Remove PyTorch and Pyro-PPL from workflows (#689)

* Remove PyTorch and Pyro-PPL from workflows

- Remove torch, torchvision, torchaudio installation from all workflows
- Remove pyro-ppl installation from all workflows
- Update step names from 'Install JAX, Numpyro, PyTorch' to 'Install JAX and Numpyro'
- Disable build cache in ci.yml for full test run
- Keep JAX and NumPyro installations

* Fix JAX CUDA installation to use cuda12_pip instead of cuda12-local

Changes jax[cuda12-local] to jax[cuda12_pip] to avoid cuDNN version
compatibility issues. The cuda12_pip variant includes compatible
CUDA and cuDNN libraries bundled with JAX, preventing runtime errors
from mismatched local CUDA installations.

Fixes cuDNN 9.10.0 backward-compatibility error.

* Update RunsOn AMI to latest CUDA/cuDNN image

Updates the quantecon_ubuntu2404 AMI from ami-09baf66e396fa7cfd
to ami-0edec81935264b6d3 which includes the latest CUDA and cuDNN
libraries for improved compatibility with JAX.

* Use cuda12-local for JAX installation across all workflows

Changes all workflows to use jax[cuda12-local] to leverage the CUDA
and cuDNN libraries pre-installed in the new AMI (ami-0edec81935264b6d3).
This is faster than cuda12_pip and uses the system libraries.

Also removes version pin (==0.6.2) from cache.yml and publish.yml
to make all workflows consistent.

* Update JAX to use cuda13-local for CUDA 13 support

Changes all workflows from cuda12-local to cuda13-local to match
the CUDA 13 installation in the new AMI (ami-0edec81935264b6d3).

* Update JAX installation to use cuda13 (recommended syntax)

Changes from 'jax[cuda13-local]' to 'jax[cuda13]' following the
official JAX documentation for CUDA 13 support. Also uses -U flag
instead of --upgrade for consistency with JAX docs.

* Move nvidia-smi check before JAX installation in ci.yml

* Update runs-on.yml with region and use AMI ID directly in ci.yml

* Rename image to quantecon_lecture_build to avoid cache issues

* Revert image name back to quantecon_ubuntu2404

* Re-enable build cache

* Revert instance name back to quantecon_ubuntu2404

Author: mmcky

.github/workflows/cache.yml

@@ -20,12 +20,11 @@ jobs:
           python-version: "3.13"
           environment-file: environment.yml
           activate-environment: quantecon
-      - name: Install JAX, Numpyro, PyTorch
+      - name: Install JAX and Numpyro
         shell: bash -l {0}
         run: |
-          pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu128
-          pip install --upgrade "jax[cuda12-local]==0.6.2"
-          pip install numpyro  
+          pip install -U "jax[cuda13]"
+          pip install numpyro
           python scripts/test-jax-install.py
       - name: Check nvidia drivers
         shell: bash -l {0}

.github/workflows/ci.yml

@@ -28,17 +28,15 @@ jobs:
           python-version: "3.13"
           environment-file: environment.yml
           activate-environment: quantecon
-      - name: Install JAX, Numpyro, PyTorch
-        shell: bash -l {0}
-        run: |
-          # pip install torch torchvision --index-url https://download.pytorch.org/whl/cu121
-          # pip install pyro-ppl
-          pip install "jax[cuda12-local]==0.6.2"
-          pip install numpyro pyro-ppl
-          python scripts/test-jax-install.py
       - name: Check nvidia Drivers
         shell: bash -l {0}
         run: nvidia-smi
+      - name: Install JAX and Numpyro
+        shell: bash -l {0}
+        run: |
+          pip install -U "jax[cuda13]"
+          pip install numpyro
+          python scripts/test-jax-install.py
       - name: Display Conda Environment Versions
         shell: bash -l {0}
         run: conda list

.github/workflows/publish.yml

@@ -19,12 +19,11 @@ jobs:
           python-version: "3.13"
           environment-file: environment.yml
           activate-environment: quantecon
-      - name: Install JAX, Numpyro, PyTorch
+      - name: Install JAX and Numpyro
         shell: bash -l {0}
         run: |
-          pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu128
-          pip install --upgrade "jax[cuda12-local]==0.6.2"
-          pip install numpyro  
+          pip install -U "jax[cuda13]"
+          pip install numpyro
           python scripts/test-jax-install.py
       - name: Check nvidia drivers
         shell: bash -l {0}