08T22:01:45.537Z",539 "cooked": "<p>Is there a class in Pytorch which can save the best accurate model of all the epochs automatically ?</p>\n<p>For example : <a href=\"https://keras.io/api/callbacks/model_checkpoint/\" rel=\"noopener nofollow ugc\">https://keras.io/api/callbacks/model_checkpoint/</a></p>\n<p>There, they have options like ModelCheckpoint which saves the best model. Also like EarlyStopping.</p>\n<p>Just wondering if there is any alternative in Pytorch.</p>",540 "post_number": 1,541 "post_type": 1,542 "posts_count": 2,543 "updated_at": "2020-11-08T22:01:45.537Z",544 "reply_count": 0,545 "reply_to_post_number": null,546 "quote_count": 0,547 "incoming_link_count": 165,548 "reads": 11,549 "readers_count": 10,550 "score": 827.2,551 "yours": false,552 "topic_id": 102099,553 "topic_slug": "class-equivalent-to-modelcheckpoint-keras-in-pytorch",554 "display_username": "Sandeep Pathania",555 "primary_group_name": null,556 "flair_name": null,557 "flair_url": null,558 "flair_bg_color": null,559 "flair_color": null,560 "flair_group_id": null,561 "badges_granted": [],562 "version": 1,563 "can_edit": false,564 "can_delete": false,565 "can_recover": false,566 "can_see_hidden_post": false,567 "can_wiki": false,568 "link_counts": [569 {570 "url": "https://keras.io/api/callbacks/model_checkpoint/",571 "internal": false,572 "reflection": false,573 "title": "ModelCheckpoint",574 "clicks": 4575 }576 ],577 "read": true,578 "user_title": null,579 "bookmarked": false,580 "actions_summary": [],581 "moderator": false,582 "admin": false,583 "staff": false,584 "user_id": 37274,585 "hidden": false,586 "trust_level": 1,587 "deleted_at": null,588 "user_deleted": false,589 "edit_reason": null,590 "can_view_edit_history": true,591 "wiki": false,592 "post_url": "/t/class-equivalent-to-modelcheckpoint-keras-in-pytorch/102099/1",593 "can_accept_answer": false,594 "can_unaccept_answer": false,595 "accepted_answer": false,596 "topic_accepted_answer": null,597 "can_vote": false598 },599 {600 "id": 242554,601 "name": "",602 "username": "ptrblck",603 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",604 "created_at": "2020-11-09T07:14:28.944Z",605 "cooked": "<p>You should be able to find these higher-level APIs and hooks in wrappers such as <code>Ignite</code>, <code>PyTorch Lightning</code>, <code>Catalyst</code> etc.</p>",606 "post_number": 2,607 "post_type": 1,608 "posts_count": 2,609 "updated_at": "2020-11-09T07:14:28.944Z",610 "reply_count": 0,611 "reply_to_post_number": null,612 "quote_count": 0,613 "incoming_link_count": 6,614 "reads": 9,615 "readers_count": 8,616 "score": 46.8,617 "yours": false,618 "topic_id": 102099,619 "topic_slug": "class-equivalent-to-modelcheckpoint-keras-in-pytorch",620 "display_username": "",621 "primary_group_name": null,622 "flair_name": null,623 "flair_url": null,624 "flair_bg_color": null,625 "flair_color": null,626 "flair_group_id": null,627 "badges_granted": [],628 "version": 1,629 "can_edit": false,630 "can_delete": false,631 "can_recover": false,632 "can_see_hidden_post": false,633 "can_wiki": false,634 "read": true,635 "user_title": "",636 "bookmarked": false,637 "actions_summary": [638 {639 "id": 2,640 "count": 1641 }642 ],643 "moderator": true,644 "admin": true,645 "staff": true,646 "user_id": 3534,647 "hidden": false,648 "trust_level": 2,649 "deleted_at": null,650 "user_deleted": false,651 "edit_reason": null,652 "can_view_edit_history": true,653 "wiki": false,654 "post_url": "/t/class-equivalent-to-modelcheckpoint-keras-in-pytorch/102099/2",655 "can_accept_answer": false,656 "can_unaccept_answer": false,657 "accepted_answer": false,658 "topic_accepted_answer": null659 }660 ],661 "stream": [662 242462,663 242554664 ]665 },666 "timeline_lookup": [667 [668 1,669 1812670 ]671 ],672 "suggested_topics": [673 {674 "fancy_title": "Non blocking copy from CPU to GPU",675 "id": 213522,676 "title": "Non blocking copy from CPU to GPU",677 "slug": "non-blocking-copy-from-cpu-to-gpu",678 "posts_count": 2,679 "reply_count": 0,680 "highest_post_number": 2,681 "image_url": null,682 "created_at": "2024-11-27T13:17:50.447Z",683 "last_posted_at": "2024-11-28T09:42:41.192Z",684 "bumped": true,685 "bumped_at": "2024-11-28T09:42:41.192Z",686 "archetype": "regular",687 "unseen": false,688 "pinned": false,689 "unpinned": null,690 "visible": true,691 "closed": false,692 "archived": false,693 "bookmarked": null,694 "liked": null,695 "tags_descriptions": {},696 "like_count": 0,697 "views": 162,698 "category_id": 1,699 "featured_link": null,700 "has_accepted_answer": false,701 "posters": [702 {703 "extras": "latest single",704 "description": "Original Poster, Most Recent Poster",705 "user": {706 "id": 81162,707 "username": "shira",708 "name": "shira",709 "avatar_template": "/user_avatar/discuss.pytorch.org/shira/{size}/74225_2.png",710 "trust_level": 1711 }712 }713 ]714 },715 {716 "fancy_title": "Pytorch support SM_87 cuda architecture",717 "id": 214103,718 "title": "Pytorch support SM_87 cuda architecture",719 "slug": "pytorch-support-sm-87-cuda-architecture",720 "posts_count": 5,721 "reply_count": 3,722 "highest_post_number": 5,723 "image_url": null,724 "created_at": "2024-12-11T12:01:08.814Z",725 "last_posted_at": "2024-12-12T03:07:12.873Z",726 "bumped": true,727 "bumped_at": "2024-12-12T03:07:12.873Z",728 "archetype": "regular",729 "unseen": false,730 "pinned": false,731 "unpinned": null,732 "visible": true,733 "closed": false,734 "archived": false,735 "bookmarked": null,736 "liked": null,737 "tags_descriptions": {},738 "like_count": 0,739 "views": 479,740 "category_id": 1,741 "featured_link": null,742 "has_accepted_answer": true,743 "posters": [744 {745 "extras": "latest",746 "description": "Original Poster, Most Recent Poster",747 "user": {748 "id": 81448,749 "username": "yajuvendra",750 "name": "Yajuvendra Rawat",751 "avatar_template": "/user_avatar/discuss.pytorch.org/yajuvendra/{size}/74470_2.png",752 "trust_level": 0753 }754 },755 {756 "extras": null,757 "description": "Frequent Poster, Accepted Answer",758 "user": {759 "id": 3534,760 "username": "ptrblck",761 "name": "",762 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",763 "admin": true,764 "moderator": true,765 "trust_level": 2766 }767 }768 ]769 },770 {771 "fancy_title": "Problems with fork() in DataLoader with num_workers > 1",772 "id": 214210,773 "title": "Problems with fork() in DataLoader with num_workers > 1",774 "slug": "problems-with-fork-in-dataloader-with-num-workers-1",775 "posts_count": 3,776 "reply_count": 0,777 "highest_post_number": 3,778 "image_url": null,779 "created_at": "2024-12-14T02:50:38.012Z",780 "last_posted_at": "2024-12-17T11:59:06.598Z",781 "bumped": true,782 "bumped_at": "2024-12-17T12:49:17.791Z",783 "archetype": "regular",784 "unseen": false,785 "pinned": false,786 "unpinned": null,787 "visible": true,788 "closed": false,789 "archived": false,790 "bookmarked": null,791 "liked": null,792 "tags_descriptions": {},793 "like_count": 0,794 "views": 588,795 "category_id": 1,796 "featured_link": null,797 "has_accepted_answer": false,798 "posters": [799 {800 "extras": "latest",801 "description": "Original Poster, Most Recent Poster",802 "user": {803 "id": 33086,804 "username": "tanhevg",805 "name": "Evgeny Tankhilevich",806 "avatar_template": "/user_avatar/discuss.pytorch.org/tanhevg/{size}/62710_2.png",807 "trust_level": 1808 }809 },810 {811 "extras": null,812 "description": "Frequent Poster",813 "user": {814 "id": 3534,815 "username": "ptrblck",816 "name": "",817 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",818 "admin": true,819 "moderator": true,820 "trust_level": 2821 }822 }823 ]824 },825 {826 "fancy_title": "Custom kernels for Intel GPUs",827 "id": 214471,828 "title": "Custom kernels for Intel GPUs",829 "slug": "custom-kernels-for-intel-gpus",830 "posts_count": 7,831 "reply_count": 4,832 "highest_post_number": 7,833 "image_url": null,834 "created_at": "2024-12-20T20:49:12.270Z",835 "last_posted_at": "2025-01-06T12:40:19.114Z",836 "bumped": true,837 "bumped_at": "2025-01-06T12:40:19.114Z",838 "archetype": "regular",839 "unseen": false,840 "pinned": false,841 "unpinned": null,842 "visible": true,843 "closed": false,844 "archived": false,845 "bookmarked": null,846 "liked": null,847 "tags_descriptions": {},848 "like_count": 0,849 "views": 292,850 "category_id": 1,851 "featured_link": null,852 "has_accepted_answer": false,853 "posters": [854 {855 "extras": null,856 "description": "Original Poster",857 "user": {858 "id": 50704,859 "username": "Matthias_Moller",860 "name": "Matthias Möller",861 "avatar_template": "/user_avatar/discuss.pytorch.org/matthias_moller/{size}/43979_2.png",862 "trust_level": 2863 }864 },865 {866 "extras": "latest",867 "description": "Most Recent Poster",868 "user": {869 "id": 72061,870 "username": "EikanWang",871 "name": "Eikan Wang",872 "avatar_template": "/user_avatar/discuss.pytorch.org/eikanwang/{size}/66555_2.png",873 "trust_level": 1874 }875 }876 ]877 },878 {879 "fancy_title": "Accessing weights and biases without detaching them",880 "id": 213355,881 "title": "Accessing weights and biases without detaching them",882 "slug": "accessing-weights-and-biases-without-detaching-them",883 "posts_count": 2,884 "reply_count": 0,885 "highest_post_number": 2,886 "image_url": null,887 "created_at": "2024-11-23T23:22:59.303Z",888 "last_posted_at": "2024-11-24T04:46:01.646Z",889 "bumped": true,890 "bumped_at": "2024-11-24T04:46:01.646Z",891 "archetype": "regular",892 "unseen": false,893 "pinned": false,894 "unpinned": null,895 "visible": true,896 "closed": false,897 "archived": false,898 "bookmarked": null,899 "liked": null,900 "tags_descriptions": {},901 "like_count": 1,902 "views": 29,903 "category_id": 1,904 "featured_link": null,905 "has_accepted_answer": false,906 "posters": [907 {908 "extras": "latest single",909 "description": "Original Poster, Most Recent Poster",910 "user": {911 "id": 76827,912 "username": "dmeister",913 "name": "meistdan",914 "avatar_template": "/letter_avatar_proxy/v4/letter/d/65b543/{size}.png",915 "trust_level": 0916 }917 }918 ]919 }920 ],921 "tags_descriptions": {},922 "fancy_title": "Class equivalent to ModelCheckpoint(Keras) in Pytorch?",923 "id": 102099,924 "title": "Class equivalent to ModelCheckpoint(Keras) in Pytorch?",925 "posts_count": 2,926 "created_at": "2020-11-08T22:01:45.457Z",927 "views": 759,928 "reply_count": 0,929 "like_count": 1,930 "last_posted_at": "2020-11-09T07:14:28.944Z",931 "visible": true,932 "closed": false,933 "archived": false,934 "has_summary": false,935 "archetype": "regular",936 "slug": "class-equivalent-to-modelcheckpoint-keras-in-pytorch",937 "category_id": 1,938 "word_count": 70,939 "deleted_at": null,940 "user_id": 37274,941 "featured_link": null,942 "pinned_globally": false,943 "pinned_at": null,944 "pinned_until": null,945 "image_url": null,946 "slow_mode_seconds": 0,947 "draft": null,948 "draft_key": "topic_102099",949 "draft_sequence": null,950 "unpinned": null,951 "pinned": false,952 "current_post_number": 1,953 "highest_post_number": 2,954 "deleted_by": null,955 "actions_summary": [956 {957 "id": 4,958 "count": 0,959 "hidden": false,960 "can_act": false961 },962 {963 "id": 8,964 "count": 0,965 "hidden": false,966 "can_act": false967 },968 {969 "id": 10,970 "count": 0,971 "hidden": false,972 "can_act": false973 },974 {975 "id": 7,976 "count": 0,977 "hidden": false,978 "can_act": false979 }980 ],981 "chunk_size": 20,982 "bookmarked": false,983 "topic_timer": null,984 "message_bus_last_id": 0,985 "participant_count": 2,986 "show_read_indicator": false,987 "thumbnails": null,988 "slow_mode_enabled_until": null,989 "can_vote": false,990 "vote_count": 0,991 "user_voted": false,992 "discourse_zendesk_plugin_zendesk_id": null,993 "discourse_zendesk_plugin_zendesk_url": "https://your-url.zendesk.com/agent/tickets/",994 "details": {995 "can_edit": false,996 "notification_level": 1,997 "participants": [998 {999 "id": 3534,1000 "username": "ptrblck",1001 "name": "",1002 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",1003 "post_count": 1,1004 "primary_group_name": null,1005 "flair_name": null,1006 "flair_url": null,1007 "flair_color": null,1008 "flair_bg_color": null,1009 "flair_group_id": null,1010 "admin": true,1011 "moderator": true,1012 "trust_level": 21013 },1014 {1015 "id": 37274,1016 "username": "Sandeep_Pathania",1017 "name": "Sandeep Pathania",1018 "avatar_template": "/user_avatar/discuss.pytorch.org/sandeep_pathania/{size}/29350_2.png",1019 "post_count": 1,1020 "primary_group_name": null,1021 "flair_name": null,1022 "flair_url": null,1023 "flair_color": null,1024 "flair_bg_color": null,1025 "flair_group_id": null,1026 "trust_level": 11027 }1028 ],1029 "created_by": {1030 "id": 37274,1031 "username": "Sandeep_Pathania",1032 "name": "Sandeep Pathania",1033 "avatar_template": "/user_avatar/discuss.pytorch.org/sandeep_pathania/{size}/29350_2.png"1034 },1035 "last_poster": {1036 "id": 3534,1037 "username": "ptrblck",1038 "name": "",1039 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png"1040 },1041 "links": [1042 {1043 "url": "https://keras.io/api/callbacks/model_checkpoint/",1044 "title": "ModelCheckpoint",1045 "internal": false,1046 "attachment": false,1047 "reflection": false,1048 "clicks": 4,1049 "user_id": 37274,1050 "domain": "keras.io",1051 "root_domain": "keras.io"1052 }1053 ]1054 },1055 "bookmarks": []1056 },1057 {1058 "post_stream": {1059 "posts": [1060 {1061 "id": 242419,1062 "name": "Rayhanul Rumel",1063 "username": "Rayhanul_Rumel",1064 "avatar_template": "/user_avatar/discuss.pytorch.org/rayhanul_rumel/{size}/30772_2.png",1065 "created_at": "2020-11-08T15:40:58.935Z",1066 "cooked": "<p>Hi,</p>\n<p>I was playing around with MNIST and I came up with the following concept:</p>\n<ol>\n<li>I will create 5 subsets of the training set e.g. a,b,c,d,e where b will have 25% common data of a and rest will be unique, c will have 50% common data of b and the rest will be unique, and so on.</li>\n<li>Each subset will have the same length (11000)</li>\n<li>As trainset_a will be unique, it was easier for me to take the subset of length 11,000 from the main trainset of MNIST.</li>\n<li>For trainset_b, I created two subsets: one contains 25% date of trainset_a, other contains 75% unique data.</li>\n</ol>\n<p>Now I want to combine these two subsets in such a way that it treats as a single subset. What I want to say is if I call the data loader upon trainset_b (trainset_b_loader), then by calling trainset_b_load.dataset I can access all of the data from two subsets without creating any subfolder as dataset/ index for two different subsets under dataset. As I am a newbie, I am stuck at this point and unable to find a way to achieve the goal.</p>\n<p>My code is given below. Any help would be highly appreciated.</p>\n<pre><code class=\"lang-auto\">from __future__ import print_function, division\n\nimport torch\nimport torch.nn as nn\nimport torch.optim as optim\nfrom torch.optim import lr_scheduler\nimport numpy as np\nimport torchvision\nfrom torchvision import datasets, models, transforms\nfrom torch.utils.data import Subset, Dataset, DataLoader\nimport matplotlib.pyplot as plt\nimport time\nimport os\nimport copy\nimport pandas as pd\nimport random\n\nfrom torch.utils.data import Subset\nfrom PIL import Image\nfrom torchvision.datasets import MNIST, FashionMNIST\n\nimport torchvision.transforms as transforms\n\n#plt.ion() # interactive mode\ndevice = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n# torch.cuda.set_device(device)\n\ndef get_target_label_idx(labels, targets, shots=5, test=False):\n \"\"\"\n Get the indices of labels that are included in targets.\n :param labels: array of labels\n :param targets: list/tuple of target labels\n :return: list with indices of target labels\n \"\"\"\n final_list = []\n # Both if and else operations seem to be the same, what would be the purpose of this?\n for t in targets:\n if test:\n final_list += np.argwhere(np.isin(labels, t)).flatten().tolist()\n else:\n final_list += np.argwhere(np.isin(labels, t)).flatten().tolist()\n\n return final_list\n\n\ndef convert_label(x):\n if x >= 5:\n return x - 5\n else:\n return x\n\n\nnormal_classes = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]\n\ntransform = transforms.Compose([transforms.ToTensor()])\n\ntrain_set = torchvision.datasets.MNIST(root='./data', train=True,\n download=True, transform=transform)\n\ntrainloader_test = torch.utils.data.DataLoader(train_set, batch_size=4,\n shuffle=True, num_workers=2)\n\ntrain_index = get_target_label_idx(train_set.train_labels.clone().data.cpu().numpy(), normal_classes)\nrandom.shuffle(train_index)\n\n# Split train_index into two batch:\n\ntrain_index_unseen = train_index[0:5000]\ntrain_index_rest = train_index[5000:]\n\n# print(type(train_index_rest))\n\n# This data will be used in the attack model\nmnist_unseen = Subset(trainloader_test.dataset, train_index_unseen) # -----------unseen\n\n# This data will be splitted into 5 batches\nmnist_trainset_rest = Subset(trainloader_test.dataset, train_index_rest)\nmnist_trainset_rest_loader = torch.utils.data.DataLoader(mnist_trainset_rest.dataset, shuffle=True, num_workers=2)\n\n# This will return the length of each batch\ntrain_set_split_length = int(len(train_index_rest) / 5) # --------Each trainset size: 11000\n\n# train_set_b: To choose 25% trainset from train_set_a\n# Each train set will contain 11000 datapoints\n\ncommon_portion_b = int((train_set_split_length * (25 / 100))) # ----25% common: 2750\nunique_portion_b = train_set_split_length - common_portion_b\n# print(unique_portion_b)\nrest_portion_b = train_set_split_length + unique_portion_b\n# print(rest_portion_b)\n\n# train_set_c: To choose 50% trainset from train_set_b\n# Each train set will contain 11000 datapoints\n\ncommon_portion_c = int((train_set_split_length * (50 / 100))) # ----50% common: 5500\nunique_portion_c = train_set_split_length - common_portion_c\n# print(unique_portion_c)\nrest_portion_c = rest_portion_b + unique_portion_c\n# print(rest_portion_c)\n\n\n# train_set_d: To choose 75% trainset from train_set_c\n# Each train set will contain 11000 datapoints\n\ncommon_portion_d = int((train_set_split_length * (75 / 100))) # ----50% common: 8250\nunique_portion_d = train_set_split_length - common_portion_d\nrest_portion_d = rest_portion_c + unique_portion_d\n# print(rest_portion_d)\n\n# First trainset- Unique Trainset\ntrain_set_a = Subset(mnist_trainset_rest_loader.dataset, train_index_rest[0:train_set_split_length])\n# train_set_a_df = PandasDataset(train_set_a)\ntrain_set_a_loader = torch.utils.data.DataLoader(train_set_a.dataset, batch_size=4,\n shuffle=True, num_workers=2)\n\n# Second trainset- 25% common of first Trainset\ntrain_set_b_1 = Subset(train_set_a_loader.dataset, train_index_rest[0:common_portion_b])\ntrain_set_b_2 = Subset(mnist_trainset_rest.dataset, train_index_rest[train_set_split_length:rest_portion_b])\ntrain_set_b = ????