Automated tutorials push

Author: pytorchbot

_downloads/562d6bd0e2a429f010fcf8007f6a7cac/pinmem_nonblock.py

@@ -547,7 +547,7 @@ def pin_copy_to_device_nonblocking(*tensors):
 
 i = -1
 for i in range(100):
-    # Create a tensor in pin-memory
+    # Create a tensor in pageable memory
     cpu_tensor = torch.ones(1024, 1024)
     torch.cuda.synchronize()
     # Send the tensor to CUDA

_downloads/6a760a243fcbf87fb3368be3d4d860ee/pinmem_nonblock.ipynb

@@ -758,7 +758,7 @@
    "source": [
     "i = -1\n",
     "for i in range(100):\n",
-    "    # Create a tensor in pin-memory\n",
+    "    # Create a tensor in pageable memory\n",
     "    cpu_tensor = torch.ones(1024, 1024)\n",
     "    torch.cuda.synchronize()\n",
     "    # Send the tensor to CUDA\n",

_images/sphx_glr_char_rnn_classification_tutorial_001.png

@@ -1634,26 +1634,26 @@ modules we need.
 
 
       0%|          | 0/10000 [00:00<?, ?it/s]
-      8%|8         | 800/10000 [00:00<00:05, 1706.86it/s]
-     16%|#6        | 1600/10000 [00:02<00:16, 494.56it/s]
-     24%|##4       | 2400/10000 [00:03<00:10, 714.14it/s]
-     32%|###2      | 3200/10000 [00:03<00:07, 927.36it/s]
-     40%|####      | 4000/10000 [00:04<00:05, 1108.64it/s]
-     48%|####8     | 4800/10000 [00:04<00:04, 1263.34it/s]
-     56%|#####6    | 5600/10000 [00:05<00:03, 1386.67it/s]
-    reward: -2.53 (r0 = -2.38), reward eval: reward:  0.00, reward normalized=-2.87/6.26, grad norm= 172.44, loss_value= 372.48, loss_actor= 13.77, target value: -17.13:  56%|#####6    | 5600/10000 [00:06<00:03, 1386.67it/s]
-    reward: -2.53 (r0 = -2.38), reward eval: reward:  0.00, reward normalized=-2.87/6.26, grad norm= 172.44, loss_value= 372.48, loss_actor= 13.77, target value: -17.13:  64%|######4   | 6400/10000 [00:06<00:03, 914.00it/s] 
-    reward: -0.10 (r0 = -2.38), reward eval: reward:  0.00, reward normalized=-1.77/5.75, grad norm= 83.74, loss_value= 243.26, loss_actor= 12.18, target value: -11.92:  64%|######4   | 6400/10000 [00:07<00:03, 914.00it/s] 
-    reward: -0.10 (r0 = -2.38), reward eval: reward:  0.00, reward normalized=-1.77/5.75, grad norm= 83.74, loss_value= 243.26, loss_actor= 12.18, target value: -11.92:  72%|#######2  | 7200/10000 [00:08<00:03, 732.98it/s]
-    reward: -3.06 (r0 = -2.38), reward eval: reward:  0.00, reward normalized=-2.10/6.37, grad norm= 175.25, loss_value= 410.52, loss_actor= 14.70, target value: -14.73:  72%|#######2  | 7200/10000 [00:09<00:03, 732.98it/s]
-    reward: -3.06 (r0 = -2.38), reward eval: reward:  0.00, reward normalized=-2.10/6.37, grad norm= 175.25, loss_value= 410.52, loss_actor= 14.70, target value: -14.73:  80%|########  | 8000/10000 [00:09<00:03, 648.25it/s]
-    reward: -5.12 (r0 = -2.38), reward eval: reward:  0.00, reward normalized=-2.76/5.11, grad norm= 204.29, loss_value= 256.34, loss_actor= 13.80, target value: -18.21:  80%|########  | 8000/10000 [00:10<00:03, 648.25it/s]
-    reward: -5.12 (r0 = -2.38), reward eval: reward:  0.00, reward normalized=-2.76/5.11, grad norm= 204.29, loss_value= 256.34, loss_actor= 13.80, target value: -18.21:  88%|########8 | 8800/10000 [00:11<00:01, 603.43it/s]
-    reward: -4.34 (r0 = -2.38), reward eval: reward: -3.72, reward normalized=-2.41/4.96, grad norm= 83.94, loss_value= 184.70, loss_actor= 12.14, target value: -16.55:  88%|########8 | 8800/10000 [00:14<00:01, 603.43it/s] 
-    reward: -4.34 (r0 = -2.38), reward eval: reward: -3.72, reward normalized=-2.41/4.96, grad norm= 83.94, loss_value= 184.70, loss_actor= 12.14, target value: -16.55:  96%|#########6| 9600/10000 [00:14<00:00, 408.81it/s]
-    reward: -12.30 (r0 = -2.38), reward eval: reward: -3.72, reward normalized=-3.23/6.55, grad norm= 181.45, loss_value= 321.95, loss_actor= 15.06, target value: -22.91:  96%|#########6| 9600/10000 [00:15<00:00, 408.81it/s]
