Spaces:
Runtime error
Runtime error
Upload 4 files
Browse files- app.py +8 -23
- requirements.txt +11 -4
app.py
CHANGED
@@ -1,4 +1,4 @@
|
|
1 |
-
import
|
2 |
from PIL import Image
|
3 |
import cv2 as cv
|
4 |
|
@@ -303,10 +303,8 @@ def visualize(model, data, dims):
|
|
303 |
# t_img = transforms.Resize((dims[0], dims[1]))(t_img)
|
304 |
img = Image.fromarray(np.uint8(fake_imgs[i]))
|
305 |
img = cv.resize(fake_imgs[i], dsize=(dims[1], dims[0]), interpolation=cv.INTER_CUBIC)
|
306 |
-
return img
|
307 |
# st.text(f"Size of fake image {fake_imgs[i].shape} \n Type of image = {type(fake_imgs[i])}")
|
308 |
-
|
309 |
-
|
310 |
|
311 |
def log_results(loss_meter_dict):
|
312 |
for loss_name, loss_meter in loss_meter_dict.items():
|
@@ -354,27 +352,14 @@ def make_dataloaders2(batch_size=16, n_workers=4, pin_memory=True, **kwargs): #
|
|
354 |
pin_memory=pin_memory)
|
355 |
return dataloader
|
356 |
|
357 |
-
|
358 |
-
|
359 |
-
|
360 |
-
|
361 |
a = im.shape
|
362 |
-
|
363 |
test_dl = make_dataloaders2(img_list=[im])
|
364 |
for data in test_dl:
|
365 |
model.setup_input(data)
|
366 |
model.optimize()
|
367 |
-
|
368 |
-
return (size_text,img)
|
369 |
-
|
370 |
-
title = "PicSum"
|
371 |
-
description = "Gradio demo for PicSum project. You can give an image as input on the left side and then click on the submit button. The generated text, summary, important sentences and fill in the gaps would be generated on the right side."
|
372 |
-
gr.Interface(
|
373 |
-
extract,
|
374 |
-
[gr.inputs.Image(type="filepath", label="Input"),gr.inputs.CheckboxGroup(choices, type="value", default=['Generate text'], label='Options') ],
|
375 |
-
[gr.outputs.Textbox(label="Generated Text"),"image"],
|
376 |
-
title=title,
|
377 |
-
description=description,
|
378 |
-
# examples=[['a.png', ['Generate text']], ['b.png', ['Generate text','Summary','Important Sentences']], ]
|
379 |
-
).launch(enable_queue=True)
|
380 |
-
|
|
|
1 |
+
import streamlit as st
|
2 |
from PIL import Image
|
3 |
import cv2 as cv
|
4 |
|
|
|
303 |
# t_img = transforms.Resize((dims[0], dims[1]))(t_img)
|
304 |
img = Image.fromarray(np.uint8(fake_imgs[i]))
|
305 |
img = cv.resize(fake_imgs[i], dsize=(dims[1], dims[0]), interpolation=cv.INTER_CUBIC)
|
|
|
306 |
# st.text(f"Size of fake image {fake_imgs[i].shape} \n Type of image = {type(fake_imgs[i])}")
|
307 |
+
st.image(img, caption="Output image", use_column_width='auto', clamp=True)
|
|
|
308 |
|
309 |
def log_results(loss_meter_dict):
|
310 |
for loss_name, loss_meter in loss_meter_dict.items():
|
|
|
352 |
pin_memory=pin_memory)
|
353 |
return dataloader
|
354 |
|
355 |
+
file_up = st.file_uploader("Upload an jpg image", type="jpg")
|
356 |
+
if file_up is not None:
|
357 |
+
im = Image.open(file_up)
|
358 |
+
st.text(body=f"Size of uploaded image {im.shape}")
|
359 |
a = im.shape
|
360 |
+
st.image(im, caption="Uploaded Image.", use_column_width='auto')
|
361 |
test_dl = make_dataloaders2(img_list=[im])
|
362 |
for data in test_dl:
|
363 |
model.setup_input(data)
|
364 |
model.optimize()
|
365 |
+
visualize(model, data, a)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
requirements.txt
CHANGED
@@ -1,15 +1,22 @@
|
|
1 |
-
|
|
|
2 |
glob2
|
3 |
numpy
|
4 |
pathlib
|
5 |
tqdm
|
6 |
-
matplotlib
|
7 |
-
matplotlib-venn
|
8 |
scikit-image
|
9 |
-
torch
|
10 |
torchvision
|
11 |
torchsummary
|
12 |
fastai==2.4
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
opencv-contrib-python==4.6.0.66
|
14 |
opencv-python==4.6.0.66
|
15 |
opencv-python-headless==4.6.0.66
|
|
|
1 |
+
streamlit
|
2 |
+
Pillow
|
3 |
glob2
|
4 |
numpy
|
5 |
pathlib
|
6 |
tqdm
|
7 |
+
matplotlib==3.2.2
|
8 |
+
matplotlib-venn==0.11.7
|
9 |
scikit-image
|
10 |
+
torch
|
11 |
torchvision
|
12 |
torchsummary
|
13 |
fastai==2.4
|
14 |
+
fastcore==1.3.29
|
15 |
+
fastdownload==0.0.7
|
16 |
+
fastdtw==0.3.4
|
17 |
+
fastjsonschema==2.16.2
|
18 |
+
fastprogress==1.0.3
|
19 |
+
fastrlock==0.8.1
|
20 |
opencv-contrib-python==4.6.0.66
|
21 |
opencv-python==4.6.0.66
|
22 |
opencv-python-headless==4.6.0.66
|