def handle_movie_review(movieReview):
path = 'sgd/static/trained/natural-language-processor'
learn = learner.load_learner(path)
predictions = dict()
predictions['raw'] = learn.predict(movieReview)
predictions['originalReview'] = movieReview
return predictions
def _classification(image_rgb, weights):
"""Helper function. Classifies the input image according to `weights`.
Parameters
----------
image_rgb : 3D array
RGB image of the lepidopteran (ruler and tags cropped out).
weights : str or pathlib.Path
Path of the file containing weights.
Returns
-------
prediction : int
Prediction obtained with the given weights.
Notes
-----
If a string is given in `weights`, it will be converted into a pathlib.Path
object.
"""
if isinstance(weights, str):
weights = Path(weights)
connection.download_weights(weights)
# parameters here were defined when training the networks.
learner = load_learner(fname=weights)
_, prediction, _ = learner.predict(image_rgb)
return int(prediction)
def handle_uploaded_file(image):
imageLocation = save(image)
image = open_image(imageLocation)
path = 'sgd/static/trained/comp-vision'
learn = learner.load_learner(path)
predictions = dict()
predictions['raw'] = learn.predict(image)
predictions['friendly'] = getFriendlyProbabilities(predictions['raw'])
return predictions
def predict(url):
response = requests.get(url)
img = Image.open(BytesIO(response.content)).resize((512, 384),
Image.ANTIALIAS)
filename = "./ml-model/data/img.jpg"
img.save(filename)
learn = load_learner("./ml-model/")
category = str(learn.predict(open_image(filename))[0])
tensor_probs = learn.predict(open_image(filename))[2]
pred_probs = {
"cardboard": round(float(tensor_probs[0]), 2),
"glass": round(float(tensor_probs[1]), 2),
"metal": round(float(tensor_probs[2]), 2),
"paper": round(float(tensor_probs[3]), 2),
"plastic": round(float(tensor_probs[4]), 2),
"trash": round(float(tensor_probs[5]), 2)
}
ans = {"url": url, "category": category, "pred_probs": pred_probs}
return ans
def predicting_classes(image_rgb, weights=WEIGHTS_CLASSES):
"""Predicts position and gender of the lepidopteran in `image_rgb`,
according to `weights`.
Parameters
----------
image_rgb : 3D array
RGB image of the lepidopteran (ruler and tags cropped out).
weights : str or pathlib.Path
Path of the file containing weights.
Returns
-------
prediction : string
Prediction obtained with the given weights, between the classes
`female`, `male`, or `upside_down`.
probabilities : 1D array
Probabilities for the prediction returned by the network for each
class.
Notes
-----
If a string is given in `weights`, it will be converted into a pathlib.Path
object.
"""
if isinstance(weights, str):
weights = Path(weights)
connection.download_weights(weights)
# parameters here were defined when training the networks.
learner = load_learner(fname=weights)
prediction, _, probabilities = learner.predict(image_rgb)
return prediction, probabilities
def binarization(image_rgb, weights=WEIGHTS_BIN):
"""Extract the shape of the elements in an input image using the U-net
deep learning architecture.
Parameters
----------
image_rgb : (M, N, 3) ndarray
Input RGB image of a lepidopteran, with ruler and tags.
weights : str or pathlib.Path
Path of the file containing weights for segmentation.
Returns
-------
tags_bin : (M, N) ndarray
Binary image containing tags in the input image.
ruler_bin : (M, N) ndarray
Binary image containing the ruler in the input image.
lepidop_bin : (M, N) ndarray
Binary image containing the lepidopteran in the input image.
"""
if isinstance(weights, str):
weights = Path(weights)
connection.download_weights(weights)
learner = load_learner(fname=weights)
print('Processing U-net...')
_, _, classes = learner.predict(image_rgb)
_, tags_bin, ruler_bin, lepidop_bin = classes[:4]
# rescale the predicted images back up and binarize them.
tags_bin = img_as_bool(_rescale_image(image_rgb, tags_bin))
ruler_bin = img_as_bool(_rescale_image(image_rgb, ruler_bin))
lepidop_bin = img_as_bool(_rescale_image(image_rgb, lepidop_bin))
return tags_bin, ruler_bin, lepidop_bin
from fastai.vision import open_image
from fastai.vision.learner import load_learner
import sys
#defaults.device = torch.device('cpu')
if (len(sys.argv) != 2):
print("USAGE: classify.py <path to image>")
sys.exit(1)
img = open_image(sys.argv[1])
learn = load_learner(".", fname="its_raining_men.pkl")
pred_class,pred_idx,outputs = learn.predict(img)
print("p(ramen) = %1.2f, p(soba) = %1.2f, p(udon) = %1.2f. I think this is %s." % (outputs[0],outputs[1],outputs[2],pred_class))
# In[ ]:
learn.save("/content/drive/My Drive/data/models/model3")
# In[ ]:
learn.export("/content/drive/My Drive/data/models/model3.pkl")
# In[ ]:
learn = load_learner("/content/drive/My Drive/data/models/")
# In[ ]:
learn.load("/content/drive/My Drive/data/models/model3")
# In[49]:
learn.predict(open_image("stack-of-paper.jpg"))
# In[ ]: