Update params.json

cfn/params.json

@@ -1,22 +1,22 @@
 [
     {
       "ParameterKey": "GitHubRepo",
-      "ParameterValue": "codepipeline-ecr-build-sf-execution"
+      "ParameterValue": "enter cloned repo name here for codepipeline-ecr-build-sf-execution"
     },
     {
         "ParameterKey": "GitHubBranch",
-        "ParameterValue": "master"
-      },
-      {
-        "ParameterKey": "GitHubToken",
-        "ParameterValue": ""
-      },
-      {
-        "ParameterKey": "GitHubUser",
-        "ParameterValue": ""
-      },
-      {
-        "ParameterKey": "MlOpsStepFunctionArn",
-        "ParameterValue": ""
-      }
-]
+        "ParameterValue": "enter branch - main or xxx"
+    },
+    {
+      "ParameterKey": "GitHubToken",
+      "ParameterValue": "enter github token here"
+    },
+    {
+      "ParameterKey": "GitHubUser",
+      "ParameterValue": "enter github user here"
+    },
+    {
+      "ParameterKey": "MlOpsStepFunctionArn",
+      "ParameterValue": "enter MLOpsStep Functions Arn here"
+    }
+]

container/Dockerfile

@@ -3,24 +3,44 @@
 # for serving inferences in a stable way.
 
 FROM public.ecr.aws/lts/ubuntu:20.04
+RUN apt update
+RUN apt install -y software-properties-common
+RUN add-apt-repository ppa:deadsnakes/ppa
+RUN apt install -y python3.7
 
+# install pip 
+RUN apt install -y python3-pip
 
-RUN apt-get -y update && apt-get install -y --no-install-recommends \
-         wget \
-         python \
-         nginx \
-         ca-certificates \
-    && rm -rf /var/lib/apt/lists/*
+RUN apt-get install -y --no-install-recommends nginx ca-certificates
+RUN rm -rf /var/lib/apt/lists/*
 
-# Here we get all python packages.
-# There's substantial overlap between scipy and numpy that we eliminate by
-# linking them together. Likewise, pip leaves the install caches populated which uses
-# a significant amount of space. These optimizations save a fair amount of space in the
-# image, which reduces start up time.
-RUN wget https://bootstrap.pypa.io/pip/2.7/get-pip.py && python get-pip.py && \
-    pip install numpy==1.16.2 scipy==1.2.1 scikit-learn==0.20.2 pandas flask gevent gunicorn && \
-        (cd /usr/local/lib/python2.7/dist-packages/scipy/.libs; rm *; ln ../../numpy/.libs/* .) && \
-        rm -rf /root/.cache
+RUN python3.7 -m pip install langid
+RUN python3.7 -m pip install langdetect
+RUN python3.7 -m pip install emoji
+RUN python3.7 -m pip install nltk==3.2.5
+RUN python3.7 -m pip install pandas==1.1.5
+RUN python3.7 -m pip install numpy==1.19.5
+RUN python3.7 -m pip install torch
+RUN python3.7 -m pip install transformers
+RUN python3.7 -m pip install flask
+#previos versions for package gevent and gunicorn
+#RUN python3.7 -m pip install gevent==1.5.0
+#RUN python3.7 -m pip install gunicorn
+
+#RUN python3.7 -m pip install gevent==1.3.4
+#RUN python3.7 -m pip install gunicorn==19.9.0
+
+
+#RUN python3.7 -m pip install gevent==20.6.2
+#RUN python3.7 -m pip install gunicorn==20.0.4
+
+RUN python3.7 -m pip install gevent==20.9.0
+RUN python3.7 -m pip install gunicorn==20.0.4
+
+RUN python3.7 -m pip install sklearn
+RUN python3.7 -m pip install gensim
+RUN python3.7 -m pip install xlrd
+RUN python3.7 -m pip install openpyxl
 
 # Set some environment variables. PYTHONUNBUFFERED keeps Python from buffering our standard
 # output stream, which means that logs can be delivered to the user quickly. PYTHONDONTWRITEBYTECODE
@@ -37,4 +57,3 @@ WORKDIR /opt/program
 
 RUN chmod +x /opt/program/train
 RUN chmod +x /opt/program/serve
-

container/decision_trees/predictor.py

@@ -6,14 +6,16 @@
 import os
 import json
 import pickle
-import StringIO
 import sys
 import signal
 import traceback
-
 import flask
+import io
+
+import numpy as np
 
 import pandas as pd
+import torch
 
 prefix = '/opt/ml/'
 model_path = os.path.join(prefix, 'model')
@@ -27,28 +29,73 @@ class ScoringService(object):
     @classmethod
     def get_model(cls):
         """Get the model object for this instance, loading it if it's not already loaded."""
+        print('test get model -- test+1')
         if cls.model == None:
-            with open(os.path.join(model_path, 'decision-tree-model.pkl'), 'r') as inp:
-                cls.model = pickle.load(inp)
+            #with open(os.path.join(model_path, 'decision-tree-model.pkl'), 'r') as inp:
+        #    with open(os.path.join(model_path, 'best-LSTM-model-parameters.pkl'), 'rb') as inp:
+        #       cls.model = pickle.load(inp)
+            #the_model = torch.load(os.path.join(model_path, 'best-LSTM-model-parameters.pt'))
+        #return cls.model
+
+            cls.model = torch.jit.load(os.path.join(model_path, 'best-LSTM-model-parameters.pth'))
+
         return cls.model
 
     @classmethod
     def predict(cls, input):
+        print('test predict')
         """For the input, do the predictions and return them.
 
         Args:
             input (a pandas dataframe): The data on which to do the predictions. There will be
                 one prediction per row in the dataframe"""
-        clf = cls.get_model()
-        return clf.predict(input)
-
+
+        try:
+            clf = cls.get_model()
+        except:
+            print('did not get the model')
+
+        print('@@@@@@@@@@@@@@@@@@@')
+        print(type(input))
+        print(input)
+        print(input[0])
+        print(input[1])
+        print('@@@@@@@@@@@@@@@@@@@')
+
+
+        test_input=[[2779, 4496, 1744, 3480,  674,   24,   24,   24,   24,   24,   24,
+          24,   24,   24,   24,   24,   24,   24,   24,   24,   24,   24,
+          24,   24,   24,   24,   24,   24,   24,   24,   24,   24,   24,
+          24,   24,   24,   24,   24,   24,   24,   24,   24,   24,   24,
+          24,   24,   24,   24,   24,   24]]
+
+        #return clf.predict(input)
+
+
+        try:
+            device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+            x=torch.from_numpy(np.array(input)).to(device)
+        except Exception as error:
+            print('Caught this error: ' + repr(error))
+            print('error when generating input')
+
+        try:
+            with torch.no_grad():
+                result = clf(x)
+        except Exception as error:
+            print('Caught this error: ' + repr(error))
+            print('error when returing results for prediction')
+
+        return result
+
 # The flask app for serving predictions
 app = flask.Flask(__name__)
 
 @app.route('/ping', methods=['GET'])
 def ping():
     """Determine if the container is working and healthy. In this sample container, we declare
     it healthy if we can load the model successfully."""
+    print('testing ping()')
     health = ScoringService.get_model() is not None  # You can insert a health check here
 
     status = 200 if health else 404
@@ -61,23 +108,44 @@ def transformation():
     just means one prediction per line, since there's a single column.
     """
     data = None
-
+    print('testing transformation')
     # Convert from CSV to pandas
     if flask.request.content_type == 'text/csv':
         data = flask.request.data.decode('utf-8')
-        s = StringIO.StringIO(data)
+        s = io.StringIO(data)
         data = pd.read_csv(s, header=None)
     else:
         return flask.Response(response='This predictor only supports CSV data', status=415, mimetype='text/plain')
 
+
+    print('#############')
+    print('aaa')
+    print('#############1')
+    print(data)
+    print('#############2')
+    print(data[0])
+    print('#############3')
+    print(data.head() )
+    print(type(data))
+    print('############4')
+    print('start')
+    print(len(data[0]))
+    for x in range(len(data[0])):
+        print(x)
+
+    print('aaa')
     print('Invoked with {} records'.format(data.shape[0]))
 
     # Do the prediction
     predictions = ScoringService.predict(data)
-
+
+    print('*******')
+    print(predictions)
+    print('*******')
+
     # Convert from numpy back to CSV
-    out = StringIO.StringIO()
-    pd.DataFrame({'results':predictions}).to_csv(out, header=False, index=False)
+    out = io.StringIO()
+    pd.DataFrame({'results':predictions[0]}).to_csv(out, header=False, index=False)
     result = out.getvalue()
 
     return flask.Response(response=result, status=200, mimetype='text/csv')

container/decision_trees/serve

@@ -1,4 +1,5 @@
-#!/usr/bin/env python
+#!/usr/bin/python3.7
+#######!/usr/bin/env python
 
 # This file implements the scoring service shell. You don't necessarily need to modify it for various
 # algorithms. It starts nginx and gunicorn with the correct configurations and then simply waits until