Modelnet40 eval

[missbook520]

Hello, thank you very much for your work！
But I used the pospool_sin_cos_avg pre-trained model you provided to predict modelnet40, and the results are as follows:
[11/29 13:58:30] modelnet40_eval INFO: Full config saved to log_eval/modelnet40/pospool_sin_cos_avg_1606629510/config.json
[11/29 14:25:04] modelnet40_eval INFO: length of testing dataset: 2468
[11/29 14:25:41] modelnet40_eval INFO: => loading checkpoint 'weights/pospool_sin_cos_avg.pth'
[11/29 14:25:42] modelnet40_eval INFO: => loaded successfully 'weights/pospool_sin_cos_avg.pth' (epoch 270)
[11/29 14:25:43] modelnet40_eval INFO: ==> checking loaded ckpt
[11/29 14:25:49] modelnet40_eval INFO: Test: [0/155] Time 6.572 (6.572) Loss 3.9528 (3.9528) Acc@1 0.000% (0.000%)
[11/29 14:25:58] modelnet40_eval INFO: Test: [10/155] Time 0.912 (1.387) Loss 3.8535 (3.8445) Acc@1 0.000% (0.000%)
[11/29 14:26:06] modelnet40_eval INFO: Test: [20/155] Time 0.738 (1.133) Loss 3.9266 (3.8434) Acc@1 0.000% (0.000%)
[11/29 14:26:15] modelnet40_eval INFO: Test: [30/155] Time 0.819 (1.049) Loss 3.6361 (3.8057) Acc@1 0.000% (0.000%)
[11/29 14:26:24] modelnet40_eval INFO: Test: [40/155] Time 0.923 (1.007) Loss 4.3326 (3.8251) Acc@1 0.000% (0.000%)
[11/29 14:26:32] modelnet40_eval INFO: Test: [50/155] Time 0.875 (0.977) Loss 2.8819 (3.7561) Acc@1 56.250% (5.760%)
[11/29 14:26:42] modelnet40_eval INFO: Test: [60/155] Time 0.936 (0.968) Loss 3.6222 (3.7104) Acc@1 12.500% (7.480%)
[11/29 14:26:50] modelnet40_eval INFO: Test: [70/155] Time 0.844 (0.955) Loss 3.8727 (3.7256) Acc@1 0.000% (6.426%)
[11/29 14:26:58] modelnet40_eval INFO: Test: [80/155] Time 0.809 (0.935) Loss 3.5683 (3.7146) Acc@1 0.000% (5.864%)
[11/29 14:27:07] modelnet40_eval INFO: Test: [90/155] Time 0.954 (0.931) Loss 3.8353 (3.7245) Acc@1 0.000% (5.220%)
[11/29 14:27:16] modelnet40_eval INFO: Test: [100/155] Time 0.737 (0.928) Loss 3.6307 (3.7281) Acc@1 0.000% (4.703%)
[11/29 14:27:24] modelnet40_eval INFO: Test: [110/155] Time 0.922 (0.918) Loss 3.7987 (3.7046) Acc@1 0.000% (4.617%)
[11/29 14:27:33] modelnet40_eval INFO: Test: [120/155] Time 0.924 (0.915) Loss 3.7480 (3.7105) Acc@1 0.000% (4.236%)
[11/29 14:27:43] modelnet40_eval INFO: Test: [130/155] Time 1.049 (0.917) Loss 3.9255 (3.7225) Acc@1 0.000% (3.912%)
[11/29 14:27:52] modelnet40_eval INFO: Test: [140/155] Time 0.821 (0.919) Loss 4.0577 (3.7144) Acc@1 0.000% (3.635%)
[11/29 14:28:01] modelnet40_eval INFO: Test: [150/155] Time 0.955 (0.914) Loss 4.4066 (3.7499) Acc@1 0.000% (3.394%)
[11/29 14:28:04] modelnet40_eval INFO: * Acc@1 3.323%
[11/29 14:28:04] modelnet40_eval INFO: * Vote0 Acc@1 3.323%
