Automated tutorials push

pytorch · Jan 30, 2025 · 0c2da3c · 0c2da3c
1 parent fed6e56
commit 0c2da3c
Show file tree

Hide file tree

Showing 196 changed files with 13,556 additions and 12,892 deletions.
diff --git a/_downloads/13cdb386a4b0dc48c626f32e6cf8681d/amp_recipe.ipynb b/_downloads/13cdb386a4b0dc48c626f32e6cf8681d/amp_recipe.ipynb
@@ -261,7 +261,7 @@
     "# The same ``GradScaler`` instance should be used for the entire convergence run.\n",
     "# If you perform multiple convergence runs in the same script, each run should use\n",
     "# a dedicated fresh ``GradScaler`` instance. ``GradScaler`` instances are lightweight.\n",
-    "scaler = torch.cuda.amp.GradScaler()\n",
+    "scaler = torch.amp.GradScaler(\"cuda\")\n",
     "\n",
     "for epoch in range(0): # 0 epochs, this section is for illustration only\n",
     "    for input, target in zip(data, targets):\n",
@@ -308,7 +308,7 @@
     "\n",
     "net = make_model(in_size, out_size, num_layers)\n",
     "opt = torch.optim.SGD(net.parameters(), lr=0.001)\n",
-    "scaler = torch.cuda.amp.GradScaler(enabled=use_amp)\n",
+    "scaler = torch.amp.GradScaler(\"cuda\" ,enabled=use_amp)\n",
     "\n",
     "start_timer()\n",
     "for epoch in range(epochs):\n",

diff --git a/_downloads/cadb3a57e7a6d7c149b5ae377caf36a8/amp_recipe.py b/_downloads/cadb3a57e7a6d7c149b5ae377caf36a8/amp_recipe.py
@@ -150,7 +150,7 @@ def make_model(in_size, out_size, num_layers):
 # The same ``GradScaler`` instance should be used for the entire convergence run.
 # If you perform multiple convergence runs in the same script, each run should use
 # a dedicated fresh ``GradScaler`` instance. ``GradScaler`` instances are lightweight.
-scaler = torch.cuda.amp.GradScaler()
+scaler = torch.amp.GradScaler("cuda")
 
 for epoch in range(0): # 0 epochs, this section is for illustration only
     for input, target in zip(data, targets):
@@ -182,7 +182,7 @@ def make_model(in_size, out_size, num_layers):
 
 net = make_model(in_size, out_size, num_layers)
 opt = torch.optim.SGD(net.parameters(), lr=0.001)
-scaler = torch.cuda.amp.GradScaler(enabled=use_amp)
+scaler = torch.amp.GradScaler("cuda" ,enabled=use_amp)
 
 start_timer()
 for epoch in range(epochs):

diff --git a/_images/sphx_glr_char_rnn_classification_tutorial_001.png b/_images/sphx_glr_char_rnn_classification_tutorial_001.png
diff --git a/_images/sphx_glr_char_rnn_classification_tutorial_002.png b/_images/sphx_glr_char_rnn_classification_tutorial_002.png
diff --git a/_images/sphx_glr_coding_ddpg_001.png b/_images/sphx_glr_coding_ddpg_001.png
diff --git a/_images/sphx_glr_dqn_with_rnn_tutorial_001.png b/_images/sphx_glr_dqn_with_rnn_tutorial_001.png
diff --git a/_images/sphx_glr_neural_style_tutorial_004.png b/_images/sphx_glr_neural_style_tutorial_004.png
diff --git a/_images/sphx_glr_pinmem_nonblock_001.png b/_images/sphx_glr_pinmem_nonblock_001.png
diff --git a/_images/sphx_glr_pinmem_nonblock_002.png b/_images/sphx_glr_pinmem_nonblock_002.png
diff --git a/_images/sphx_glr_pinmem_nonblock_003.png b/_images/sphx_glr_pinmem_nonblock_003.png
diff --git a/_images/sphx_glr_pinmem_nonblock_004.png b/_images/sphx_glr_pinmem_nonblock_004.png
diff --git a/_images/sphx_glr_reinforcement_ppo_001.png b/_images/sphx_glr_reinforcement_ppo_001.png
diff --git a/_images/sphx_glr_reinforcement_q_learning_001.png b/_images/sphx_glr_reinforcement_q_learning_001.png
diff --git a/_images/sphx_glr_semi_structured_sparse_001.png b/_images/sphx_glr_semi_structured_sparse_001.png
diff --git a/_images/sphx_glr_semi_structured_sparse_002.png b/_images/sphx_glr_semi_structured_sparse_002.png
diff --git a/_images/sphx_glr_spatial_transformer_tutorial_001.png b/_images/sphx_glr_spatial_transformer_tutorial_001.png
diff --git a/_images/sphx_glr_torchvision_tutorial_002.png b/_images/sphx_glr_torchvision_tutorial_002.png
diff --git a/_sources/advanced/coding_ddpg.rst.txt b/_sources/advanced/coding_ddpg.rst.txt
@@ -1634,26 +1634,26 @@ modules we need.
 
