Spaces:
Running
on
Zero
Running
on
Zero
wjbmattingly
commited on
Commit
•
3fc0241
1
Parent(s):
546d56f
Update app.py
Browse files
app.py
CHANGED
@@ -7,6 +7,7 @@ from PIL import Image, ImageDraw
|
|
7 |
import os
|
8 |
import tempfile
|
9 |
import numpy as np
|
|
|
10 |
# Dictionary of model names and their corresponding HuggingFace model IDs
|
11 |
MODEL_OPTIONS = {
|
12 |
"Microsoft Handwritten": "microsoft/trocr-base-handwritten",
|
@@ -36,12 +37,12 @@ def load_model(model_name):
|
|
36 |
current_model = VisionEncoderDecoderModel.from_pretrained(model_id)
|
37 |
current_model_name = model_name
|
38 |
|
39 |
-
# Move model to GPU
|
40 |
-
|
|
|
41 |
|
42 |
return current_processor, current_model
|
43 |
|
44 |
-
|
45 |
def process_image(image, model_name):
|
46 |
# Save the uploaded image to a temporary file
|
47 |
with tempfile.NamedTemporaryFile(suffix=".jpg", delete=False) as temp_img:
|
@@ -59,6 +60,9 @@ def process_image(image, model_name):
|
|
59 |
|
60 |
processor, model = load_model(model_name)
|
61 |
|
|
|
|
|
|
|
62 |
# Process each line
|
63 |
transcriptions = []
|
64 |
for line in lines_data['lines']:
|
@@ -79,7 +83,7 @@ def process_image(image, model_name):
|
|
79 |
|
80 |
# Prepare image for TrOCR
|
81 |
pixel_values = processor(images=line_image_np, return_tensors="pt").pixel_values
|
82 |
-
pixel_values = pixel_values.to(
|
83 |
|
84 |
# Generate (no beam search)
|
85 |
with torch.no_grad():
|
@@ -117,4 +121,4 @@ with gr.Blocks() as iface:
|
|
117 |
submit_button = gr.Button("Transcribe")
|
118 |
submit_button.click(fn=process_image, inputs=[input_image, model_dropdown], outputs=[output_image, transcription_output])
|
119 |
|
120 |
-
iface.launch()
|
|
|
7 |
import os
|
8 |
import tempfile
|
9 |
import numpy as np
|
10 |
+
|
11 |
# Dictionary of model names and their corresponding HuggingFace model IDs
|
12 |
MODEL_OPTIONS = {
|
13 |
"Microsoft Handwritten": "microsoft/trocr-base-handwritten",
|
|
|
37 |
current_model = VisionEncoderDecoderModel.from_pretrained(model_id)
|
38 |
current_model_name = model_name
|
39 |
|
40 |
+
# Move model to GPU if available, else use CPU
|
41 |
+
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
|
42 |
+
current_model = current_model.to(device)
|
43 |
|
44 |
return current_processor, current_model
|
45 |
|
|
|
46 |
def process_image(image, model_name):
|
47 |
# Save the uploaded image to a temporary file
|
48 |
with tempfile.NamedTemporaryFile(suffix=".jpg", delete=False) as temp_img:
|
|
|
60 |
|
61 |
processor, model = load_model(model_name)
|
62 |
|
63 |
+
# Determine device
|
64 |
+
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
|
65 |
+
|
66 |
# Process each line
|
67 |
transcriptions = []
|
68 |
for line in lines_data['lines']:
|
|
|
83 |
|
84 |
# Prepare image for TrOCR
|
85 |
pixel_values = processor(images=line_image_np, return_tensors="pt").pixel_values
|
86 |
+
pixel_values = pixel_values.to(device)
|
87 |
|
88 |
# Generate (no beam search)
|
89 |
with torch.no_grad():
|
|
|
121 |
submit_button = gr.Button("Transcribe")
|
122 |
submit_button.click(fn=process_image, inputs=[input_image, model_dropdown], outputs=[output_image, transcription_output])
|
123 |
|
124 |
+
iface.launch(share=True)
|