1. gallery
  2. pytorch
  3. fashion-mnist

BentoML Example: PyTorch Fashion MNIST Classification

BentoML is an open source platform for machine learning model serving and deployment. In this project we will use BentoML to package the image classifier model, and build a containerized REST API model server.

This notebook demonstrates how to use BentoML to turn a PyTorch model into a docker image containing a REST API server serving this model, how to use your ML service built with BentoML as a CLI tool, and how to distribute it a pypi package.

This example was built based on https://github.com/baldassarreFe/zalando-pytorch/blob/master/notebooks/4.0-fb-autoencoder.ipynb, if you are familiar with this, jump start to Model Serving using BentoML

Impression

In [1]:
%reload_ext autoreload
%autoreload 2
%matplotlib inline
In [ ]:
!pip install bentoml
!pip install torch torchvision sklearn pillow pandas numpy
In [2]:
import bentoml
import matplotlib.pyplot as plt

import torch
import torch.nn as nn
import torch.optim as optim

from torchvision import transforms
from torch.autograd import Variable

from sklearn.manifold import TSNE
from sklearn.metrics import accuracy_score

Prepare Dataset

PyTorch supports FashionMNIST now, so we can import it directly.

In [3]:
from torchvision.datasets import FashionMNIST
FashionMNIST.classes
Out[3]:
['T-shirt/top',
 'Trouser',
 'Pullover',
 'Dress',
 'Coat',
 'Sandal',
 'Shirt',
 'Sneaker',
 'Bag',
 'Ankle boot']

Load train and test set in batches of 1000.

The 28x28 images are scaled up to 29x29 so that combining convolutions and transposed convolutions would not chop off pixels from the reconstructed images.

In [4]:
batch_size = 1000

train_dataset = FashionMNIST(
    '../data', train=True, download=True, 
    transform=transforms.Compose([transforms.CenterCrop((29, 29)), transforms.ToTensor()]))
train_loader = torch.utils.data.DataLoader(train_dataset, batch_size=batch_size, shuffle=True)

test_dataset = FashionMNIST(
    '../data', train=False, download=True, 
    transform=transforms.Compose([transforms.CenterCrop((29, 29)), transforms.ToTensor()]))
test_loader = torch.utils.data.DataLoader(test_dataset, batch_size=batch_size, shuffle=True)

Unsupervised reconstruction

Note that in this section we'll never use the image labels, the whole training is unsupervised.

Autoencoder

The two components of the autoencoder are defined subclassing nn.Module, that gives more flexibility than nn.Sequential.

Encoder

A series of convolutions with kernel_size=5 and stride=2 is used to squeeze the images into a volume of 40x1x1, then a fully connected layer turns this vector in a vector of size embedding_size, that can be specified externally.

Decoder

The decoder takes up from where the encoder left, first transforming back the embedding of size embedding_size into a volume of size 40x1x1, then applying a series of Transposed Convolutions to yield an image of the same size of the original input.

At this time we can show some images in this Dataloader.

In [5]:
class Encoder(nn.Module):
    def __init__(self, embedding_size):
        super(Encoder, self).__init__()
        self.conv1 = nn.Conv2d(1, 10, kernel_size=5, stride=2)
        self.conv2 = nn.Conv2d(10, 20, kernel_size=5, stride=2)
        self.conv3 = nn.Conv2d(20, 40, kernel_size=5, stride=2)
        self.fully = nn.Linear(40, embedding_size)

    def forward(self, x):
        # 1x29x29
        x = torch.relu(self.conv1(x))
        # 10x13x13
        x = torch.relu(self.conv2(x))
        # 20x5x5
        x = torch.relu(self.conv3(x))
        # 40x1x1
        x = x.view(x.data.shape[0], 40)
        # 40
        x = self.fully(x)
        # output_size
        return x

class Decoder(nn.Module):
    def __init__(self, input_size):
        super(Decoder, self).__init__()
        self.fully = nn.Linear(input_size, 40)
        self.conv1 = nn.ConvTranspose2d(40, 20, kernel_size=5, stride=2)
        self.conv2 = nn.ConvTranspose2d(20, 10, kernel_size=5, stride=2)
        self.conv3 = nn.ConvTranspose2d(10, 1, kernel_size=5, stride=2)
    
    def forward(self, x):
        x = self.fully(x)
        x = x.view(x.data.shape[0], 40, 1, 1)
        x = torch.relu(self.conv1(x))
        x = torch.relu(self.conv2(x))
        x = torch.sigmoid(self.conv3(x))
        return x

We are going to use an embedding size of 20, this number has no particular reason, except that it is in the same range of the number of classes. Naively, the network could learn to encode coarse-grained information (i.e. the kind of dress) in half of the embedding vector and then use the other half for fine-grained information.

