adding backend folder

backend/config.json

@@ -0,0 +1,41 @@
+{
+  "id2label": {
+    "0": "Abyssinian",
+    "1": "Bengal",
+    "2": "Birman",
+    "3": "Bombay",
+    "4": "British_Shorthair",
+    "5": "Egyptian_Mau",
+    "6": "Maine_Coon",
+    "7": "Persian",
+    "8": "Ragdoll",
+    "9": "Russian_Blue",
+    "10": "Siamese",
+    "11": "Sphynx",
+    "12": "american_bulldog",
+    "13": "american_pit_bull_terrier",
+    "14": "basset_hound",
+    "15": "beagle",
+    "16": "boxer",
+    "17": "chihuahua",
+    "18": "english_cocker_spaniel",
+    "19": "english_setter",
+    "20": "german_shorthaired",
+    "21": "great_pyrenees",
+    "22": "havanese",
+    "23": "japanese_chin",
+    "24": "keeshond",
+    "25": "leonberger",
+    "26": "miniature_pinscher",
+    "27": "newfoundland",
+    "28": "pomeranian",
+    "29": "pug",
+    "30": "saint_bernard",
+    "31": "samoyed",
+    "32": "scottish_terrier",
+    "33": "shiba_inu",
+    "34": "staffordshire_bull_terrier",
+    "35": "wheaten_terrier",
+    "36": "yorkshire_terrier"
+  }
+}

backend/pipeline.py

@@ -0,0 +1,73 @@
+from typing import Dict, List, Any
+from PIL import Image
+import os
+import json
+import numpy as np
+from fastai.learner import load_learner
+
+
+class PreTrainedPipeline:
+    def __init__(self, path=""):
+        # IMPLEMENT_THIS
+        # Preload all the elements you are going to need at inference.
+        # For instance your model, processors, tokenizer that might be needed.
+        # This function is only called once, so do all the heavy processing I/O here"""
+        self.model = load_learner(os.path.join(path, "export.pkl"))
+        with open(os.path.join(path, "config.json")) as config:
+            config = json.load(config)
+        self.id2label = config["id2label"]
+
+    def __call__(self, inputs: "Image.Image") -> List[Dict[str, Any]]:
+        """
+        Args:
+            inputs (:obj:`PIL.Image`):
+                The raw image representation as PIL.
+                No transformation made whatsoever from the input. Make all necessary transformations here.
+        Return:
+            A :obj:`list`:. The list contains items that are dicts should be liked {"label": "XXX", "score": 0.82}
+                It is preferred if the returned list is in decreasing `score` order
+        """
+        # IMPLEMENT_THIS
+        # FastAI expects a np array, not a PIL Image.
+        _, _, preds = self.model.predict(np.array(inputs))
+        preds = preds.tolist()
+        labels = [
+            {"label": str(self.id2label["0"]), "score": preds[0]},
+            {"label": str(self.id2label["1"]), "score": preds[1]},
+            {"label": str(self.id2label["2"]), "score": preds[2]},
+            {"label": str(self.id2label["3"]), "score": preds[3]},
+            {"label": str(self.id2label["4"]), "score": preds[4]},
+            {"label": str(self.id2label["5"]), "score": preds[5]},
+            {"label": str(self.id2label["6"]), "score": preds[6]},
+            {"label": str(self.id2label["7"]), "score": preds[7]},
+            {"label": str(self.id2label["8"]), "score": preds[8]},
+            {"label": str(self.id2label["9"]), "score": preds[9]},
+            {"label": str(self.id2label["10"]), "score": preds[10]},
+            {"label": str(self.id2label["11"]), "score": preds[11]},
+            {"label": str(self.id2label["12"]), "score": preds[12]},
+            {"label": str(self.id2label["13"]), "score": preds[13]},
+            {"label": str(self.id2label["14"]), "score": preds[14]},
+            {"label": str(self.id2label["15"]), "score": preds[15]},
+            {"label": str(self.id2label["16"]), "score": preds[16]},
+            {"label": str(self.id2label["17"]), "score": preds[17]},
+            {"label": str(self.id2label["18"]), "score": preds[18]},
+            {"label": str(self.id2label["19"]), "score": preds[19]},
+            {"label": str(self.id2label["20"]), "score": preds[20]},
+            {"label": str(self.id2label["21"]), "score": preds[21]},
+            {"label": str(self.id2label["22"]), "score": preds[22]},
+            {"label": str(self.id2label["23"]), "score": preds[23]},
+            {"label": str(self.id2label["24"]), "score": preds[24]},
+            {"label": str(self.id2label["25"]), "score": preds[25]},
+            {"label": str(self.id2label["26"]), "score": preds[26]},
+            {"label": str(self.id2label["27"]), "score": preds[27]},
+            {"label": str(self.id2label["28"]), "score": preds[28]},
+            {"label": str(self.id2label["29"]), "score": preds[29]},
+            {"label": str(self.id2label["30"]), "score": preds[30]},
+            {"label": str(self.id2label["31"]), "score": preds[31]},
+            {"label": str(self.id2label["32"]), "score": preds[32]},
+            {"label": str(self.id2label["33"]), "score": preds[33]},
+            {"label": str(self.id2label["34"]), "score": preds[34]},
+            {"label": str(self.id2label["35"]), "score": preds[35]},
+            {"label": str(self.id2label["36"]), "score": preds[36]},
+        ]
+        return labels

backend/util.py

@@ -0,0 +1,40 @@
+import streamlit as st
+from PIL import Image
+from backend.pipeline import PreTrainedPipeline
+import pandas as pd
+import io
+import matplotlib.pyplot as plt
+import numpy as np
+
+
+def import_fig():
+    image = st.file_uploader("Upload your picture.", type=["png", "jpg", "jpeg"])
+    if image:
+        bytes_image = image.getvalue()
+        image = Image.open(io.BytesIO(bytes_image))
+        st.image(image, caption=["We are classifying this image..."])
+    return image
+
+
+def plot(data=None):
+
+    fig = plt.figure()
+    ax = fig.add_axes([0, 0, 1, 1])
+    breeds = data.head(3)["label"].tolist()
+    labels = data.head(3)["score"].tolist()
+    ax.bar(breeds, labels)
+    ax.set_ylabel("Probability that your pet is breed X")
+    ax.grid("on")
+
+    st.pyplot(fig)
+
+
+@st.cache(allow_output_mutation=True)
+def fastai_model(image):
+    if image:
+        model = PreTrainedPipeline(path="backend")
+        outputs = model(image)
+
+        outputs_df = pd.DataFrame(outputs)
+
+        return outputs_df.sort_values(by=["score"], ascending=False)