ValueError: The shape of the target variable and the shape of the target value in `variable.assign(value)` must match

42 views Asked by At
ValueError: The shape of the target variable and the shape of the target value in `variable.assign(value)` must match. variable.shape=(1, 1, 3, 3), Received: value.shape=(1, 1, 32, 3). Target variable: <KerasVariable shape=(1, 1, 3, 3), dtype=float32, path=conv2d_13/kernel>

Model: "functional_52"
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓
┃ Layer (type)                  ┃ Output Shape              ┃         Param # ┃ Connected to               ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩
│ input_layer (InputLayer)      │ (None, 32, 32, 3)         │               0 │ -                          │
├───────────────────────────────┼───────────────────────────┼─────────────────┼────────────────────────────┤
│ conv2d (Conv2D)               │ (None, 30, 30, 32)        │             896 │ input_layer[0][0]          │
├───────────────────────────────┼───────────────────────────┼─────────────────┼────────────────────────────┤
│ flatten (Flatten)             │ (None, 28800)             │               0 │ conv2d[0][0]               │
├───────────────────────────────┼───────────────────────────┼─────────────────┼────────────────────────────┤
│ dense (Dense)                 │ (None, 32)                │         921,632 │ flatten[0][0]              │
├───────────────────────────────┼───────────────────────────┼─────────────────┼────────────────────────────┤
│ dense_1 (Dense)               │ (None, 6)                 │             198 │ dense[0][0]                │
├───────────────────────────────┼───────────────────────────┼─────────────────┼────────────────────────────┤
│ spatial_transformer           │ (None, 32, 32, 3)         │               0 │ conv2d[0][0],              │
│ (SpatialTransformer)          │                           │                 │ dense_1[0][0]              │
├───────────────────────────────┼───────────────────────────┼─────────────────┼────────────────────────────┤
│ conv2d_1 (Conv2D)             │ (None, 32, 32, 3)         │              12 │ spatial_transformer[0][0]  │
├───────────────────────────────┼───────────────────────────┼─────────────────┼────────────────────────────┤
│ sequential (Sequential)       │ (None, 3)                 │       8,963,395 │ conv2d_1[0][0]             │
├───────────────────────────────┼───────────────────────────┼─────────────────┼────────────────────────────┤
│ dense_5 (Dense)               │ (None, 3)                 │              12 │ sequential[0][0]           │
└───────────────────────────────┴───────────────────────────┴─────────────────┴────────────────────────────┘
 Total params: 9,886,145 (37.71 MB)
 Trainable params: 9,878,721 (37.68 MB)
 Non-trainable params: 7,424 (29.00 KB)

I'm not able to load the model. I don't understand which layer is causing the trouble. I tried debugging not understanding what is causing the trouble.

0

There are 0 answers