committing the pet classifier

app.py

@@ -1,7 +1,26 @@
-import gradio as gr
+# AUTOGENERATED! DO NOT EDIT! File to edit: drive/MyDrive/Colab Notebooks/fastbook/fastcourse/ch2minimal/tmabrahamgradio/gradiotest/gradiotest.ipynb.
+
+# %% auto 0
+__all__ = ['learn', 'labels', 'predict']
+
+# %% drive/MyDrive/Colab Notebooks/fastbook/fastcourse/ch2minimal/tmabrahamgradio/gradiotest/gradiotest.ipynb 1
+from fastai.vision.all import *
+import skimage
+
+# %% drive/MyDrive/Colab Notebooks/fastbook/fastcourse/ch2minimal/tmabrahamgradio/gradiotest/gradiotest.ipynb 4
+learn = load_learner('export.pkl')
+
+labels = learn.dls.vocab
 
-def greet(name):
-    return "Hello " + name + "!!"
+def predict(img):
+  img = PILImage.create(img)
+  pred, pred_idx, probs = learn.predict(img)
+  return {labels[i]: float(probs[i]) for i in range(len(labels))}
+
+
+
+
+# %% drive/MyDrive/Colab Notebooks/fastbook/fastcourse/ch2minimal/tmabrahamgradio/gradiotest/gradiotest.ipynb 7
+import gradio as gr
+gr.Interface(fn=predict, inputs=gr.Image(shape=(512,512)), outputs=gr.outputs.Label(num_top_classes=3), title="Pet Breed Classifier", description="A pet breed classifier trained on the Oxford Pets dataset using the fastai library (5 epochs) as a proof of concept for Gradio.", article="<p style='text-align: center'><a href='https://gameologist.com/portfolio' target='_blank'>see more of my things here</a></p>", examples=['abys.jpg', 'download (3).jpg'], enable_queue=True).launch(share=True)
 
-iface = gr.Interface(fn=greet, inputs="text", outputs="text")
-iface.launch()

export.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37b1f46c87b7f1257d475ecc7205bdf8f283dc55b1be48192e91c11e3babd7f6
+size 103072981

requirements.txt

@@ -0,0 +1,2 @@
+fastai
+scikit-image