Edit model card

segformer-b0-finetuned-warehouse-part-1-V2

This model is a fine-tuned version of nvidia/mit-b5 on the jakka/warehouse_part1 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2737
  • Mean Iou: 0.7224
  • Mean Accuracy: 0.8119
  • Overall Accuracy: 0.9668
  • Per Category Iou: [0.0, 0.9392313580983768, 0.9322932027111482, 0.9772249946988713, 0.8749950826812657, 0.9591121585348171, 0.9803780030124933, 0.8554852055380204, 0.9661475962866876, 0.5609089467958914, 0.0, 0.8095003013989066, 0.7113799121381718, 0.8927260044840537, 0.6133653057361015, 0.8420100377966416, 0.33841086205511367, 0.553361761785151, 0.8141592920353983, 0.8270316181708587]
  • Per Category Accuracy: [nan, 0.9727824725573769, 0.9676994291705018, 0.9882968957337019, 0.9679484011220059, 0.9772700079950366, 0.9882492205666621, 0.9252107983136135, 0.9825945071781523, 0.6062795795494159, 0.0, 0.894776445179671, 0.7968855332344613, 0.9522349792248335, 0.6544510171692397, 0.9276157710790738, 0.42203029817249116, 0.5863404454740788, 0.8963814834175524, 0.9193914381006046]

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 6e-05
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 20

Training results

Training Loss Epoch Step Validation Loss Mean Iou Mean Accuracy Overall Accuracy Per Category Iou Per Category Accuracy
0.7008 1.0 787 0.2473 0.5595 0.6448 0.9325 [0.0, 0.8572456184869756, 0.8403481284744914, 0.9524827531570127, 0.7992052152702355, 0.9196710216877864, 0.9471503664300267, 0.6193304552041781, 0.9133086982125345, 0.17558267725303728, 0.0, 0.6344520667741999, 0.3360920970752956, 0.7642426437536942, 0.510575871022846, 0.6056988833269157, 0.021209386281588447, 0.27355691497341356, 0.6138181818181818, 0.40645271873846317] [nan, 0.9155298033269351, 0.9463379226245591, 0.978836265135544, 0.9240214201112357, 0.9448111967681583, 0.9643622308798924, 0.6930912552699579, 0.9497575640760723, 0.18632531152693993, 0.0, 0.7500919033177098, 0.36409599568558715, 0.8900647437729461, 0.5728964730263244, 0.6549871668851026, 0.02166159025328631, 0.2902301645548354, 0.7353197421153511, 0.4694729147312794]
0.1321 2.0 1574 0.2331 0.6221 0.7115 0.9457 [0.0, 0.8970560279823083, 0.8791120244598839, 0.9603620467193393, 0.8160602187615088, 0.934767875213888, 0.9616837752836253, 0.7419391385825133, 0.9351874201394574, 0.26717521084051926, 0.0, 0.6985475965645938, 0.43481867741170893, 0.8134984418163408, 0.5459611126448698, 0.7401712453141447, 0.13175924760380514, 0.355121624272543, 0.7060811650388926, 0.6229231428877693] [nan, 0.951233770160613, 0.9409053657605947, 0.9843213861494523, 0.9219686102230917, 0.9665968250506056, 0.9829729958024298, 0.8238168094655243, 0.9620596605954946, 0.29986351309033543, 0.0, 0.8030913978494624, 0.49467439665633006, 0.909599171191769, 0.5931253087796156, 0.8208142201834863, 0.14682189804424495, 0.3841705499014086, 0.8251147122030551, 0.70800907664895]
0.1085 3.0 2361 0.2457 0.6542 0.7530 0.9521 [0.0, 0.9079405116712079, 0.8959028018194484, 0.9654330936322201, 0.8358564096747072, 0.942169826126924, 0.967131589172387, 0.7785683188874377, 0.942506044201895, 0.3544242514524058, 0.0, 0.7247706422018348, 0.5044915351836923, 0.8273089178892802, 0.5630444261421442, 0.7399785788281565, 0.21738423517169614, 0.46725284186024263, 0.7218755768875762, 0.7280122150607375] [nan, 0.9545620491089126, 0.9497321958018098, 0.9837544714508515, 0.9402501375924134, 0.9686463320401577, 0.9809467909731419, 0.8694886440908473, 0.9735407105395524, 0.3936199755387097, 0.0, 0.8558151824280856, 0.5906026695429419, 0.9157369138435157, 0.6097401660523865, 0.8630406290956749, 0.2679143956396281, 0.5182902566913956, 0.8517163268862171, 0.8205229733639949]