[11/29 14:28:07] modelnet40_eval INFO: Test: [0/155] Time 3.408 (0.929) Loss 4.0015 (3.7566) Acc@1 0.000% (0.000%)
[11/29 14:28:20] modelnet40_eval INFO: Test: [10/155] Time 1.210 (0.948) Loss 3.8147 (3.7612) Acc@1 0.000% (0.000%)
[11/29 14:28:31] modelnet40_eval INFO: Test: [20/155] Time 1.014 (0.958) Loss 4.0274 (3.7657) Acc@1 0.000% (0.000%)
[11/29 14:28:42] modelnet40_eval INFO: Test: [30/155] Time 0.981 (0.964) Loss 3.6712 (3.7635) Acc@1 0.000% (0.000%)
[11/29 14:28:52] modelnet40_eval INFO: Test: [40/155] Time 1.137 (0.969) Loss 4.2934 (3.7681) Acc@1 0.000% (0.000%)
[11/29 14:29:03] modelnet40_eval INFO: Test: [50/155] Time 1.154 (0.972) Loss 2.7808 (3.7535) Acc@1 81.250% (6.740%)
[11/29 14:29:14] modelnet40_eval INFO: Test: [60/155] Time 0.986 (0.977) Loss 3.5801 (3.7426) Acc@1 12.500% (7.889%)
[11/29 14:29:24] modelnet40_eval INFO: Test: [70/155] Time 1.054 (0.981) Loss 3.8196 (3.7444) Acc@1 0.000% (6.778%)
[11/29 14:29:35] modelnet40_eval INFO: Test: [80/155] Time 1.089 (0.985) Loss 3.5128 (3.7382) Acc@1 0.000% (6.404%)
[11/29 14:29:45] modelnet40_eval INFO: Test: [90/155] Time 1.170 (0.987) Loss 3.8592 (3.7419) Acc@1 0.000% (5.769%)
[11/29 14:29:56] modelnet40_eval INFO: Test: [100/155] Time 1.064 (0.992) Loss 3.6984 (3.7430) Acc@1 0.000% (5.198%)
[11/29 14:30:07] modelnet40_eval INFO: Test: [110/155] Time 1.231 (0.993) Loss 3.7226 (3.7339) Acc@1 0.000% (4.955%)
[11/29 14:30:17] modelnet40_eval INFO: Test: [120/155] Time 1.026 (0.995) Loss 3.7505 (3.7345) Acc@1 0.000% (4.545%)
[11/29 14:30:28] modelnet40_eval INFO: Test: [130/155] Time 0.915 (0.998) Loss 3.9110 (3.7391) Acc@1 0.000% (4.198%)
[11/29 14:30:38] modelnet40_eval INFO: Test: [140/155] Time 0.968 (0.999) Loss 4.0316 (3.7338) Acc@1 0.000% (3.901%)
[11/29 14:30:49] modelnet40_eval INFO: Test: [150/155] Time 1.129 (1.003) Loss 4.4362 (3.7506) Acc@1 0.000% (3.642%)
[11/29 14:30:53] modelnet40_eval INFO: * Acc@1 3.566%
[11/29 14:30:53] modelnet40_eval INFO: * Vote1 Acc@1 3.606%
[11/29 14:30:56] modelnet40_eval INFO: Test: [0/155] Time 2.464 (1.006) Loss 3.9860 (3.7538) Acc@1 0.000% (0.000%)
[11/29 14:31:06] modelnet40_eval INFO: Test: [10/155] Time 1.089 (1.009) Loss 3.7555 (3.7566) Acc@1 0.000% (0.000%)
[11/29 14:31:17] modelnet40_eval INFO: Test: [20/155] Time 1.149 (1.011) Loss 3.9376 (3.7588) Acc@1 0.000% (0.000%)
[11/29 14:31:28] modelnet40_eval INFO: Test: [30/155] Time 1.099 (1.012) Loss 3.6077 (3.7575) Acc@1 6.250% (0.202%)
[11/29 14:31:38] modelnet40_eval INFO: Test: [40/155] Time 1.014 (1.014) Loss 4.2509 (3.7600) Acc@1 0.000% (0.152%)
[11/29 14:31:49] modelnet40_eval INFO: Test: [50/155] Time 1.146 (1.015) Loss 2.8229 (3.7521) Acc@1 50.000% (6.250%)
