[DO NOT MERGE][Reproducing issue] : View converter , bs >1

CLA.md

@@ -0,0 +1,58 @@
+## Individual Contributor License Agreement (CLA)
+
+**Thank you for submitting your contributions to this project.**
+
+By signing this CLA, you agree that the following terms apply to all of your past, present and future contributions
+to the project.
+
+### License.
+
+You hereby represent that all present, past and future contributions are governed by the
+[MIT License](https://opensource.org/licenses/MIT)
+copyright statement.
+
+This entails that to the extent possible under law, you transfer all copyright and related or neighboring rights
+of the code or documents you contribute to the project itself or its maintainers.
+Furthermore you also represent that you have the authority to perform the above waiver
+with respect to the entirety of you contributions.
+
+### Moral Rights.
+
+To the fullest extent permitted under applicable law, you hereby waive, and agree not to
+assert, all of your “moral rights” in or relating to your contributions for the benefit of the project.
+
+### Third Party Content.
+
+If your Contribution includes or is based on any source code, object code, bug fixes, configuration changes, tools,
+specifications, documentation, data, materials, feedback, information or other works of authorship that were not
+authored by you (“Third Party Content”) or if you are aware of any third party intellectual property or proprietary
+rights associated with your Contribution (“Third Party Rights”),
+then you agree to include with the submission of your Contribution full details respecting such Third Party
+Content and Third Party Rights, including, without limitation, identification of which aspects of your
+Contribution contain Third Party Content or are associated with Third Party Rights, the owner/author of the
+Third Party Content and Third Party Rights, where you obtained the Third Party Content, and any applicable
+third party license terms or restrictions respecting the Third Party Content and Third Party Rights. For greater
+certainty, the foregoing obligations respecting the identification of Third Party Content and Third Party Rights
+do not apply to any portion of a Project that is incorporated into your Contribution to that same Project.
+
+### Representations.
+
+You represent that, other than the Third Party Content and Third Party Rights identified by
+you in accordance with this Agreement, you are the sole author of your Contributions and are legally entitled
+to grant the foregoing licenses and waivers in respect of your Contributions. If your Contributions were
+created in the course of your employment with your past or present employer(s), you represent that such
+employer(s) has authorized you to make your Contributions on behalf of such employer(s) or such employer
+(s) has waived all of their right, title or interest in or to your Contributions.
+
+### Disclaimer.
+
+To the fullest extent permitted under applicable law, your Contributions are provided on an "as is"
+basis, without any warranties or conditions, express or implied, including, without limitation, any implied
+warranties or conditions of non-infringement, merchantability or fitness for a particular purpose. You are not
+required to provide support for your Contributions, except to the extent you desire to provide support.
+
+### No Obligation.
+
+You acknowledge that the maintainers of this project are under no obligation to use or incorporate your contributions
+into the project. The decision to use or incorporate your contributions into the project will be made at the
+sole discretion of the maintainers or their authorized delegates.

CONTRIBUTORS.md

@@ -3,6 +3,7 @@
 Below is a list of developers who have contributed to torch2trt.  This is also used to track contributors
 who have agreed to torch2trt's Contributor License Agreement.
 
+- [John Welsh](https://github.com/jaybdub) (CLA)
 - John Welsh
 
 ## Becoming a Contributor
@@ -42,6 +43,6 @@ In some instances, you may be requested to sign torch2trt's Contributor License
 4. Make a signed commit with the following text
 
    ```md
-   git commit -S -m "I have read and agree to the Contributor License Agreement as written in the file CLA.pdf of this project.  Signed, <Full Name>"
+   git commit -S -m "I have read and agree to the Contributor License Agreement as written in the file CLA.md of this project.  Signed, <Full Name>"
    ```
 

torch2trt/converters/view.py

@@ -12,10 +12,14 @@
 @tensorrt_converter('torch.unsqueeze')
 def convert_view(ctx):
     input = ctx.method_args[0]
+    print("input_shape",input.shape)
     input_trt = add_missing_trt_tensors(ctx.network, [input])[0]
+    print("input_trt",input_trt.shape)
     output = ctx.method_return
+    print("o/p",output.shape)
     layer = ctx.network.add_shuffle(input_trt)
     layer.reshape_dims = tuple(output.shape[1:])
+    print("trt layer o/p shape",layer.get_output(0).shape)
     output._trt = layer.get_output(0)
 
 
@@ -45,13 +49,13 @@ def forward(self, x):
         return x.unsqueeze(dim=self.dim)        
 
 
-@add_module_test(torch.float32, torch.device('cuda'), [(1, 3)])
-@add_module_test(torch.float32, torch.device('cuda'), [(1, 3, 3)])
-@add_module_test(torch.float32, torch.device('cuda'), [(1, 3, 3, 3)])
+#@add_module_test(torch.float32, torch.device('cuda'), [(1, 3)])
+#@add_module_test(torch.float32, torch.device('cuda'), [(1, 3, 3)])
+@add_module_test(torch.float32, torch.device('cuda'), [(2, 3, 3, 3)],max_batch_size=3)
 def test_view_1d():
     return View(1, -1)
 
-
+'''
 @add_module_test(torch.float32, torch.device('cuda'), [(1, 3)])
 @add_module_test(torch.float32, torch.device('cuda'), [(1, 3, 3)])
 @add_module_test(torch.float32, torch.device('cuda'), [(1, 3, 3, 3)])
@@ -73,5 +77,5 @@ def test_unsqueeze():
 @add_module_test(torch.float32, torch.device('cuda'), [(1, 3, 1, 3)])
 def test_squeeze():
     return Squeeze(2)    
-
+'''
 

torch2trt/test.py

@@ -45,6 +45,8 @@ def run(self):
         if outputs[i].dtype == torch.bool:
             max_error_i = torch.sum(outputs[i] ^ outputs_trt[i])
         else:
+            print("under test.py")
+            print("pytorch op",outputs[i].size(),"trt op",outputs_trt[i].size())
             max_error_i = torch.max(torch.abs(outputs[i] - outputs_trt[i]))
 
         if max_error_i > max_error:
@@ -144,4 +146,4 @@ def run(self):
     print('NUM_TESTS: %d' % num_tests)
     print('NUM_SUCCESSFUL_CONVERSION: %d' % num_success)
     print('NUM_FAILED_CONVERSION: %d' % num_error)
-    print('NUM_ABOVE_TOLERANCE: %d' % num_tolerance)
+    print('NUM_ABOVE_TOLERANCE: %d' % num_tolerance)

torch2trt/torch2trt.py

@@ -146,15 +146,18 @@ def add_missing_trt_tensors(network, tensors):
         # get tensor w/ _trt
         # or... add constant for scalar primitive
         if isinstance(t, float) or isinstance(t, int):
+            print("if")
             shape = (1,)
             scalar = t * torch.ones(shape, dtype=dtype).cpu().numpy()
             trt_tensor = network.add_constant(shape, scalar).get_output(0)
         elif hasattr(t, "_trt"):
+            print("elif")
+            print(t.shape,t._trt.shape)
             trt_tensor = t._trt
 
         # or... add constant for leaf tensor w/o _trt
         else:
-
+            print("else condition") 
             # remove all preceding ones, these can be re-inserted later when broadcasting
             num_preceding_ones = 0
             for j in range(len(t.shape)):
@@ -163,14 +166,13 @@ def add_missing_trt_tensors(network, tensors):
                 else:
                     break
             shape = tuple(t.shape[num_preceding_ones:])
-
             weight = t.detach().cpu().numpy()
             t._trt = network.add_constant(shape, weight).get_output(0)
             trt_tensor = t._trt
 
 
         assert trt_tensor is not None
-
+        print(trt_tensor.shape)
         trt_tensors[i] = trt_tensor
 
     return trt_tensors
@@ -514,8 +516,9 @@ def torch2trt(module,
     inputs_in = inputs
 
     # copy inputs to avoid modifications to source data
-    inputs = [tensor.clone()[0:1] for tensor in inputs]  # only run single entry
-
+    inputs = [tensor.clone() for tensor in inputs]  # only run single entry
+    for tensor in inputs:
+        print(tensor.shape)
     logger = trt.Logger(log_level)
     builder = trt.Builder(logger)