0.8409 4.0 3148 0.2533 0.6749 0.7760 0.9559 [0.0, 0.912375840411698, 0.904072054206276, 0.9676067299522242, 0.900289256120933, 0.9448264254043457, 0.9706472863960092, 0.7942658684379895, 0.9498265874428659, 0.5556284571729604, 0.0, 0.743214707471828, 0.529188361408882, 0.7269154778675782, 0.5697874335729916, 0.7702618169892564, 0.2288491765188273, 0.5089612784265519, 0.757448678510892, 0.7646070737475812] [nan, 0.9601569621727435, 0.9525397945710891, 0.9830820784511696, 0.9462795897530819, 0.9732812778343284, 0.9810361205428978, 0.8895280837753298, 0.9743959070958451, 0.6854951638729194, 0.0, 0.8531327543424317, 0.5823783200755023, 0.9177828280607646, 0.6184135395216047, 0.8657506006989952, 0.26841535748637385, 0.5491586570344761, 0.8759801359121798, 0.8665306184609293]
0.0655 5.0 3935 0.2164 0.6815 0.7909 0.9577 [0.0, 0.9195724102825147, 0.8817887152896982, 0.9692666162636345, 0.90446655617651, 0.9477266300807918, 0.972197851990263, 0.8006212298550464, 0.9526181996158507, 0.48675750740382695, 0.0, 0.7544064333927534, 0.589975775752682, 0.8568833610473964, 0.5739430151581254, 0.7804109001873066, 0.2738491187715644, 0.46180522107696753, 0.7493122891746226, 0.754828899421902] [nan, 0.9629768162749704, 0.9511904548979574, 0.9855793956741679, 0.9532853326979632, 0.9705567416728694, 0.9856702233410021, 0.9070277437780497, 0.9761803883026475, 0.7497090051817757, 0.0, 0.8653903593419723, 0.689564513954429, 0.9349779882164135, 0.6119830537374903, 0.9072670926168632, 0.3530779095864059, 0.5086786980626564, 0.8741215078120462, 0.8391483788434887]
0.0568 6.0 4722 0.2803 0.6876 0.7839 0.9591 [0.0, 0.9166100071412383, 0.913602419181271, 0.9710201737288663, 0.8563050555469198, 0.9497657746314072, 0.9730697054916811, 0.8143688646719719, 0.9549812903957364, 0.460486150973965, 0.0, 0.7634781269254467, 0.6136748147716002, 0.8542174198928293, 0.5922937831600485, 0.8066394260877113, 0.28399126278134795, 0.5207639813581891, 0.7629174644376197, 0.7438457521999924] [nan, 0.9601927982852421, 0.9660710264704008, 0.982455068550298, 0.957830657460364, 0.9688535013815731, 0.9819961506837456, 0.893842649258806, 0.9749506995826178, 0.5071640856263331, 0.0, 0.8540977391783844, 0.7091141971147364, 0.9317785850902456, 0.653052819349169, 0.8880378986456968, 0.35953029817249116, 0.553305686470427, 0.862098507289307, 0.8895268263710157]
0.8994 7.0 5509 0.2743 0.6868 0.7764 0.9606 [0.0, 0.92180556388016, 0.9171201062365498, 0.9721111956032598, 0.8587950800137758, 0.9513526631552707, 0.9756092701000854, 0.819792597945916, 0.9576544961199075, 0.4512109977539036, 0.0, 0.7723053199691596, 0.61351217088922, 0.8696959538394335, 0.5947007494875557, 0.8068989910272162, 0.2400942828140323, 0.49048112386556714, 0.772383338067815, 0.7496112574696395] [nan, 0.9644998510561574, 0.9609472275076806, 0.9854828942497743, 0.9565172529563908, 0.9753485051500238, 0.9840922427646661, 0.8947674418604651, 0.974328764760461, 0.49258184783186704, 0.0, 0.8630410807830162, 0.6660374814615073, 0.9410600831006661, 0.6446391486645419, 0.8876351572739187, 0.2796369028534787, 0.5232773027508334, 0.8685891851077423, 0.8883389427836073]