\n</code></pre>",1067 "post_number": 1,1068 "post_type": 1,1069 "posts_count": 2,1070 "updated_at": "2020-11-08T15:40:58.935Z",1071 "reply_count": 0,1072 "reply_to_post_number": null,1073 "quote_count": 0,1074 "incoming_link_count": 2150,1075 "reads": 40,1076 "readers_count": 39,1077 "score": 10748.0,1078 "yours": false,1079 "topic_id": 102077,1080 "topic_slug": "combining-subsets-pytorch",1081 "display_username": "Rayhanul Rumel",1082 "primary_group_name": null,1083 "flair_name": null,1084 "flair_url": null,1085 "flair_bg_color": null,1086 "flair_color": null,1087 "flair_group_id": null,1088 "badges_granted": [],1089 "version": 1,1090 "can_edit": false,1091 "can_delete": false,1092 "can_recover": false,1093 "can_see_hidden_post": false,1094 "can_wiki": false,1095 "read": true,1096 "user_title": null,1097 "bookmarked": false,1098 "actions_summary": [],1099 "moderator": false,1100 "admin": false,1101 "staff": false,1102 "user_id": 38671,1103 "hidden": false,1104 "trust_level": 1,1105 "deleted_at": null,1106 "user_deleted": false,1107 "edit_reason": null,1108 "can_view_edit_history": true,1109 "wiki": false,1110 "post_url": "/t/combining-subsets-pytorch/102077/1",1111 "can_accept_answer": false,1112 "can_unaccept_answer": false,1113 "accepted_answer": false,1114 "topic_accepted_answer": null,1115 "can_vote": false1116 },1117 {1118 "id": 242552,1119 "name": "",1120 "username": "ptrblck",1121 "avatar_template": "/user_avatar/discuss.pytorch.org/ptrblck/{size}/1823_2.png",1122 "created_at": "2020-11-09T07:05:49.092Z",1123 "cooked": "<p>If I understand the use case correctly, you would like to create a new <code>Dataset</code> by concatenating <code>train_set_b_1</code> and <code>train_set_b_2</code>?<br>\nIf so, you could use <code>ConcatDataset</code> and pass both datasets to it.</p>",1124 "post_number": 2,1125 "post_type": 1,1126 "posts_count": 2,1127 "updated_at": "2020-11-09T07:05:49.092Z",1128 "reply_count": 0,1129 "reply_to_post_number": null,1130 "quote_count": 0,1131 "incoming_link_count": 29,1132 "reads": 34,1133 "readers_count": 33,1134 "score": 166.8,1135 "yours": false,1136 "topic_id": 102077,1137 "topic_slug": "combining-subsets-pytorch",1138 "display_username": "",1139 "primary_group_name": null,1140 "flair_name": null,1141 "flair_url": null,1142 "flair_bg_color": null,1143 "flair_color": null,1144 "flair_group_id": null,1145 "badges_granted": [],1146 "version": 1,1147 "can_edit": false,1148 "can_delete": false,1149 "can_recover": false,1150 "can_see_hidden_post": false,1151 "can_wiki": false,1152 "read": true,1153 "user_title": "",1154 "bookmarked": false,1155 "actions_summary": [1156 {1157 "id": 2,1158 "count": 11159 }1160 ],1161 "moderator": true,1162 "admin": true,1163 "staff": true,1164 "user_id": 3534,1165 "hidden": false,1166 "trust_level": 2,1167 "deleted_at": null,1168 "user_deleted": false,1169 "edit_reason": null,1170 "can_view_edit_history": true,1171 "wiki": false,1172 "post_url": "/t/combining-subsets-pytorch/102077/2",1173 "can_accept_answer": false,1174 "can_unaccept_answer": false,1175 "accepted_answer": false,1176 "topic_accepted_answer": null1177 }1178 ],1179 "stream": [1180 242419,1181 2425521182 ]1183 },1184 "timeline_lookup": [1185 [1186 1,1187 18121188 ]1189 ],1190 "suggested_topics": [1191 {1192 "fancy_title": "No errors before saving, but an error during testing",1193 "id": 212388,1194 "title": "No errors before saving, but an error during testing",1195 "slug": "no-errors-before-saving-but-an-error-during-testing",1196 "posts_count": 8,1197 "reply_count": 2,1198 "highest_post_number": 8,1199 "image_url": null,1200 "created_at": "2024-11-01T03:04:03.812Z",