Upd info about several models (HRNetV2-W30, RegNetY-16GF, SE-ResNeXt-…

…101 (32x4d))
osmr · Mar 5, 2021 · 7086458 · 7086458
1 parent 4209ba3
commit 7086458
Show file tree

Hide file tree

Showing 8 changed files with 24 additions and 24 deletions.
diff --git a/chainer_/README.md b/chainer_/README.md
@@ -225,7 +225,7 @@ Some remarks:
 | SE-PreResNet-BC-38b | 21.46 | 5.66 | 24,019,064 | 3,236.07M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.409/sepreresnetbc38b-0566-4b9ce096.npz.log)) |
 | SE-PreResNet-50b | 20.75 | 5.31 | 28,080,472 | 4,113.27M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.461/sepreresnet50b-0531-fde03b26.npz.log)) |
 | SE-ResNeXt-50 (32x4d) | 18.73 | 4.35 | 27,559,896 | 4,261.16M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.505/seresnext50_32x4d-0435-52a11b61.npz.log)) |
-| SE-ResNeXt-101 (32x4d) | 19.03 | 4.59 | 48,955,416 | 8,012.73M | From [dmlc/gluon-cv] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.418/seresnext101_32x4d-0459-df43a39e.npz.log)) |
+| SE-ResNeXt-101 (32x4d) | 19.40 | 4.60 | 48,955,416 | 8,012.73M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.525/seresnext101_32x4d-0460-22aca06b.npz.log)) |
 | SE-ResNeXt-101 (64x4d) | 18.98 | 4.68 | 88,232,984 | 15,509.54M | From [dmlc/gluon-cv] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.418/seresnext101_64x4d-0468-ae28d0b4.npz.log)) |
 | SENet-16 | 25.39 | 8.07 | 31,366,168 | 5,081.30M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.341/senet16-0807-f45aa3ff.npz.log)) |
 | SENet-28 | 21.65 | 5.91 | 36,453,768 | 5,732.71M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.356/senet28-0591-7e7bf250.npz.log)) |
@@ -267,7 +267,7 @@ Some remarks:
 | RegNetY-6.4GF | 18.91 | 4.46 | 30,583,252 | 6,388.91M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.513/regnety064-0446-22d3ab05.npz.log)) |
 | RegNetY-8.0GF | 18.78 | 4.37 | 39,180,068 | 7,996.54M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.516/regnety080-0437-f324b046.npz.log)) |
 | RegNetY-12GF | 19.13 | 4.53 | 51,822,544 | 12,132.55M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.524/regnety120-0453-d35fa1b0.npz.log)) |
-| RegNetY-16GF | 19.63 | 4.95 | 83,590,140 | 15,944.53M | From [rwightman/pyt...models] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.473/regnety160-0495-a8102b65.npz.log)) |
+| RegNetY-16GF | 20.21 | 4.83 | 83,590,140 | 15,944.53M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.525/regnety160-0483-af0774ec.npz.log)) |
 | RegNetY-32GF | 19.14 | 4.58 | 145,046,770 | 32,317.66M | From [rwightman/pyt...models] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.473/regnety320-0458-e26048d1.npz.log)) |
 | PyramidNet-101 (a=360) | 20.43 | 5.20 | 42,455,070 | 8,743.54M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.507/pyramidnet101_a360-0520-e37960c4.npz.log)) |
 | DiracNetV2-18 | 30.60 | 11.13 | 11,511,784 | 1,796.62M | From [szagoruyko/diracnets] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.111/diracnet18v2-1113-b85b43d1.npz.log)) |
@@ -324,7 +324,7 @@ Some remarks:
 | HRNet-W18 Small V1 | 26.24 | 8.73 | 13,187,464 | 1,615.00M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.492/hrnet_w18_small_v1-0873-96476e4b.npz.log)) |
 | HRNet-W18 Small V2 | 21.73 | 6.03 | 15,597,464 | 2,618.84M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.499/hrnet_w18_small_v2-0603-87be5731.npz.log)) |