0.0757 8.0 6296 0.2245 0.7038 0.8009 0.9625 [0.0, 0.9246349181813107, 0.9204571437331909, 0.9735757462990084, 0.8677796689121399, 0.9529629595462734, 0.9762280475446855, 0.8249549577060494, 0.9591099123245741, 0.6276133447390932, 0.0, 0.7755030368136181, 0.6490189248809939, 0.8729206918730364, 0.598100700980074, 0.8000277974172574, 0.27374031814774713, 0.5049971433066432, 0.7770387696167466, 0.7981819415236415] [nan, 0.964623037692871, 0.9637122903759715, 0.9863849456780516, 0.9537638293913148, 0.974798022498043, 0.985726579790157, 0.9184958520331837, 0.980103295010109, 0.7586190597174544, 0.0, 0.8624896608767576, 0.7536739921801268, 0.9379994558884956, 0.6446181625809385, 0.9037175076452599, 0.32931227957678744, 0.5392729877180727, 0.863477957832375, 0.8959383518876689]
0.0638 9.0 7083 0.2660 0.7091 0.8064 0.9632 [0.0, 0.9247942993361187, 0.9227547653133065, 0.9737952169757659, 0.8675395458562903, 0.954005651357167, 0.9771936329793919, 0.832432130071599, 0.960664758331238, 0.6439555818513429, 0.0, 0.7800093558353167, 0.6503190735050816, 0.8771838558892437, 0.6000063410406786, 0.8135397086825815, 0.29345229389108285, 0.5278915956856804, 0.7979207701237885, 0.7849771726504039] [nan, 0.9696983271254734, 0.9626331855239437, 0.9865491477141318, 0.9580933383611586, 0.9736782563602464, 0.9877136372491695, 0.9107507139942881, 0.9774734570720269, 0.778129006717992, 0.0, 0.8715651135005974, 0.7419441822839423, 0.9522322311869326, 0.6453719127503574, 0.9070076998689384, 0.36183472266752165, 0.5638987382066087, 0.8882354649474357, 0.8850494190030915]
0.1028 10.0 7870 0.2753 0.7045 0.7986 0.9632 [0.0, 0.9310677916035094, 0.9231154731835156, 0.9742966471140867, 0.8659672807905657, 0.9548025101399095, 0.9761885400996432, 0.8359586760218701, 0.9606324687638941, 0.536304571449891, 0.0, 0.7861687315154533, 0.6648749707875672, 0.8782393648813203, 0.6028230645967004, 0.8034017821150734, 0.2798240884275797, 0.5292981433685788, 0.7976529535864979, 0.7897882016975595] [nan, 0.9671696414372969, 0.9640722977320454, 0.9864307028133905, 0.9566418983913256, 0.9766712626661613, 0.984078186494131, 0.917516659866721, 0.9804665003157427, 0.5945275248601157, 0.0, 0.8886304108078301, 0.7671565322906836, 0.945889759711566, 0.6500072139662386, 0.9114992900830057, 0.33277893555626803, 0.5621391244374099, 0.8784050647615729, 0.9097665351872439]
0.098 11.0 8657 0.2029 0.7052 0.8014 0.9640 [0.0, 0.9288737885707921, 0.9265083379180753, 0.9747097980123621, 0.8738478537660755, 0.9558379241305062, 0.9781696214462526, 0.8391837240652649, 0.9626716931455067, 0.507780252899168, 0.0, 0.7878061172645057, 0.6769843155893536, 0.8815102118136605, 0.6056046400027283, 0.8269347543218291, 0.3132485690006253, 0.5154277002618235, 0.7927511930865472, 0.7569567975718071] [nan, 0.9711631282238503, 0.964815472153087, 0.9853689377873769, 0.9652020663968313, 0.9754185940822899, 0.9867780413729902, 0.9206854345165238, 0.9811350296034029, 0.5495104787677182, 0.0, 0.8906350519253745, 0.7681677227989753, 0.9430888220810342, 0.65217140383783, 0.9110078090869376, 0.3914916639948702, 0.5500605696196935, 0.8924609397688331, 0.9267167202229566]
0.0734 12.0 9444 0.2171 0.7126 0.8001 0.9648 [0.0, 0.9309643707918894, 0.9277494647914695, 0.9750904306170505, 0.8777832954332417, 0.9566409475731096, 0.9780693213049435, 0.8436550838167809, 0.9635515941347027, 0.527304314900299, 0.0, 0.7909202018197202, 0.6909584834347133, 0.8836639196984207, 0.6084447805077513, 0.8287813112544289, 0.31069205419260343, 0.5403587067765045, 0.7955642033577429, 0.8211277996631356] [nan, 0.9680901815771025, 0.9655377799057193, 0.9852963747008175, 0.9662340833391586, 0.9756774116913669, 0.9890014280908129, 0.9132224942200462, 0.9813789993824062, 0.5595195188097869, 0.0, 0.8697959746346843, 0.7887285964675745, 0.9477302580957196, 0.6557731404362482, 0.9149260048055919, 0.374058191728118, 0.5695666398450833, 0.8786809548701865, 0.8983598068927706]
0.0839 13.0 10231 0.2606 0.7139 0.8056 0.9651 [0.0, 0.932934590872574, 0.928599894716927, 0.9759876131918817, 0.8695983139625728, 0.9571779321732448, 0.979228463067019, 0.8446447574729073, 0.9630766038435438, 0.47072541703248466, 0.0, 0.7968195631480623, 0.6967972782731112, 0.8867456411969523, 0.6076684496270689, 0.8274634197517912, 0.3560522933191209, 0.5582305522639651, 0.8036840005319856, 0.8219356251968073] [nan, 0.970161956830923, 0.9673467595439784, 0.9869340313021197, 0.9654732145230638, 0.9756083312329464, 0.9874815117348184, 0.9121141030871753, 0.9832381474966617, 0.50686275089071, 0.0, 0.8991361088135281, 0.8007954698665228, 0.9482970409127882, 0.6487891466970965, 0.9152673110528615, 0.4551538954793203, 0.5915043371384613, 0.8774612301794738, 0.914289630385453]
0.0797 14.0 11018 0.2504 0.7153 0.8044 0.9655 [0.0, 0.9353593794015038, 0.9288667661318105, 0.9762064564453578, 0.8718886319160292, 0.9576685946960725, 0.9788546612617008, 0.8472608735210976, 0.9642969355331718, 0.5361721760842425, 0.0, 0.8004189668257286, 0.696640611014977, 0.8853084044449696, 0.6099045788314064, 0.8344863725117123, 0.3254310344827586, 0.5323734971095841, 0.8050435956126539, 0.8204823185898129] [nan, 0.9668112803123117, 0.9681903691382433, 0.9879581433175818, 0.9650443397090228, 0.9762644155033261, 0.9866578405548627, 0.9181626546987625, 0.9814820281384267, 0.5836381147080894, 0.0, 0.8844717856814631, 0.7870432789537549, 0.9470982093785038, 0.6547561898016377, 0.9131239078200087, 0.39335524206476435, 0.5610603662472479, 0.8835162920369403, 0.9243561823249014]
0.0606 15.0 11805 0.2363 0.7209 0.8122 0.9661 [0.0, 0.9354450021238048, 0.9300759788666999, 0.9766100423179009, 0.8739351769905989, 0.9580569741305669, 0.9795622398211299, 0.8496875639431477, 0.9646763306438436, 0.6043151650835981, 0.0, 0.8018012422360249, 0.7004677380666826, 0.889289794511031, 0.610767874342205, 0.8325289843013258, 0.33953698039089414, 0.5566040090865972, 0.7993623498974272, 0.8161583186067531] [nan, 0.966786642984969, 0.965287953144928, 0.9879603875367537, 0.9664012618135025, 0.9766460508200225, 0.9889968302453108, 0.9177070583435333, 0.9825186826442273, 0.650711681743251, 0.0, 0.8897849462365591, 0.7874477551570715, 0.9497445698771078, 0.655411130494091, 0.9220183486238532, 0.42261141391471624, 0.5914689680174724, 0.8883080676075972, 0.9213864733563804]
0.0532 16.0 12592 0.2531 0.7201 0.8074 0.9662 [0.0, 0.9383203952011292, 0.9288414046194093, 0.9769141389017822, 0.8756205335515858, 0.9582358666094781, 0.979632260873732, 0.8522102747909199, 0.9655114623669192, 0.6115704722763623, 0.0, 0.8053745416448402, 0.7045095417527653, 0.8906375387790608, 0.6007837805741991, 0.8399368744136342, 0.33049747893639037, 0.5151462046865611, 0.8091001625973271, 0.8195206947575124] [nan, 0.9678438083036752, 0.9684728717259394, 0.9879746009248427, 0.9684402878462824, 0.9766889829923047, 0.9883229174617107, 0.9215762273901809, 0.9820408723178519, 0.6655775287006565, 0.0, 0.8831104677878872, 0.7814480248078738, 0.9439503319629784, 0.6414396453351872, 0.9228033529925732, 0.40323420968259055, 0.5458428019417647, 0.8887436835685659, 0.9025173994487001]
