Update app.py

app.py

@@ -6,12 +6,13 @@ import jax
 import jax.numpy as jnp
 import numpy as np
 import flax.linen as nn
-from inference import PaliGemmaModel
+from inference import PaliGemmaModel, VAEModel
 
 COLORS = ['#4285f4', '#db4437', '#f4b400', '#0f9d58', '#e48ef1']
 
-# Instantiate the model
+# Instantiate the models
 pali_gemma_model = PaliGemmaModel()
+vae_model = VAEModel('vae-oid.npz')
 
 ##### Parse segmentation output tokens into masks
 ##### Also returns bounding boxes with their labels
@@ -120,118 +121,6 @@ with gr.Blocks(css="style.css") as demo:
 ### Postprocessing Utils for Segmentation Tokens
 ### Segmentation tokens are passed to another VAE which decodes them to a mask
 
-_MODEL_PATH = 'vae-oid.npz'
-
-_SEGMENT_DETECT_RE = re.compile(
-    r'(.*?)' +
-    r'<loc(\d{4})>' * 4 + r'\s*' +
-    '(?:%s)?' % (r'<seg(\d{3})>' * 16) +
-    r'\s*([^;<>]+)? ?(?:; )?',
-)
-
-def _get_params(checkpoint):
-    """Converts PyTorch checkpoint to Flax params."""
-
-    def transp(kernel):
-        return np.transpose(kernel, (2, 3, 1, 0))
-
-    def conv(name):
-        return {
-            'bias': checkpoint[name + '.bias'],
-            'kernel': transp(checkpoint[name + '.weight']),
-        }
-
-    def resblock(name):
-        return {
-            'Conv_0': conv(name + '.0'),
-            'Conv_1': conv(name + '.2'),
-            'Conv_2': conv(name + '.4'),
-        }
-
-    return {
-        '_embeddings': checkpoint['_vq_vae._embedding'],
-        'Conv_0': conv('decoder.0'),
-        'ResBlock_0': resblock('decoder.2.net'),
-        'ResBlock_1': resblock('decoder.3.net'),
-        'ConvTranspose_0': conv('decoder.4'),
-        'ConvTranspose_1': conv('decoder.6'),
-        'ConvTranspose_2': conv('decoder.8'),
-        'ConvTranspose_3': conv('decoder.10'),
-        'Conv_1': conv('decoder.12'),
-    }
-
-def _quantized_values_from_codebook_indices(codebook_indices, embeddings):
-    batch_size, num_tokens = codebook_indices.shape
-    assert num_tokens == 16, codebook_indices.shape
-    unused_num_embeddings, embedding_dim = embeddings.shape
-
-    encodings = jnp.take(embeddings, codebook_indices.reshape((-1)), axis=0)
-    encodings = encodings.reshape((batch_size, 4, 4, embedding_dim))
-    return encodings
-
-@functools.cache
-def _get_reconstruct_masks():
-    """Reconstructs masks from codebook indices.
-    Returns:
-        A function that expects indices shaped `[B, 16]` of dtype int32, each
-        ranging from 0 to 127 (inclusive), and that returns a decoded masks sized
-        `[B, 64, 64, 1]`, of dtype float32, in range [-1, 1].
-    """
-
-    class ResBlock(nn.Module):
-        features: int
-
-        @nn.compact
-        def __call__(self, x):
-            original_x = x
-            x = nn.Conv(features=self.features, kernel_size=(3, 3), padding=1)(x)
-            x = nn.relu(x)
-            x = nn.Conv(features=self.features, kernel_size=(3, 3), padding=1)(x)
-            x = nn.relu(x)
-            x = nn.Conv(features=self.features, kernel_size=(1, 1), padding=0)(x)
-            return x + original_x
-
-    class Decoder(nn.Module):
-        """Upscales quantized vectors to mask."""
-
-        @nn.compact
-        def __call__(self, x):
-            num_res_blocks = 2
-            dim = 128
-            num_upsample_layers = 4
-
-            x = nn.Conv(features=dim, kernel_size=(1, 1), padding=0)(x)
-            x = nn.relu(x)
-
-            for _ in range(num_res_blocks):
-                x = ResBlock(features=dim)(x)
-
-            for _ in range(num_upsample_layers):
-                x = nn.ConvTranspose(
-                    features=dim,
-                    kernel_size=(4, 4),
-                    strides=(2, 2),
-                    padding=2,
-                    transpose_kernel=True,
-                )(x)
-                x = nn.relu(x)
-                dim //= 2
-
-            x = nn.Conv(features=1, kernel_size=(1, 1), padding=0)(x)
-
-            return x
-
-    def reconstruct_masks(codebook_indices):
-        quantized = _quantized_values_from_codebook_indices(
-            codebook_indices, params['_embeddings']
-        )
-        return Decoder().apply({'params': params}, quantized)
-
-    with open(_MODEL_PATH, 'rb') as f:
-        params = _get_params(dict(np.load(f)))
-
-    return jax.jit(reconstruct_masks, backend='cpu')
-
 def extract_objs(text, width, height, unique_labels=False):
     """Returns objs for a string with "<loc>" and "<seg>" tokens."""
     objs = []
@@ -252,7 +141,7 @@ def extract_objs(text, width, height, unique_labels=False):
             mask = None
         else:
             seg_indices = np.array([int(x) for x in seg_indices], dtype=np.int32)
-            m64, = _get_reconstruct_masks()(seg_indices[None])[..., 0]
+            m64, = vae_model.reconstruct_masks(seg_indices[None])[..., 0]
             m64 = np.clip(np.array(m64) * 0.5 + 0.5, 0, 1)
             m64 = PIL.Image.fromarray((m64 * 255).astype('uint8'))
             mask = np.zeros([height, width])
@@ -275,7 +164,12 @@ def extract_objs(text, width, height, unique_labels=False):
 
     return objs
 
-#########
+_SEGMENT_DETECT_RE = re.compile(
+    r'(.*?)' +
+    r'<loc(\d{4})>' * 4 + r'\s*' +
+    '(?:%s)?' % (r'<seg(\d{3})>' * 16) +
+    r'\s*([^;<>]+)? ?(?:; )?',
+)
 
 if __name__ == "__main__":
     demo.queue(max_size=10).launch(debug=True)