[11/29 14:32:00] modelnet40_eval INFO: Test: [60/155] Time 1.005 (1.017) Loss 3.6266 (3.7457) Acc@1 12.500% (7.582%)
[11/29 14:32:11] modelnet40_eval INFO: Test: [70/155] Time 1.010 (1.019) Loss 3.8605 (3.7469) Acc@1 0.000% (6.514%)
[11/29 14:32:21] modelnet40_eval INFO: Test: [80/155] Time 1.179 (1.019) Loss 3.5332 (3.7433) Acc@1 0.000% (6.096%)
[11/29 14:32:32] modelnet40_eval INFO: Test: [90/155] Time 1.017 (1.020) Loss 3.8547 (3.7454) Acc@1 0.000% (5.426%)
[11/29 14:32:42] modelnet40_eval INFO: Test: [100/155] Time 0.947 (1.021) Loss 3.6940 (3.7461) Acc@1 0.000% (4.889%)
[11/29 14:32:53] modelnet40_eval INFO: Test: [110/155] Time 0.944 (1.022) Loss 3.7454 (3.7411) Acc@1 0.000% (4.673%)
[11/29 14:33:03] modelnet40_eval INFO: Test: [120/155] Time 1.006 (1.023) Loss 3.7019 (3.7411) Acc@1 0.000% (4.287%)
[11/29 14:33:14] modelnet40_eval INFO: Test: [130/155] Time 0.940 (1.023) Loss 4.0530 (3.7445) Acc@1 0.000% (3.960%)
[11/29 14:33:24] modelnet40_eval INFO: Test: [140/155] Time 1.174 (1.024) Loss 4.0547 (3.7407) Acc@1 0.000% (3.723%)
[11/29 14:33:35] modelnet40_eval INFO: Test: [150/155] Time 1.107 (1.025) Loss 4.4534 (3.7519) Acc@1 0.000% (3.477%)
[11/29 14:33:39] modelnet40_eval INFO: * Acc@1 3.404%
[11/29 14:33:39] modelnet40_eval INFO: * Vote2 Acc@1 3.849%
[11/29 14:33:41] modelnet40_eval INFO: Test: [0/155] Time 2.628 (1.028) Loss 3.9853 (3.7541) Acc@1 0.000% (0.000%)
[11/29 14:33:52] modelnet40_eval INFO: Test: [10/155] Time 1.153 (1.028) Loss 3.7854 (3.7556) Acc@1 0.000% (0.000%)
[11/29 14:34:03] modelnet40_eval INFO: Test: [20/155] Time 0.946 (1.029) Loss 4.0336 (3.7574) Acc@1 0.000% (0.000%)
[11/29 14:34:13] modelnet40_eval INFO: Test: [30/155] Time 1.095 (1.030) Loss 3.6432 (3.7567) Acc@1 0.000% (0.000%)
[11/29 14:34:24] modelnet40_eval INFO: Test: [40/155] Time 1.017 (1.031) Loss 4.3070 (3.7586) Acc@1 0.000% (0.000%)
[11/29 14:34:35] modelnet40_eval INFO: Test: [50/155] Time 0.964 (1.031) Loss 2.8068 (3.7530) Acc@1 62.500% (6.373%)
[11/29 14:34:45] modelnet40_eval INFO: Test: [60/155] Time 1.162 (1.032) Loss 3.6481 (3.7488) Acc@1 6.250% (7.787%)
[11/29 14:34:56] modelnet40_eval INFO: Test: [70/155] Time 1.064 (1.032) Loss 3.8415 (3.7494) Acc@1 0.000% (6.690%)
[11/29 14:35:07] modelnet40_eval INFO: Test: [80/155] Time 1.031 (1.033) Loss 3.5724 (3.7470) Acc@1 0.000% (6.250%)
[11/29 14:35:17] modelnet40_eval INFO: Test: [90/155] Time 1.100 (1.034) Loss 3.8922 (3.7488) Acc@1 0.000% (5.563%)
[11/29 14:35:28] modelnet40_eval INFO: Test: [100/155] Time 1.120 (1.034) Loss 3.6246 (3.7488) Acc@1 0.000% (5.012%)
