Using LIBERO constants: NUM_ACTIONS_CHUNK = 8 ACTION_DIM = 7 PROPRIO_DIM = 8 ACTION_PROPRIO_NORMALIZATION_TYPE = bounds_q99 If needed, manually set the correct constants in `prismatic/vla/constants.py`! 2025-09-19 05:23:03.628892: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`. 2025-09-19 05:23:03.659621: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered 2025-09-19 05:23:03.659666: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered 2025-09-19 05:23:03.660575: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered 2025-09-19 05:23:03.665568: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: AVX2 AVX512F AVX512_VNNI AVX512_BF16 AVX_VNNI AMX_TILE AMX_INT8 AMX_BF16 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. 2025-09-19 05:23:04.336851: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT 2025-09-19 05:23:04.997003: W tensorflow/core/common_runtime/gpu/gpu_device.cc:2348] TensorFlow was not built with CUDA kernel binaries compatible with compute capability 9.0. CUDA kernels will be jit-compiled from PTX, which could take 30 minutes or longer. 09/19 [05:23:15] INFO | >> Global seed set to 7 seed.py:54 Instantiating pretrained VLA policy... Loaded model: pybullet build time: Jan 29 2025 23:16:28 argv[0]=--width=200 argv[1]=--height=200 09/19 [05:23:22] INFO | >> Loading EGL plugin (may play_table_env.py:106 segfault on misconfigured systems)... EGL device choice: -1 of 17. Loaded EGL 1.5 after reload. GL_VENDOR=NVIDIA Corporation GL_RENDERER=NVIDIA H100 80GB HBM3/PCIe/SSE2 GL_VERSION=3.3.0 NVIDIA 555.42.02 GL_SHADING_LANGUAGE_VERSION=3.30 NVIDIA via Cg compiler Version = 3.3.0 NVIDIA 555.42.02 Vendor = NVIDIA Corporation Renderer = NVIDIA H100 80GB HBM3/PCIe/SSE2 INFO | >> Successfully loaded egl play_table_env.py:116 plugin INFO | >> Connected to server with play_table_env.py:122 id: 0 INFO | >> Connected to server with play_table_env.py:128 id: 0 INFO | >> Loading robot robot.py:40 INFO | >> Resetting simulation play_table_env.py:133 INFO | >> Setting gravity play_table_env.py:135 09/19 [05:23:23] INFO | >> Using calvin_env with commit play_table_env.py:75 1431a46bd36bde5903fb6345e68b5ccc3 0def666. INFO | >> Initialized PlayTableEnv calvin_env_wrapper.py:31 for device cuda:0 logging to evaluation_results/calvin/CALVIN-ABC-Pro INFO | >> Start generating multistep_sequences.py:368 evaluation sequences. 09/19 [05:23:28] INFO | >> Done generating multistep_sequences.py:380 evaluation sequences. 0%| | 0/1000 [00:00