-    reward: -12.30 (r0 = -2.38), reward eval: reward: -3.72, reward normalized=-3.23/6.55, grad norm= 181.45, loss_value= 321.95, loss_actor= 15.06, target value: -22.91: : 10400it [00:17, 371.73it/s]                        
-    reward: -3.25 (r0 = -2.38), reward eval: reward: -3.72, reward normalized=-4.02/6.12, grad norm= 131.26, loss_value= 237.20, loss_actor= 23.33, target value: -27.38: : 10400it [00:18, 371.73it/s] 
+      8%|8         | 800/10000 [00:00<00:05, 1686.02it/s]
+     16%|#6        | 1600/10000 [00:03<00:17, 474.11it/s]
+     24%|##4       | 2400/10000 [00:03<00:11, 671.52it/s]
+     32%|###2      | 3200/10000 [00:04<00:07, 855.29it/s]
+     40%|####      | 4000/10000 [00:04<00:05, 1017.39it/s]
+     48%|####8     | 4800/10000 [00:05<00:04, 1163.51it/s]
+     56%|#####6    | 5600/10000 [00:05<00:03, 1272.13it/s]
+    reward: -2.71 (r0 = -3.45), reward eval: reward:  0.01, reward normalized=-2.89/6.36, grad norm= 139.40, loss_value= 452.21, loss_actor= 15.36, target value: -17.24:  56%|#####6    | 5600/10000 [00:06<00:03, 1272.13it/s]
+    reward: -2.71 (r0 = -3.45), reward eval: reward:  0.01, reward normalized=-2.89/6.36, grad norm= 139.40, loss_value= 452.21, loss_actor= 15.36, target value: -17.24:  64%|######4   | 6400/10000 [00:07<00:04, 870.02it/s] 
+    reward: -0.14 (r0 = -3.45), reward eval: reward:  0.01, reward normalized=-2.44/5.99, grad norm= 161.20, loss_value= 324.10, loss_actor= 12.66, target value: -15.55:  64%|######4   | 6400/10000 [00:07<00:04, 870.02it/s]
+    reward: -0.14 (r0 = -3.45), reward eval: reward:  0.01, reward normalized=-2.44/5.99, grad norm= 161.20, loss_value= 324.10, loss_actor= 12.66, target value: -15.55:  72%|#######2  | 7200/10000 [00:08<00:03, 703.58it/s]
+    reward: -2.45 (r0 = -3.45), reward eval: reward:  0.01, reward normalized=-2.42/5.98, grad norm= 130.59, loss_value= 312.04, loss_actor= 12.01, target value: -15.33:  72%|#######2  | 7200/10000 [00:09<00:03, 703.58it/s]
+    reward: -2.45 (r0 = -3.45), reward eval: reward:  0.01, reward normalized=-2.42/5.98, grad norm= 130.59, loss_value= 312.04, loss_actor= 12.01, target value: -15.33:  80%|########  | 8000/10000 [00:10<00:03, 624.50it/s]
+    reward: -4.12 (r0 = -3.45), reward eval: reward:  0.01, reward normalized=-2.73/5.00, grad norm= 205.76, loss_value= 217.68, loss_actor= 16.33, target value: -18.26:  80%|########  | 8000/10000 [00:11<00:03, 624.50it/s]
+    reward: -4.12 (r0 = -3.45), reward eval: reward:  0.01, reward normalized=-2.73/5.00, grad norm= 205.76, loss_value= 217.68, loss_actor= 16.33, target value: -18.26:  88%|########8 | 8800/10000 [00:12<00:02, 581.06it/s]
+    reward: -5.10 (r0 = -3.45), reward eval: reward: -7.09, reward normalized=-2.55/5.13, grad norm= 123.25, loss_value= 197.78, loss_actor= 19.07, target value: -17.02:  88%|########8 | 8800/10000 [00:14<00:02, 581.06it/s]
+    reward: -5.10 (r0 = -3.45), reward eval: reward: -7.09, reward normalized=-2.55/5.13, grad norm= 123.25, loss_value= 197.78, loss_actor= 19.07, target value: -17.02:  96%|#########6| 9600/10000 [00:15<00:01, 391.08it/s]
+    reward: -5.12 (r0 = -3.45), reward eval: reward: -7.09, reward normalized=-3.09/5.14, grad norm= 269.24, loss_value= 261.51, loss_actor= 18.04, target value: -22.19:  96%|#########6| 9600/10000 [00:16<00:01, 391.08it/s]
+    reward: -5.12 (r0 = -3.45), reward eval: reward: -7.09, reward normalized=-3.09/5.14, grad norm= 269.24, loss_value= 261.51, loss_actor= 18.04, target value: -22.19: : 10400it [00:18, 354.47it/s]                        
+    reward: -5.27 (r0 = -3.45), reward eval: reward: -7.09, reward normalized=-3.08/5.03, grad norm= 66.19, loss_value= 194.93, loss_actor= 20.81, target value: -22.46: : 10400it [00:19, 354.47it/s] 
 
 
 
@@ -1723,7 +1723,7 @@ To iterate further on this loss module we might consider:
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  28.061 seconds)
+   **Total running time of the script:** ( 0 minutes  29.218 seconds)
 
 
 .. _sphx_glr_download_advanced_coding_ddpg.py:

_images/sphx_glr_char_rnn_classification_tutorial_002.png

@@ -517,9 +517,9 @@ models run single threaded.
  .. code-block:: none
 
     loss: 5.167
-    elapsed time (seconds): 205.9
+    elapsed time (seconds): 199.5
     loss: 5.168
-    elapsed time (seconds): 117.1
+    elapsed time (seconds): 118.4
 
 
 
@@ -541,7 +541,7 @@ Thanks for reading! As always, we welcome any feedback, so please create an issu
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 5 minutes  31.533 seconds)
+   **Total running time of the script:** ( 5 minutes  26.291 seconds)
 
 
 .. _sphx_glr_download_advanced_dynamic_quantization_tutorial.py:

_images/sphx_glr_coding_ddpg_001.png

@@ -410,33 +410,33 @@ network to evaluation mode using ``.eval()``.
     Downloading: "https://download.pytorch.org/models/vgg19-dcbb9e9d.pth" to /var/lib/ci-user/.cache/torch/hub/checkpoints/vgg19-dcbb9e9d.pth
 
