Merge pull request #308 from sryu1/master

Updated keras and opencv keras snippets attempt

README.md

@@ -26,13 +26,12 @@ This repository contains two components of [Teachable Machine](https://teachable
 
 You have a few options:
 
-- [Use this form](https://forms.gle/uthe2C4tZNPA11GX7).
 - Share your projects using [#teachablemachine](https://twitter.com/hashtag/teachablemachine) on Twitter or on the [Experiments with Google](https://experiments.withgoogle.com/submit) page.
 - Open an issue in this repository.
 
 ## Community Contributions and Projects
 
-- [Teachable Machine Node Library for image models](https://github.com/tr7zw/teachablemachine-node-example)
+- [Teachable Machine Node Library for image models](https://github.com/tr7zw/teachablemachine-node-example) (Archived and now continued [here](https://github.com/drinkspiller/teachablemachine-node-example/))
 - [Teachable Machine Mobile for image models](https://github.com/mstale007/Teachable_Machine_Mobile/tree/master)
 
 ## Disclaimer

libraries/README.md

@@ -1,13 +1,13 @@
 # Teachable Machine Libraries
 
-This section contains support libraries for the new version of Teachable Machine. For more info go to: [Teachable Machine](https://teachablemachine.withgoogle.com/io19).
+This section contains support libraries for the new version of Teachable Machine.
 
 ## Model Libraries
 
 | Library | Based on model  | Details                                                 | Install | CDN | 
 |---------|-----------------|---------------------------------------------------------|---------|-----|
 | [Image](./image/) | [MobileNet](https://github.com/tensorflow/tfjs-models/tree/master/mobilenet)       | Use a model trained to classify your own images         | `npm i @teachablemachine/image` | [![](https://data.jsdelivr.com/v1/package/npm/@teachablemachine/image/badge)](https://www.jsdelivr.com/package/npm/@teachablemachine/image) |
-| [Audio](./audio/)   | [Speech Commands](https://github.com/tensorflow/tfjs-models/tree/master/speech-commands) | Use a model trained to classify your own audio snippets | npm i @tensorflow-models/speech-commands     | [![](https://data.jsdelivr.com/v1/package/npm/@tensorflow-models/speech-commands/badge)](https://github.com/tensorflow/tfjs-models/tree/master/speech-commands) | 
+| [Audio](./audio/)   | [Speech Commands](https://github.com/tensorflow/tfjs-models/tree/master/speech-commands) | Use a model trained to classify your own audio snippets | `npm i @tensorflow-models/speech-commands`     | [![](https://data.jsdelivr.com/v1/package/npm/@tensorflow-models/speech-commands/badge)](https://github.com/tensorflow/tfjs-models/tree/master/speech-commands) | 
 | [Pose](./pose/)   | [PoseNet](https://github.com/tensorflow/tfjs-models/tree/master/posenet) | Use a model trained to classify body poses | `npm i @teachablemachine/pose`     | [![](https://data.jsdelivr.com/v1/package/npm/@teachablemachine/pose/badge)](https://www.jsdelivr.com/package/npm/@teachablemachine/pose) |
 
 ## Development

snippets/markdown/image/tensorflow/keras.md

@@ -1,45 +1,45 @@
 ```python
-from keras.models import load_model
-from PIL import Image, ImageOps #Install pillow instead of PIL
+from keras.models import load_model  # TensorFlow is required for Keras to work
+from PIL import Image, ImageOps  # Install pillow instead of PIL
 import numpy as np
 
 # Disable scientific notation for clarity
 np.set_printoptions(suppress=True)
 
 # Load the model
-model = load_model('keras_Model.h5', compile=False)
+model = load_model("keras_Model.h5", compile=False)
 
 # Load the labels
-class_names = open('labels.txt', 'r').readlines()
+class_names = open("labels.txt", "r").readlines()
 
 # Create the array of the right shape to feed into the keras model
 # The 'length' or number of images you can put into the array is
-# determined by the first position in the shape tuple, in this case 1.
+# determined by the first position in the shape tuple, in this case 1
 data = np.ndarray(shape=(1, 224, 224, 3), dtype=np.float32)
 
 # Replace this with the path to your image
-image = Image.open('<IMAGE_PATH>').convert('RGB')
+image = Image.open("<IMAGE_PATH>").convert("RGB")
 
-#resize the image to a 224x224 with the same strategy as in TM2:
-#resizing the image to be at least 224x224 and then cropping from the center
+# resizing the image to be at least 224x224 and then cropping from the center
 size = (224, 224)
 image = ImageOps.fit(image, size, Image.Resampling.LANCZOS)
 
-#turn the image into a numpy array
+# turn the image into a numpy array
 image_array = np.asarray(image)
 
 # Normalize the image
-normalized_image_array = (image_array.astype(np.float32) / 127.0) - 1
+normalized_image_array = (image_array.astype(np.float32) / 127.5) - 1
 
 # Load the image into the array
 data[0] = normalized_image_array
 
-# run the inference
+# Predicts the model
 prediction = model.predict(data)
 index = np.argmax(prediction)
 class_name = class_names[index]
 confidence_score = prediction[0][index]
 
-print('Class:', class_name, end='')
-print('Confidence score:', confidence_score)
+# Print prediction and confidence score
+print("Class:", class_name[2:], end="")
+print("Confidence Score:", confidence_score)
 ```

snippets/markdown/image/tensorflow/opencv-keras.md

@@ -1,36 +1,49 @@
-```Python
-import cv2
+```python
+from keras.models import load_model  # TensorFlow is required for Keras to work
+import cv2  # Install opencv-python
 import numpy as np
-from keras.models import load_model
+
+# Disable scientific notation for clarity
+np.set_printoptions(suppress=True)
 
 # Load the model
-model = load_model('keras_model.h5')
+model = load_model("keras_Model.h5", compile=False)
 
-# CAMERA can be 0 or 1 based on default camera of your computer.
-camera = cv2.VideoCapture(0)
+# Load the labels
+class_names = open("labels.txt", "r").readlines()
 
-# Grab the labels from the labels.txt file. This will be used later.
-labels = open('labels.txt', 'r').readlines()
+# CAMERA can be 0 or 1 based on default camera of your computer
+camera = cv2.VideoCapture(0)
 
 while True:
-    # Grab the webcameras image.
+    # Grab the webcamera's image.
     ret, image = camera.read()
-    # Resize the raw image into (224-height,224-width) pixels.
+
+    # Resize the raw image into (224-height,224-width) pixels
     image = cv2.resize(image, (224, 224), interpolation=cv2.INTER_AREA)
+
     # Show the image in a window
-    cv2.imshow('Webcam Image', image)
+    cv2.imshow("Webcam Image", image)
+
     # Make the image a numpy array and reshape it to the models input shape.
     image = np.asarray(image, dtype=np.float32).reshape(1, 224, 224, 3)
+
     # Normalize the image array
     image = (image / 127.5) - 1
-    # Have the model predict what the current image is. Model.predict
-    # returns an array of percentages. Example:[0.2,0.8] meaning its 20% sure
-    # it is the first label and 80% sure its the second label.
-    probabilities = model.predict(image)
-    # Print what the highest value probabilitie label
-    print(labels[np.argmax(probabilities)])
+
+    # Predicts the model
+    prediction = model.predict(image)
+    index = np.argmax(prediction)
+    class_name = class_names[index]
+    confidence_score = prediction[0][index]
+
+    # Print prediction and confidence score
+    print("Class:", class_name[2:], end="")
+    print("Confidence Score:", str(np.round(confidence_score * 100))[:-2], "%")
+
     # Listen to the keyboard for presses.
     keyboard_input = cv2.waitKey(1)
+
     # 27 is the ASCII for the esc key on your keyboard.
     if keyboard_input == 27:
         break

snippets/package.json

@@ -1,12 +1,12 @@
 {
     "dependencies": {
-        "http-server": "^0.11.1"
+        "http-server": "^14.1.1"
     },
     "scripts": {
         "start": "http-server --cors snippets",
         "lint": "markdownlint ./markdown ./converter"
     },
     "devDependencies": {
-        "markdownlint-cli": "^0.18.0"
+        "markdownlint-cli": "^0.32.2"
     }
 }