rakam-eval-sdk 0.2.0__tar.gz → 0.2.0rc1__tar.gz

{rakam_eval_sdk-0.2.0 → rakam_eval_sdk-0.2.0rc1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: rakam-eval-sdk
-Version: 0.2.0
+Version: 0.2.0rc1
 Summary: Evaluation Framework SDK
 Author: Mohamed Bachar Touil
 License: MIT

{rakam_eval_sdk-0.2.0 → rakam_eval_sdk-0.2.0rc1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "uv_build"
 
 [project]
 name = "rakam-eval-sdk"
-version = "0.2.0"
+version = "0.2.0rc1"
 description = "Evaluation Framework SDK"
 readme = "README.md"
 requires-python = ">=3.8"

{rakam_eval_sdk-0.2.0 → rakam_eval_sdk-0.2.0rc1}/src/rakam_eval_sdk/cli.py

@@ -6,7 +6,7 @@ import uuid
 from datetime import datetime
 from pathlib import Path
 from pprint import pprint
-from typing import Any, Dict, Optional, Sequence
+from typing import Any, Optional
 
 import typer
 from dotenv import load_dotenv
@@ -29,64 +29,9 @@ console = Console()
 PROJECT_ROOT = os.path.abspath(".")
 if PROJECT_ROOT not in sys.path:
     sys.path.insert(0, PROJECT_ROOT)
-list_app = typer.Typer(help="List resources")
-app.add_typer(list_app, name="list")
-metrics_app = typer.Typer(help="Metrics utilities")
-app.add_typer(metrics_app, name="metrics")
-
 
-@metrics_app.command("list")
-def list_metrics(
-    limit: int = typer.Option(
-        20,
-        "--limit",
-        help="Number of testcases to inspect for metrics",
-    ),
-):
-    """
-    List unique metric names found in evaluation testcases.
-    """
-    client = DeepEvalClient()
 
-    testcases = client.list_evaluation_testcases(
-        limit=limit,
-        offset=0,
-        raise_exception=True,
-    )
-
-    if not testcases:
-        typer.echo("No evaluation testcases found.")
-        return
-
-    metric_names: set[str] = set()
-
-    def collect_metrics(entries: Sequence[Dict] | None):
-        if not entries:
-            return
-        for entry in entries:
-            for metric in entry.get("metrics", []) or []:
-                name = metric.get("name")
-                if name:
-                    metric_names.add(name)
-
-    for tc in testcases:
-
-        collect_metrics(tc.get("result"))
-
-    if not metric_names:
-        typer.echo("No metrics found.")
-        return
-
-    typer.echo(
-        f"📊 Found {len(metric_names)} unique metrics "
-        f"(from latest {limit} testcases)\n"
-    )
-
-    for name in sorted(metric_names):
-        typer.echo(f"- {name}")
-
-
-@list_app.command("eval")
+@app.command()
 def list(
     directory: Path = typer.Argument(
         Path("./eval"),
@@ -120,10 +65,17 @@ def list(
         typer.echo(f"No @{TARGET_DECORATOR} functions found.")
 
 
+list_app = typer.Typer(help="List resources")
+app.add_typer(list_app, name="list")
+
+
 @list_app.command("runs")
 def list_runs(
     limit: int = typer.Option(20, help="Max number of runs"),
     offset: int = typer.Option(0, help="Pagination offset"),
+    status: Optional[str] = typer.Option(
+        None, help="Filter by status (running, completed, failed)"
+    ),
 ):
     """
     List evaluation runs (newest first).
@@ -140,7 +92,19 @@ def list_runs(
         typer.echo("No evaluation runs found.")
         return
 
-    typer.echo(f"[id] " f"{'unique_id':<20}" f"{'label':<20}" f"created_at")
+    # optional status filtering (client-side for now)
+    if status:
+        runs = [
+            r for r in runs
+            if r.get("result", {}).get("status") == status
+        ]
+
+    typer.echo(
+        f"[id] "
+        f"{'unique_id':<20}"
+        f"{'label':<20}"
+        f"created_at"
+    )
     # pretty CLI output
     for run in runs:
         run_id = run.get("id")
@@ -156,7 +120,12 @@ def list_runs(
             except ValueError:
                 pass
 
-        typer.echo(f"[{run_id}] " f"{uid:<20} " f"{label:<20} " f"{created_at}")
+        typer.echo(
+            f"[{run_id}] "
+            f"{uid:<20} "
+            f"{label:<20} "
+            f"{created_at}"
+        )
 
 
 @list_app.command("show")
@@ -376,7 +345,8 @@ def _print_and_save(
         return
 
     if out.exists() and not overwrite:
-        typer.echo(f"❌ File already exists: {out} (use --overwrite to replace)")
+        typer.echo(
+            f"❌ File already exists: {out} (use --overwrite to replace)")
         raise typer.Exit(code=1)
 
     out.parent.mkdir(parents=True, exist_ok=True)

{rakam_eval_sdk-0.2.0 → rakam_eval_sdk-0.2.0rc1}/src/rakam_eval_sdk/client.py

@@ -38,7 +38,8 @@ class DeepEvalClient:
         )
         self.base_url = raw_url.rstrip("/")
         self.api_token = (
-            api_token or settings_token or os.getenv("EVALFRAMWORK_API_KEY", "")
+            api_token or settings_token or os.getenv(
+                "EVALFRAMWORK_API_KEY", "")
         )
         self.timeout = timeout
 

{rakam_eval_sdk-0.2.0 → rakam_eval_sdk-0.2.0rc1}/src/rakam_eval_sdk/schema.py

@@ -94,7 +94,8 @@ MetricConfig = Annotated[
 ]
 
 SchemaMetricConfig = Annotated[
-    Union[JsonCorrectnessConfig, FieldsPresenceConfig], Field(discriminator="type")
+    Union[JsonCorrectnessConfig, FieldsPresenceConfig], Field(
+        discriminator="type")
 ]