Update app.py (#1)

- Update app.py (0171e240f90330af7f0c3d9549a2f7b0db4b59f1)

app.py

@@ -6,139 +6,13 @@ Automatically generated by Colaboratory.
 Original file is located at
     https://colab.research.google.com/drive/1L-z1dtcO8Co-TSZyGVYHNUAX3wvUSswD
 """
-
-#hide
-
-'''
-!pip install fastbook
-!pip install gradio
-
-
-
-import fastbook
-fastbook.setup_book()
-'''
-
 #hide
 from fastbook import *
 from fastai.vision.widgets import *
 from fastai.vision.utils import *
 
-ims = search_images_ddg('satellite imagery')
-len(ims)
-
-#Image.open(dest)
-
-
-
-def get_failed_images(path):
-  fns = get_image_files(path)
-  failed = verify_images(fns)
-  failed.map(Path.unlink)
-
-def download_single_type_image(image_of):
-   if not isinstance(image_of, str):
-    return "Please input a string as a file path"
-   for count in range(0, len(ims) - 1):
-    dest = 'images/' + image_of + str(count) + '.jpg'
-    try:
-      download_url(ims[count], dest)
-      #print('this is the file name' + str(dest))
-    except:
-      print("The file at count " + str(count))
-
-def create_data_block():
-  data = DataBlock(
-  blocks=(ImageBlock, CategoryBlock),
-  get_items=get_image_files,
-  splitter=RandomSplitter(valid_pct=0.2, seed=42),
-  get_y=parent_label,
-  item_tfms=Resize(128)
-  )
-  return data
-'''
-def show_batch(path, data, n_items=9, n_rows=3):
-  dls = data.dataloaders('/content/' + path)
-  dls.valid.show_batch(max_n=n_items, nrows=n_rows)
-'''
-
-def train_model(path, data, num_epochs):
-  data = data.new(
-      item_tfms=RandomResizedCrop(224, min_scale=0.5),
-      batch_tfms=aug_transforms())
-  dls = data.dataloaders(path)
-  learn = vision_learner(dls, resnet18, metrics=error_rate)
-  #the cbs=ShowGraphCallback should graph the loss
-  learn.fine_tune(num_epochs, cbs=ShowGraphCallback())
-  return learn
-
-def download_multi_type_image(file_path, categories):
-  if not isinstance(file_path, str):
-    return "Please input a string as a file path"
-  if not isinstance(categories, list):
-    return "Please input a list of strings a categories"
-  
-  #urban_types = 'city','town','village'
-  path = Path(file_path)
-
-  if not path.exists():
-    path.mkdir()
-  """
-  for o in categories:
-    dest = (path/o)
-    dest.mkdir(exist_ok=True)
-    results = search_images_ddg(f'{o}')
-    download_images(dest, urls=results)
-  """
-  for o in categories:
-    dest = (path/o)
-    print(dest)
-    dest.mkdir(exist_ok=True)
-    results = search_images_ddg(f'{o} ')
-    print(results)
-    download_images(dest, urls=results)
-    print('an image is downlaoding')
-
-def show_confusion_matrix(learn):
-  interp = ClassificationInterpretation.from_learner(learn)
-  interp.plot_confusion_matrix()
-
-  interp.plot_top_losses(5, nrows=3)
-
-def clean_data(learn):
-  cleaner = ImageClassifierCleaner(learn)
-  return cleaner
-
-directory_name = 'satellite imagery'
-classes =  ['satellite imagery tokyo', 'satellite imagery New York City', 'satellite imagery baltimore']
-
-#dest = 'images/city.jpg'
-  #download_url(ims[0], dest)
-
-  #im = Image.open(dest)
-  #im.to_thumb(1000,1000)
-
-#step 1 download the data the I want to build a classifier for
-# the first parameter is the name of the directory that the data will be stored in
-# the second paramteter is a list of all the of the classes the the image classifier will be trained on
-download_multi_type_image(directory_name,classes)
-
-data = create_data_block()
-
-get_image_files(directory_name)
-
-a  = get_failed_images(directory_name)
-
-#show_batch(directory_name, data)
-
-learn = train_model(directory_name, data, 10)
-
-learn
-
-show_confusion_matrix(learn)
-
-get_failed_images(directory_name)
 
+learn = load_learner('model.pkl')
 labels = learn.dls.vocab
 def predict(img):
     img = PILImage.create(img)
@@ -148,42 +22,3 @@ def predict(img):
 import gradio as gr
 gr.Interface(fn=predict, inputs=gr.inputs.Image(shape=(512, 512)), outputs=gr.outputs.Label(num_top_classes=3)).launch(share=False)
 
-#dls = data.dataloaders('/content/satellite imagery')
-#dls
-#dls.valid.show_batch(max_n=10, nrows=4)
-
-"""
-data = adatart.new(item_tfms=Resize(128, ResizeMethod.Squish))
-dls = data.dataloaders(path)
-dls.valid.show_batch(max_n=10, nrows=2)
-
-dls.valid.show_batch(max_n=16, nrows=4)
-
-art = data.new(item_tfms=Resize(128, ResizeMethod.Pad, pad_mode='zeros'))
-dls = art.dataloaders(path)
-dls.valid.show_batch(max_n=4, nrows=1)
-
-art = art.new(item_tfms=RandomResizedCrop(128, min_scale=0.3))
-dls = art.dataloaders(path)
-dls.train.show_batch(max_n=16, nrows=4, unique=True)
-
-art = art.new(item_tfms=Resize(128), batch_tfms=aug_transforms(mult=2))
-dls = art.dataloaders(path)
-dls.train.show_batch(max_n=8, nrows=2, unique=True)
-
-art = art.new(
-item_tfms=RandomResizedCrop(224, min_scale=0.5),
-batch_tfms=aug_transforms())
-dls = art.dataloaders(path)
-
-learn = vision_learner(dls, resnet18, metrics=error_rate)
-learn.fine_tune(4)
-
-interp = ClassificationInterpretation.from_learner(learn)
-interp.plot_confusion_matrix()
-
-interp.plot_top_losses(5, nrows=3)
-
-cleaner = ImageClassifierCleaner(learn)
-cleaner
-"""