Evaluation code problem，我只想使用训练好的策略进行评价，我在脚本文件中增加了test，和model_dir，为什么程序还是从零开始训练呢？

[RexDrac]

Importing module 'gym_38' (/home/nuc/Downloads/IsaacGym_Preview_4_Package/isaacgym/python/isaacgym/_bindings/linux-x86_64/gym_38.so)
Setting GYM_USD_PLUG_INFO_PATH to /home/nuc/Downloads/IsaacGym_Preview_4_Package/isaacgym/python/isaacgym/_bindings/linux-x86_64/usd/plugInfo.json
PyTorch version 1.13.0+cu117
Device count 1
/home/nuc/Downloads/IsaacGym_Preview_4_Package/isaacgym/python/isaacgym/_bindings/src/gymtorch
Using /home/nuc/.cache/torch_extensions/py38_cu117 as PyTorch extensions root...
Emitting ninja build file /home/nuc/.cache/torch_extensions/py38_cu117/gymtorch/build.ninja...
Building extension module gymtorch...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
ninja: no work to do.
Loading extension module gymtorch...
/home/nuc/miniforge3/envs/dexgrasp/lib/python3.8/site-packages/pointnet2_ops/pointnet2_utils.py:15: UserWarning: Unable to load pointnet2_ops cpp extension. JIT Compiling.
warnings.warn("Unable to load pointnet2_ops cpp extension. JIT Compiling.")
Setting seed: 0
Algorithm: ppo
Python
Averaging factor: 0.01
Obs type: full_state
self.device_id: 0
self.graphics_device_id 0
Not connected to PVD
+++ Using GPU PhysX
Physics Engine: PhysX
Physics Device: cuda:0
GPU Pipeline: enabled
JointSpec type free not yet supported!
self.num_shadow_hand_bodies: 24
self.num_shadow_hand_shapes: 20
self.num_shadow_hand_dofs: 22
self.num_shadow_hand_actuators: 18
self.num_shadow_hand_tendons: 4
Using VHACD cache directory '/home/nuc/.isaacgym/vhacd'
Found existing convex decomposition for mesh '/home/nuc/xiaomi/UniDexGrasp2/assets/meshdatav3_scaled/sem/Car-669043a8ce40d9d78781f76a6db4ab62/coacd/decomposed_006.obj'
/home/nuc/miniforge3/envs/dexgrasp/lib/python3.8/site-packages/gym/spaces/box.py:127: UserWarning: WARN: Box bound precision lowered by casting to float32
logger.warn(f"Box bound precision lowered by casting to {self.dtype}")
RL device: cuda:0
Sequential(
(0): Linear(in_features=300, out_features=1024, bias=True)
(1): ELU(alpha=1.0)
(2): Linear(in_features=1024, out_features=1024, bias=True)
(3): ELU(alpha=1.0)
(4): Linear(in_features=1024, out_features=512, bias=True)
(5): ELU(alpha=1.0)
(6): Linear(in_features=512, out_features=512, bias=True)
(7): ELU(alpha=1.0)
(8): Linear(in_features=512, out_features=24, bias=True)
)
Sequential(
(0): Linear(in_features=300, out_features=1024, bias=True)
(1): ELU(alpha=1.0)
(2): Linear(in_features=1024, out_features=1024, bias=True)
(3): ELU(alpha=1.0)
(4): Linear(in_features=1024, out_features=512, bias=True)
(5): ELU(alpha=1.0)
(6): Linear(in_features=512, out_features=512, bias=True)
(7): ELU(alpha=1.0)
(8): Linear(in_features=512, out_features=1, bias=True)
)
################################################################################
Learning iteration 0/10000

                   Computation: 10170 steps/s (collection: 0.697s, learning 0.090s)
           Value function loss: 14.2616
                Surrogate loss: 0.0191
         Mean action noise std: 0.80
                   Mean reward: -2.95
           Mean episode length: 4.79
              Mean reward/step: -0.63
   Mean episode length/episode: 7.74
        Mean episode successes: 0.0000
Mean episode current_successes: 0.0000

Mean episode consecutive_successes: 0.0000

               Total timesteps: 8000
                Iteration time: 0.79s
                    Total time: 0.79s
                           ETA: 7865.6s

################################################################################
Learning iteration 1/10000

                   Computation: 19142 steps/s (collection: 0.334s, learning 0.084s)
           Value function loss: 4.0435
                Surrogate loss: -0.0087
         Mean action noise std: 0.80
                   Mean reward: -4.97
           Mean episode length: 7.98
              Mean reward/step: -0.63
   Mean episode length/episode: 7.74

[ZenGrasper]

按说不会呀，看看是不是model_dir后面少了个 \ 。只要取消--test就会进入测试模型不会训练了

[huamo555]

How do I test it? I can't find the entry for inputting model weights. How can I add the model weights given by the author?