| {"class_name": "PytorchONNXTensorRTInferenceLearner", "module_name": "nebullvm.operations.inference_learners.tensor_rt", "network_parameters": {"batch_size": 2, "input_infos": [{"dtype": "float16", "size": [2, 4, 64, 64]}, {"dtype": "int64", "size": [1]}, {"dtype": "float16", "size": [2, 77, 768]}], "output_sizes": [[2, 4, 64, 64]], "output_types": ["float16"], "dynamic_info": {"inputs": [{"0": {"name": "2B", "min_val": 2, "opt_val": 2, "max_val": 2}, "2": {"name": "H", "min_val": 64, "opt_val": 64, "max_val": 96}, "3": {"name": "W", "min_val": 64, "opt_val": 64, "max_val": 96}}, {}, {"0": {"name": "2B", "min_val": 2, "opt_val": 2, "max_val": 2}}], "outputs": [{"0": "2B", "2": "H", "3": "W"}]}}, "input_tfms": {"tfms": [{"module": "nebullvm.tools.transformations", "name": "VerifyContiguity"}]}, "device": "gpu", "input_names": ["input_0", "input_1", "input_2"], "output_names": ["output_0"]} |