[11/29 14:35:38] modelnet40_eval INFO: Test: [110/155] Time 0.994 (1.034) Loss 3.7005 (3.7446) Acc@1 0.000% (4.730%)
[11/29 14:35:49] modelnet40_eval INFO: Test: [120/155] Time 1.144 (1.035) Loss 3.6256 (3.7444) Acc@1 0.000% (4.339%)
[11/29 14:36:00] modelnet40_eval INFO: Test: [130/155] Time 1.052 (1.035) Loss 3.9319 (3.7460) Acc@1 0.000% (4.008%)
[11/29 14:36:10] modelnet40_eval INFO: Test: [140/155] Time 0.971 (1.036) Loss 4.0477 (3.7433) Acc@1 0.000% (3.723%)
[11/29 14:36:21] modelnet40_eval INFO: Test: [150/155] Time 1.064 (1.036) Loss 4.4978 (3.7514) Acc@1 0.000% (3.477%)
[11/29 14:36:25] modelnet40_eval INFO: * Acc@1 3.404%
[11/29 14:36:25] modelnet40_eval INFO: * Vote3 Acc@1 3.809%
[11/29 14:36:27] modelnet40_eval INFO: Test: [0/155] Time 2.517 (1.038) Loss 4.0088 (3.7531) Acc@1 0.000% (0.000%)
[11/29 14:36:38] modelnet40_eval INFO: Test: [10/155] Time 0.997 (1.039) Loss 3.7598 (3.7546) Acc@1 0.000% (0.000%)
[11/29 14:36:49] modelnet40_eval INFO: Test: [20/155] Time 1.153 (1.039) Loss 3.9763 (3.7559) Acc@1 0.000% (0.000%)
[11/29 14:36:59] modelnet40_eval INFO: Test: [30/155] Time 0.986 (1.040) Loss 3.6495 (3.7556) Acc@1 0.000% (0.000%)
[11/29 14:37:10] modelnet40_eval INFO: Test: [40/155] Time 1.209 (1.040) Loss 4.2321 (3.7570) Acc@1 0.000% (0.000%)
[11/29 14:37:21] modelnet40_eval INFO: Test: [50/155] Time 0.933 (1.040) Loss 2.7845 (3.7523) Acc@1 75.000% (6.250%)
[11/29 14:37:32] modelnet40_eval INFO: Test: [60/155] Time 1.171 (1.041) Loss 3.6392 (3.7490) Acc@1 12.500% (7.582%)
[11/29 14:37:42] modelnet40_eval INFO: Test: [70/155] Time 0.942 (1.041) Loss 3.8255 (3.7495) Acc@1 0.000% (6.514%)
[11/29 14:37:53] modelnet40_eval INFO: Test: [80/155] Time 1.103 (1.042) Loss 3.5287 (3.7477) Acc@1 0.000% (6.096%)
[11/29 14:38:03] modelnet40_eval INFO: Test: [90/155] Time 1.008 (1.042) Loss 3.8621 (3.7487) Acc@1 0.000% (5.495%)
[11/29 14:38:14] modelnet40_eval INFO: Test: [100/155] Time 1.002 (1.043) Loss 3.6615 (3.7489) Acc@1 0.000% (4.950%)
[11/29 14:38:25] modelnet40_eval INFO: Test: [110/155] Time 1.145 (1.043) Loss 3.7387 (3.7459) Acc@1 0.000% (4.899%)
[11/29 14:38:36] modelnet40_eval INFO: Test: [120/155] Time 1.161 (1.043) Loss 3.6656 (3.7459) Acc@1 0.000% (4.494%)
[11/29 14:38:46] modelnet40_eval INFO: Test: [130/155] Time 1.067 (1.043) Loss 3.9450 (3.7475) Acc@1 0.000% (4.151%)
[11/29 14:38:56] modelnet40_eval INFO: Test: [140/155] Time 0.956 (1.043) Loss 4.0510 (3.7452) Acc@1 0.000% (3.945%)
[11/29 14:39:07] modelnet40_eval INFO: Test: [150/155] Time 1.051 (1.044) Loss 4.3686 (3.7517) Acc@1 0.000% (3.684%)
[11/29 14:39:11] modelnet40_eval INFO: * Acc@1 3.606%
[11/29 14:39:11] modelnet40_eval INFO: * Vote4 Acc@1 3.890%
