Generate input tensor constraints (#401)

* support checking model redundancy

* revert change of vision_model_test

* reformat python code.

* reformat bert_model_test.py and utils.py

* minor fix

* fix failed check by comparing directories after os.path.realpath()

* fix bugs in check_validate.sh

* set dynamic=False in single_device_runner.py

* reset graph hash

* add robustness code for generating input tensor constraints

* Introduce input_tensor_constraints.py using shape propagation logic.

* support dimension generalization for torch.Tensor.view and torch.Tensor.reshape

* 1) support dimension generalization for torch.Tensor.expand(); 2) fix bugs in generalization for torch.Tensor.view and torch.Tensor.reshape

* dimension_generalization_passes

* Refactored DimensionGeneralizationPass.__init__ to accept argument dim_axes_pairs, enabling targeted configuration for specific use cases

* save dimension generalization pass names into graph_net.json

* Generalize sequence dimension

* more dimension generalization passes for token dimension

* refactor parse_sole_graph_module

* Enhance the performance of the input_tensor_constraints.py file generation process.

* minor fix

* more dimension generalization pass

* fix some hint bugs

* generate input_tensor_constraints.py

Author: lixinqi

.pre-commit-config.yaml

@@ -9,7 +9,7 @@ repos:
   rev: v0.14.4
   hooks:
     - id: ruff-check
-      args: [--fix, --exit-non-zero-on-fix, --no-cache]
+      args: [--fix, --exit-non-zero-on-fix, --no-cache, --exclude=samples]
 
 - repo: https://github.com/Lucas-C/pre-commit-hooks.git
   rev: v1.5.1

samples/cosyvoice/CosyVoice-300M/graph_net.json

@@ -1,8 +1 @@
-{
-    "framework": "torch",
-    "num_devices_required": 1,
-    "num_nodes_required": 1,
-    "dynamic": false,
-    "source": "cosyvoice",
-    "heuristic_tag": "audio"
-}
+{"framework": "torch", "num_devices_required": 1, "num_nodes_required": 1, "dynamic": false, "source": "cosyvoice", "heuristic_tag": "audio", "dimension_generalization_passes": ["naive_call_method_view_pass", "naive_call_method_reshape_pass", "naive_call_method_expand_pass", "non_batch_call_method_expand_pass", "non_batch_call_function_arange_pass", "non_batch_call_function_getitem_slice_pass", "non_batch_call_function_full_pass", "non_batch_call_function_full_plus_one_pass", "non_batch_call_function_zeros_pass", "non_batch_call_function_arange_plus_one_pass"]}