In [6]:
embedding_size = 20
encoder = Encoder(embedding_size)
decoder = Decoder(embedding_size)

autoencoder = nn.Sequential(encoder, decoder)

Sanity check

A 29x29 black and white image passed through the autoencoder should give the same output dimension

In [7]:
x = Variable(torch.ones(1, 1, 29, 29))
e = encoder(x)
d = decoder(e)

print('Input\t ', list(x.data.shape))
print('Embedding', list(e.data.shape))
print('Output\t ', list(d.data.shape))

Input	  [1, 1, 29, 29]
Embedding [1, 20]
Output	  [1, 1, 29, 29]

Training

In [8]:
autoencoder.train()

loss_fn = nn.MSELoss()
optimizer = optim.Adam(autoencoder.parameters())
epoch_loss = []

for epoch in range(15):
    batch_loss = []
    for batch_num, (data, _) in enumerate(train_loader):
        data = Variable(data)
        optimizer.zero_grad()
        output = autoencoder(data)
        loss = loss_fn(output, data)
        loss.backward()
        optimizer.step()
        batch_loss.append(loss.item())
    epoch_loss.append(sum(batch_loss) / len(batch_loss))
    print('Epoch {}:\tloss {:.4f}'.format(epoch, epoch_loss[-1]))

Epoch 0:	loss 0.1243
Epoch 1:	loss 0.0749
Epoch 2:	loss 0.0489
Epoch 3:	loss 0.0344
Epoch 4:	loss 0.0286
Epoch 5:	loss 0.0264
Epoch 6:	loss 0.0250
Epoch 7:	loss 0.0238
Epoch 8:	loss 0.0229
Epoch 9:	loss 0.0222
Epoch 10:	loss 0.0215
Epoch 11:	loss 0.0209
Epoch 12:	loss 0.0205
Epoch 13:	loss 0.0201
Epoch 14:	loss 0.0198
In [9]:
plt.plot(epoch_loss)
plt.title('Final value {:.4f}'.format(epoch_loss[-1]))
plt.xlabel('Epoch')
plt.grid(True)

Evaluation

Reconsruction evaluation on a single batch

In [10]:
autoencoder.eval()
data, targets = next(test_loader.__iter__())
encodings = encoder(Variable(data))
outputs = decoder(encodings)

print('Test loss: {:.4f}'.format(loss_fn(outputs, Variable(data)).item()))

Test loss: 0.0196
In [11]:
fig, axes = plt.subplots(8, 8, figsize=(16, 16))
axes = axes.ravel()

zip_these = axes[::2], axes[1::2], data.numpy().squeeze(), outputs.data.numpy().squeeze(), targets
for ax1, ax2, original, reconstructed, target in zip(*zip_these):
    ax1.imshow(original, cmap='gray')
    ax1.axis('off')
    ax1.set_title(FashionMNIST.classes[target])
    ax2.imshow(reconstructed, cmap='gray')
    ax2.axis('off')

Embeddings

The embeddings are 20-dimensional, t-SNE is used to visualize them as clusters in 2D space.

Even though the autoencoder learned the embeddings in a completely unsupervised way we can observe the emergence of clusters:

  • shoes (sandals, sneakers and ankle boot) are clustered together
  • bags form a group on their own (they are the only images with a clear squared-rectangular shape)
  • same goes for trousers, that form their own group
  • all the others are quite mixed together, meaning that the network has learned the concept of clothes for the upper body, but is not able to tell a coat from a pullover
In [12]:
pca = TSNE(n_components=2)
encodings_2 = pca.fit_transform(encodings.data.numpy())
plt.figure(figsize=(10, 10))
for k in range(len(FashionMNIST.classes)):
    class_indexes = (targets.numpy() == k)
    plt.scatter(encodings_2[class_indexes, 0], encodings_2[class_indexes, 1], label=FashionMNIST.classes[k])
plt.legend();

Supervised classification

Once trained in an unsupervised fashion, the encoder module can be used to generate fashion embeddings (see what I did here?), that can then be used to train a simple classifier on the original labels.

Model

The weights of the encoder are freezed, so only the classifier will be trained.

(later on, when the classifier starts performing decently, we could unfreeze them and do some fine-tuning)

In [13]:
for param in encoder.parameters():
    param.requires_grad = False

classifier = nn.Sequential(
    encoder, 
    nn.Linear(embedding_size, 15),
    nn.ReLU(),
    nn.Linear(15, len(FashionMNIST.classes)),
    nn.LogSoftmax()
)

Training

In [14]:
classifier.train()

loss_fn = nn.NLLLoss()
optimizer = optim.Adam([p for p in classifier.parameters() if p.requires_grad])
epoch_loss = []

for epoch in range(15):
    batch_loss = []
    for batch_num, (data, targets) in enumerate(train_loader):
        data, targets = Variable(data), Variable(targets)
        optimizer.zero_grad()
        output = classifier(data)
        loss = loss_fn(output, targets)
        loss.backward()
        optimizer.step()
        batch_loss.append(loss.item())
    epoch_loss.append(sum(batch_loss) / len(batch_loss))
    accuracy = accuracy_score(targets.data.numpy(), output.data.numpy().argmax(axis=1))
    print('Epoch {}:\tloss {:.4f}\taccuracy {:.2%}'.format(epoch, epoch_loss[-1], accuracy))

Epoch 0:	loss 2.9452	accuracy 42.70%
Epoch 1:	loss 1.2041	accuracy 64.70%
Epoch 2:	loss 0.9187	accuracy 68.10%
Epoch 3:	loss 0.8261	accuracy 70.00%
Epoch 4:	loss 0.7806	accuracy 73.40%
Epoch 5:	loss 0.7525	accuracy 72.00%
Epoch 6:	loss 0.7336	accuracy 72.90%
Epoch 7:	loss 0.7184	accuracy 71.70%
Epoch 8:	loss 0.7082	accuracy 75.30%
Epoch 9:	loss 0.6996	accuracy 71.30%
Epoch 10:	loss 0.6918	accuracy 73.40%
Epoch 11:	loss 0.6859	accuracy 74.40%
Epoch 12:	loss 0.6815	accuracy 73.50%
Epoch 13:	loss 0.6766	accuracy 74.00%
Epoch 14:	loss 0.6736	accuracy 76.30%
In [15]:
plt.plot(epoch_loss)
plt.title('Final value {:.4f}'.format(epoch_loss[-1]))
plt.xlabel('Epoch')
plt.grid(True)

Evaluation

Reconsruction evaluation on a single batch

In [16]:
classifier.eval()
data, targets = next(test_loader.__iter__())
outputs = classifier(Variable(data))
log_probs, output_classes = outputs.max(dim=1)

accuracy = accuracy_score(targets.numpy(), output_classes.data.numpy())
print('Accuracy: {:.2%}'.format(accuracy))

Accuracy: 72.80%
In [17]:
fig, axex = plt.subplots(8, 8, figsize=(16, 16))

zip_these = axex.ravel(), log_probs.data.exp(), output_classes.data, targets, data.numpy().squeeze()
for ax, prob, output_class, target, img in zip(*zip_these):
    ax.imshow(img, cmap='gray' if output_class == target else 'autumn')
    ax.axis('off')
    ax.set_title('{} {:.1%}'.format(FashionMNIST.classes[output_class], prob))

Define BentoService for model serving

In [18]:
%%writefile pytorch_fashion_mnist.py

import bentoml
from PIL import Image
from torchvision import transforms

from bentoml.artifact import PytorchModelArtifact
from bentoml.handlers import ImageHandler


FASHION_MNIST_CLASSES = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat', 'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']


@bentoml.env(pip_dependencies=['torch', 'numpy', 'torchvision', 'scikit-learn'])
@bentoml.artifacts([PytorchModelArtifact('classifier')])
class FashionMNISTServicer(bentoml.BentoService):

    @bentoml.api(ImageHandler, pilmode='L')
    def predict(self, image):
        img = Image.fromarray(image).resize((28, 28))
        transform=transforms.Compose([transforms.CenterCrop((29, 29)), transforms.ToTensor()])
        img_tensor = transform(img)
        outputs = self.artifacts.classifier(img_tensor.unsqueeze(0))        
        _, output_classes = outputs.max(dim=1)
        
        return FASHION_MNIST_CLASSES[output_classes]

Overwriting pytorch_fashion_mnist.py

Save BentoService to file archive

In [19]:
# 1) import the custom BentoService defined above
from pytorch_fashion_mnist import FashionMNISTServicer

# 2) `pack` it with required artifacts
bento_svc = FashionMNISTServicer.pack(classifier=classifier)

# 3) save your BentoSerivce
saved_path = bento_svc.save()

[2019-09-19 12:20:55,162] INFO - Successfully saved Bento 'FashionMNISTServicer:2019_09_19_8a0375b7' to path: /Users/chaoyuyang/bentoml/repository/FashionMNISTServicer/2019_09_19_8a0375b7

Model Serving via REST API

In your termnial, run the following command to start the REST API server:### Run REST API server

In [20]:
!bentoml serve {saved_path}

 * Serving Flask app "FashionMNISTServicer" (lazy loading)
 * Environment: production
   WARNING: This is a development server. Do not use it in a production deployment.
   Use a production WSGI server instead.
 * Debug mode: off
 * Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)
127.0.0.1 - - [19/Sep/2019 12:52:50] "GET / HTTP/1.1" 200 -
/Users/chaoyuyang/anaconda3/envs/bentoml-dev/lib/python3.7/site-packages/torch/nn/modules/container.py:92: UserWarning: Implicit dimension choice for log_softmax has been deprecated. Change the call to include dim=X as an argument.
  input = module(input)
127.0.0.1 - - [19/Sep/2019 12:52:58] "POST /predict HTTP/1.1" 200 -
127.0.0.1 - - [19/Sep/2019 12:53:01] "GET /docs.json HTTP/1.1" 200 -
127.0.0.1 - - [19/Sep/2019 12:53:04] "POST /predict HTTP/1.1" 200 -
^C

Call REST API from a client

Sending POST request from termnial:

curl -X POST "http://127.0.0.1:5000/predict" -F image=@sample_image.png

curl -X POST "http://127.0.0.1:5000/predict" -H "Content-Type: image/png" --data-binary @sample_image.png

Go visit http://127.0.0.1:5000/ from your browser, click /predict -> Try it out -> Choose File -> Execute to sumbit an image from your computer

image.png

Containerize REST API server with Docker

** Make sure you have docker installed, note that it is not available when running in Google Colaboratory

In [21]:
!cd {saved_path} && docker build . -t pytorch-fashion-mnist

Sending build context to Docker daemon  133.6kB
Step 1/11 : FROM continuumio/miniconda3
 ---> ae46c364060f
Step 2/11 : ENTRYPOINT [ "/bin/bash", "-c" ]
 ---> Using cache
 ---> 2f135ada8e2d
Step 3/11 : EXPOSE 5000
 ---> Using cache
 ---> 738f652d09ae
Step 4/11 : RUN set -x      && apt-get update      && apt-get install --no-install-recommends --no-install-suggests -y libpq-dev build-essential      && rm -rf /var/lib/apt/lists/*
 ---> Using cache
 ---> 70c62a45013a
Step 5/11 : RUN conda update conda -y       && conda install pip numpy scipy       && pip install gunicorn six
 ---> Using cache
 ---> fe5d966ecc35
Step 6/11 : COPY . /bento
 ---> d00c6768ebbe
Step 7/11 : WORKDIR /bento
 ---> Running in 65e4f1f8e11f
Removing intermediate container 65e4f1f8e11f
 ---> d0a6e7065765
Step 8/11 : RUN conda env update -n base -f /bento/environment.yml
 ---> Running in 6166f7e1de32
Collecting package metadata: ...working... done
Solving environment: ...working... 
The environment is inconsistent, please check the package plan carefully
The following packages are causing the inconsistency:

  - defaults/linux-64::pyopenssl==18.0.0=py37_0
  - defaults/linux-64::idna==2.8=py37_0
  - defaults/linux-64::six==1.12.0=py37_0
  - defaults/linux-64::cryptography==2.4.2=py37h1ba5d50_0
  - defaults/linux-64::ruamel_yaml==0.15.46=py37h14c3975_0
  - defaults/linux-64::pycparser==2.19=py37_0
  - defaults/linux-64::libedit==3.1.20170329=h6b74fdf_2
  - defaults/linux-64::zlib==1.2.11=h7b6447c_3
  - defaults/linux-64::readline==7.0=h7b6447c_5
  - defaults/linux-64::tk==8.6.8=hbc83047_0
  - defaults/linux-64::urllib3==1.24.1=py37_0
  - defaults/linux-64::pycosat==0.6.3=py37h14c3975_0
  - defaults/linux-64::ncurses==6.1=he6710b0_1
  - defaults/linux-64::requests==2.21.0=py37_0
  - defaults/linux-64::wheel==0.32.3=py37_0
  - defaults/linux-64::asn1crypto==0.24.0=py37_0
  - defaults/linux-64::chardet==3.0.4=py37_1
  - defaults/linux-64::setuptools==40.6.3=py37_0
  - defaults/linux-64::cffi==1.11.5=py37he75722e_1
  - defaults/linux-64::sqlite==3.26.0=h7b6447c_0
  - defaults/linux-64::xz==5.2.4=h14c3975_4
  - defaults/linux-64::libffi==3.2.1=hd88cf55_4
  - defaults/linux-64::libgcc-ng==8.2.0=hdf63c60_1
  - defaults/linux-64::yaml==0.1.7=had09818_2
  - defaults/linux-64::pysocks==1.6.8=py37_0
  - defaults/linux-64::python==3.7.1=h0371630_7
  - defaults/linux-64::pip==19.0.3=py37_0
  - defaults/linux-64::mkl_random==1.0.2=py37hd81dba3_0
  - defaults/linux-64::numpy-base==1.16.3=py37hde5b4d6_0
  - defaults/linux-64::mkl_fft==1.0.10=py37ha843d7b_0
  - defaults/linux-64::openssl==1.1.1b=h7b6447c_1
  - defaults/linux-64::conda==4.6.14=py37_0
  - defaults/linux-64::certifi==2019.3.9=py37_0
  - defaults/linux-64::scipy==1.2.1=py37h7c811a0_0
  - defaults/linux-64::numpy==1.16.3=py37h7e9f1db_0
done

Downloading and Extracting Packages
libedit-3.1.20181209 | 188 KB    | ########## | 100% 
sqlite-3.29.0        | 1.9 MB    | ########## | 100% 
certifi-2019.6.16    | 156 KB    | ########## | 100% 
ca-certificates-2019 | 134 KB    | ########## | 100% 
openssl-1.1.1d       | 3.7 MB    | ########## | 100% 
wheel-0.33.4         | 39 KB     | ########## | 100% 
libgcc-ng-9.1.0      | 8.1 MB    | ########## | 100% 
pip-19.2.2           | 1.9 MB    | ########## | 100% 
python-3.7.3         | 36.7 MB   | ########## | 100% 
_libgcc_mutex-0.1    | 3 KB      | ########## | 100% 
setuptools-41.0.1    | 648 KB    | ########## | 100% 
Preparing transaction: ...working... done
Verifying transaction: ...working... done
Executing transaction: ...working... done


==> WARNING: A newer version of conda exists. <==
  current version: 4.6.14
  latest version: 4.7.12

Please update conda by running

    $ conda update -n base -c defaults conda


Ran pip subprocess with arguments:
['/opt/conda/bin/python', '-m', 'pip', 'install', '-U', '-r', '/bento/condaenv.zv4o07od.requirements.txt']
Pip subprocess output:
Collecting bentoml[api_server]==0.4.1 (from -r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/45/7e/bef20c3d36ae7455c5ff3b7629c93b176df22f6131872857a3c746c031bb/BentoML-0.4.1-py3-none-any.whl (159kB)
Collecting flask (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/9b/93/628509b8d5dc749656a9641f4caf13540e2cdec85276964ff8f43bbb1d3b/Flask-1.1.1-py2.py3-none-any.whl (94kB)
Collecting ruamel.yaml>=0.15.0 (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/fa/90/ecff85a2e9c497e2fa7142496e10233556b5137db5bd46f3f3b006935ca8/ruamel.yaml-0.16.5-py2.py3-none-any.whl (123kB)
Collecting pathlib2 (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/67/c6/4dbf5dfdbe1140cadf765c3896acc098578626c35721bc7d3eb35f6a8fc1/pathlib2-2.3.4-py2.py3-none-any.whl
Collecting grpcio (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/e5/27/1f908ebb99c8d48a5ba4eb9d7997f5633b920d98fe712f67aaa0663f1307/grpcio-1.23.0-cp37-cp37m-manylinux1_x86_64.whl (2.2MB)
Collecting python-json-logger (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/80/9d/1c3393a6067716e04e6fcef95104c8426d262b4adaf18d7aa2470eab028d/python-json-logger-0.1.11.tar.gz
Collecting protobuf>=3.6.0 (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/30/fd/60ce148d8e4205bdf6da4ffec31348fd33f710c20a882b44319d54fd51ae/protobuf-3.9.1-cp37-cp37m-manylinux1_x86_64.whl (1.2MB)
Collecting prometheus-client (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/b3/23/41a5a24b502d35a4ad50a5bb7202a5e1d9a0364d0c12f56db3dbf7aca76d/prometheus_client-0.7.1.tar.gz
Collecting sqlalchemy (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/fc/49/82d64d705ced344ba458197dadab30cfa745f9650ee22260ac2b275d288c/SQLAlchemy-1.3.8.tar.gz (5.9MB)
Requirement already satisfied, skipping upgrade: gunicorn in /opt/conda/lib/python3.7/site-packages (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1)) (19.9.0)
Collecting dill (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/39/7a/70803635c850e351257029089d38748516a280864c97cbc73087afef6d51/dill-0.3.0.tar.gz (151kB)
Collecting pandas (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/7e/ab/ea76361f9d3e732e114adcd801d2820d5319c23d0ac5482fa3b412db217e/pandas-0.25.1-cp37-cp37m-manylinux1_x86_64.whl (10.4MB)
Collecting configparser (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/7a/2a/95ed0501cf5d8709490b1d3a3f9b5cf340da6c433f896bbe9ce08dbe6785/configparser-4.0.2-py2.py3-none-any.whl
Collecting boto3 (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/7a/2d/023459434edb679dea2c1547a75a5cb64d6c5c93ab601d664fde88c0bd83/boto3-1.9.231-py2.py3-none-any.whl (128kB)
Requirement already satisfied, skipping upgrade: numpy in /opt/conda/lib/python3.7/site-packages (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1)) (1.16.3)
Requirement already satisfied, skipping upgrade: six in /opt/conda/lib/python3.7/site-packages (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1)) (1.12.0)
Collecting click (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/fa/37/45185cb5abbc30d7257104c434fe0b07e5a195a6847506c074527aa599ec/Click-7.0-py2.py3-none-any.whl (81kB)
Requirement already satisfied, skipping upgrade: requests in /opt/conda/lib/python3.7/site-packages (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1)) (2.21.0)
Collecting packaging (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/cf/94/9672c2d4b126e74c4496c6b3c58a8b51d6419267be9e70660ba23374c875/packaging-19.2-py2.py3-none-any.whl
Collecting cerberus (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/c9/0e/f78e23b778c2234972d364d0f8bea2de0a09f450f65d3f05ce091dd0f104/Cerberus-1.3.1.tar.gz (52kB)
Collecting docker (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/95/47/5560c9cf0c92b50da24216f0e7733250fbed5a497f69e3c70e1be62143fe/docker-4.0.2-py2.py3-none-any.whl (138kB)
Collecting Werkzeug; extra == "api_server" (from bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/ce/42/3aeda98f96e85fd26180534d36570e4d18108d62ae36f87694b476b83d6f/Werkzeug-0.16.0-py2.py3-none-any.whl (327kB)
Collecting Jinja2>=2.10.1 (from flask->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/1d/e7/fd8b501e7a6dfe492a433deb7b9d833d39ca74916fa8bc63dd1a4947a671/Jinja2-2.10.1-py2.py3-none-any.whl (124kB)
Collecting itsdangerous>=0.24 (from flask->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/76/ae/44b03b253d6fade317f32c24d100b3b35c2239807046a4c953c7b89fa49e/itsdangerous-1.1.0-py2.py3-none-any.whl
Collecting ruamel.yaml.clib>=0.1.2; platform_python_implementation == "CPython" and python_version < "3.8" (from ruamel.yaml>=0.15.0->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/f2/bd/ff0d632e1b248c28427a23bd741f5b8f954acba82df86661e5daf7cc79a6/ruamel.yaml.clib-0.1.2-cp37-cp37m-manylinux1_x86_64.whl (547kB)
Requirement already satisfied, skipping upgrade: setuptools in /opt/conda/lib/python3.7/site-packages (from protobuf>=3.6.0->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1)) (41.0.1)
Collecting pytz>=2017.2 (from pandas->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/87/76/46d697698a143e05f77bec5a526bf4e56a0be61d63425b68f4ba553b51f2/pytz-2019.2-py2.py3-none-any.whl (508kB)
Collecting python-dateutil>=2.6.1 (from pandas->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/41/17/c62faccbfbd163c7f57f3844689e3a78bae1f403648a6afb1d0866d87fbb/python_dateutil-2.8.0-py2.py3-none-any.whl (226kB)
Collecting botocore<1.13.0,>=1.12.231 (from boto3->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/b1/18/90fa120b459ef691a19d2972790834b7eaf843e944d99d798005cb14e8fa/botocore-1.12.231-py2.py3-none-any.whl (5.7MB)
Collecting jmespath<1.0.0,>=0.7.1 (from boto3->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/83/94/7179c3832a6d45b266ddb2aac329e101367fbdb11f425f13771d27f225bb/jmespath-0.9.4-py2.py3-none-any.whl
Collecting s3transfer<0.3.0,>=0.2.0 (from boto3->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/16/8a/1fc3dba0c4923c2a76e1ff0d52b305c44606da63f718d14d3231e21c51b0/s3transfer-0.2.1-py2.py3-none-any.whl (70kB)
Requirement already satisfied, skipping upgrade: chardet<3.1.0,>=3.0.2 in /opt/conda/lib/python3.7/site-packages (from requests->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1)) (3.0.4)
Requirement already satisfied, skipping upgrade: urllib3<1.25,>=1.21.1 in /opt/conda/lib/python3.7/site-packages (from requests->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1)) (1.24.1)
Requirement already satisfied, skipping upgrade: certifi>=2017.4.17 in /opt/conda/lib/python3.7/site-packages (from requests->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1)) (2019.6.16)
Requirement already satisfied, skipping upgrade: idna<2.9,>=2.5 in /opt/conda/lib/python3.7/site-packages (from requests->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1)) (2.8)
Collecting pyparsing>=2.0.2 (from packaging->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/11/fa/0160cd525c62d7abd076a070ff02b2b94de589f1a9789774f17d7c54058e/pyparsing-2.4.2-py2.py3-none-any.whl (65kB)
Collecting websocket-client>=0.32.0 (from docker->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/29/19/44753eab1fdb50770ac69605527e8859468f3c0fd7dc5a76dd9c4dbd7906/websocket_client-0.56.0-py2.py3-none-any.whl (200kB)
Collecting MarkupSafe>=0.23 (from Jinja2>=2.10.1->flask->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/98/7b/ff284bd8c80654e471b769062a9b43cc5d03e7a615048d96f4619df8d420/MarkupSafe-1.1.1-cp37-cp37m-manylinux1_x86_64.whl
Collecting docutils<0.16,>=0.10 (from botocore<1.13.0,>=1.12.231->boto3->bentoml[api_server]==0.4.1->-r /bento/condaenv.zv4o07od.requirements.txt (line 1))
  Downloading https://files.pythonhosted.org/packages/22/cd/a6aa959dca619918ccb55023b4cb151949c64d4d5d55b3f4ffd7eee0c6e8/docutils-0.15.2-py3-none-any.whl (547kB)
Building wheels for collected packages: python-json-logger, prometheus-client, sqlalchemy, dill, cerberus
  Building wheel for python-json-logger (setup.py): started
  Building wheel for python-json-logger (setup.py): finished with status 'done'
  Created wheel for python-json-logger: filename=python_json_logger-0.1.11-py2.py3-none-any.whl size=5077 sha256=b6be66e346371d286ecd9e89c98867aeda49c57e978779718f8f7e6b0ececf83
  Stored in directory: /root/.cache/pip/wheels/97/f7/a1/752e22bb30c1cfe38194ea0070a5c66e76ef4d06ad0c7dc401
  Building wheel for prometheus-client (setup.py): started
  Building wheel for prometheus-client (setup.py): finished with status 'done'
  Created wheel for prometheus-client: filename=prometheus_client-0.7.1-cp37-none-any.whl size=41405 sha256=c70c6f2c47a9daff675335b505dbb9056b504fe915d6db3f376ef20f4cc8f8c7
  Stored in directory: /root/.cache/pip/wheels/1c/54/34/fd47cd9b308826cc4292b54449c1899a30251ef3b506bc91ea
  Building wheel for sqlalchemy (setup.py): started
  Building wheel for sqlalchemy (setup.py): finished with status 'done'
  Created wheel for sqlalchemy: filename=SQLAlchemy-1.3.8-cp37-cp37m-linux_x86_64.whl size=1197985 sha256=298499650bcbceae0dda2404f393a0709ea18495e8d137e1554b5ef02bfcc562
  Stored in directory: /root/.cache/pip/wheels/97/b6/66/de2064d40c920adc2984ff3b8fd4f11494c8ab9e48ba87e8a2
  Building wheel for dill (setup.py): started
  Building wheel for dill (setup.py): finished with status 'done'
  Created wheel for dill: filename=dill-0.3.0-cp37-none-any.whl size=77512 sha256=1388231f916366300a3449c95674a3a96979637333384eafae9bf179591af69d
  Stored in directory: /root/.cache/pip/wheels/c9/de/a4/a91eec4eea652104d8c81b633f32ead5eb57d1b294eab24167
  Building wheel for cerberus (setup.py): started
  Building wheel for cerberus (setup.py): finished with status 'done'
  Created wheel for cerberus: filename=Cerberus-1.3.1-cp37-none-any.whl size=54129 sha256=8788b3c5839d1466149ca68564a94af3043a447be2e96e1d0d9f876652cbdcfb
  Stored in directory: /root/.cache/pip/wheels/4f/b4/2b/83fc9a02bdbb49350a9eeb6d34753e4d7df7f77cb027e1d4b3
Successfully built python-json-logger prometheus-client sqlalchemy dill cerberus
Installing collected packages: MarkupSafe, Jinja2, click, Werkzeug, itsdangerous, flask, ruamel.yaml.clib, ruamel.yaml, pathlib2, grpcio, python-json-logger, protobuf, prometheus-client, sqlalchemy, dill, pytz, python-dateutil, pandas, configparser, jmespath, docutils, botocore, s3transfer, boto3, pyparsing, packaging, cerberus, websocket-client, docker, bentoml
Successfully installed Jinja2-2.10.1 MarkupSafe-1.1.1 Werkzeug-0.16.0 bentoml-0.4.1 boto3-1.9.231 botocore-1.12.231 cerberus-1.3.1 click-7.0 configparser-4.0.2 dill-0.3.0 docker-4.0.2 docutils-0.15.2 flask-1.1.1 grpcio-1.23.0 itsdangerous-1.1.0 jmespath-0.9.4 packaging-19.2 pandas-0.25.1 pathlib2-2.3.4 prometheus-client-0.7.1 protobuf-3.9.1 pyparsing-2.4.2 python-dateutil-2.8.0 python-json-logger-0.1.11 pytz-2019.2 ruamel.yaml-0.16.5 ruamel.yaml.clib-0.1.2 s3transfer-0.2.1 sqlalchemy-1.3.8 websocket-client-0.56.0