[11/29 14:39:13] modelnet40_eval INFO: Test: [0/155] Time 2.532 (1.045) Loss 4.0175 (3.7530) Acc@1 0.000% (0.000%)
[11/29 14:39:24] modelnet40_eval INFO: Test: [10/155] Time 1.107 (1.045) Loss 3.7895 (3.7543) Acc@1 0.000% (0.000%)
[11/29 14:39:35] modelnet40_eval INFO: Test: [20/155] Time 1.217 (1.046) Loss 3.9840 (3.7553) Acc@1 0.000% (0.000%)
[11/29 14:39:46] modelnet40_eval INFO: Test: [30/155] Time 1.033 (1.046) Loss 3.6501 (3.7550) Acc@1 0.000% (0.000%)
[11/29 14:39:56] modelnet40_eval INFO: Test: [40/155] Time 0.967 (1.046) Loss 4.2784 (3.7561) Acc@1 0.000% (0.000%)
[11/29 14:40:07] modelnet40_eval INFO: Test: [50/155] Time 0.979 (1.046) Loss 2.7975 (3.7524) Acc@1 56.250% (5.760%)
[11/29 14:40:16] modelnet40_eval INFO: Test: [60/155] Time 0.748 (1.045) Loss 3.6368 (3.7493) Acc@1 12.500% (7.582%)
[11/29 14:40:25] modelnet40_eval INFO: Test: [70/155] Time 0.865 (1.043) Loss 3.8345 (3.7498) Acc@1 0.000% (6.514%)
[11/29 14:40:37] modelnet40_eval INFO: Test: [80/155] Time 1.562 (1.045) Loss 3.5329 (3.7482) Acc@1 0.000% (6.327%)
[11/29 14:40:48] modelnet40_eval INFO: Test: [90/155] Time 0.888 (1.046) Loss 3.8715 (3.7492) Acc@1 0.000% (5.632%)
[11/29 14:40:59] modelnet40_eval INFO: Test: [100/155] Time 1.173 (1.047) Loss 3.6670 (3.7493) Acc@1 0.000% (5.074%)
[11/29 14:41:12] modelnet40_eval INFO: Test: [110/155] Time 1.302 (1.049) Loss 3.7494 (3.7467) Acc@1 0.000% (4.842%)
[11/29 14:41:24] modelnet40_eval INFO: Test: [120/155] Time 1.062 (1.051) Loss 3.7491 (3.7466) Acc@1 0.000% (4.442%)
[11/29 14:41:33] modelnet40_eval INFO: Test: [130/155] Time 0.823 (1.049) Loss 4.0204 (3.7480) Acc@1 0.000% (4.103%)
[11/29 14:41:45] modelnet40_eval INFO: Test: [140/155] Time 1.304 (1.050) Loss 4.0211 (3.7463) Acc@1 0.000% (3.812%)
[11/29 14:41:58] modelnet40_eval INFO: Test: [150/155] Time 1.058 (1.053) Loss 4.3938 (3.7515) Acc@1 0.000% (3.560%)
[11/29 14:42:02] modelnet40_eval INFO: * Acc@1 3.485%
[11/29 14:42:02] modelnet40_eval INFO: * Vote5 Acc@1 3.930%
[11/29 14:42:05] modelnet40_eval INFO: Test: [0/155] Time 3.213 (1.055) Loss 3.9715 (3.7526) Acc@1 0.000% (0.000%)
[11/29 14:42:16] modelnet40_eval INFO: Test: [10/155] Time 1.184 (1.056) Loss 3.7984 (3.7535) Acc@1 0.000% (0.000%)
[11/29 14:42:27] modelnet40_eval INFO: Test: [20/155] Time 1.061 (1.056) Loss 4.0312 (3.7547) Acc@1 0.000% (0.000%)
[11/29 14:42:37] modelnet40_eval INFO: Test: [30/155] Time 0.991 (1.056) Loss 3.5954 (3.7543) Acc@1 0.000% (0.000%)
[11/29 14:42:48] modelnet40_eval INFO: Test: [40/155] Time 1.063 (1.056) Loss 4.3076 (3.7554) Acc@1 0.000% (0.000%)
[11/29 14:42:58] modelnet40_eval INFO: Test: [50/155] Time 1.079 (1.056) Loss 2.7333 (3.7522) Acc@1 75.000% (6.495%)
[11/29 14:43:09] modelnet40_eval INFO: Test: [60/155] Time 1.106 (1.056) Loss 3.6089 (3.7501) Acc@1 12.500% (7.787%)
[11/29 14:43:19] modelnet40_eval INFO: Test: [70/155] Time 0.992 (1.056) Loss 3.8443 (3.7505) Acc@1 0.000% (6.690%)
[11/29 14:43:30] modelnet40_eval INFO: Test: [80/155] Time 0.996 (1.056) Loss 3.5482 (3.7490) Acc@1 0.000% (6.173%)
[11/29 14:43:41] modelnet40_eval INFO: Test: [90/155] Time 1.055 (1.056) Loss 3.8434 (3.7499) Acc@1 0.000% (5.495%)
[11/29 14:43:51] modelnet40_eval INFO: Test: [100/155] Time 1.121 (1.056) Loss 3.6470 (3.7500) Acc@1 0.000% (4.950%)
[11/29 14:44:02] modelnet40_eval INFO: Test: [110/155] Time 1.174 (1.056) Loss 3.7180 (3.7476) Acc@1 0.000% (4.842%)
[11/29 14:44:12] modelnet40_eval INFO: Test: [120/155] Time 1.036 (1.056) Loss 3.6894 (3.7476) Acc@1 0.000% (4.442%)
[11/29 14:44:23] modelnet40_eval INFO: Test: [130/155] Time 1.149 (1.056) Loss 4.0758 (3.7487) Acc@1 0.000% (4.103%)
[11/29 14:44:34] modelnet40_eval INFO: Test: [140/155] Time 1.118 (1.056) Loss 4.0272 (3.7470) Acc@1 0.000% (3.856%)
[11/29 14:44:45] modelnet40_eval INFO: Test: [150/155] Time 0.970 (1.057) Loss 4.3587 (3.7517) Acc@1 0.000% (3.601%)
[11/29 14:44:49] modelnet40_eval INFO: * Acc@1 3.525%
[11/29 14:44:49] modelnet40_eval INFO: * Vote6 Acc@1 3.849%
[11/29 14:44:51] modelnet40_eval INFO: Test: [0/155] Time 2.593 (1.057) Loss 4.0289 (3.7526) Acc@1 0.000% (0.000%)
[11/29 14:45:02] modelnet40_eval INFO: Test: [10/155] Time 1.182 (1.058) Loss 3.7716 (3.7534) Acc@1 0.000% (0.568%)
[11/29 14:45:12] modelnet40_eval INFO: Test: [20/155] Time 1.012 (1.058) Loss 3.9894 (3.7542) Acc@1 0.000% (0.298%)
[11/29 14:45:23] modelnet40_eval INFO: Test: [30/155] Time 1.071 (1.058) Loss 3.6501 (3.7540) Acc@1 0.000% (0.202%)
[11/29 14:45:33] modelnet40_eval INFO: Test: [40/155] Time 1.060 (1.058) Loss 4.3594 (3.7550) Acc@1 0.000% (0.152%)
[11/29 14:45:44] modelnet40_eval INFO: Test: [50/155] Time 0.974 (1.058) Loss 2.7839 (3.7524) Acc@1 81.250% (6.127%)
[11/29 14:45:54] modelnet40_eval INFO: Test: [60/155] Time 0.930 (1.057) Loss 3.6214 (3.7503) Acc@1 12.500% (7.582%)
[11/29 14:46:05] modelnet40_eval INFO: Test: [70/155] Time 1.096 (1.057) Loss 3.8087 (3.7506) Acc@1 0.000% (6.514%)
[11/29 14:46:16] modelnet40_eval INFO: Test: [80/155] Time 1.296 (1.057) Loss 3.4563 (3.7495) Acc@1 6.250% (6.019%)
[11/29 14:46:26] modelnet40_eval INFO: Test: [90/155] Time 1.184 (1.058) Loss 3.8604 (3.7502) Acc@1 0.000% (5.357%)
[11/29 14:46:37] modelnet40_eval INFO: Test: [100/155] Time 1.098 (1.057) Loss 3.6172 (3.7502) Acc@1 0.000% (4.827%)