 
       0%|          | 0/10000 [00:00<?, ?it/s]
-      8%|8         | 800/10000 [00:00<00:05, 1660.50it/s]
-     16%|#6        | 1600/10000 [00:02<00:17, 477.55it/s]
-     24%|##4       | 2400/10000 [00:03<00:11, 674.49it/s]
-     32%|###2      | 3200/10000 [00:04<00:07, 850.24it/s]
-     40%|####      | 4000/10000 [00:04<00:06, 995.16it/s]
-     48%|####8     | 4800/10000 [00:05<00:04, 1116.81it/s]
-     56%|#####6    | 5600/10000 [00:05<00:03, 1208.77it/s]
-    reward: -2.34 (r0 = -2.00), reward eval: reward: -0.00, reward normalized=-2.59/6.17, grad norm= 147.92, loss_value= 295.62, loss_actor= 13.96, target value: -15.91:  56%|#####6    | 5600/10000 [00:06<00:03, 1208.77it/s]
-    reward: -2.34 (r0 = -2.00), reward eval: reward: -0.00, reward normalized=-2.59/6.17, grad norm= 147.92, loss_value= 295.62, loss_actor= 13.96, target value: -15.91:  64%|######4   | 6400/10000 [00:07<00:04, 841.61it/s] 
-    reward: -0.11 (r0 = -2.00), reward eval: reward: -0.00, reward normalized=-1.53/5.46, grad norm= 118.19, loss_value= 194.76, loss_actor= 10.63, target value: -10.14:  64%|######4   | 6400/10000 [00:08<00:04, 841.61it/s]
-    reward: -0.11 (r0 = -2.00), reward eval: reward: -0.00, reward normalized=-1.53/5.46, grad norm= 118.19, loss_value= 194.76, loss_actor= 10.63, target value: -10.14:  72%|#######2  | 7200/10000 [00:08<00:03, 700.19it/s]
-    reward: -2.33 (r0 = -2.00), reward eval: reward: -0.00, reward normalized=-2.42/5.58, grad norm= 182.04, loss_value= 220.44, loss_actor= 13.69, target value: -16.09:  72%|#######2  | 7200/10000 [00:09<00:03, 700.19it/s]
-    reward: -2.33 (r0 = -2.00), reward eval: reward: -0.00, reward normalized=-2.42/5.58, grad norm= 182.04, loss_value= 220.44, loss_actor= 13.69, target value: -16.09:  80%|########  | 8000/10000 [00:10<00:03, 623.68it/s]
-    reward: -4.44 (r0 = -2.00), reward eval: reward: -0.00, reward normalized=-2.44/4.89, grad norm= 111.35, loss_value= 211.11, loss_actor= 15.74, target value: -15.42:  80%|########  | 8000/10000 [00:11<00:03, 623.68it/s]
-    reward: -4.44 (r0 = -2.00), reward eval: reward: -0.00, reward normalized=-2.44/4.89, grad norm= 111.35, loss_value= 211.11, loss_actor= 15.74, target value: -15.42:  88%|########8 | 8800/10000 [00:12<00:02, 588.63it/s]
-    reward: -4.96 (r0 = -2.00), reward eval: reward: -5.98, reward normalized=-2.32/4.85, grad norm= 54.44, loss_value= 165.27, loss_actor= 16.38, target value: -16.11:  88%|########8 | 8800/10000 [00:14<00:02, 588.63it/s] 
-    reward: -4.96 (r0 = -2.00), reward eval: reward: -5.98, reward normalized=-2.32/4.85, grad norm= 54.44, loss_value= 165.27, loss_actor= 16.38, target value: -16.11:  96%|#########6| 9600/10000 [00:15<00:01, 399.47it/s]
-    reward: -4.86 (r0 = -2.00), reward eval: reward: -5.98, reward normalized=-3.02/4.89, grad norm= 173.10, loss_value= 234.27, loss_actor= 13.70, target value: -21.43:  96%|#########6| 9600/10000 [00:16<00:01, 399.47it/s]
-    reward: -4.86 (r0 = -2.00), reward eval: reward: -5.98, reward normalized=-3.02/4.89, grad norm= 173.10, loss_value= 234.27, loss_actor= 13.70, target value: -21.43: : 10400it [00:18, 364.88it/s]                        
-    reward: -4.93 (r0 = -2.00), reward eval: reward: -5.98, reward normalized=-3.38/3.91, grad norm= 120.25, loss_value= 129.44, loss_actor= 15.23, target value: -23.98: : 10400it [00:18, 364.88it/s]
+      8%|8         | 800/10000 [00:00<00:05, 1683.66it/s]
