website: Use pyscript fetch instead of js fetch

* Also add some CSS styling
tedgravlin · Dec 2, 2023 · 98b33e3 · 98b33e3
1 parent 98de729
commit 98b33e3
Show file tree

Hide file tree

Showing 5 changed files with 78 additions and 49 deletions.
diff --git a/.gitattributes b/.gitattributes
diff --git a/docs/index.html b/docs/index.html
@@ -3,6 +3,9 @@
 
 <head>
     <title>Domain Validator</title>
+    <!-- Import CSS -->
+    <link rel="stylesheet" href="main.css">
+    <!-- Import pyscript -->
     <script type="module" src="https://pyscript.net/releases/2023.11.2/core.js"></script>
 </head>
 
@@ -13,9 +16,9 @@ <h1>Neural Network Domain Validator</h1>
     <label for="url-input">Website URL</label>
     <input type="text" id="url-input" placeholder="Website URL">
     <button py-click="get_url">Check URL</button>
-    <p id="progress-text"></p>
+    <p id="progress-text">Loading pyscript. Please wait...</p>
     <p id="result-text"></p>
-    <script type="py" src="main.py" config="./pyscript.json"></script>
+    <script type="py" src="./main.py" config="./pyscript.json"></script>
 </body>
 
 </html>
diff --git a/docs/main.css b/docs/main.css
@@ -0,0 +1,3 @@
+body {
+    text-align: center;
+}
diff --git a/docs/main.py b/docs/main.py
@@ -1,6 +1,7 @@
 import pandas as pd
 import joblib
 from pyscript import document
+from pyscript import display
 from sklearn.feature_extraction.text import TfidfVectorizer
 from sklearn.model_selection import train_test_split
 from sklearn.neural_network import MLPClassifier
@@ -9,72 +10,86 @@
 from sklearn.preprocessing import StandardScaler
 from scipy.sparse import hstack
 from js import XMLHttpRequest
-from js import window
+#from js import window
 import io
 from os.path import exists
 import os
-import re
+#import re
+import csv
 
 # Get the progress text HTML element
 progress_text = document.querySelector("#progress-text")
+progress_text.innerText = "Pyscipt loaded."
 
 # Gets the model and returns it
 def load_model():
 
     progress_text.innerText = "Removing old model..."
 
     # Delete the model files if they already exist
-    if (exists('model.pkl')):
-        os.remove('model.pkl')
-    if (exists('tfidf.pkl')):
-        os.remove('tfidf.pkl')
-
-    # Check if the current domain is an IP address (hosted locally) or a real domain (hosted on GH pages)
-    if (re.search("^((25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\\.){3}(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)$", window.location.hostname)):
-        model_url = 'http://127.0.0.1:5500/models/model.pkl'
-        tfidf_url = 'http://127.0.0.1:5500/models/tfidf.pkl'
-    else:
-        model_url = 'https://github.com/tedgravlin/neural-network-domain-validator/blob/a792b3c04d16be88eca95f3d6948d25f020722e2/models/model.pkl'
-        tfidf_url = 'https://github.com/tedgravlin/neural-network-domain-validator/blob/a792b3c04d16be88eca95f3d6948d25f020722e2/models/tfidf.pkl'
+    # if (exists('model.pkl')):
+    #     os.remove('model.pkl')
+    # if (exists('tfidf.pkl')):
+    #     os.remove('tfidf.pkl')
+
+    # model_url = 'http://127.0.0.1:5500/models/model.pkl'
+    # tfidf_url = 'http://127.0.0.1:5500/models/tfidf.pkl'
 
-    progress_text.innerText = "Fetching model.pkl..."
+    # progress_text.innerText = "Fetching model.pkl..."
+
+    # # Get model.pkl
+    # model_req = XMLHttpRequest.new()
+    # model_req.open("GET", model_url, False)
+    # model_req.send(None)
+    # model_response = model_req.response.encode()
 
-    # Get model.pkl
-    model_req = XMLHttpRequest.new()
-    model_req.open("GET", model_url, False)
-    model_req.send(None)
-    model_response = (model_req.response)
+    # progress_text.innerText = "Writing to model.pkl..."
 
