Upload 3 files
Browse files- .gitattributes +1 -0
- Pokemon_transfer_learning.keras +3 -0
- app.ipynb +125 -0
- requirements.txt +1 -0
.gitattributes
CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
Pokemon_transfer_learning.keras filter=lfs diff=lfs merge=lfs -text
|
Pokemon_transfer_learning.keras
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:afe180bd42b6f226d6a8554dcc9231f01e4d0fa85604f534581b321b6e9b86af
|
3 |
+
size 250560250
|
app.ipynb
ADDED
@@ -0,0 +1,125 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cells": [
|
3 |
+
{
|
4 |
+
"cell_type": "code",
|
5 |
+
"execution_count": 7,
|
6 |
+
"metadata": {},
|
7 |
+
"outputs": [],
|
8 |
+
"source": [
|
9 |
+
"import gradio as gr\n",
|
10 |
+
"import tensorflow as tf\n",
|
11 |
+
"import numpy as np\n",
|
12 |
+
"from PIL import Image"
|
13 |
+
]
|
14 |
+
},
|
15 |
+
{
|
16 |
+
"cell_type": "code",
|
17 |
+
"execution_count": 8,
|
18 |
+
"metadata": {},
|
19 |
+
"outputs": [],
|
20 |
+
"source": [
|
21 |
+
"model_path = \"Pokemon_transfer_learning.keras\"\n",
|
22 |
+
"model = tf.keras.models.load_model(model_path)"
|
23 |
+
]
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"cell_type": "code",
|
27 |
+
"execution_count": 9,
|
28 |
+
"metadata": {},
|
29 |
+
"outputs": [],
|
30 |
+
"source": [
|
31 |
+
"# Define the core prediction function\n",
|
32 |
+
"def predict_pokemon(image):\n",
|
33 |
+
" # Preprocess image\n",
|
34 |
+
" print(type(image))\n",
|
35 |
+
" image = Image.fromarray(image.astype('uint8')) # Convert numpy array to PIL image\n",
|
36 |
+
" image = image.resize((150, 150)) #resize the image to 28x28 and converts it to gray scale\n",
|
37 |
+
" image = np.array(image)\n",
|
38 |
+
" image = np.expand_dims(image, axis=0) # same as image[None, ...]\n",
|
39 |
+
" \n",
|
40 |
+
" # Predict\n",
|
41 |
+
" prediction = model.predict(image)\n",
|
42 |
+
" \n",
|
43 |
+
" # No need to apply sigmoid, as the output layer already uses softmax\n",
|
44 |
+
" # Convert the probabilities to rounded values\n",
|
45 |
+
" prediction = np.round(prediction, 2)\n",
|
46 |
+
" \n",
|
47 |
+
" # Separate the probabilities for each class\n",
|
48 |
+
" p_pikachu = prediction[0][0] # Probability for class 'articuno'\n",
|
49 |
+
" p_rattata = prediction[0][1] # Probability for class 'moltres'\n",
|
50 |
+
" p_zubat = prediction[0][2] # Probability for class 'zapdos'\n",
|
51 |
+
" \n",
|
52 |
+
" return {'pikachu': p_pikachu, 'rattata': p_rattata, 'zubat': p_zubat}"
|
53 |
+
]
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"cell_type": "code",
|
57 |
+
"execution_count": 13,
|
58 |
+
"metadata": {},
|
59 |
+
"outputs": [
|
60 |
+
{
|
61 |
+
"name": "stdout",
|
62 |
+
"output_type": "stream",
|
63 |
+
"text": [
|
64 |
+
"Running on local URL: http://127.0.0.1:7861\n",
|
65 |
+
"\n",
|
66 |
+
"To create a public link, set `share=True` in `launch()`.\n"
|
67 |
+
]
|
68 |
+
},
|
69 |
+
{
|
70 |
+
"data": {
|
71 |
+
"text/html": [
|
72 |
+
"<div><iframe src=\"http://127.0.0.1:7861/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
|
73 |
+
],
|
74 |
+
"text/plain": [
|
75 |
+
"<IPython.core.display.HTML object>"
|
76 |
+
]
|
77 |
+
},
|
78 |
+
"metadata": {},
|
79 |
+
"output_type": "display_data"
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"data": {
|
83 |
+
"text/plain": []
|
84 |
+
},
|
85 |
+
"execution_count": 13,
|
86 |
+
"metadata": {},
|
87 |
+
"output_type": "execute_result"
|
88 |
+
}
|
89 |
+
],
|
90 |
+
"source": [
|
91 |
+
"# Create the Gradio interface\n",
|
92 |
+
"input_image = gr.Image()\n",
|
93 |
+
"iface = gr.Interface(\n",
|
94 |
+
" fn=predict_pokemon,\n",
|
95 |
+
" inputs=input_image,\n",
|
96 |
+
" outputs=gr.Label(),\n",
|
97 |
+
" examples=[\"images/pikachu1.jpg\", \"images/pikachu2.jpg\", \"images/pikachu3.jpg\", \"images/rattata1.jpg\", \"images/rattata2.jpg\", \"images/rattata3.jpg\", \"images/zubat1.jpg\", \"images/zubat2.jpg\", \"images/zubat3.jpg\"], \n",
|
98 |
+
" description=\"TEST.\")\n",
|
99 |
+
"\n",
|
100 |
+
"iface.launch()\n"
|
101 |
+
]
|
102 |
+
}
|
103 |
+
],
|
104 |
+
"metadata": {
|
105 |
+
"kernelspec": {
|
106 |
+
"display_name": "venv_new",
|
107 |
+
"language": "python",
|
108 |
+
"name": "python3"
|
109 |
+
},
|
110 |
+
"language_info": {
|
111 |
+
"codemirror_mode": {
|
112 |
+
"name": "ipython",
|
113 |
+
"version": 3
|
114 |
+
},
|
115 |
+
"file_extension": ".py",
|
116 |
+
"mimetype": "text/x-python",
|
117 |
+
"name": "python",
|
118 |
+
"nbconvert_exporter": "python",
|
119 |
+
"pygments_lexer": "ipython3",
|
120 |
+
"version": "3.11.8"
|
121 |
+
}
|
122 |
+
},
|
123 |
+
"nbformat": 4,
|
124 |
+
"nbformat_minor": 2
|
125 |
+
}
|
requirements.txt
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
tensorflow
|