Merge branch 'master' into update_profiler_tutorial

Author: brianjo

beginner_source/basics/optimization_tutorial.py

@@ -167,6 +167,7 @@ def train_loop(dataloader, model, loss_fn, optimizer):
 
 def test_loop(dataloader, model, loss_fn):
     size = len(dataloader.dataset)
+    num_batches = len(dataloader)
     test_loss, correct = 0, 0
 
     with torch.no_grad():
@@ -175,7 +176,7 @@ def test_loop(dataloader, model, loss_fn):
             test_loss += loss_fn(pred, y).item()
             correct += (pred.argmax(1) == y).type(torch.float).sum().item()
 
-    test_loss /= size
+    test_loss /= num_batches
     correct /= size
     print(f"Test Error: \n Accuracy: {(100*correct):>0.1f}%, Avg loss: {test_loss:>8f} \n")
 

beginner_source/basics/quickstart_tutorial.py

@@ -158,8 +158,9 @@ def train(dataloader, model, loss_fn, optimizer):
 ##############################################################################
 # We also check the model's performance against the test dataset to ensure it is learning.
 
-def test(dataloader, model):
+def test(dataloader, model, loss_fn):
     size = len(dataloader.dataset)
+    num_batches = len(dataloader)
     model.eval()
     test_loss, correct = 0, 0
     with torch.no_grad():
@@ -168,7 +169,7 @@ def test(dataloader, model):
             pred = model(X)
             test_loss += loss_fn(pred, y).item()
             correct += (pred.argmax(1) == y).type(torch.float).sum().item()
-    test_loss /= size
+    test_loss /= num_batches
     correct /= size
     print(f"Test Error: \n Accuracy: {(100*correct):>0.1f}%, Avg loss: {test_loss:>8f} \n")
 
@@ -181,7 +182,7 @@ def test(dataloader, model):
 for t in range(epochs):
     print(f"Epoch {t+1}\n-------------------------------")
     train(train_dataloader, model, loss_fn, optimizer)
-    test(test_dataloader, model)
+    test(test_dataloader, model, loss_fn)
 print("Done!")
 
 ######################################################################

beginner_source/basics/transforms_tutorial.py

@@ -26,6 +26,7 @@
 To make these transformations, we use ``ToTensor`` and ``Lambda``.
 """
 
+import torch
 from torchvision import datasets
 from torchvision.transforms import ToTensor, Lambda
 

intermediate_source/char_rnn_generation_tutorial.py

@@ -97,8 +97,8 @@ def unicodeToAscii(s):
 
 # Read a file and split into lines
 def readLines(filename):
-    lines = open(filename, encoding='utf-8').read().strip().split('\n')
-    return [unicodeToAscii(line) for line in lines]
+    with open(filename, encoding='utf-8') as some_file:
+        return [unicodeToAscii(line.strip()) for line in some_file]
 
 # Build the category_lines dictionary, a list of lines per category
 category_lines = {}
@@ -347,7 +347,6 @@ def timeSince(since):
 #
 
 import matplotlib.pyplot as plt
-import matplotlib.ticker as ticker
 
 plt.figure()
 plt.plot(all_losses)

intermediate_source/dist_tuto.rst

@@ -233,10 +233,10 @@ of 6 collectives currently implemented in PyTorch.
    ``tensor`` and stores the result in ``dst``.
 -  ``dist.all_reduce(tensor, op, group)``: Same as reduce, but the
    result is stored in all processes.
--  ``dist.scatter(tensor, src, scatter_list, group)``: Copies the
+-  ``dist.scatter(tensor, scatter_list, src, group)``: Copies the
    :math:`i^{\text{th}}` tensor ``scatter_list[i]`` to the
    :math:`i^{\text{th}}` process.
--  ``dist.gather(tensor, dst, gather_list, group)``: Copies ``tensor``
+-  ``dist.gather(tensor, gather_list, dst, group)``: Copies ``tensor``
    from all processes in ``dst``.
 -  ``dist.all_gather(tensor_list, tensor, group)``: Copies ``tensor``
    from all processes to ``tensor_list``, on all processes.

recipes_source/recipes/dynamic_quantization.py

@@ -220,9 +220,6 @@ def print_size_of_model(model, label=""):
 q=print_size_of_model(quantized_lstm,"int8")
 print("{0:.2f} times smaller".format(f/q))
 
-# note that this value is wrong in PyTorch 1.4 due to https://github.com/pytorch/pytorch/issues/31468
-# this will be fixed in 1.5 with https://github.com/pytorch/pytorch/pull/31540
-
 
 ######################################################################
 # 4. Look at Latency