-    progress_text.innerText = "Writing to model.pkl..."
+    # # Write the contents of the model response to model.pkl
+    # with open('model.pkl', 'wb') as model_file:
+    #     model_file.write(model_response)
 
-    # Write the contents of the model response to model.pkl
-    with open('model.pkl', 'w') as model_file:
-        model_file.write(model_response)
+    # progress_text.innerText = "Fetching tfidf.pkl..."
 
-    progress_text.innerText = "Fetching tfidf.pkl..."
+    # # Get tfidfk.pkl
+    # tfidf_req = XMLHttpRequest.new()
+    # tfidf_req.open("GET", tfidf_url, False)
+    # tfidf_req.send(None)
+    # tfidf_response = tfidf_req.response.encode()
 
-    # Get tfidfk.pkl
-    tfidf_req = XMLHttpRequest.new()
-    tfidf_req.open("GET", tfidf_url, False)
-    tfidf_req.send(None)
-    tfidf_response = (tfidf_req.response)
+    # progress_text.innerText = "Writing to tfidf.pkl..."
 
-    progress_text.innerText = "Writing to tfidf.pkl..."
+    # # Write the contents of the tfidf response to tfidf.pkl
+    # with open('tfidf.pkl', 'wb') as tfidf_file:
+    #     tfidf_file.write(tfidf_response)
 
-    # Write the contents of the tfidf response to tfidf.pkl
-    with open('tfidf.pkl', 'w') as tfidf_file:
-        tfidf_file.write(tfidf_response)
+    display("model file size", os.path.getsize("model.pkl"))
+    display("tfidf file size", os.path.getsize("tfidf.pkl"))
 
-    #model = joblib.load(model_file)
-    #tfidf = joblib.load(tfidf_file)
+    model = joblib.load('model.pkl')
+    tfidf = joblib.load('tfidf.pkl')
 
     progress_text.innerText = "Model load complete."
 
-    return exists('model.pkl'), exists('tfidf.pkl')
+    return model, tfidf
 
 def test_model(model, tfidf):
+    # Delete the test dataset if  it already exists
+    if (exists('testdataset.csv')):
+        os.remove('testdataset.csv')
+
+    # Fetch testdataset.csv
+    req = XMLHttpRequest.new()
+    req.open("GET", 'http://127.0.0.1:5500/dataset/testdataset.csv',True)
+    req.send(None)
+    response = (req.response)
+
+    # Write the contents of the testdataset response to testdataset.csv
+    with open('testdataset.csv', 'w') as testdataset_file:
+        testdataset_file.write(response)
+
     # Get the test dataset CSV file
-    test_dataset = pd.read_csv("./dataset/testdataset.csv")
+    test_dataset = pd.read_csv("testdataset.csv")
 
     # Turn the test dataset into a pandas data frame
     dataframe = pd.DataFrame(test_dataset)
@@ -107,18 +122,19 @@ def test_model(model, tfidf):
         if (prediction[label] == y.to_list()[label]):
             correct_count = correct_count + 1
         count = count + 1
-    print("ACTUAL:", y.to_list())
-    print("CORRECT PREDICTIONS:", correct_count,"/",count)
-    print("ACCURACY OF THIS TEST:", (correct_count/count) * 100)
+    display("ACTUAL:", y.to_list())
+    display("CORRECT PREDICTIONS:", correct_count,"/",count)
+    display("ACCURACY OF THIS TEST:", (correct_count/count) * 100)
 
 def get_url(event):
     # Load the model from storage
-    model = load_model()
-    #test_model(model, tfidf)
+    model, tfidf = load_model()
+    # Test the model with the test dataset
+    test_model(model, tfidf)
     input_text = document.querySelector("#url-input")
     url = input_text.value
     output_text = document.querySelector("#result-text")
-    output_text.innerText = model
+    #output_text.innerText = model
 
 
 

diff --git a/docs/pyscript.json b/docs/pyscript.json
@@ -1,4 +1,12 @@
 {
-    "packages": ["pandas", "joblib", "scikit-learn"]
-}
+  "packages": [
+    "pandas",
+    "joblib",
+    "scikit-learn"
+  ],
+  "files": {
+    "../models/model.pkl": "model.pkl",
+    "../models/tfidf.pkl": "tfidf.pkl"
+  }
 
+}