Update README.md

README.md

@@ -2,16 +2,17 @@
 license: apache-2.0
 ---
 
-# Tensorflow
-
-* Version: 2.7.0
-* TensorRT version: 7.2.2.1
-* Docker image: nvcr.io/nvidia/tensorflow:20.12-tf2-py3
 * GPU: NVIDIA GeForce 3090
 * CUDA: 11.6
 * Driver version: 510.54
+* Input shape (200,66,3)
 
+# Tensorflow
 
+* Tensorflow version: 2.7.0
+* TensorRT version: 7.2.2.1
+* Docker image: nvcr.io/nvidia/tensorflow:20.12-tf2-py3
+* nvidia-tensorrt: 7.2.2.1
 
 | Optimization                      | Model size   (MB)             |        MSE             | Inference time  (s/frame)       | Filename                          |
 | --------------------------------- | ----------------------   | ---------------------- | ---------------------- | --------------------------------- |
@@ -27,7 +28,17 @@ license: apache-2.0
 | Pruning preserving Quantization Aware       |  1.5446319580078125     | 0.010758002372154884   | 0.0008252830505371093  | 28_04_pilotnet_pqat_model.tflite       |
 | Sparsity and cluster preserving quantization aware training (PCQAT)       |  1.5446319580078125     | 0.008262857163545972   | 0.0008286898136138916  | 28_04_pilotnet_pcqat_model.tflite       |
 
-# TensorRT-Tensorflow
+TensorRT-Tensorflow:
+
+To do inference:
+
+```
+  pip install nvidia-tensorrt===7.2.2.1
+  python3 -c "import tensorrt; print(tensorrt.__version__); assert tensorrt.Builder(tensorrt.Logger())"
+  export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/python3.8/site-packages/tensorrt
+  python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
+```
+
 
 | Optimization                      | Model size   (MB)             |        MSE             | Inference time  (s/frame)       | Folder                          |
 | --------------------------------- | ----------------------   | ---------------------- | ---------------------- | --------------------------------- |
@@ -35,6 +46,51 @@ license: apache-2.0
 | Float16 Quantization                          |  0.00390625      | 0.010798278900279191   | 0.00042218327522277834  | 24_04_pilotnet_tftrt_fp16       |
 | Int8 Quantization                          |  0.00390625      | 0.04791482252948612   | 0.0003384373188018799  | 14_06_pilotnet_tftrt_fp16       |
 
+---
+
+# PyTorch
+
+
+* PyTorch version: 1.13.1+cu116
+* TensorRT version: 8.5.5
+* Docker image: nvcr.io/nvidia/pytorch:22.12-py3
+* torch-tensorrt: 1.3.0
+
+
+| Optimization                      | Model size   (MB)             |        MSE             | Inference time  (s/frame)       | Filename                          |
+| --------------------------------- | ----------------------   | ---------------------- | ---------------------- | --------------------------------- |
+| Dynamic Range Quantization      |  1.9493608474731445      | 0.012065857842182075   | 0.001480283498764038  | 28_04_dynamic_quan.pth      |
+| Static Quantization      |  1.6071176528930664      | 0.012072610909984047   | 0.0007314345836639404  | 28_04_static_quan.pth      |
+| Quantization Aware Training      |  1.6069536209106445      | 0.01109830549109022   | 0.0011710402965545653  | 28_04_quan_aware.pth      |
+| Local Prune      |  6.122584342956543      | 0.010850968803449539   | 0.0014387350082397461  | 28_04_local_prune.pth      |
+| Global Prune      |  6.122775077819824      | 0.010964057565769462   | 0.0014179635047912597  | 28_04_global_prune.pth      |
+| Prune + Quantization      |  1.6067094802856445      | 0.010949893930274941   | 0.0011728739738464356  | 28_04_prune_quan.pth      |
+
+TensorRT-PyTorch:
+
+To do inference:
+
+```
+  pip install torch-tensorrt==1.3.0
+```
+
+| Optimization                      | Model size   (MB)             |        MSE             | Inference time  (s/frame)       | Filename                          |
+| --------------------------------- | ----------------------   | ---------------------- | ---------------------- | --------------------------------- |
+| Float32 Quantization          |  -                      | -                    | -                      | 28_04_trt_mod_float_28_04_float.jit.pt      |
+| Float16 Quantization          |  -                      | -                    | -                      | 28_04_trt_mod_float_28_04_half.jit.pt      |
+| Int8 Quantization          |  -                      | -                    | -                      | 28_04_trt_mod_float_28_04_int8.jit.pt      |
+
+
+
+
+
+
+
+
+
+
+
+