![]() ![]() I still need to mark gun ports and jig line. ![]() Overall, I am extreamly satisfied with CAD drawing. Once the body plan was imported and I got up to speed, I estimate that I can knock out a full set of frame drawings in a fraction of the time it would take by hand. ![]() ![]() In addition, the CAD drawing is extreamly accurate as compared to lofting by hand. The current layer, the layer AM0, and layers used by objects or external references, cannot be deleted.' I cannot see anything from this statement that would prevent the deletion of the layer that was the subject of THIS post. This exercise has enabled me to better understand the relationship between all the parts of a ship. There is this statement in 'Help' for the Delete Layer dialog box, 'Delete> Deletes layers. And, I havent even scratched the surface of what is possible. Next is to finish these drawings, redo the sheer plan, half-breadth, framing plan, keel plan and then make a jig plan. If you are interested in learning CAD drawing, I am happy to help as I am sure Bill and Dave are as well. This is within reach of the average modeler, so long as you are patient and determined. Regarding CAD drawing, I view it as a hobby within a hobby. I probably have a solid 100 hours easily invested in drawings which means that my other projects have stalled for the time being. As for tutorials, Dave has a few nice segments regarding the GH that peaked my interest and our new member Bill Edgin recommended this program and answered questions along the way. When I get a little further along, I may consider a tutorial, but for now need to hone my skills. I will go back and redo the Essex framing plan using CAD, since I found a few errors in my early calculations.įor those considering CAD drawing, I would recommend starting with a ship where there is already a good set of plans published, such as the AOTS series. Among other things, the rising wood and stern deadwood are already done for you. But then again,with the Peacock, I get to experiment to see what was logical and what will fit. Particularly when you are modifying existing files or deleting files, you want those steps to run in the same command where they were first created. The solution is often to merge multiple commands into a single RUN command. I continue to make progress using DeltaCAD. One downside of the layers is building images that duplicate files or ship files that are deleted in a later layer. I have finished all the frames and am now onto the Sheer Plan. model.get_submodule('.'.join(parent)) = nn.At any rate, below is a screen shot showing some framing detail. Finally, we can fetch the layer reference in order to overwrite their value: > for *parent, k in maxpools: Here they all come from the same parent module model.features. We can extract the parent module name for each of those layers: > maxpools = [k.split('.') for k, m in model.named_modules() A canonical approach is to filter the layers of model.modules and only keep the max pool layers, then replace those with average pool layers: > maxpools = [k for k, m in model.named_modules() Also do note: not all layers are registered inside the nn.Module, non-parametric functions such as most activation functions can be applied via the functional approach directly in the forward of the module.įor a given nn.Module m you can extract its layer name by using type(m)._name_. model.features), however this is not always the case, for instance nn.Sequential use indices: model.features to select one of the relu activations. For example, submodule are often accessible via attributes ( e.g. This will depend on your model's implementation. I would like to assign the assembly pieces to a specific bolt, so when I turn that bolt off the assembly pieces also turn off. I have three different bolts, each bolt has four different assembly pieces that go with it. Select a submodule and interact with it as you would with any other nn.Module. there anyway to create sub-layers within one main layer, so when you turn the main layer off, the sub-layers also turn off i.e. Assuming you know the structure of your model, you can: > model = torchvision.models(pretrained=True) ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |