AKASH2907 / deepfakes_video_classification

Deepfakes Video classification via CNN, LSTM, C3D and triplets [IWBF'20]

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

demo error

ss880426 opened this issue · comments

I run 10.evaluate_triplets.py and got this error
ValueError: Input 0 of layer Conv2d_1a_3x3 is incompatible with the layer: : expected min_ndim=4, found ndim=3. Full shape received: (32, 160, 160)
It seems to be related to embeddings = embedder.embeddings(batches)
How to solve this?
thanks!

i use batches = batches.values.reshape(-1,160,160,3) to solved it
but got new error from
x_test = testing_embeddings.predict(x_test_after)
ValueError: Input 0 is incompatible with layer model: expected shape=(None, 512), found shape=(None, 64)