René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"confusing results, the same usage memory in loops besides negative values","articleBody":"@fabianp  How could I get true results or interpret them? `memory-profiler` shows same usage values for lines in the loop.\r\n```\r\nimport numpy as np\r\nfrom scipy.spatial import cKDTree, distance\r\nimport os\r\nfrom memory_profiler import profile\r\n\r\nradii = np.loadtxt(os.path.join(os.path.join(os.environ['USERPROFILE']), 'Desktop', \"data\", 'radii.csv'))\r\nposs = np.loadtxt(os.path.join(os.path.join(os.environ['USERPROFILE']), 'Desktop', \"data\", 'coordinates.csv'), delimiter=\",\")\r\nprint(len(radii))\r\nrad_max = np.amax(np.hstack(radii))\r\ndia_max = 2 * rad_max\r\n\r\n@profile\r\ndef ends_gap_opt(poss, dia_max):\r\n    particle_corsp_overlaps = []\r\n    ends_ind = [np.empty([1, 2], dtype=np.int64)]\r\n\r\n    kdtree = cKDTree(poss)\r\n\r\n    for particle_idx in range(len(poss)):\r\n        cur_point = poss[particle_idx]\r\n        nears_i_ind = np.array(kdtree.query_ball_point(cur_point, r=dia_max), dtype=np.int64)\r\n        assert len(nears_i_ind) \u003e 0\r\n\r\n        if len(nears_i_ind) \u003c= 1:\r\n            continue\r\n\r\n        nears_i_ind = nears_i_ind[nears_i_ind != particle_idx]\r\n        dist_i = distance.cdist(poss[nears_i_ind], cur_point[None, :]).squeeze()\r\n\r\n        contact_check = dist_i - (radii[nears_i_ind] + radii[particle_idx])\r\n        connected = contact_check[contact_check \u003c= 0]\r\n\r\n        particle_corsp_overlaps.append(connected)\r\n```\r\n![11](https://user-images.githubusercontent.com/63285654/155914051-429988c0-fa0c-4c8f-b8c3-cd08b3f3a17a.PNG)\r\n[radii.csv](https://github.com/pythonprofilers/memory_profiler/files/8150100/radii.csv)\r\n[coordinates.csv](https://github.com/pythonprofilers/memory_profiler/files/8150101/coordinates.csv)\r\n\r\nI have modified the `memory_profiler.py` as pull requests [Fix: Large negative increments #350](https://github.com/pythonprofilers/memory_profiler/pull/350/commits/aa95e34cf6be3c51ba4aa3ccd014f6d40fc32392) and also, in another test using [large negative increment values in line profiler #195](https://github.com/pythonprofilers/memory_profiler/issues/195#issuecomment-401619708), . Both solutions remove previous negative values from *increments*. Which of them is the true one?  \r\nThe same *mem usage*s are doubtful and seem to be wrong:\r\n\r\n![111](https://user-images.githubusercontent.com/63285654/155916732-d97295cf-e6e4-4b15-81d9-76c4873bf256.PNG)","author":{"url":"https://github.com/alisheikholeslam","@type":"Person","name":"alisheikholeslam"},"datePublished":"2022-02-28T02:27:16.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":3},"url":"https://github.com/353/memory_profiler/issues/353"}