[11/29 14:46:47] modelnet40_eval INFO: Test: [110/155] Time 1.039 (1.057) Loss 3.7107 (3.7481) Acc@1 0.000% (4.673%)
[11/29 14:46:58] modelnet40_eval INFO: Test: [120/155] Time 0.946 (1.057) Loss 3.6744 (3.7481) Acc@1 0.000% (4.287%)
[11/29 14:47:08] modelnet40_eval INFO: Test: [130/155] Time 1.033 (1.057) Loss 3.8788 (3.7488) Acc@1 0.000% (3.960%)
[11/29 14:47:19] modelnet40_eval INFO: Test: [140/155] Time 1.138 (1.057) Loss 4.0242 (3.7475) Acc@1 0.000% (3.679%)
[11/29 14:47:29] modelnet40_eval INFO: Test: [150/155] Time 0.968 (1.057) Loss 4.4955 (3.7516) Acc@1 0.000% (3.435%)
[11/29 14:47:33] modelnet40_eval INFO: * Acc@1 3.363%
[11/29 14:47:33] modelnet40_eval INFO: * Vote7 Acc@1 3.809%
[11/29 14:47:36] modelnet40_eval INFO: Test: [0/155] Time 2.357 (1.058) Loss 3.9977 (3.7524) Acc@1 0.000% (0.000%)
[11/29 14:47:46] modelnet40_eval INFO: Test: [10/155] Time 1.186 (1.058) Loss 3.7635 (3.7531) Acc@1 0.000% (0.000%)
[11/29 14:47:56] modelnet40_eval INFO: Test: [20/155] Time 1.049 (1.058) Loss 3.9756 (3.7537) Acc@1 0.000% (0.000%)
[11/29 14:48:07] modelnet40_eval INFO: Test: [30/155] Time 1.118 (1.058) Loss 3.6174 (3.7533) Acc@1 0.000% (0.000%)
[11/29 14:48:18] modelnet40_eval INFO: Test: [40/155] Time 0.936 (1.058) Loss 4.2586 (3.7540) Acc@1 0.000% (0.000%)
[11/29 14:48:29] modelnet40_eval INFO: Test: [50/155] Time 1.249 (1.058) Loss 2.7986 (3.7516) Acc@1 81.250% (6.618%)
[11/29 14:48:40] modelnet40_eval INFO: Test: [60/155] Time 1.045 (1.058) Loss 3.6148 (3.7500) Acc@1 12.500% (7.992%)
[11/29 14:48:50] modelnet40_eval INFO: Test: [70/155] Time 0.935 (1.058) Loss 3.8245 (3.7503) Acc@1 0.000% (6.866%)
[11/29 14:49:01] modelnet40_eval INFO: Test: [80/155] Time 1.050 (1.058) Loss 3.5344 (3.7494) Acc@1 0.000% (6.096%)
[11/29 14:49:11] modelnet40_eval INFO: Test: [90/155] Time 1.072 (1.058) Loss 3.8560 (3.7501) Acc@1 0.000% (5.426%)
[11/29 14:49:22] modelnet40_eval INFO: Test: [100/155] Time 1.159 (1.058) Loss 3.6145 (3.7500) Acc@1 0.000% (4.889%)
[11/29 14:49:32] modelnet40_eval INFO: Test: [110/155] Time 1.093 (1.058) Loss 3.6729 (3.7483) Acc@1 0.000% (4.617%)
[11/29 14:49:43] modelnet40_eval INFO: Test: [120/155] Time 0.978 (1.058) Loss 3.6947 (3.7483) Acc@1 0.000% (4.236%)
[11/29 14:49:53] modelnet40_eval INFO: Test: [130/155] Time 1.012 (1.058) Loss 3.8309 (3.7491) Acc@1 0.000% (3.912%)
[11/29 14:50:04] modelnet40_eval INFO: Test: [140/155] Time 1.066 (1.058) Loss 4.0289 (3.7480) Acc@1 0.000% (3.635%)
[11/29 14:50:14] modelnet40_eval INFO: Test: [150/155] Time 1.081 (1.058) Loss 4.3691 (3.7515) Acc@1 0.000% (3.394%)
[11/29 14:50:18] modelnet40_eval INFO: * Acc@1 3.323%
[11/29 14:50:18] modelnet40_eval INFO: * Vote8 Acc@1 3.849%
