Add SAM decoder & output masks as png (#418)

* Add loading of decoder layers in Model

* Multiply by hypernet_layer_cnt for ctx_size on model load

* Add decoder layers to py conversion script

* Fix wrong and reversed tensor sizes for decoder

* Add decoder transformer implementation

* Add decoder hypernet and iou prediction mlps

* Add transpose convolution operation and unit test

* Finish mask decoder and write the decoder output in the model state

* Output masks to png after removing padding and upsampling to original size

- Also filter based on the iou treshold
- Additionally filtering based on the stability score and crop boxes
should be done

* Add stb image write in order to output masks from SAM

* Add transpose convolution 2d name and symbol to ggml ops static arrays

* Comment out debug print in transpose convolution test to fix compilation

ggml-ci

examples/sam/convert-pth-to-ggml.py

@@ -71,11 +71,6 @@
     name = k
     shape = v.shape
 
-    # TODO: export only the Encoder -- after it works we will export the other stuff
-    if name[:13] != "image_encoder" and \
-       name[:14] != "prompt_encoder":
-        continue
-
     if name[:19] == "prompt_encoder.mask":
         continue
 
@@ -104,7 +99,9 @@
     ftype_cur = 1
     if ftype == 0 or n_dims == 1 or \
             name == "image_encoder.pos_embed" or \
-            name.startswith("prompt_encoder"):
+            name.startswith("prompt_encoder") or \
+            name.startswith("mask_decoder.iou_token") or \
+            name.startswith("mask_decoder.mask_tokens"):
         print("  Converting to float32")
         data = data.astype(np.float32)
         ftype_cur = 0