Add model

.gitignore

@@ -0,0 +1,2 @@
+__pycache__
+.local

README.md

@@ -1,11 +1,60 @@
----
-title: ImageTransfer Server
-emoji: 🐠
-colorFrom: indigo
-colorTo: indigo
-sdk: docker
-pinned: false
-license: openrail
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Image Transfer and Upscaling API
+This service provides APIs to perform image upscaling and style transfer (Monet style) using Machine Learning models. The project is built using Flask, a lightweight WSGI web application framework.
+
+
+```bash
+git clone <repository-url>
+cd <project-directory>
+Install the dependencies using pip.
+```
+```bash
+pip install -r requirements.txt
+```
+## Running the Application
+Run the application with the following command:
+
+```bash
+ python3 -m flask run --host=0.0.0.0
+```
+The application should start and be accessible at localhost:5000.
+
+## API Endpoints
+The application exposes two API endpoints:
+
+```/monet```: POST endpoint which accepts an image file and returns the image stylized in the style of Monet paintings.
+
+```/upscale```: POST endpoint which accepts an image file and returns the upscaled version of the image.
+#File Structure
+```app.py```: This is the main application file which runs the Flask application and defines the API endpoints.
+```models/models.py```: This file contains the model loading and prediction functions.
+## models/models.py
+```tensor_to_image(tensor)```: Converts a TensorFlow tensor to an image and saves it to 'upscaled_image.jpg'.
+
+```request_to_image(request)```: Converts a Flask request object containing an image to a numpy array.
+
+```_monet(image, upscale=False)```: Accepts an image as input and returns the image stylized in the style of Monet paintings.
+
+```_upscale(image)```: Accepts an image as input and returns the upscaled version of the image.
+## Sending Requests to the API
+For both endpoints, the image file should be included in the request as form data with the key 'image'. Here is an example using curl:
+
+```bash
+Copy code
+curl -X POST -F "image=@<image-file-path>" http://localhost:5000/monet
+```
+```bash
+Copy code
+curl -X POST -F "image=@<image-file-path>" http://localhost:5000/upscale
+Replace <image-file-path> with the path to your image file.
+```
+## Notes
+The model files for the Monet style transfer (models/monet_generator) are not included in this repository. You need to download them separately and place them in the correct location.
+Contributing
+Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
+
+## License
+MIT
+
+docker buildx build --platform linux/amd64,linux/arm64 -t joshuapeddle/imagetransfer-server:0.0.2 --push .
+
+docker run -p 5000:5000 joshuapeddle/imagetransfer-server:0.0.2

app.py

@@ -0,0 +1,19 @@
+from flask import Flask , request, send_file
+from flask_cors import CORS
+from PIL import Image
+from models.models import _upscale, _monet, request_to_image
+from numpy import asarray
+
+app = Flask(__name__)
+CORS(app)
+@app.route("/monet", methods=['POST'])
+def monet():
+    return send_file(_monet(request_to_image(request) ,True), mimetype='image/jpg')
+
+@app.route("/upscale", methods=['POST'])
+def upscale():
+    return send_file(_upscale(request_to_image(request)), mimetype='image/jpg')
+
+@app.route("/", methods=['GET'])
+def hello():
+    return "Hello World"

dockerfile

@@ -0,0 +1,10 @@
+FROM python:3.9.16-buster
+
+WORKDIR /usr/src/app
+
+COPY requirements.txt ./
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+CMD [ "python3", "-m" , "flask", "run", "--host=0.0.0.0", "--debug"]

