Fix out of date model configuration parameters (#51)

Author: gregorybchris

docs/batch/batch-job-info.md docs/batch/batch-job-details.mddocs/batch/batch-job-info.md renamed to docs/batch/batch-job-details.md

@@ -38,10 +38,12 @@
     "print()\n",
     "\n",
     "with open(\"predictions.json\", \"r\") as f:\n",
-    "    predictions = json.load(f)\n",
-    "    for prediction in predictions:\n",
-    "        for file_data in prediction[\"files\"]:\n",
-    "            face_predictions = file_data[\"models\"][\"face\"]\n",
+    "    full_predictions = json.load(f)\n",
+    "    for source in full_predictions:\n",
+    "        source_name = source[\"source\"][\"url\"]\n",
+    "        predictions = source[\"results\"][\"predictions\"]\n",
+    "        for prediction in predictions:\n",
+    "            face_predictions = prediction[\"models\"][\"face\"][\"grouped_predictions\"]\n",
     "            for face_prediction in face_predictions:\n",
     "                for frame in face_prediction[\"predictions\"]:\n",
     "                    print_emotions(frame[\"emotions\"])"

examples/batch-facial-action-coding-system/batch-facial-action-coding-system.ipynb

@@ -38,10 +38,12 @@
         "print()\n",
         "\n",
         "with open(\"predictions.json\", \"r\") as f:\n",
-        "    predictions = json.load(f)\n",
-        "    for prediction in predictions:\n",
-        "        for file_data in prediction[\"files\"]:\n",
-        "            ner_predictions = file_data[\"models\"][\"ner\"]\n",
+        "    full_predictions = json.load(f)\n",
+        "    for source in full_predictions:\n",
+        "        source_name = source[\"source\"][\"url\"]\n",
+        "        predictions = source[\"results\"][\"predictions\"]\n",
+        "        for prediction in predictions:\n",
+        "            ner_predictions = prediction[\"models\"][\"ner\"][\"grouped_predictions\"]\n",
         "            for ner_prediction in ner_predictions:\n",
         "                for entity_data in ner_prediction[\"predictions\"]:\n",
         "                    print(entity_data[\"entity\"])\n",

examples/batch-text-entity-recognition/batch-text-entity-recognition.ipynb

@@ -38,13 +38,15 @@
         "print()\n",
         "\n",
         "with open(\"predictions.json\", \"r\") as f:\n",
-        "    predictions = json.load(f)\n",
-        "    for prediction in predictions:\n",
-        "        for file_data in prediction[\"files\"]:\n",
-        "            language_predictions = file_data[\"models\"][\"language\"]\n",
+        "    full_predictions = json.load(f)\n",
+        "    for source in full_predictions:\n",
+        "        source_name = source[\"source\"][\"url\"]\n",
+        "        predictions = source[\"results\"][\"predictions\"]\n",
+        "        for prediction in predictions:\n",
+        "            language_predictions = prediction[\"models\"][\"language\"][\"grouped_predictions\"]\n",
         "            for language_prediction in language_predictions:\n",
         "                for chunk in language_prediction[\"predictions\"]:\n",
-        "                    print(chunk[\"word\"])\n",
+        "                    print(chunk[\"text\"])\n",
         "                    print_emotions(chunk[\"emotions\"])\n",
         "                    print(\"~ ~ ~\")\n",
         "                    print_sentiment(chunk[\"sentiment\"])\n",

examples/batch-text-sentiment-analysis/batch-text-sentiment-analysis.ipynb

@@ -38,13 +38,15 @@
         "print()\n",
         "\n",
         "with open(\"predictions.json\", \"r\") as f:\n",
-        "    predictions = json.load(f)\n",
-        "    for prediction in predictions:\n",
-        "        for file_data in prediction[\"files\"]:\n",
-        "            language_predictions = file_data[\"models\"][\"language\"]\n",
+        "    full_predictions = json.load(f)\n",
+        "    for source in full_predictions:\n",
+        "        source_name = source[\"source\"][\"url\"]\n",
+        "        predictions = source[\"results\"][\"predictions\"]\n",
+        "        for prediction in predictions:\n",
+        "            language_predictions = prediction[\"models\"][\"language\"][\"grouped_predictions\"]\n",
         "            for language_prediction in language_predictions:\n",
         "                for chunk in language_prediction[\"predictions\"]:\n",
-        "                    print(chunk[\"word\"])\n",
+        "                    print(chunk[\"text\"])\n",
         "                    print_emotions(chunk[\"emotions\"])\n",
         "                    print(\"~ ~ ~\")\n",
         "                    print_toxicity(chunk[\"toxicity\"])\n",

examples/batch-text-toxicity-detection/batch-text-toxicity-detection.ipynb

@@ -39,19 +39,21 @@
         "print(\"Predictions ready!\")\n",
         "\n",
         "with open(\"predictions.json\", \"r\") as f:\n",
-        "    predictions = json.load(f)\n",
-        "    for prediction in predictions:\n",
-        "        for file_data in prediction[\"files\"]:\n",
+        "    full_predictions = json.load(f)\n",
+        "    for source in full_predictions:\n",
+        "        source_name = source[\"source\"][\"url\"]\n",
+        "        predictions = source[\"results\"][\"predictions\"]\n",
+        "        for prediction in predictions:\n",
         "            print()\n",
         "            print(\"Speech prosody\")\n",
-        "            prosody_predictions = file_data[\"models\"][\"prosody\"]\n",
+        "            prosody_predictions = prediction[\"models\"][\"prosody\"][\"grouped_predictions\"]\n",
         "            for prosody_prediction in prosody_predictions:\n",
         "                for segment in prosody_prediction[\"predictions\"][:1]:\n",
         "                    print_emotions(segment[\"emotions\"])\n",
         "\n",
         "            print()\n",
         "            print(\"Vocal burst\")\n",
-        "            burst_predictions = file_data[\"models\"][\"burst\"]\n",
+        "            burst_predictions = prediction[\"models\"][\"burst\"][\"grouped_predictions\"]\n",
         "            for burst_prediction in burst_predictions:\n",
         "                for segment in burst_prediction[\"predictions\"][:1]:\n",
         "                    print_emotions(segment[\"emotions\"])"

examples/batch-voice-expression/batch-voice-expression.ipynb

@@ -1,5 +1,6 @@
 """Batch job details."""
 import json
+from datetime import datetime
 from typing import Any, Dict, List, Optional
 
 from hume._batch.batch_job_state import BatchJobState
@@ -99,3 +100,55 @@ def _get_invalid_response_message(cls, response: Any) -> str:
                 message = "HumeBatchClient initialized with invalid API key."
 
         return message
+
+    def get_status(self) -> BatchJobStatus:
+        """Get the status of the job.
+
+        Returns:
+            BatchJobStatus: The status of the `BatchJob`.
+        """
+        return self.state.status
+
+    def get_run_time_ms(self) -> Optional[int]:
+        """Get the total time in milliseconds it took for the job to run if the job is in a terminal state.
+
+        Returns:
+            Optional[int]: Time in milliseconds it took for the job to run. If the job is not in a terminal
+                state then `None` is returned.
+        """
+        if self.state.started_timestamp_ms is not None and self.state.ended_timestamp_ms is not None:
+            return self.state.ended_timestamp_ms - self.state.started_timestamp_ms
+        return None
+
+    def get_created_time(self) -> Optional[datetime]:
+        """Get the time the job was created.
+
+        Returns:
+            Optional[datetime]: Datetime when the job was created. If the job has not started
+                then `None` is returned.
+        """
+        if self.state.created_timestamp_ms is None:
+            return None
+        return datetime.utcfromtimestamp(self.state.created_timestamp_ms / 1000)
+
+    def get_started_time(self) -> Optional[datetime]:
+        """Get the time the job started running.
+
+        Returns:
+            Optional[datetime]: Datetime when the job started running. If the job has not started
+                then `None` is returned.
+        """
+        if self.state.started_timestamp_ms is None:
+            return None
+        return datetime.utcfromtimestamp(self.state.started_timestamp_ms / 1000)
+
+    def get_ended_time(self) -> Optional[datetime]:
+        """Get the time the job stopped running if the job is in a terminal state.
+
+        Returns:
+            Optional[datetime]: Datetime when the job started running. If the job is not in a terminal
+                state then `None` is returned.
+        """
+        if self.state.ended_timestamp_ms is None:
+            return None
+        return datetime.utcfromtimestamp(self.state.ended_timestamp_ms / 1000)

hume/_batch/batch_job_details.py

@@ -1,6 +1,5 @@
 """Batch job state."""
 from dataclasses import dataclass
-from datetime import datetime
 from typing import Optional
 
 from hume._batch.batch_job_status import BatchJobStatus
@@ -21,47 +20,3 @@ class BatchJobState:
     created_timestamp_ms: Optional[int]
     started_timestamp_ms: Optional[int]
     ended_timestamp_ms: Optional[int]
-
-    def get_run_time_ms(self) -> Optional[int]:
-        """Get the total time in milliseconds it took for the job to run if the job is in a terminal state.
-
-        Returns:
-            Optional[int]: Time in milliseconds it took for the job to run. If the job is not in a terminal
-                state then `None` is returned.
-        """
-        if self.started_timestamp_ms is not None and self.ended_timestamp_ms is not None:
-            return self.ended_timestamp_ms - self.started_timestamp_ms
-        return None
-
-    def get_created_time(self) -> Optional[datetime]:
-        """Get the time the job was created.
-
-        Returns:
-            Optional[datetime]: Datetime when the job was created. If the job has not started
-                then `None` is returned.
-        """
-        if self.created_timestamp_ms is None:
-            return None
-        return datetime.utcfromtimestamp(self.created_timestamp_ms / 1000)
-
-    def get_started_time(self) -> Optional[datetime]:
-        """Get the time the job started running.
-
-        Returns:
-            Optional[datetime]: Datetime when the job started running. If the job has not started
-                then `None` is returned.
-        """
-        if self.started_timestamp_ms is None:
-            return None
-        return datetime.utcfromtimestamp(self.started_timestamp_ms / 1000)
-
-    def get_ended_time(self) -> Optional[datetime]:
-        """Get the time the job stopped running if the job is in a terminal state.
-
-        Returns:
-            Optional[datetime]: Datetime when the job started running. If the job is not in a terminal
-                state then `None` is returned.
-        """
-        if self.ended_timestamp_ms is None:
-            return None
-        return datetime.utcfromtimestamp(self.ended_timestamp_ms / 1000)

hume/_batch/batch_job_state.py

@@ -14,12 +14,6 @@ class TranscriptionConfig(ConfigBase["TranscriptionConfig"]):
             If missing or null, it will be automatically detected. Values are `zh`, `da`, `nl`, `en`, `en-AU`,
             `en-IN`, `en-NZ`, `en-GB`, `fr`, `fr-CA`, `de`, `hi`, `hi-Latn`, `id`, `it`, `ja`, `ko`, `no`,
             `pl`, `pt`, `pt-BR`, `pt-PT`, `ru`, `es`, `es-419`, `sv`, `ta`, `tr`, or `uk`.
-            This configuration is not available for the streaming API.
-        identify_speakers (Optional[bool]): Whether to return identifiers for speakers over time.
-            If true, unique identifiers will be assigned to spoken words to differentiate different speakers.
-            If false, all speakers will be tagged with an "unknown" ID.
-            This configuration is not available for the streaming API.
     """
 
     language: Optional[str] = None
-    identify_speakers: Optional[bool] = None

hume/_batch/transcription_config.py

@@ -13,24 +13,28 @@ class FaceConfig(ModelConfigBase["FaceConfig"]):
     Args:
         fps_pred (Optional[float]): Number of frames per second to process. Other frames will be omitted
             from the response.
-            This configuration is not available for the streaming API.
+            This configuration is only available for the batch API.
         prob_threshold (Optional[float]): Face detection probability threshold. Faces detected with a
             probability less than this threshold will be omitted from the response.
-            This configuration is not available for the streaming API.
+            This configuration is only available for the batch API.
         identify_faces (Optional[bool]): Whether to return identifiers for faces across frames.
             If true, unique identifiers will be assigned to face bounding boxes to differentiate different faces.
             If false, all faces will be tagged with an "unknown" ID.
         min_face_size (Optional[float]): Minimum bounding box side length in pixels to treat as a face.
             Faces detected with a bounding box side length in pixels less than this threshold will be
             omitted from the response.
-            This configuration is not available for the streaming API.
+            This configuration is only available for the batch API.
         save_faces (Optional[bool]): Whether to extract and save the detected faces to the artifacts
             directory included in the response.
-            This configuration is not available for the streaming API.
+            This configuration is only available for the batch API.
         descriptions (Optional[Dict[str, Any]]): Configuration for Descriptions predictions.
-            If missing or null, no Descriptions predictions will be generated.
+            Descriptions prediction can be enabled by setting "descriptions": {}.
+            Currently, Descriptions prediction cannot be further configured with any parameters.
+            If missing or null, no descriptions predictions will be generated.
         facs (Optional[Dict[str, Any]]): Configuration for FACS predictions.
-            If missing or null, no FACS predictions will be generated.
+            FACS prediction can be enabled by setting "facs": {}.
+            Currently, FACS prediction cannot be further configured with any parameters.
+            If missing or null, no facs predictions will be generated.
     """
 
     fps_pred: Optional[float] = None