Fix fractional_max_pool3d_with_indices implementation

[yf225]

Stack from ghstack:

• Fix F::interpolate and torch::nn::Upsample implementation #35025 Fix F::interpolate and torch::nn::Upsample implementation
• Fix fractional_max_pool3d_with_indices implementation #35024 Fix fractional_max_pool3d_with_indices implementation
• Fix Conv and ConvTranspose implementation #35023 Fix Conv and ConvTranspose implementation
• Fix AdaptiveAvgPool{2,3}d and AdaptiveMaxPool{2,3}d implementation #35022 Fix AdaptiveAvgPool{2,3}d and AdaptiveMaxPool{2,3}d implementation

TODO: cherry-pick this PR into v1.5 release branch.

Differential Revision: D20559891

[dr-ci]

💊 CircleCI build failures summary and remediations

As of commit d197434 (more details on the Dr. CI page):

---

• 1/1 failures introduced in this PR

---

🕵️ 1 new failure recognized by patterns

The following build failures do not appear to be due to upstream breakages:

pytorch_xla_linux_xenial_py3_6_clang7_test (1/1)

Step: "Test" (full log | pattern match details)

Mar 20 22:46:16 ERROR: test_accurracy (__main__.TrainMnist)

Mar 20 22:45:41 2020-03-20 22:45:41.635063: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x56085790c030 initialized for platform Host (this does not guarantee that XLA will be used). Devices: 
Mar 20 22:45:41 2020-03-20 22:45:41.635131: I tensorflow/compiler/xla/service/service.cc:176]   StreamExecutor device (0): Host, Default Version 
Mar 20 22:45:41 2020-03-20 22:45:41.640305: I tensorflow/core/distributed_runtime/rpc/grpc_channel.cc:301] Initialize GrpcChannelCache for job localservice -> {0 -> localhost:40795} 
Mar 20 22:45:41 2020-03-20 22:45:41.640681: I tensorflow/core/distributed_runtime/rpc/grpc_server_lib.cc:390] Started server with target: grpc://localhost:40795 
Mar 20 22:45:41 2020-03-20 22:45:41.786302: W tensorflow/compiler/jit/xla_device.cc:398] XLA_GPU and XLA_CPU devices are deprecated and will be removed in subsequent releases. Instead, use either @tf.function(experimental_compile=True) for must-compile semantics, or run with TF_XLA_FLAGS=--tf_xla_auto_jit=2 for auto-clustering best-effort compilation. 
Mar 20 22:45:44 Running MNIST Test 
Mar 20 22:45:44 + echo 'Running MNIST Test' 
Mar 20 22:45:44 + python test/test_train_mnist.py --tidy 
Mar 20 22:46:16  0it [00:00, ?it/s]E 
Mar 20 22:46:16 ====================================================================== 
Mar 20 22:46:16 ERROR: test_accurracy (__main__.TrainMnist) 
Mar 20 22:46:16 ---------------------------------------------------------------------- 
Mar 20 22:46:16 Traceback (most recent call last): 
Mar 20 22:46:16   File "test/test_train_mnist.py", line 186, in test_accurracy 
Mar 20 22:46:16     self.assertGreaterEqual(train_mnist(), FLAGS.target_accuracy) 
Mar 20 22:46:16   File "test/test_train_mnist.py", line 74, in train_mnist 
Mar 20 22:46:16     transforms.Normalize((0.1307,), (0.3081,))])) 
Mar 20 22:46:16   File "/var/lib/jenkins/.local/lib/python3.6/site-packages/torchvision/datasets/mnist.py", line 71, in __init__ 
Mar 20 22:46:16     self.download() 
Mar 20 22:46:16   File "/var/lib/jenkins/.local/lib/python3.6/site-packages/torchvision/datasets/mnist.py", line 138, in download 
Mar 20 22:46:16     download_and_extract_archive(url, download_root=self.raw_folder, filename=filename, md5=md5) 

---

This comment was automatically generated by Dr. CI (expand for details).

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions on the GitHub issue tracker.

This comment has been revised 11 times.

[facebook-github-bot]

@yf225 merged this pull request in c0958c8.