Automated tutorials push

Author: pytorchbot

_downloads/0ccffddcfee1f815c02241b985844376/torch_compile_user_defined_triton_kernel_tutorial.py

@@ -227,12 +227,6 @@ def sin_kernel(
     output = tl.sin(x)
     tl.store(out_ptr + offsets, output, mask=mask)
 
-def sin_triton(x):
-    out = torch.empty_like(x)
-    n_elements = x.numel()
-    sin_kernel[(n_elements,)](x, out, n_elements, BLOCK_SIZE=4)
-    return out
-
 ######################################################################
 # You can invoke the ``triton_op`` in one of the following two ways.
 

_downloads/3195443a0ced3cabc0ad643537bdb5cd/introyt1_tutorial.ipynb

@@ -34,7 +34,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "fd93c4dd",
+   "id": "08cb6c00",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -50,7 +50,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "13993efb",
+   "id": "76b87d77",
    "metadata": {},
    "source": [
     "\n",

_downloads/4355e2cef7d17548f1e25f97a62828c4/template_tutorial.ipynb

@@ -31,7 +31,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "0b7bd3cd",
+   "id": "c26dff5f",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -47,7 +47,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "4c9e9395",
+   "id": "24e76ea7",
    "metadata": {},
    "source": [
     "\n",

_downloads/63a0f0fc7b3ffb15d3a5ac8db3d521ee/tensors_deeper_tutorial.ipynb

@@ -34,7 +34,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "88e7740c",
+   "id": "a44f97a5",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -50,7 +50,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "6aaaa644",
+   "id": "a9601b89",
    "metadata": {},
    "source": [
     "\n",

_downloads/770632dd3941d2a51b831c52ded57aa2/trainingyt.ipynb

@@ -35,7 +35,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "c0f2ffb6",
+   "id": "73319286",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -51,7 +51,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "58b8afc8",
+   "id": "f4c949b1",
    "metadata": {},
    "source": [
     "\n",

_downloads/c28f42852d456daf9af72da6c6909556/captumyt.ipynb

@@ -37,7 +37,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "c49ae6d6",
+   "id": "e14c1480",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -53,7 +53,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "0fec58ed",
+   "id": "0c23d28f",
    "metadata": {},
    "source": [
     "\n",

_downloads/e2e556f6b4693c2cef716dd7f40caaf6/tensorboardyt_tutorial.ipynb

@@ -35,7 +35,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "e8788757",
+   "id": "8c4b1dc8",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -51,7 +51,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "a71e84dd",
+   "id": "de347df5",
    "metadata": {},
    "source": [
     "\n",

_downloads/ed9d4f94afb79f7dada6742a06c486a5/autogradyt_tutorial.ipynb

@@ -34,7 +34,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "29285e10",
+   "id": "71e2b25d",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -50,7 +50,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "125f6aa8",
+   "id": "2b11e675",
    "metadata": {},
    "source": [
     "\n",

_downloads/f827f181506a79226f4ffbcf7c9a5a50/torch_compile_user_defined_triton_kernel_tutorial.ipynb

@@ -283,13 +283,7 @@
     "    mask = offsets < n_elements\n",
     "    x = tl.load(in_ptr0 + offsets, mask=mask)\n",
     "    output = tl.sin(x)\n",
-    "    tl.store(out_ptr + offsets, output, mask=mask)\n",
-    "\n",
-    "def sin_triton(x):\n",
-    "    out = torch.empty_like(x)\n",
-    "    n_elements = x.numel()\n",
-    "    sin_kernel[(n_elements,)](x, out, n_elements, BLOCK_SIZE=4)\n",
-    "    return out"
+    "    tl.store(out_ptr + offsets, output, mask=mask)"
    ]
   },
   {