 | HRNetV2-W18 | 20.12 | 5.04 | 21,299,004 | 4,323.07M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.508/hrnetv2_w18-0504-27b854bf.npz.log)) |
-| HRNetV2-W30 | 22.28 | 6.07 | 37,712,220 | 8,156.82M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w30-0607-f685319f.npz.log)) |
+| HRNetV2-W30 | 20.26 | 5.06 | 37,712,220 | 8,156.82M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.525/hrnetv2_w30-0506-04453d82.npz.log)) |
 | HRNetV2-W32 | 22.32 | 6.07 | 41,232,680 | 8,974.04M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w32-0607-0b9c71a6.npz.log)) |
 | HRNetV2-W40 | 21.72 | 5.73 | 57,557,160 | 12,752.26M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w40-0573-340d594a.npz.log)) |
 | HRNetV2-W44 | 21.74 | 5.93 | 67,064,984 | 14,946.96M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w44-0593-8426d89a.npz.log)) |

diff --git a/chainer_/chainercv2/models/model_store.py b/chainer_/chainercv2/models/model_store.py
@@ -91,7 +91,7 @@
     ('sepreresnetbc38b', '0566', '4b9ce0969cc05e5a710fed7073132ad025789091', 'v0.0.409'),
     ('sepreresnet50b', '0531', 'fde03b262da6696c27d1de9188b0d4f35854b92b', 'v0.0.461'),
     ('seresnext50_32x4d', '0435', '52a11b61d67ede6016b5169d10294bce351e3862', 'v0.0.505'),
-    ('seresnext101_32x4d', '0459', 'df43a39efbfe351fbb4a605ef7ab2b76ed86b76b', 'v0.0.418'),
+    ('seresnext101_32x4d', '0460', '22aca06b01ff751164e605f14270d12efbc7ff76', 'v0.0.525'),
     ('seresnext101_64x4d', '0468', 'ae28d0b429fd96c3a102c499ed37b4cac0d0206d', 'v0.0.418'),
     ('senet16', '0807', 'f45aa3fffb8ea5148c53d031e50a3f93ab00ede0', 'v0.0.341'),
     ('senet28', '0591', '7e7bf250ab1bb4842f6dd32ceb93967a7c02239b', 'v0.0.356'),
@@ -133,7 +133,7 @@
     ('regnety064', '0446', '22d3ab05b95da9402dd8aaf08970c7fb13120b76', 'v0.0.513'),
     ('regnety080', '0437', 'f324b0464dc045e411606b6da836d13d5e456ce8', 'v0.0.516'),
     ('regnety120', '0453', 'd35fa1b00b083184e1a444662f709a4f2f65d47c', 'v0.0.524'),
-    ('regnety160', '0495', 'a8102b65d744bc15fcb106f6df3d8d80276e23c8', 'v0.0.473'),
+    ('regnety160', '0483', 'af0774ecc04c142586bb88cc7c8d8192318f0049', 'v0.0.525'),
     ('regnety320', '0458', 'e26048d11a550420d8c9a7c78df801ebc7eb06df', 'v0.0.473'),
     ('pyramidnet101_a360', '0520', 'e37960c468df05477caaf2fec90d9a81ff675cf3', 'v0.0.507'),
     ('diracnet18v2', '1113', 'b85b43d13697dfbddbea6e46dea4766359fff7e5', 'v0.0.111'),
@@ -190,7 +190,7 @@
     ('hrnet_w18_small_v1', '0873', '96476e4b15451b62abf443101a5152e34be552e2', 'v0.0.492'),
     ('hrnet_w18_small_v2', '0603', '87be573162d6f21af4c816a7b389a93ef9f7af85', 'v0.0.499'),
     ('hrnetv2_w18', '0504', '27b854bf5ec915e735ef9deb7b28049db3f32e14', 'v0.0.508'),
-    ('hrnetv2_w30', '0607', 'f685319f78ea409cadd98e2b43600f6bee40cbbb', 'v0.0.421'),
+    ('hrnetv2_w30', '0506', '04453d82f1903ef7bae24928089eee31e96b7501', 'v0.0.525'),
     ('hrnetv2_w32', '0607', '0b9c71a6535054ef4f9f3b4c4e1b7592400f1e70', 'v0.0.421'),
     ('hrnetv2_w40', '0573', '340d594a4f83a8e0f129b857512eba2f99d0208d', 'v0.0.421'),
     ('hrnetv2_w44', '0593', '8426d89ac900c33640a57b301f4c47db4c9a782a', 'v0.0.421'),

diff --git a/gluon/README.md b/gluon/README.md
@@ -240,7 +240,7 @@ bottleneck block. Respectively a network without b-suffix has the stride in the
 | SE-PreResNet-BC-38b | 21.42 | 5.63 | 24,019,064 | 3,233.81M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.409/sepreresnetbc38b-0563-d8f0fbd3.params.log)) |
 | SE-PreResNet-50b | 20.67 | 5.32 | 28,080,472 | 4,110.51M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.461/sepreresnet50b-0532-5b620ff7.params.log)) |