+     16%|#6        | 1600/10000 [00:02<00:17, 485.23it/s]
+     24%|##4       | 2400/10000 [00:03<00:11, 687.00it/s]
+     32%|###2      | 3200/10000 [00:04<00:07, 877.15it/s]
+     40%|####      | 4000/10000 [00:04<00:05, 1045.73it/s]
+     48%|####8     | 4800/10000 [00:05<00:04, 1188.29it/s]
+     56%|#####6    | 5600/10000 [00:05<00:03, 1288.59it/s]
+    reward: -2.38 (r0 = -3.75), reward eval: reward: -0.01, reward normalized=-2.90/6.27, grad norm= 184.34, loss_value= 346.66, loss_actor= 15.34, target value: -18.57:  56%|#####6    | 5600/10000 [00:06<00:03, 1288.59it/s]
+    reward: -2.38 (r0 = -3.75), reward eval: reward: -0.01, reward normalized=-2.90/6.27, grad norm= 184.34, loss_value= 346.66, loss_actor= 15.34, target value: -18.57:  64%|######4   | 6400/10000 [00:07<00:04, 876.74it/s] 
+    reward: -0.14 (r0 = -3.75), reward eval: reward: -0.01, reward normalized=-2.44/5.85, grad norm= 49.14, loss_value= 239.65, loss_actor= 14.06, target value: -14.04:  64%|######4   | 6400/10000 [00:07<00:04, 876.74it/s] 
+    reward: -0.14 (r0 = -3.75), reward eval: reward: -0.01, reward normalized=-2.44/5.85, grad norm= 49.14, loss_value= 239.65, loss_actor= 14.06, target value: -14.04:  72%|#######2  | 7200/10000 [00:08<00:03, 713.38it/s]
+    reward: -2.36 (r0 = -3.75), reward eval: reward: -0.01, reward normalized=-2.26/5.79, grad norm= 128.31, loss_value= 254.09, loss_actor= 12.59, target value: -14.58:  72%|#######2  | 7200/10000 [00:09<00:03, 713.38it/s]
+    reward: -2.36 (r0 = -3.75), reward eval: reward: -0.01, reward normalized=-2.26/5.79, grad norm= 128.31, loss_value= 254.09, loss_actor= 12.59, target value: -14.58:  80%|########  | 8000/10000 [00:10<00:03, 627.39it/s]
+    reward: -4.64 (r0 = -3.75), reward eval: reward: -0.01, reward normalized=-3.11/5.26, grad norm= 115.52, loss_value= 239.16, loss_actor= 18.21, target value: -19.89:  80%|########  | 8000/10000 [00:11<00:03, 627.39it/s]
+    reward: -4.64 (r0 = -3.75), reward eval: reward: -0.01, reward normalized=-3.11/5.26, grad norm= 115.52, loss_value= 239.16, loss_actor= 18.21, target value: -19.89:  88%|########8 | 8800/10000 [00:11<00:02, 582.30it/s]
+    reward: -4.99 (r0 = -3.75), reward eval: reward: -5.59, reward normalized=-3.11/5.35, grad norm= 68.47, loss_value= 219.28, loss_actor= 19.75, target value: -19.78:  88%|########8 | 8800/10000 [00:14<00:02, 582.30it/s] 
+    reward: -4.99 (r0 = -3.75), reward eval: reward: -5.59, reward normalized=-3.11/5.35, grad norm= 68.47, loss_value= 219.28, loss_actor= 19.75, target value: -19.78:  96%|#########6| 9600/10000 [00:15<00:00, 404.39it/s]
+    reward: -5.39 (r0 = -3.75), reward eval: reward: -5.59, reward normalized=-3.21/5.31, grad norm= 204.79, loss_value= 298.20, loss_actor= 19.41, target value: -22.63:  96%|#########6| 9600/10000 [00:15<00:00, 404.39it/s]
+    reward: -5.39 (r0 = -3.75), reward eval: reward: -5.59, reward normalized=-3.21/5.31, grad norm= 204.79, loss_value= 298.20, loss_actor= 19.41, target value: -22.63: : 10400it [00:17, 366.15it/s]                        
+    reward: -4.68 (r0 = -3.75), reward eval: reward: -5.59, reward normalized=-3.57/4.27, grad norm= 74.57, loss_value= 203.88, loss_actor= 23.34, target value: -24.98: : 10400it [00:18, 366.15it/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  29.005 seconds)
+   **Total running time of the script:** ( 0 minutes  28.797 seconds)
 
 
 .. _sphx_glr_download_advanced_coding_ddpg.py:

diff --git a/_sources/advanced/dynamic_quantization_tutorial.rst.txt b/_sources/advanced/dynamic_quantization_tutorial.rst.txt
@@ -517,9 +517,9 @@ models run single threaded.
  .. code-block:: none
 
     loss: 5.167
-    elapsed time (seconds): 208.2
+    elapsed time (seconds): 202.0
     loss: 5.168
-    elapsed time (seconds): 115.8
+    elapsed time (seconds): 113.8
 
 
 
@@ -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  34.880 seconds)
+   **Total running time of the script:** ( 5 minutes  26.458 seconds)
 
 
 .. _sphx_glr_download_advanced_dynamic_quantization_tutorial.py:

diff --git a/_sources/advanced/neural_style_tutorial.rst.txt b/_sources/advanced/neural_style_tutorial.rst.txt
@@ -410,32 +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]
-      4%|3         | 20.5M/548M [00:00<00:02, 215MB/s]
-      8%|7         | 41.5M/548M [00:00<00:02, 217MB/s]
-     11%|#1        | 62.6M/548M [00:00<00:02, 219MB/s]
-     15%|#5        | 83.9M/548M [00:00<00:02, 220MB/s]
-     19%|#9        | 105M/548M [00:00<00:02, 221MB/s] 
-     23%|##3       | 126M/548M [00:00<00:02, 221MB/s]
-     27%|##6       | 148M/548M [00:00<00:01, 221MB/s]
-     31%|###       | 169M/548M [00:00<00:01, 221MB/s]
-     35%|###4      | 190M/548M [00:00<00:01, 221MB/s]
-     39%|###8      | 211M/548M [00:01<00:01, 221MB/s]
-     42%|####2     | 232M/548M [00:01<00:01, 221MB/s]
-     46%|####6     | 254M/548M [00:01<00:01, 222MB/s]
-     50%|#####     | 275M/548M [00:01<00:01, 222MB/s]
-     54%|#####4    | 296M/548M [00:01<00:01, 221MB/s]
-     58%|#####7    | 317M/548M [00:01<00:01, 221MB/s]
-     62%|######1   | 338M/548M [00:01<00:00, 221MB/s]
-     66%|######5   | 360M/548M [00:01<00:00, 221MB/s]
-     69%|######9   | 381M/548M [00:01<00:00, 221MB/s]
-     73%|#######3  | 402M/548M [00:01<00:00, 221MB/s]
-     77%|#######7  | 423M/548M [00:02<00:00, 221MB/s]
-     81%|########1 | 444M/548M [00:02<00:00, 221MB/s]
-     85%|########4 | 465M/548M [00:02<00:00, 221MB/s]
-     89%|########8 | 486M/548M [00:02<00:00, 221MB/s]
-     93%|#########2| 507M/548M [00:02<00:00, 221MB/s]
-     96%|#########6| 528M/548M [00:02<00:00, 221MB/s]
-    100%|##########| 548M/548M [00:02<00:00, 221MB/s]
+      4%|3         | 20.4M/548M [00:00<00:02, 213MB/s]
+      8%|7         | 41.1M/548M [00:00<00:02, 216MB/s]
+     11%|#1        | 62.0M/548M [00:00<00:02, 217MB/s]
+     15%|#5        | 82.9M/548M [00:00<00:02, 217MB/s]
+     19%|#8        | 104M/548M [00:00<00:02, 218MB/s] 
+     23%|##2       | 125M/548M [00:00<00:02, 218MB/s]
+     27%|##6       | 146M/548M [00:00<00:01, 218MB/s]
+     30%|###       | 166M/548M [00:00<00:01, 218MB/s]
+     34%|###4      | 187M/548M [00:00<00:01, 218MB/s]
+     38%|###8      | 208M/548M [00:01<00:01, 219MB/s]
+     42%|####1     | 229M/548M [00:01<00:01, 219MB/s]
+     46%|####5     | 250M/548M [00:01<00:01, 219MB/s]
+     49%|####9     | 271M/548M [00:01<00:01, 219MB/s]
+     53%|#####3    | 292M/548M [00:01<00:01, 219MB/s]
+     57%|#####7    | 313M/548M [00:01<00:01, 219MB/s]
+     61%|######    | 334M/548M [00:01<00:01, 219MB/s]
+     65%|######4   | 355M/548M [00:01<00:00, 219MB/s]
+     69%|######8   | 376M/548M [00:01<00:00, 219MB/s]
+     72%|#######2  | 397M/548M [00:01<00:00, 219MB/s]
+     76%|#######6  | 418M/548M [00:02<00:00, 219MB/s]
+     80%|########  | 439M/548M [00:02<00:00, 219MB/s]
+     84%|########3 | 460M/548M [00:02<00:00, 219MB/s]
+     88%|########7 | 481M/548M [00:02<00:00, 219MB/s]
+     92%|#########1| 502M/548M [00:02<00:00, 219MB/s]
+     95%|#########5| 523M/548M [00:02<00:00, 219MB/s]
+     99%|#########9| 544M/548M [00:02<00:00, 219MB/s]
+    100%|##########| 548M/548M [00:02<00:00, 219MB/s]
 
 
 
@@ -756,22 +757,22 @@ Finally, we can run the algorithm.
 
     Optimizing..
     run [50]:
-    Style Loss : 4.243999 Content Loss: 4.230177
+    Style Loss : 4.124988 Content Loss: 4.180835
 
     run [100]:
-    Style Loss : 1.153091 Content Loss: 3.027826
+    Style Loss : 1.149009 Content Loss: 3.025290
 
     run [150]:
-    Style Loss : 0.714814 Content Loss: 2.653670
+    Style Loss : 0.722513 Content Loss: 2.658751
 
     run [200]:
-    Style Loss : 0.479303 Content Loss: 2.491420
+    Style Loss : 0.484897 Content Loss: 2.496019
 
     run [250]:
-    Style Loss : 0.347053 Content Loss: 2.402259
+    Style Loss : 0.348076 Content Loss: 2.404442
 
     run [300]:
-    Style Loss : 0.262986 Content Loss: 2.348989
+    Style Loss : 0.265312 Content Loss: 2.350962
 
 
 
@@ -780,7 +781,7 @@ Finally, we can run the algorithm.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  38.572 seconds)
+   **Total running time of the script:** ( 0 minutes  38.498 seconds)
 
 
 .. _sphx_glr_download_advanced_neural_style_tutorial.py:

diff --git a/_sources/advanced/numpy_extensions_tutorial.rst.txt b/_sources/advanced/numpy_extensions_tutorial.rst.txt
@@ -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  0.602 seconds)
+   **Total running time of the script:** ( 0 minutes  0.568 seconds)
 
 
 .. _sphx_glr_download_advanced_numpy_extensions_tutorial.py: