SqueezeNet num_classes=1 no WebGL

[desrius]

When I try to convert my SqueezeNet Keras Model with one output Class I get this assertion error for webgl:

[convert_keras.py] Failed generating descriptor for webgl backend
Traceback (most recent call last):
  File "bin/convert_keras.py", line 97, in main
    graph_exec_data = generate_descriptor(backend, graph, constant_encoder_name=args.encoding)
  File "/home/*/anaconda3/lib/python3.6/site-packages/webdnn-1.2.6-py3.6.egg/webdnn/backend/interface/generator.py", line 107, in generate_descriptor
    return generator(graph, **kwargs)
  File "/home/*/anaconda3/lib/python3.6/site-packages/webdnn-1.2.6-py3.6.egg/webdnn/backend/webgl/generator.py", line 92, in generate
    return WebGLDescriptorGenerator.generate(graph, **kwargs)
  File "/home/*/anaconda3/lib/python3.6/site-packages/webdnn-1.2.6-py3.6.egg/webdnn/backend/webgl/generator.py", line 61, in generate
    kernels = cls.generate_kernels(graph)
  File "/home/*/anaconda3/lib/python3.6/site-packages/webdnn-1.2.6-py3.6.egg/webdnn/backend/webgl/generator.py", line 86, in generate_kernels
    kernels += cls._handler_map[cls.__name__][key](op)
  File "/home/*/anaconda3/lib/python3.6/site-packages/webdnn-1.2.6-py3.6.egg/webdnn/backend/webgl/kernels/reshape.py", line 77, in reshape
    convert_position("gl_FragCoord.yx", texture_shape(y)[:2], texture_stride(y)[:2], shapes_y_dy[y], strides_y_dy[y]),
  File "/home/*/anaconda3/lib/python3.6/site-packages/webdnn-1.2.6-py3.6.egg/webdnn/backend/webgl/kernels/util.py", line 53, in convert_position
    ivec(out_stride), ", ",
  File "/home/*/anaconda3/lib/python3.6/site-packages/webdnn-1.2.6-py3.6.egg/webdnn/backend/webgl/kernels/util.py", line 100, in ivec
    assert 2 <= len(sequence) <= 4
AssertionError

When I increase the number of classes to 2, it works. I use pytorch2keras to convert the squeezenet model from pytorch to keras.