clint-greene commited on
Commit
9f23c08
·
1 Parent(s): 2b2d98d

Upload folder using huggingface_hub

Browse files
.ipynb_checkpoints/app-checkpoint.py CHANGED
@@ -25,7 +25,7 @@ magic_model.decoder.compile(jit_compile=True)
25
  magic_model.text_encoder.compile(jit_compile=True)
26
 
27
  # Warm-up the model
28
- _ = magic_model.text_to_image("flying dragons", batch_size=num_images_to_gen)
29
 
30
  def generate_image_fn(prompt: str, steps: int) -> list:
31
  start_time = time.time()
@@ -43,7 +43,7 @@ def generate_image_fn(prompt: str, steps: int) -> list:
43
  return [image for image in images]
44
 
45
 
46
- description = "This Space demonstrates a fine-tuned Stable Diffusion model. You can use it for generating custom Magic the Gathering cards. To get started, either enter a prompt or pick one from the examples below. For details on the fine-tuning procedure, refer to [this repository](https://gpuopen.com/)."
47
  article = "We use mixed-precision and XLA to speed up the inference latency."
48
  gr.Interface(
49
  generate_image_fn,
@@ -53,12 +53,12 @@ gr.Interface(
53
  max_lines=1,
54
  placeholder="Jedi",
55
  ),
56
- gr.Slider(value=70, minimum=10, maximum=100, step=1),
57
  ],
58
- outputs=gr.Gallery().style(grid=[2], height="auto"),
59
  title="Generate custom magic the gathering cards",
60
  description=description,
61
  article=article,
62
- examples=[["Yoda", 70], ["Lisa Su", 70]],
63
  allow_flagging=False,
64
  ).launch(enable_queue=True)
 
25
  magic_model.text_encoder.compile(jit_compile=True)
26
 
27
  # Warm-up the model
28
+ _ = magic_model.text_to_image("flying dragons", batch_size=num_images_to_gen, num_steps=15)
29
 
30
  def generate_image_fn(prompt: str, steps: int) -> list:
31
  start_time = time.time()
 
43
  return [image for image in images]
44
 
45
 
46
+ description = "This Space demonstrates a fine-tuned Stable Diffusion model. You can use it for generating custom Magic the Gathering cards. To get started, either enter a prompt or pick one from the examples below. For details on the fine-tuning procedure, refer to [this tutorial](https://gpuopen.com/)."
47
  article = "We use mixed-precision and XLA to speed up the inference latency."
48
  gr.Interface(
49
  generate_image_fn,
 
53
  max_lines=1,
54
  placeholder="Jedi",
55
  ),
56
+ gr.Slider(value=30, minimum=10, maximum=100, step=1),
57
  ],
58
+ outputs=gr.Gallery().style(height="auto"),
59
  title="Generate custom magic the gathering cards",
60
  description=description,
61
  article=article,
62
+ examples=[["Yoda", 30], ["Lisa Su", 30]],
63
  allow_flagging=False,
64
  ).launch(enable_queue=True)
.ipynb_checkpoints/constants-checkpoint.py CHANGED
@@ -122,7 +122,7 @@ css = """
122
  .image_duplication{position: absolute; width: 100px; left: 50px}
123
  """
124
 
125
- num_images_to_gen = 3
126
  unconditional_guidance_scale = 30
127
  img_height = 512
128
  img_width = 512
 
122
  .image_duplication{position: absolute; width: 100px; left: 50px}
123
  """
124
 
125
+ num_images_to_gen = 1
126
  unconditional_guidance_scale = 30
127
  img_height = 512
128
  img_width = 512
app.py CHANGED
@@ -25,7 +25,7 @@ magic_model.decoder.compile(jit_compile=True)
25
  magic_model.text_encoder.compile(jit_compile=True)
26
 
27
  # Warm-up the model
28
- _ = magic_model.text_to_image("flying dragons", batch_size=num_images_to_gen)
29
 
30
  def generate_image_fn(prompt: str, steps: int) -> list:
31
  start_time = time.time()
@@ -43,7 +43,7 @@ def generate_image_fn(prompt: str, steps: int) -> list:
43
  return [image for image in images]
44
 
45
 
46
- description = "This Space demonstrates a fine-tuned Stable Diffusion model. You can use it for generating custom Magic the Gathering cards. To get started, either enter a prompt or pick one from the examples below. For details on the fine-tuning procedure, refer to [this repository](https://gpuopen.com/)."
47
  article = "We use mixed-precision and XLA to speed up the inference latency."
48
  gr.Interface(
49
  generate_image_fn,
@@ -53,12 +53,12 @@ gr.Interface(
53
  max_lines=1,
54
  placeholder="Jedi",
55
  ),
56
- gr.Slider(value=70, minimum=10, maximum=100, step=1),
57
  ],
58
- outputs=gr.Gallery().style(grid=[2], height="auto"),
59
  title="Generate custom magic the gathering cards",
60
  description=description,
61
  article=article,
62
- examples=[["Yoda", 70], ["Lisa Su", 70]],
63
  allow_flagging=False,
64
  ).launch(enable_queue=True)
 
25
  magic_model.text_encoder.compile(jit_compile=True)
26
 
27
  # Warm-up the model
28
+ _ = magic_model.text_to_image("flying dragons", batch_size=num_images_to_gen, num_steps=15)
29
 
30
  def generate_image_fn(prompt: str, steps: int) -> list:
31
  start_time = time.time()
 
43
  return [image for image in images]
44
 
45
 
46
+ description = "This Space demonstrates a fine-tuned Stable Diffusion model. You can use it for generating custom Magic the Gathering cards. To get started, either enter a prompt or pick one from the examples below. For details on the fine-tuning procedure, refer to [this tutorial](https://gpuopen.com/)."
47
  article = "We use mixed-precision and XLA to speed up the inference latency."
48
  gr.Interface(
49
  generate_image_fn,
 
53
  max_lines=1,
54
  placeholder="Jedi",
55
  ),
56
+ gr.Slider(value=30, minimum=10, maximum=100, step=1),
57
  ],
58
+ outputs=gr.Gallery().style(height="auto"),
59
  title="Generate custom magic the gathering cards",
60
  description=description,
61
  article=article,
62
+ examples=[["Yoda", 30], ["Lisa Su", 30]],
63
  allow_flagging=False,
64
  ).launch(enable_queue=True)
constants.py CHANGED
@@ -122,7 +122,7 @@ css = """
122
  .image_duplication{position: absolute; width: 100px; left: 50px}
123
  """
124
 
125
- num_images_to_gen = 3
126
  unconditional_guidance_scale = 30
127
  img_height = 512
128
  img_width = 512
 
122
  .image_duplication{position: absolute; width: 100px; left: 50px}
123
  """
124
 
125
+ num_images_to_gen = 1
126
  unconditional_guidance_scale = 30
127
  img_height = 512
128
  img_width = 512