0.0862 17.0 13379 0.2458 0.7201 0.8087 0.9665 [0.0, 0.9368370402512427, 0.9309393106006786, 0.9769932787053442, 0.8747985979138234, 0.95879411739136, 0.9800136137207117, 0.8526248910947767, 0.9651962916423883, 0.5741264468224503, 0.0, 0.8066815029500052, 0.7084107667406031, 0.8910943581653369, 0.6137487567405265, 0.843379759286757, 0.32885159559677446, 0.5243792475829478, 0.8126121336965911, 0.8231331714477782] [nan, 0.9768073159423666, 0.9678409097683983, 0.9877789798203552, 0.9673405331004518, 0.977145821644341, 0.9876622727465598, 0.9216680266557867, 0.9832398839363699, 0.6213226822336585, 0.0, 0.8952934013417885, 0.7966158824322502, 0.946850198957944, 0.6577528276561605, 0.9188715050240279, 0.4028735171529336, 0.5553570954877843, 0.887857931114596, 0.9137413764220337]
0.057 18.0 14166 0.2807 0.7169 0.8024 0.9665 [0.0, 0.9391255338059006, 0.9316246290236013, 0.9771178536356643, 0.8736374236266327, 0.9587095139235466, 0.9802820999385629, 0.8534991833144867, 0.965491782119557, 0.5173244886677723, 0.0, 0.8079528780010615, 0.7036495460915129, 0.8919428858888571, 0.6128251272343798, 0.8423749359527112, 0.3030539267193167, 0.5387041043962495, 0.8154057368308808, 0.8249477907232359] [nan, 0.9703254590941974, 0.967385397276143, 0.9883638482723315, 0.9660909281555922, 0.9783173801174915, 0.987878896953218, 0.9238406092751258, 0.9828454227159885, 0.5529433313441302, 0.0, 0.8918872346291701, 0.7785492786841041, 0.9525571866687186, 0.6544903660759959, 0.9202435561380515, 0.3583279897403014, 0.5679750294005819, 0.8882935470755648, 0.9144114645995461]
0.27 19.0 14953 0.2799 0.7210 0.8089 0.9668 [0.0, 0.9392661644355319, 0.932096490765189, 0.9772444850416163, 0.8748583460799624, 0.959030800837604, 0.9803660417493171, 0.8549763601588193, 0.9661359625948338, 0.5489573339508828, 0.0, 0.8082856800928263, 0.707609022556391, 0.8930480213758131, 0.6125057936760998, 0.8439663143164156, 0.3240623821315535, 0.5560068921314832, 0.813374539715939, 0.8289533147998521] [nan, 0.9703971313191945, 0.9680462515437895, 0.9881404237858805, 0.9683475421909045, 0.9777759016962746, 0.988822374850258, 0.9210152318781449, 0.9816258632275899, 0.588252672130082, 0.0, 0.8922778237294366, 0.7930430093029527, 0.9508458460659089, 0.6517263239814098, 0.9221548711227611, 0.3959802821417121, 0.5906377936742327, 0.8980803856653308, 0.9218433516592297]
0.0369 20.0 15740 0.2737 0.7224 0.8119 0.9668 [0.0, 0.9392313580983768, 0.9322932027111482, 0.9772249946988713, 0.8749950826812657, 0.9591121585348171, 0.9803780030124933, 0.8554852055380204, 0.9661475962866876, 0.5609089467958914, 0.0, 0.8095003013989066, 0.7113799121381718, 0.8927260044840537, 0.6133653057361015, 0.8420100377966416, 0.33841086205511367, 0.553361761785151, 0.8141592920353983, 0.8270316181708587] [nan, 0.9727824725573769, 0.9676994291705018, 0.9882968957337019, 0.9679484011220059, 0.9772700079950366, 0.9882492205666621, 0.9252107983136135, 0.9825945071781523, 0.6062795795494159, 0.0, 0.894776445179671, 0.7968855332344613, 0.9522349792248335, 0.6544510171692397, 0.9276157710790738, 0.42203029817249116, 0.5863404454740788, 0.8963814834175524, 0.9193914381006046]

Framework versions

  • Transformers 4.19.2
  • Pytorch 1.11.0
  • Datasets 2.2.2
  • Tokenizers 0.12.1
Downloads last month
17
Hosted inference API
Drag image file here or click to browse from your device
This model can be loaded on the Inference API on-demand.