Spaces:
Sleeping
Sleeping
Upload 2 files
Browse files
app.py
CHANGED
@@ -75,14 +75,9 @@ def main():
|
|
75 |
|
76 |
with tabs[1]:
|
77 |
st.write('## Find Order')
|
78 |
-
st.markdown("""<p>The pretrained model has been trained with
|
79 |
-
|
80 |
-
|
81 |
-
This layer is followed by another Conv2D layer with the same filter size, activation function, and batch normalization.
|
82 |
-
MaxPooling2D is used to reduce the dimensionality of the data, with a pool size of (2,2). Dropout is then applied, with a rate of 0.25.
|
83 |
-
This process is repeated with two more sets of Conv2D, BatchNormalization, MaxPooling2D and Dropout layers, increasing the filter size to 64 and 128 for each set respectively.
|
84 |
-
Finally, the model is flattened to transform the multidimensional output into a one-dimensional array, and a Dense layer with 512 nodes is added with an activation function of ReLU.
|
85 |
-
Batch normalization and dropout are also included in this layer. Finally, a last dense layer with a softmax activation function is added, which outputs the predicted class probabilities.</p>
|
86 |
""", unsafe_allow_html=True)
|
87 |
|
88 |
with st.sidebar:
|
|
|
75 |
|
76 |
with tabs[1]:
|
77 |
st.write('## Find Order')
|
78 |
+
st.markdown("""<p>The pretrained model has been trained with 84,000 synthetic generated images. The goal is to detect the font type from a given list of 21 font types.
|
79 |
+
Each image starts with a capital letter followed by 3-10 random simple letters. images are created with random horizontal reslutions and resized to 64x64 at the end.
|
80 |
+
random amount of noise and rotation is also added.</p>
|
|
|
|
|
|
|
|
|
|
|
81 |
""", unsafe_allow_html=True)
|
82 |
|
83 |
with st.sidebar:
|
model.h5
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 16325952
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:77f193b4deb96cc63657c376dd877bcd37f1afba40996f002857c71082744b15
|
3 |
size 16325952
|