_downloads/fe726e041160526cf828806536922cf6/modelsyt_tutorial.ipynb

@@ -34,7 +34,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "6f1ec145",
+   "id": "2ebda2a4",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -50,7 +50,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "d3b40f65",
+   "id": "a3d22bbd",
    "metadata": {},
    "source": [
     "\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<00:06, 1410.71it/s]
-     16%|#6        | 1600/10000 [00:03<00:22, 378.10it/s]
-     24%|##4       | 2400/10000 [00:04<00:13, 545.68it/s]
-     32%|###2      | 3200/10000 [00:05<00:09, 722.60it/s]
-     40%|####      | 4000/10000 [00:05<00:06, 878.99it/s]
-     48%|####8     | 4800/10000 [00:06<00:05, 1015.48it/s]
-     56%|#####6    | 5600/10000 [00:06<00:04, 1098.42it/s]
-    reward: -2.34 (r0 = -2.37), reward eval: reward: -0.01, reward normalized=-2.61/6.06, grad norm= 57.52, loss_value= 275.17, loss_actor= 15.28, target value: -16.49:  56%|#####6    | 5600/10000 [00:07<00:04, 1098.42it/s]
-    reward: -2.34 (r0 = -2.37), reward eval: reward: -0.01, reward normalized=-2.61/6.06, grad norm= 57.52, loss_value= 275.17, loss_actor= 15.28, target value: -16.49:  64%|######4   | 6400/10000 [00:08<00:04, 794.01it/s] 
-    reward: -0.13 (r0 = -2.37), reward eval: reward: -0.01, reward normalized=-1.97/5.52, grad norm= 41.54, loss_value= 239.26, loss_actor= 12.43, target value: -12.86:  64%|######4   | 6400/10000 [00:09<00:04, 794.01it/s]
-    reward: -0.13 (r0 = -2.37), reward eval: reward: -0.01, reward normalized=-1.97/5.52, grad norm= 41.54, loss_value= 239.26, loss_actor= 12.43, target value: -12.86:  72%|#######2  | 7200/10000 [00:09<00:04, 673.52it/s]
-    reward: -2.83 (r0 = -2.37), reward eval: reward: -0.01, reward normalized=-2.63/6.00, grad norm= 197.69, loss_value= 303.71, loss_actor= 14.14, target value: -17.00:  72%|#######2  | 7200/10000 [00:10<00:04, 673.52it/s]
-    reward: -2.83 (r0 = -2.37), reward eval: reward: -0.01, reward normalized=-2.63/6.00, grad norm= 197.69, loss_value= 303.71, loss_actor= 14.14, target value: -17.00:  80%|########  | 8000/10000 [00:11<00:03, 607.26it/s]
-    reward: -4.58 (r0 = -2.37), reward eval: reward: -0.01, reward normalized=-2.40/4.96, grad norm= 87.82, loss_value= 180.65, loss_actor= 17.24, target value: -15.56:  80%|########  | 8000/10000 [00:12<00:03, 607.26it/s] 
-    reward: -4.58 (r0 = -2.37), reward eval: reward: -0.01, reward normalized=-2.40/4.96, grad norm= 87.82, loss_value= 180.65, loss_actor= 17.24, target value: -15.56:  88%|########8 | 8800/10000 [00:13<00:02, 572.64it/s]
-    reward: -4.56 (r0 = -2.37), reward eval: reward: -5.33, reward normalized=-2.60/4.73, grad norm= 134.09, loss_value= 156.18, loss_actor= 18.81, target value: -17.27:  88%|########8 | 8800/10000 [00:16<00:02, 572.64it/s]
-    reward: -4.56 (r0 = -2.37), reward eval: reward: -5.33, reward normalized=-2.60/4.73, grad norm= 134.09, loss_value= 156.18, loss_actor= 18.81, target value: -17.27:  96%|#########6| 9600/10000 [00:17<00:01, 364.19it/s]
-    reward: -4.35 (r0 = -2.37), reward eval: reward: -5.33, reward normalized=-2.95/5.17, grad norm= 188.52, loss_value= 288.01, loss_actor= 13.67, target value: -20.60:  96%|#########6| 9600/10000 [00:17<00:01, 364.19it/s]
-    reward: -4.35 (r0 = -2.37), reward eval: reward: -5.33, reward normalized=-2.95/5.17, grad norm= 188.52, loss_value= 288.01, loss_actor= 13.67, target value: -20.60: : 10400it [00:19, 340.11it/s]                        
-    reward: -4.58 (r0 = -2.37), reward eval: reward: -5.33, reward normalized=-3.41/4.08, grad norm= 113.89, loss_value= 176.51, loss_actor= 16.60, target value: -23.97: : 10400it [00:20, 340.11it/s]
+      8%|8         | 800/10000 [00:00<00:06, 1421.82it/s]
+     16%|#6        | 1600/10000 [00:03<00:21, 392.95it/s]
+     24%|##4       | 2400/10000 [00:04<00:13, 569.57it/s]
+     32%|###2      | 3200/10000 [00:04<00:09, 735.92it/s]
+     40%|####      | 4000/10000 [00:05<00:06, 873.28it/s]
+     48%|####8     | 4800/10000 [00:06<00:05, 988.60it/s]
+     56%|#####6    | 5600/10000 [00:06<00:04, 1075.15it/s]
+    reward: -2.13 (r0 = -2.83), reward eval: reward:  0.00, reward normalized=-2.20/6.15, grad norm= 88.70, loss_value= 325.66, loss_actor= 14.03, target value: -14.01:  56%|#####6    | 5600/10000 [00:07<00:04, 1075.15it/s]
+    reward: -2.13 (r0 = -2.83), reward eval: reward:  0.00, reward normalized=-2.20/6.15, grad norm= 88.70, loss_value= 325.66, loss_actor= 14.03, target value: -14.01:  64%|######4   | 6400/10000 [00:08<00:04, 776.79it/s] 
+    reward: -0.12 (r0 = -2.83), reward eval: reward:  0.00, reward normalized=-1.99/5.54, grad norm= 76.65, loss_value= 190.16, loss_actor= 12.20, target value: -12.53:  64%|######4   | 6400/10000 [00:09<00:04, 776.79it/s]
+    reward: -0.12 (r0 = -2.83), reward eval: reward:  0.00, reward normalized=-1.99/5.54, grad norm= 76.65, loss_value= 190.16, loss_actor= 12.20, target value: -12.53:  72%|#######2  | 7200/10000 [00:09<00:04, 665.89it/s]
+    reward: -1.84 (r0 = -2.83), reward eval: reward:  0.00, reward normalized=-1.71/5.85, grad norm= 102.12, loss_value= 245.59, loss_actor= 11.58, target value: -10.91:  72%|#######2  | 7200/10000 [00:10<00:04, 665.89it/s]
+    reward: -1.84 (r0 = -2.83), reward eval: reward:  0.00, reward normalized=-1.71/5.85, grad norm= 102.12, loss_value= 245.59, loss_actor= 11.58, target value: -10.91:  80%|########  | 8000/10000 [00:11<00:03, 603.70it/s]
+    reward: -4.98 (r0 = -2.83), reward eval: reward:  0.00, reward normalized=-2.39/4.97, grad norm= 78.56, loss_value= 187.75, loss_actor= 14.39, target value: -16.04:  80%|########  | 8000/10000 [00:12<00:03, 603.70it/s] 
+    reward: -4.98 (r0 = -2.83), reward eval: reward:  0.00, reward normalized=-2.39/4.97, grad norm= 78.56, loss_value= 187.75, loss_actor= 14.39, target value: -16.04:  88%|########8 | 8800/10000 [00:13<00:02, 567.09it/s]
+    reward: -3.58 (r0 = -2.83), reward eval: reward: -4.89, reward normalized=-2.75/4.99, grad norm= 72.07, loss_value= 206.85, loss_actor= 16.73, target value: -18.27:  88%|########8 | 8800/10000 [00:16<00:02, 567.09it/s]
+    reward: -3.58 (r0 = -2.83), reward eval: reward: -4.89, reward normalized=-2.75/4.99, grad norm= 72.07, loss_value= 206.85, loss_actor= 16.73, target value: -18.27:  96%|#########6| 9600/10000 [00:17<00:01, 366.93it/s]
+    reward: -5.36 (r0 = -2.83), reward eval: reward: -4.89, reward normalized=-2.69/5.31, grad norm= 122.46, loss_value= 278.56, loss_actor= 15.59, target value: -19.31:  96%|#########6| 9600/10000 [00:17<00:01, 366.93it/s]
+    reward: -5.36 (r0 = -2.83), reward eval: reward: -4.89, reward normalized=-2.69/5.31, grad norm= 122.46, loss_value= 278.56, loss_actor= 15.59, target value: -19.31: : 10400it [00:19, 338.76it/s]                        
+    reward: -4.41 (r0 = -2.83), reward eval: reward: -4.89, reward normalized=-3.57/4.31, grad norm= 100.75, loss_value= 213.60, loss_actor= 21.82, target value: -25.52: : 10400it [00:20, 338.76it/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  32.366 seconds)
+   **Total running time of the script:** ( 0 minutes  32.324 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): 196.7
+    elapsed time (seconds): 198.1
     loss: 5.168
-    elapsed time (seconds): 113.2
+    elapsed time (seconds): 114.6
 