 | SE-ResNeXt-50 (32x4d) | 18.74 | 4.33 | 27,559,896 | 4,258.40M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.505/seresnext50_32x4d-0433-d74d1d0a.params.log)) |
-| SE-ResNeXt-101 (32x4d) | 19.07 | 4.60 | 48,955,416 | 8,008.26M | From [dmlc/gluon-cv] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.418/seresnext101_32x4d-0460-6cb1ee64.params.log)) |
+| SE-ResNeXt-101 (32x4d) | 19.34 | 4.58 | 48,955,416 | 8,008.26M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.525/seresnext101_32x4d-0458-30eecd54.params.log)) |
 | SE-ResNeXt-101 (64x4d) | 18.98 | 4.66 | 88,232,984 | 15,505.08M | From [dmlc/gluon-cv] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.418/seresnext101_64x4d-0466-15e16730.params.log)) |
 | SENet-16 | 25.34 | 8.06 | 31,366,168 | 5,080.55M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.341/senet16-0806-ba268021.params.log)) |
 | SENet-28 | 21.68 | 5.91 | 36,453,768 | 5,731.20M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.356/senet28-0591-d5297a35.params.log)) |
@@ -288,7 +288,7 @@ bottleneck block. Respectively a network without b-suffix has the stride in the
 | RegNetY-6.4GF | 18.95 | 4.45 | 30,583,252 | 6,386.79M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.513/regnety064-0445-8ae16f97.params.log)) |
 | RegNetY-8.0GF | 18.78 | 4.36 | 39,180,068 | 7,994.33M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.516/regnety080-0436-4cd7fc61.params.log)) |
 | RegNetY-12GF | 19.10 | 4.51 | 51,822,544 | 12,129.89M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.524/regnety120-0451-e1bbafc3.params.log)) |
-| RegNetY-16GF | 19.69 | 4.97 | 83,590,140 | 15,941.65M | From [rwightman/pyt...models] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.473/regnety160-0497-e458ce58.params.log)) |
+| RegNetY-16GF | 20.23 | 4.82 | 83,590,140 | 15,941.65M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.525/regnety160-0482-b46554c0.params.log)) |
 | RegNetY-32GF | 19.10 | 4.58 | 145,046,770 | 32,313.76M | From [rwightman/pyt...models] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.473/regnety320-0458-900b9591.params.log)) |
 | PyramidNet-101 (a=360) | 20.41 | 5.20 | 42,455,070 | 8,743.54M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.507/pyramidnet101_a360-0520-3a98a2bf.params.log)) |
 | DiracNetV2-18 | 30.61 | 11.17 | 11,511,784 | 1,796.62M | From [szagoruyko/diracnets] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.111/diracnet18v2-1117-27601f6f.params.log)) |
@@ -346,7 +346,7 @@ bottleneck block. Respectively a network without b-suffix has the stride in the
 | HRNet-W18 Small V1 | 26.20 | 8.73 | 13,187,464 | 1,614.87M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.492/hrnet_w18_small_v1-0873-1060c1c5.params.log)) |
 | HRNet-W18 Small V2 | 21.71 | 6.02 | 15,597,464 | 2,618.54M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.499/hrnet_w18_small_v2-0602-1f319e1b.params.log)) |
 | HRNetV2-W18 | 20.15 | 5.00 | 21,299,004 | 4,322.66M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.508/hrnetv2_w18-0500-0a700902.params.log)) |
-| HRNetV2-W30 | 22.31 | 6.07 | 37,712,220 | 8,156.14M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w30-0607-93553fe4.params.log)) |
+| HRNetV2-W30 | 20.30 | 5.08 | 37,712,220 | 8,156.14M | Training ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.525/hrnetv2_w30-0508-eb6afc49.params.log)) |
 | HRNetV2-W32 | 22.27 | 6.07 | 41,232,680 | 8,973.31M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w32-0607-e68bcf90.params.log)) |
 | HRNetV2-W40 | 21.72 | 5.71 | 57,557,160 | 12,751.34M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w40-0571-60aa3b9d.params.log)) |
 | HRNetV2-W44 | 21.73 | 5.92 | 67,064,984 | 14,945.95M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w44-0592-ff313e29.params.log)) |

diff --git a/gluon/gluoncv2/models/model_store.py b/gluon/gluoncv2/models/model_store.py
@@ -91,7 +91,7 @@
     ('sepreresnetbc38b', '0563', 'd8f0fbd35b840743e312b9ec944c7fc141824ace', 'v0.0.409'),
     ('sepreresnet50b', '0532', '5b620ff7175c674ef41647e2e79992dceefeec19', 'v0.0.461'),
     ('seresnext50_32x4d', '0433', 'd74d1d0a5bd1449562e9f1de92b1389c647fdd13', 'v0.0.505'),
-    ('seresnext101_32x4d', '0460', '6cb1ee646fdd47b745ffa09e0e8ecfec3ce695a3', 'v0.0.418'),
+    ('seresnext101_32x4d', '0458', '30eecd54fa9ee8e1b9c433fb0346c60b3af7604b', 'v0.0.525'),
     ('seresnext101_64x4d', '0466', '15e1673077240a648027505540fc893a68401e87', 'v0.0.418'),
     ('senet16', '0806', 'ba26802160725af926e5a3217f8848bd3a6599fd', 'v0.0.341'),
     ('senet28', '0591', 'd5297a35a2e56ecc892499c1bd4373125a40b783', 'v0.0.356'),
@@ -139,7 +139,7 @@
     ('regnety064', '0445', '8ae16f971eacad369df091e68c26e63b8c6da9a5', 'v0.0.513'),
     ('regnety080', '0436', '4cd7fc61d54ec63bfeff11246e8258188351e7ed', 'v0.0.516'),
     ('regnety120', '0451', 'e1bbafc369b3b40be034ba85cc6445ebcfb1fdc3', 'v0.0.524'),
-    ('regnety160', '0497', 'e458ce58977d36ffb48043853a65e471c4a100c2', 'v0.0.473'),
+    ('regnety160', '0482', 'b46554c0a36ea354cf327f196668ecb6b2962311', 'v0.0.525'),
     ('regnety320', '0458', '900b95919cafc18d2a3404dcbe89e008f44170c2', 'v0.0.473'),
     ('pyramidnet101_a360', '0520', '3a98a2bfc4e74b00b5fff20783613281121a6d37', 'v0.0.507'),
     ('diracnet18v2', '1117', '27601f6fa54e3b10d77981f30650d7a9d4bce91e', 'v0.0.111'),
@@ -197,7 +197,7 @@
     ('hrnet_w18_small_v1', '0873', '1060c1c562770adb94115de1cad797684dbb5703', 'v0.0.492'),
     ('hrnet_w18_small_v2', '0602', '1f319e1b1990818754b97104467cfecf3401ae4e', 'v0.0.499'),
     ('hrnetv2_w18', '0500', '0a70090294c726c1f303e9114d23c9373695eb44', 'v0.0.508'),
-    ('hrnetv2_w30', '0607', '93553fe4cccd22afaac7f73937b98e8ac89850cd', 'v0.0.421'),
+    ('hrnetv2_w30', '0508', 'eb6afc492537295615dc465236934a9a970c1bef', 'v0.0.525'),
     ('hrnetv2_w32', '0607', 'e68bcf9079982540f80607d7fb1331594f721fd6', 'v0.0.421'),
     ('hrnetv2_w40', '0571', '60aa3b9d3a91d20535753fcef0d668aeae466b10', 'v0.0.421'),
     ('hrnetv2_w44', '0592', 'ff313e2963d1464251f5dbb7032502ffd028d612', 'v0.0.421'),

diff --git a/pytorch/README.md b/pytorch/README.md
@@ -244,7 +244,7 @@ OpenCV `Resize` transformation instead of PIL one quality evaluation results wil
 | SE-PreResNet-BC-38b | 21.60 | 5.78 | 24,019,064 | 3,236.07M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.409/sepreresnetbc38b-0578-12827fcd.pth.log)) |
 | SE-PreResNet-50b | 20.85 | 5.49 | 28,080,472 | 4,113.27M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.461/sepreresnet50b-0549-4628a07d.pth.log)) |
 | SE-ResNeXt-50 (32x4d) | 18.94 | 4.51 | 27,559,896 | 4,261.16M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.505/seresnext50_32x4d-0451-52029a7f.pth.log)) |
-| SE-ResNeXt-101 (32x4d) | 19.22 | 4.80 | 48,955,416 | 8,012.73M | From [dmlc/gluon-cv] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.418/seresnext101_32x4d-0480-4f6479f0.pth.log)) |
+| SE-ResNeXt-101 (32x4d) | 19.67 | 4.76 | 48,955,416 | 8,012.73M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.525/seresnext101_32x4d-0476-2a3b9d3b.pth.log)) |
 | SE-ResNeXt-101 (64x4d) | 19.28 | 4.76 | 88,232,984 | 15,509.54M | From [dmlc/gluon-cv] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.418/seresnext101_64x4d-0476-da806109.pth.log)) |
 | SENet-16 | 25.65 | 8.20 | 31,366,168 | 5,081.30M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.341/senet16-0820-373aeafd.pth.log)) |
 | SENet-28 | 21.94 | 5.98 | 36,453,768 | 5,732.71M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.356/senet28-0598-27165b63.pth.log)) |
@@ -292,7 +292,7 @@ OpenCV `Resize` transformation instead of PIL one quality evaluation results wil
 | RegNetY-6.4GF | 19.03 | 4.56 | 30,583,252 | 6,388.91M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.513/regnety064-0456-bff39135.pth.log)) |
 | RegNetY-8.0GF | 18.91 | 4.48 | 39,180,068 | 7,996.54M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.516/regnety080-0448-c084bf6a.pth.log)) |
 | RegNetY-12GF | 19.07 | 4.61 | 51,822,544 | 12,132.55M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.524/regnety120-0461-6ce18cad.pth.log)) |
-| RegNetY-16GF | 19.76 | 5.03 | 83,590,140 | 15,944.53M | From [rwightman/pyt...models] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.473/regnety160-0503-2c0ad1f9.pth.log)) |
+| RegNetY-16GF | 20.25 | 4.98 | 83,590,140 | 15,944.53M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.525/regnety160-0498-66977a86.pth.log)) |
 | RegNetY-32GF | 19.32 | 4.74 | 145,046,770 | 32,317.66M | From [rwightman/pyt...models] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.473/regnety320-0474-643155eb.pth.log)) |
 | PyramidNet-101 (a=360) | 20.53 | 5.43 | 42,455,070 | 8,743.54M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.507/pyramidnet101_a360-0543-7f1747f8.pth.log)) |
 | DiracNetV2-18 | 31.47 | 11.70 | 11,511,784 | 1,796.62M | From [szagoruyko/diracnets] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.111/diracnet18v2-1170-e0673770.pth.log)) |
@@ -349,7 +349,7 @@ OpenCV `Resize` transformation instead of PIL one quality evaluation results wil
 | HRNet-W18 Small V1 | 26.48 | 9.01 | 13,187,464 | 1,615.00M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.492/hrnet_w18_small_v1-0901-30023064.pth.log)) |
 | HRNet-W18 Small V2 | 21.93 | 6.18 | 15,597,464 | 2,618.84M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.499/hrnet_w18_small_v2-0618-ef7b1fe4.pth.log)) |
 | HRNetV2-W18 | 20.20 | 5.12 | 21,299,004 | 4,323.07M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.508/hrnetv2_w18-0512-9d2b7fbf.pth.log)) |
-| HRNetV2-W30 | 21.80 | 5.78 | 37,712,220 | 8,156.82M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w30-0578-839e57eb.pth.log)) |
+| HRNetV2-W30 | 20.35 | 5.21 | 37,712,220 | 8,156.82M | Converted from GL model ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.525/hrnetv2_w30-0521-73d7e48d.pth.log)) |
 | HRNetV2-W32 | 21.55 | 5.81 | 41,232,680 | 8,974.04M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w32-0581-bef9ada0.pth.log)) |
 | HRNetV2-W40 | 21.07 | 5.53 | 57,557,160 | 12,752.26M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w40-0553-e4b5a38a.pth.log)) |
 | HRNetV2-W44 | 21.11 | 5.63 | 67,064,984 | 14,946.96M | From [HRNet/HRNet...ation] ([log](https://github.com/osmr/imgclsmob/releases/download/v0.0.421/hrnetv2_w44-0563-9321bfd8.pth.log)) |