[11/29 14:50:21] modelnet40_eval INFO: Test: [0/155] Time 2.526 (1.059) Loss 4.0171 (3.7523) Acc@1 0.000% (0.000%)
[11/29 14:50:31] modelnet40_eval INFO: Test: [10/155] Time 1.065 (1.058) Loss 3.7515 (3.7529) Acc@1 0.000% (0.000%)
[11/29 14:50:41] modelnet40_eval INFO: Test: [20/155] Time 0.959 (1.058) Loss 3.9989 (3.7534) Acc@1 0.000% (0.000%)
[11/29 14:50:52] modelnet40_eval INFO: Test: [30/155] Time 1.050 (1.058) Loss 3.6395 (3.7531) Acc@1 0.000% (0.000%)
[11/29 14:51:03] modelnet40_eval INFO: Test: [40/155] Time 0.954 (1.058) Loss 4.2843 (3.7539) Acc@1 0.000% (0.000%)
[11/29 14:51:13] modelnet40_eval INFO: Test: [50/155] Time 1.034 (1.059) Loss 2.7649 (3.7518) Acc@1 87.500% (6.005%)
[11/29 14:51:24] modelnet40_eval INFO: Test: [60/155] Time 1.066 (1.059) Loss 3.6225 (3.7503) Acc@1 12.500% (7.172%)
[11/29 14:51:35] modelnet40_eval INFO: Test: [70/155] Time 1.119 (1.059) Loss 3.8787 (3.7506) Acc@1 0.000% (6.162%)
[11/29 14:51:45] modelnet40_eval INFO: Test: [80/155] Time 1.106 (1.059) Loss 3.5740 (3.7497) Acc@1 0.000% (5.787%)
[11/29 14:51:56] modelnet40_eval INFO: Test: [90/155] Time 0.973 (1.059) Loss 3.8728 (3.7503) Acc@1 0.000% (5.151%)
[11/29 14:52:07] modelnet40_eval INFO: Test: [100/155] Time 1.152 (1.059) Loss 3.6249 (3.7503) Acc@1 0.000% (4.641%)
[11/29 14:52:18] modelnet40_eval INFO: Test: [110/155] Time 0.924 (1.059) Loss 3.7242 (3.7488) Acc@1 0.000% (4.392%)
[11/29 14:52:29] modelnet40_eval INFO: Test: [120/155] Time 1.165 (1.059) Loss 3.7197 (3.7487) Acc@1 0.000% (4.029%)
[11/29 14:52:39] modelnet40_eval INFO: Test: [130/155] Time 1.071 (1.059) Loss 4.0915 (3.7496) Acc@1 0.000% (3.721%)
[11/29 14:52:50] modelnet40_eval INFO: Test: [140/155] Time 1.052 (1.059) Loss 4.0544 (3.7484) Acc@1 0.000% (3.457%)
[11/29 14:53:00] modelnet40_eval INFO: Test: [150/155] Time 1.138 (1.059) Loss 4.3561 (3.7516) Acc@1 0.000% (3.228%)
[11/29 14:53:04] modelnet40_eval INFO: * Acc@1 3.160%
[11/29 14:53:04] modelnet40_eval INFO: * Vote9 Acc@1 3.809%
It seems that the performance is not very good, can you help me solve it？
My calling command is python -m torch.distributed.launch --master_port 1234 --nproc_per_node 1 function/evaluate_modelnet_dist.py --cfg cfgs/modelnet/pospool_sin_cos_avg.yaml --load_path weights/pospool_sin_cos_avg.pth
THX!!

[zeliu98]

Hi @missbook520 ,
I'm sorry that the checkpoint provided here is not correspond to the latest code(we change all LeakyReLU to ReLU in the latest code), I'll re-train all models and update them soon. To use the current models, you can git checkout f388452b and then retry it.

Best,
Ze

[TaoHaozJ]

Hello, hasn't checkout been updated yet