models/models.py

@@ -0,0 +1,83 @@
+import os
+import time
+from PIL import Image
+import tensorflow as tf
+import tensorflow_hub as hub
+from numpy import asarray
+from tensorflow.python.ops.numpy_ops import np_config
+np_config.enable_numpy_behavior()
+os.environ["TFHUB_DOWNLOAD_PROGRESS"] = "True"
+from huggingface_hub import from_pretrained_keras
+def tensor_to_image(tensor):
+    tensor = tf.cast(tf.clip_by_value(tensor, 0, 255), tf.uint8)
+    tensor = Image.fromarray(tensor.numpy())
+    tensor.save('upscaled_image.jpg')
+    return 'upscaled_image.jpg'
+
+
+def request_to_image(request):
+    image = request.files.get('image', False)
+    print(image)
+    if image:
+        return asarray(Image.open(image))
+    else:return asarray(Image.open(request.files['image']))
+    
+
+def _monet(image, upscale=False):
+    IMAGE_SIZE = (256, 256)
+    def decode_image(image):
+        #image = tf.image.decode_jpeg(image, channels=3)
+        image = (tf.cast(image, tf.float32) / 127.5) - 1
+        image = tf.reshape(image, [*IMAGE_SIZE, 3])
+        return image
+
+    #new_model = tf.keras.models.load_model('models/monet_generator', compile=False)
+    new_model = from_pretrained_keras("JoshuaPeddle/MonetGenerator")
+    #new_model.summary()
+    image = tf.image.resize(image, IMAGE_SIZE)
+    image = decode_image(image)
+    
+    image = tf.expand_dims(image, 0)
+    start = time.time()
+    prediction = new_model(image, training=False)
+    prediction = tf.reshape(prediction, [256, 256, 3])
+    prediction = (prediction + 1) / 2
+    prediction = tf.image.convert_image_dtype(prediction, tf.uint8)
+    print("Time Taken: %f" % (time.time() - start))
+
+    if upscale:
+        return _upscale(prediction)
+    
+    return tensor_to_image(prediction)
+
+
+
+
+def _upscale(image):
+    SAVED_MODEL_PATH = "https://tfhub.dev/captain-pool/esrgan-tf2/1"
+    def preprocess_image(_image):
+        """ Loads image from path and preprocesses to make it model ready
+            Args:
+                image_path: Path to the image file
+        """
+        #hr_image = tf.image.decode_image(tf.io.read_file(image_path))
+        hr_image = _image
+        # If PNG, remove the alpha channel. The model only supports
+        # images with 3 color channels.
+        if hr_image.shape[-1] == 4:
+            hr_image = hr_image[...,:-1]
+        hr_size = (tf.convert_to_tensor(hr_image.shape[:-1]) // 4) * 4
+        hr_image = tf.image.crop_to_bounding_box(hr_image, 0, 0, hr_size[0], hr_size[1])
+        hr_image = tf.cast(hr_image, tf.float32)
+        return tf.expand_dims(hr_image, 0)
+
+
+    hr_image = preprocess_image(image)
+    model = hub.load(SAVED_MODEL_PATH)
+
+    start = time.time()
+    fake_image = model(hr_image)
+    fake_image = tf.squeeze(fake_image)
+    print("Time Taken: %f" % (time.time() - start))
+    fake_image = tf.image.resize(fake_image, (512,512))
+    return tensor_to_image(fake_image)

requirements.txt

@@ -0,0 +1,64 @@
+absl-py==1.4.0
+astunparse==1.6.3
+blinker==1.6.2
+cachetools==5.3.0
+certifi==2023.5.7
+charset-normalizer==3.1.0
+click==8.1.3
+contourpy==1.0.7
+cycler==0.11.0
+filelock==3.12.0
+Flask==2.3.2
+Flask-Cors==3.0.10
+flatbuffers==23.5.8
+fonttools==4.39.4
+fsspec==2023.5.0
+gast==0.4.0
+google-auth==2.18.0
+google-auth-oauthlib==1.0.0
+google-pasta==0.2.0
+grpcio==1.54.0
+h5py==3.8.0
+huggingface-hub==0.14.1
+idna==3.4
+importlib-metadata==6.6.0
+importlib-resources==5.12.0
+itsdangerous==2.1.2
+jax==0.4.9
+Jinja2==3.1.2
+keras==2.12.0
+kiwisolver==1.4.4
+libclang==16.0.0
+Markdown==3.4.3
+MarkupSafe==2.1.2
+matplotlib==3.7.1
+ml-dtypes==0.1.0
+numpy==1.23.5
+oauthlib==3.2.2
+opt-einsum==3.3.0
+packaging==23.1
+Pillow==9.5.0
+protobuf==4.23.0
+pyasn1==0.5.0
+pyasn1-modules==0.3.0
+pyparsing==3.0.9
+python-dateutil==2.8.2
+PyYAML==6.0
+requests==2.30.0
+requests-oauthlib==1.3.1
+rsa==4.9
+scipy==1.10.1
+six==1.16.0
+tensorboard==2.12.3
+tensorboard-data-server==0.7.0
+tensorflow==2.12.0
+tensorflow-estimator==2.12.0
+tensorflow-hub==0.13.0
+tensorflow-io-gcs-filesystem==0.32.0
+termcolor==2.3.0
+tqdm==4.65.0
+typing_extensions==4.5.0
+urllib3==1.26.15
+Werkzeug==2.3.4
+wrapt==1.14.1
+zipp==3.15.0