#
# To activate this environment, use:
# > conda activate base
#
# To deactivate an active environment, use:
# > conda deactivate
#

Removing intermediate container 6166f7e1de32
 ---> 666f7c81f870
Step 9/11 : RUN pip install -r /bento/requirements.txt
 ---> Running in 85f87d646259
Requirement already satisfied: bentoml==0.4.1 in /opt/conda/lib/python3.7/site-packages (from -r /bento/requirements.txt (line 1)) (0.4.1)
Collecting torch (from -r /bento/requirements.txt (line 2))
  Downloading https://files.pythonhosted.org/packages/05/65/5248be50c55ab7429dd5c11f5e2f9f5865606b80e854ca63139ad1a584f2/torch-1.2.0-cp37-cp37m-manylinux1_x86_64.whl (748.9MB)
Requirement already satisfied: numpy in /opt/conda/lib/python3.7/site-packages (from -r /bento/requirements.txt (line 3)) (1.16.3)
Collecting torchvision (from -r /bento/requirements.txt (line 4))
  Downloading https://files.pythonhosted.org/packages/51/83/2d77d040e34bd8f70dcb4770f7eb7d0aa71e07738abf6831be863ade00db/torchvision-0.4.0-cp37-cp37m-manylinux1_x86_64.whl (8.8MB)
Collecting scikit-learn (from -r /bento/requirements.txt (line 5))
  Downloading https://files.pythonhosted.org/packages/9f/c5/e5267eb84994e9a92a2c6a6ee768514f255d036f3c8378acfa694e9f2c99/scikit_learn-0.21.3-cp37-cp37m-manylinux1_x86_64.whl (6.7MB)
Collecting imageio (from -r /bento/requirements.txt (line 6))
  Downloading https://files.pythonhosted.org/packages/af/0a/943c965d372dae0b1f1482677d29030ab834351a61a9a632fd62f27f1523/imageio-2.5.0-py3-none-any.whl (3.3MB)
Requirement already satisfied: flask in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (1.1.1)
Requirement already satisfied: pathlib2 in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (2.3.4)
Requirement already satisfied: configparser in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (4.0.2)
Requirement already satisfied: python-json-logger in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.1.11)
Requirement already satisfied: grpcio in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (1.23.0)
Requirement already satisfied: click in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (7.0)
Requirement already satisfied: protobuf>=3.6.0 in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (3.9.1)
Requirement already satisfied: six in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (1.12.0)
Requirement already satisfied: ruamel.yaml>=0.15.0 in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.16.5)
Requirement already satisfied: packaging in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (19.2)
Requirement already satisfied: cerberus in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (1.3.1)
Requirement already satisfied: requests in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (2.21.0)
Requirement already satisfied: boto3 in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (1.9.231)
Requirement already satisfied: dill in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.3.0)
Requirement already satisfied: gunicorn in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (19.9.0)
Requirement already satisfied: sqlalchemy in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (1.3.8)
Requirement already satisfied: prometheus-client in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.7.1)
Requirement already satisfied: pandas in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.25.1)
Requirement already satisfied: docker in /opt/conda/lib/python3.7/site-packages (from bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (4.0.2)
Collecting pillow>=4.1.1 (from torchvision->-r /bento/requirements.txt (line 4))
  Downloading https://files.pythonhosted.org/packages/a4/da/2bd281c875686230eabc13d20ab590ea617563b0e746abfb0698c4d5b645/Pillow-6.1.0-cp37-cp37m-manylinux1_x86_64.whl (2.1MB)
Requirement already satisfied: scipy>=0.17.0 in /opt/conda/lib/python3.7/site-packages (from scikit-learn->-r /bento/requirements.txt (line 5)) (1.2.1)
Collecting joblib>=0.11 (from scikit-learn->-r /bento/requirements.txt (line 5))
  Downloading https://files.pythonhosted.org/packages/cd/c1/50a758e8247561e58cb87305b1e90b171b8c767b15b12a1734001f41d356/joblib-0.13.2-py2.py3-none-any.whl (278kB)
Requirement already satisfied: Werkzeug>=0.15 in /opt/conda/lib/python3.7/site-packages (from flask->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.16.0)
Requirement already satisfied: Jinja2>=2.10.1 in /opt/conda/lib/python3.7/site-packages (from flask->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (2.10.1)
Requirement already satisfied: itsdangerous>=0.24 in /opt/conda/lib/python3.7/site-packages (from flask->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (1.1.0)
Requirement already satisfied: setuptools in /opt/conda/lib/python3.7/site-packages (from protobuf>=3.6.0->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (41.0.1)
Requirement already satisfied: ruamel.yaml.clib>=0.1.2; platform_python_implementation == "CPython" and python_version < "3.8" in /opt/conda/lib/python3.7/site-packages (from ruamel.yaml>=0.15.0->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.1.2)
Requirement already satisfied: pyparsing>=2.0.2 in /opt/conda/lib/python3.7/site-packages (from packaging->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (2.4.2)
Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.7/site-packages (from requests->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (2019.6.16)
Requirement already satisfied: chardet<3.1.0,>=3.0.2 in /opt/conda/lib/python3.7/site-packages (from requests->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (3.0.4)
Requirement already satisfied: urllib3<1.25,>=1.21.1 in /opt/conda/lib/python3.7/site-packages (from requests->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (1.24.1)
Requirement already satisfied: idna<2.9,>=2.5 in /opt/conda/lib/python3.7/site-packages (from requests->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (2.8)
Requirement already satisfied: jmespath<1.0.0,>=0.7.1 in /opt/conda/lib/python3.7/site-packages (from boto3->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.9.4)
Requirement already satisfied: s3transfer<0.3.0,>=0.2.0 in /opt/conda/lib/python3.7/site-packages (from boto3->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.2.1)
Requirement already satisfied: botocore<1.13.0,>=1.12.231 in /opt/conda/lib/python3.7/site-packages (from boto3->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (1.12.231)
Requirement already satisfied: python-dateutil>=2.6.1 in /opt/conda/lib/python3.7/site-packages (from pandas->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (2.8.0)
Requirement already satisfied: pytz>=2017.2 in /opt/conda/lib/python3.7/site-packages (from pandas->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (2019.2)
Requirement already satisfied: websocket-client>=0.32.0 in /opt/conda/lib/python3.7/site-packages (from docker->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.56.0)
Requirement already satisfied: MarkupSafe>=0.23 in /opt/conda/lib/python3.7/site-packages (from Jinja2>=2.10.1->flask->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (1.1.1)
Requirement already satisfied: docutils<0.16,>=0.10 in /opt/conda/lib/python3.7/site-packages (from botocore<1.13.0,>=1.12.231->boto3->bentoml==0.4.1->-r /bento/requirements.txt (line 1)) (0.15.2)
Installing collected packages: torch, pillow, torchvision, joblib, scikit-learn, imageio
Successfully installed imageio-2.5.0 joblib-0.13.2 pillow-6.1.0 scikit-learn-0.21.3 torch-1.2.0 torchvision-0.4.0
Removing intermediate container 85f87d646259
 ---> b4f9c1c9445b
Step 10/11 : RUN if [ -f /bento/setup.sh ]; then /bin/bash -c /bento/setup.sh; fi
 ---> Running in 71825c516006
Removing intermediate container 71825c516006
 ---> bc37731d5d39
Step 11/11 : CMD ["bentoml serve-gunicorn /bento"]
 ---> Running in 4050988df254
Removing intermediate container 4050988df254
 ---> d6ed6efbdf93
Successfully built d6ed6efbdf93
Successfully tagged pytorch-fashion-mnist:latest
In [22]:
!docker run -p 5000:5000 pytorch-fashion-mnist

[2019-09-19 19:57:42 +0000] [1] [INFO] Starting gunicorn 19.9.0
[2019-09-19 19:57:42 +0000] [1] [INFO] Listening at: http://0.0.0.0:5000 (1)
[2019-09-19 19:57:42 +0000] [1] [INFO] Using worker: sync
[2019-09-19 19:57:42 +0000] [10] [INFO] Booting worker with pid: 10
[2019-09-19 19:57:42 +0000] [11] [INFO] Booting worker with pid: 11
[2019-09-19 19:57:42 +0000] [12] [INFO] Booting worker with pid: 12
^C
[2019-09-19 20:04:03 +0000] [1] [INFO] Handling signal: int
/opt/conda/lib/python3.7/site-packages/torch/nn/modules/container.py:92: UserWarning: Implicit dimension choice for log_softmax has been deprecated. Change the call to include dim=X as an argument.
  input = module(input)
[2019-09-19 20:04:03 +0000] [12] [INFO] Worker exiting (pid: 12)
[2019-09-19 20:04:03 +0000] [10] [INFO] Worker exiting (pid: 10)
[2019-09-19 20:04:03 +0000] [11] [INFO] Worker exiting (pid: 11)