       0%|          | 0.00/548M [00:00<?, ?B/s]
-      3%|3         | 17.5M/548M [00:00<00:03, 182MB/s]
-      6%|6         | 34.9M/548M [00:00<00:03, 150MB/s]
-     10%|#         | 55.5M/548M [00:00<00:02, 177MB/s]
-     14%|#3        | 76.1M/548M [00:00<00:02, 191MB/s]
-     18%|#7        | 96.8M/548M [00:00<00:02, 200MB/s]
-     21%|##1       | 117M/548M [00:00<00:02, 205MB/s] 
-     25%|##5       | 138M/548M [00:00<00:02, 208MB/s]
-     29%|##8       | 159M/548M [00:00<00:01, 211MB/s]
-     33%|###2      | 179M/548M [00:00<00:01, 212MB/s]
-     36%|###6      | 200M/548M [00:01<00:01, 214MB/s]
-     40%|####      | 221M/548M [00:01<00:01, 214MB/s]
-     44%|####4     | 241M/548M [00:01<00:01, 215MB/s]
-     48%|####7     | 262M/548M [00:01<00:01, 215MB/s]
-     52%|#####1    | 282M/548M [00:01<00:01, 215MB/s]
-     55%|#####5    | 303M/548M [00:01<00:01, 215MB/s]
-     59%|#####9    | 324M/548M [00:01<00:01, 215MB/s]
-     63%|######2   | 344M/548M [00:01<00:00, 215MB/s]
-     67%|######6   | 365M/548M [00:01<00:00, 216MB/s]
-     70%|#######   | 386M/548M [00:01<00:00, 216MB/s]
-     74%|#######4  | 406M/548M [00:02<00:00, 216MB/s]
-     78%|#######7  | 427M/548M [00:02<00:00, 216MB/s]
-     82%|########1 | 448M/548M [00:02<00:00, 216MB/s]
-     85%|########5 | 468M/548M [00:02<00:00, 216MB/s]
-     89%|########9 | 489M/548M [00:02<00:00, 216MB/s]
-     93%|#########3| 510M/548M [00:02<00:00, 216MB/s]
-     97%|#########6| 531M/548M [00:02<00:00, 216MB/s]
-    100%|##########| 548M/548M [00:02<00:00, 211MB/s]
+      4%|3         | 20.0M/548M [00:00<00:02, 209MB/s]
+      7%|7         | 40.4M/548M [00:00<00:02, 211MB/s]
+     11%|#1        | 60.9M/548M [00:00<00:02, 213MB/s]
+     15%|#4        | 81.5M/548M [00:00<00:02, 214MB/s]
+     19%|#8        | 102M/548M [00:00<00:02, 214MB/s] 
+     22%|##2       | 122M/548M [00:00<00:02, 214MB/s]
+     26%|##6       | 143M/548M [00:00<00:01, 215MB/s]
+     30%|##9       | 164M/548M [00:00<00:01, 215MB/s]
+     34%|###3      | 184M/548M [00:00<00:01, 215MB/s]
+     37%|###7      | 205M/548M [00:01<00:01, 215MB/s]
+     41%|####1     | 225M/548M [00:01<00:01, 215MB/s]
+     45%|####4     | 246M/548M [00:01<00:01, 215MB/s]
+     49%|####8     | 267M/548M [00:01<00:01, 215MB/s]
+     52%|#####2    | 287M/548M [00:01<00:01, 215MB/s]
+     56%|#####6    | 308M/548M [00:01<00:01, 215MB/s]
+     60%|#####9    | 328M/548M [00:01<00:01, 215MB/s]
+     64%|######3   | 349M/548M [00:01<00:00, 215MB/s]
+     67%|######7   | 370M/548M [00:01<00:00, 215MB/s]
+     71%|#######1  | 390M/548M [00:01<00:00, 215MB/s]
+     75%|#######4  | 411M/548M [00:02<00:00, 215MB/s]
+     79%|#######8  | 431M/548M [00:02<00:00, 215MB/s]
+     82%|########2 | 452M/548M [00:02<00:00, 215MB/s]
+     86%|########6 | 473M/548M [00:02<00:00, 215MB/s]
+     90%|########9 | 493M/548M [00:02<00:00, 215MB/s]
+     94%|#########3| 514M/548M [00:02<00:00, 215MB/s]
+     98%|#########7| 534M/548M [00:02<00:00, 215MB/s]
+    100%|##########| 548M/548M [00:02<00:00, 215MB/s]
 
 
 
@@ -757,22 +757,22 @@ Finally, we can run the algorithm.
 
     Optimizing..
     run [50]:
-    Style Loss : 4.183072 Content Loss: 4.165503
+    Style Loss : 3.933726 Content Loss: 4.069746
 
     run [100]:
-    Style Loss : 1.164476 Content Loss: 3.053042
+    Style Loss : 1.115454 Content Loss: 3.014786
 
     run [150]:
-    Style Loss : 0.727555 Content Loss: 2.657095
+    Style Loss : 0.703395 Content Loss: 2.641813
 
     run [200]:
-    Style Loss : 0.487632 Content Loss: 2.498033
+    Style Loss : 0.472471 Content Loss: 2.486287
 
     run [250]:
-    Style Loss : 0.352444 Content Loss: 2.405909
+    Style Loss : 0.340303 Content Loss: 2.399467
 
     run [300]:
-    Style Loss : 0.266894 Content Loss: 2.351026
+    Style Loss : 0.262586 Content Loss: 2.348407
 
 
 
@@ -781,7 +781,7 @@ Finally, we can run the algorithm.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  34.609 seconds)
+   **Total running time of the script:** ( 0 minutes  36.551 seconds)
 
 
 .. _sphx_glr_download_advanced_neural_style_tutorial.py:

_images/sphx_glr_dqn_with_rnn_tutorial_001.png

@@ -303,7 +303,7 @@ The backward pass computes the gradient ``wrt`` the input and the gradient ``wrt
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  0.592 seconds)
+   **Total running time of the script:** ( 0 minutes  0.613 seconds)
 
 
 .. _sphx_glr_download_advanced_numpy_extensions_tutorial.py: