Merge branch 'dev' into Study-to-multi-eval

Author: recursix

.github/workflows/pypi.yml

@@ -0,0 +1,125 @@
+name: Build and Publish
+# based on official doc
+# https://packaging.python.org/en/latest/guides/publishing-package-distribution-releases-using-github-actions-ci-cd-workflows/
+
+on: [push, workflow_dispatch]
+
+jobs:
+    build:
+      name: Build
+      runs-on: ubuntu-22.04
+
+      steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.x"
+
+      - name: Install pypa/build
+        run: python3 -m pip install build --user
+
+      - name: Build a binary wheel and a source tarball (agentlab)
+        run: python3 -m build . --outdir dist/
+
+      - name: Store the distribution packages
+        uses: actions/upload-artifact@v4
+        with:
+          name: python-package-distributions
+          path: dist/
+
+    publish-to-pypi:
+      name: Publish to PyPI
+      if: startsWith(github.ref, 'refs/tags/')  # only publish to PyPI on tag pushes # tweak so it doesnt publish all tags
+      needs:
+        - build
+      runs-on: ubuntu-22.04
+      environment: pypi
+      permissions:
+        id-token: write  # IMPORTANT: mandatory for trusted publishing
+
+      steps:
+        - name: Download all the distribution packages
+          uses: actions/download-artifact@v4
+          with:
+            name: python-package-distributions
+            path: dist/
+
+        - name: Publish all distribution packages to PyPI
+          uses: pypa/gh-action-pypi-publish@release/v1
+
+    github-release:
+      name: Sign packages with Sigstore and upload them to GitHub Release
+      needs:
+      - publish-to-pypi
+      runs-on: ubuntu-22.04
+
+      permissions:
+        contents: write  # IMPORTANT: mandatory for making GitHub Releases
+        id-token: write  # IMPORTANT: mandatory for sigstore
+
+      steps:
+      - name: Download all the dists
+        uses: actions/download-artifact@v4
+        with:
+          name: python-package-distributions
+          path: dist/
+
+      - name: Sign the dists with Sigstore
+        uses: sigstore/[email protected]
+        with:
+          inputs: >-
+            ./dist/*.tar.gz
+            ./dist/*.whl
+
+      - name: Create GitHub Release
+        env:
+          GITHUB_TOKEN: ${{ github.token }}
+        run: >-
+          gh release create
+          '${{ github.ref_name }}'
+          --repo '${{ github.repository }}'
+          --notes ""
+
+      - name: Upload artifact signatures to GitHub Release
+        env:
+          GITHUB_TOKEN: ${{ github.token }}
+        # Upload to GitHub Release using the `gh` CLI.
+        # `dist/` contains the built packages, and the
+        # sigstore-produced signatures and certificates.
+        run: >-
+          gh release upload
+          '${{ github.ref_name }}' dist/**
+          --repo '${{ github.repository }}'
+
+      - name: Set GitHub Release as pre-release
+        if: contains(github.ref, '.dev')  # only set tags vA.B.C.devD as pre-release
+        env:
+          GITHUB_TOKEN: ${{ github.token }}
+        run: >-
+          gh release edit
+          '${{ github.ref_name }}'
+          --repo '${{ github.repository }}'
+          --prerelease
+
+    # publish-to-testpypi:
+    #   name: Publish to TestPyPI
+    #   needs:
+    #   - build
+    #   runs-on: ubuntu-latest
+    #   environment: testpypi
+    #   permissions:
+    #     id-token: write  # IMPORTANT: mandatory for trusted publishing
+
+    #   steps:
+    #   - name: Download all the distribution packages
+    #     uses: actions/download-artifact@v4
+    #     with:
+    #       name: python-package-distributions
+    #       path: dist/
+
+    #   - name: Publish distribution packages to TestPyPI
+    #     uses: pypa/gh-action-pypi-publish@release/v1
+    #     with:
+    #       repository-url: https://test.pypi.org/legacy/

src/agentlab/experiments/exp_utils.py

@@ -1,13 +1,13 @@
-import os
-from pathlib import Path
-from browsergym.experiments.loop import _move_old_exp, yield_all_exp_results
-from tqdm import tqdm
 import logging
-from browsergym.experiments.loop import ExpArgs
-from contextlib import contextmanager
+import os
 import signal
 import sys
-from time import time, sleep
+from contextlib import contextmanager
+from pathlib import Path
+from time import sleep, time
+
+from browsergym.experiments.loop import ExpArgs, _move_old_exp, yield_all_exp_results
+from tqdm import tqdm
 
 logger = logging.getLogger(__name__)  # Get logger based on module name
 
@@ -130,6 +130,7 @@ def add_dependencies(exp_args_list: list[ExpArgs], task_dependencies: dict[str,
 class MockedExpArgs:
     def __init__(self, exp_id, depends_on=None):
         self.exp_id = exp_id
+        self.exp_name = f"exp_{exp_id}"
         self.depends_on = depends_on if depends_on else []
         self.start_time = None
         self.end_time = None

src/agentlab/experiments/graph_execution_ray.py

@@ -28,7 +28,7 @@ def get_task(exp_arg: bgym.ExpArgs):
             dependency_tasks = [get_task(exp_args_map[dep_key]) for dep_key in exp_arg.depends_on]
 
             # Create new task that depends on the dependency results
-            task_map[exp_arg.exp_id] = run_exp.remote(
+            task_map[exp_arg.exp_id] = run_exp.options(name=f"{exp_arg.exp_name}").remote(
                 exp_arg, *dependency_tasks, avg_step_timeout=avg_step_timeout
             )
         return task_map[exp_arg.exp_id]

src/agentlab/ui_assistant.py

@@ -3,6 +3,7 @@
 from browsergym.experiments.loop import EnvArgs, ExpArgs
 
 from agentlab.agents.agent_args import AgentArgs
+from agentlab.agents.generic_agent.generic_agent import GenericAgentArgs
 from agentlab.experiments.exp_utils import RESULTS_DIR
 from agentlab.experiments.launch_exp import import_object
 
@@ -14,6 +15,9 @@ def make_exp_args(agent_args: AgentArgs, start_url="https://www.google.com"):
     except AttributeError:
         pass
 
+    if isinstance(agent_args, GenericAgentArgs):
+        agent_args.flags.enable_chat = True
+
     exp_args = ExpArgs(
         agent_args=agent_args,
         env_args=EnvArgs(