 
 
@@ -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  18.914 seconds)
+   **Total running time of the script:** ( 5 minutes  21.872 seconds)
 
 
 .. _sphx_glr_download_advanced_dynamic_quantization_tutorial.py:

_images/sphx_glr_coding_ddpg_001.png

@@ -410,20 +410,22 @@ 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]
-      8%|7         | 41.4M/548M [00:00<00:01, 434MB/s]
-     15%|#5        | 84.4M/548M [00:00<00:01, 443MB/s]
-     23%|##3       | 127M/548M [00:00<00:00, 445MB/s] 
-     31%|###       | 170M/548M [00:00<00:00, 432MB/s]
-     38%|###8      | 211M/548M [00:00<00:00, 413MB/s]
-     46%|####5     | 250M/548M [00:00<00:00, 412MB/s]
-     53%|#####3    | 292M/548M [00:00<00:00, 419MB/s]
-     61%|######    | 334M/548M [00:00<00:00, 424MB/s]
-     69%|######8   | 376M/548M [00:00<00:00, 429MB/s]
-     76%|#######6  | 418M/548M [00:01<00:00, 433MB/s]
-     84%|########3 | 459M/548M [00:01<00:00, 411MB/s]
-     91%|######### | 498M/548M [00:01<00:00, 404MB/s]
-     98%|#########8| 537M/548M [00:01<00:00, 380MB/s]
-    100%|##########| 548M/548M [00:01<00:00, 411MB/s]
+      5%|5         | 29.2M/548M [00:00<00:01, 307MB/s]
+     12%|#2        | 67.2M/548M [00:00<00:01, 360MB/s]
+     19%|#9        | 104M/548M [00:00<00:01, 373MB/s] 
+     26%|##5       | 140M/548M [00:00<00:01, 334MB/s]
+     31%|###1      | 172M/548M [00:00<00:01, 333MB/s]
+     38%|###7      | 206M/548M [00:00<00:01, 338MB/s]
+     44%|####3     | 240M/548M [00:00<00:00, 343MB/s]
+     50%|#####     | 275M/548M [00:00<00:00, 352MB/s]
+     56%|#####6    | 309M/548M [00:00<00:00, 348MB/s]
+     63%|######2   | 343M/548M [00:01<00:00, 350MB/s]
+     70%|######9   | 383M/548M [00:01<00:00, 370MB/s]
+     77%|#######7  | 422M/548M [00:01<00:00, 384MB/s]
+     84%|########4 | 462M/548M [00:01<00:00, 395MB/s]
+     92%|#########1| 502M/548M [00:01<00:00, 402MB/s]
+     99%|#########8| 541M/548M [00:01<00:00, 394MB/s]
+    100%|##########| 548M/548M [00:01<00:00, 367MB/s]
 
 
 
@@ -768,7 +770,7 @@ Finally, we can run the algorithm.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  11.467 seconds)
+   **Total running time of the script:** ( 0 minutes  11.635 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  0.369 seconds)
+   **Total running time of the script:** ( 0 minutes  0.365 seconds)
 
 
 .. _sphx_glr_download_advanced_numpy_extensions_tutorial.py: