[SPARKNLP-1080] AutoGGUFEmbeddings change default pretrained model

docs/en/annotator_entries/AutoGGUFEmbeddings.md

@@ -20,7 +20,7 @@ val autoGGUFEmbeddings = AutoGGUFEmbeddings.pretrained()
   .setOutputCol("embeddings")
 ```
 
-The default model is `"nomic-embed-text-v1.5.f16.gguf"`, if no name is provided.
+The default model is `"nomic-embed-text-v1.5.Q8_0.gguf"`, if no name is provided.
 
 For available pretrained models please see the [Models Hub](https://sparknlp.org/models).
 

examples/python/llama.cpp/PromptAssember_with_AutoGGUFModel.ipynb

@@ -251,7 +251,7 @@
    "provenance": []
   },
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "sparknlp_dev",
    "language": "python",
    "name": "python3"
   },
@@ -264,7 +264,8 @@
    "mimetype": "text/x-python",
    "name": "python",
    "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3"
+   "pygments_lexer": "ipython3",
+   "version": "3.10.12"
   }
  },
  "nbformat": 4,

examples/python/llama.cpp/llama.cpp_in_Spark_NLP_AutoGGUFEmbeddings.ipynb

@@ -36,27 +36,27 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "--2024-11-02 13:42:45--  https://huggingface.co/nomic-ai/nomic-embed-text-v1.5-GGUF/resolve/main/nomic-embed-text-v1.5.f16.gguf?download=true\n",
+      "--2024-11-02 13:42:45--  https://huggingface.co/nomic-ai/nomic-embed-text-v1.5-GGUF/resolve/main/nomic-embed-text-v1.5.Q8_0.gguf?download=true\n",
       "Resolving huggingface.co (huggingface.co)... 3.160.39.87, 3.160.39.100, 3.160.39.99, ...\n",
       "Connecting to huggingface.co (huggingface.co)|3.160.39.87|:443... connected.\n",
       "HTTP request sent, awaiting response... 302 Found\n",
-      "Location: https://cdn-lfs-us-1.hf.co/repos/19/39/19396cd98fe8b02e39b1be815db29f6b251fee34fc5d6550db0b478083fdda2f/f7af6f66802f4df86eda10fe9bbcfc75c39562bed48ef6ace719a251cf1c2fdb?response-content-disposition=attachment%3B+filename*%3DUTF-8%27%27nomic-embed-text-v1.5.f16.gguf%3B+filename%3D%22nomic-embed-text-v1.5.f16.gguf%22%3B&Expires=1730810566&Policy=eyJTdGF0ZW1lbnQiOlt7IkNvbmRpdGlvbiI6eyJEYXRlTGVzc1RoYW4iOnsiQVdTOkVwb2NoVGltZSI6MTczMDgxMDU2Nn19LCJSZXNvdXJjZSI6Imh0dHBzOi8vY2RuLWxmcy11cy0xLmhmLmNvL3JlcG9zLzE5LzM5LzE5Mzk2Y2Q5OGZlOGIwMmUzOWIxYmU4MTVkYjI5ZjZiMjUxZmVlMzRmYzVkNjU1MGRiMGI0NzgwODNmZGRhMmYvZjdhZjZmNjY4MDJmNGRmODZlZGExMGZlOWJiY2ZjNzVjMzk1NjJiZWQ0OGVmNmFjZTcxOWEyNTFjZjFjMmZkYj9yZXNwb25zZS1jb250ZW50LWRpc3Bvc2l0aW9uPSoifV19&Signature=R7WUe1icdziIE4kS%7EHMcsyiySLOHJpkJ9lM2nC6EtIwPj6V12RRpjpDIyyv0%7EY9h32v2HDomyyNO6-Ry5UeIm3UjyESR9In3kmTKAqZX2zzlslTQMXDQghmLpIEQGgmh5-5RHrFgYPNxKmVICyQL1Vz9IgFQRMfdug6RBTGgmmXfLgksa9IU7TdvZcqvOb68HCdmv1hEt2U5vH4A9MF81ohMBqrvTb9389jzrlP1tZtNFb5wjNdZDmr57XIsvQRZB0ZDUIsMT1nc5QehNpWpX4jMLBSnkj1-oL9XN7%7EhAXDbB1mTH9kbrD3UUNKRm4%7ER-gVhegqsfirdSFi66sP3bg__&Key-Pair-Id=K24J24Z295AEI9 [following]\n",
-      "--2024-11-02 13:42:46--  https://cdn-lfs-us-1.hf.co/repos/19/39/19396cd98fe8b02e39b1be815db29f6b251fee34fc5d6550db0b478083fdda2f/f7af6f66802f4df86eda10fe9bbcfc75c39562bed48ef6ace719a251cf1c2fdb?response-content-disposition=attachment%3B+filename*%3DUTF-8%27%27nomic-embed-text-v1.5.f16.gguf%3B+filename%3D%22nomic-embed-text-v1.5.f16.gguf%22%3B&Expires=1730810566&Policy=eyJTdGF0ZW1lbnQiOlt7IkNvbmRpdGlvbiI6eyJEYXRlTGVzc1RoYW4iOnsiQVdTOkVwb2NoVGltZSI6MTczMDgxMDU2Nn19LCJSZXNvdXJjZSI6Imh0dHBzOi8vY2RuLWxmcy11cy0xLmhmLmNvL3JlcG9zLzE5LzM5LzE5Mzk2Y2Q5OGZlOGIwMmUzOWIxYmU4MTVkYjI5ZjZiMjUxZmVlMzRmYzVkNjU1MGRiMGI0NzgwODNmZGRhMmYvZjdhZjZmNjY4MDJmNGRmODZlZGExMGZlOWJiY2ZjNzVjMzk1NjJiZWQ0OGVmNmFjZTcxOWEyNTFjZjFjMmZkYj9yZXNwb25zZS1jb250ZW50LWRpc3Bvc2l0aW9uPSoifV19&Signature=R7WUe1icdziIE4kS%7EHMcsyiySLOHJpkJ9lM2nC6EtIwPj6V12RRpjpDIyyv0%7EY9h32v2HDomyyNO6-Ry5UeIm3UjyESR9In3kmTKAqZX2zzlslTQMXDQghmLpIEQGgmh5-5RHrFgYPNxKmVICyQL1Vz9IgFQRMfdug6RBTGgmmXfLgksa9IU7TdvZcqvOb68HCdmv1hEt2U5vH4A9MF81ohMBqrvTb9389jzrlP1tZtNFb5wjNdZDmr57XIsvQRZB0ZDUIsMT1nc5QehNpWpX4jMLBSnkj1-oL9XN7%7EhAXDbB1mTH9kbrD3UUNKRm4%7ER-gVhegqsfirdSFi66sP3bg__&Key-Pair-Id=K24J24Z295AEI9\n",
+      "Location: https://cdn-lfs-us-1.hf.co/repos/19/39/19396cd98fe8b02e39b1be815db29f6b251fee34fc5d6550db0b478083fdda2f/f7af6f66802f4df86eda10fe9bbcfc75c39562bed48ef6ace719a251cf1c2fdb?response-content-disposition=attachment%3B+filename*%3DUTF-8%27%27nomic-embed-text-v1.5.Q8_0.gguf%3B+filename%3D%22nomic-embed-text-v1.5.Q8_0.gguf%22%3B&Expires=1730810566&Policy=eyJTdGF0ZW1lbnQiOlt7IkNvbmRpdGlvbiI6eyJEYXRlTGVzc1RoYW4iOnsiQVdTOkVwb2NoVGltZSI6MTczMDgxMDU2Nn19LCJSZXNvdXJjZSI6Imh0dHBzOi8vY2RuLWxmcy11cy0xLmhmLmNvL3JlcG9zLzE5LzM5LzE5Mzk2Y2Q5OGZlOGIwMmUzOWIxYmU4MTVkYjI5ZjZiMjUxZmVlMzRmYzVkNjU1MGRiMGI0NzgwODNmZGRhMmYvZjdhZjZmNjY4MDJmNGRmODZlZGExMGZlOWJiY2ZjNzVjMzk1NjJiZWQ0OGVmNmFjZTcxOWEyNTFjZjFjMmZkYj9yZXNwb25zZS1jb250ZW50LWRpc3Bvc2l0aW9uPSoifV19&Signature=R7WUe1icdziIE4kS%7EHMcsyiySLOHJpkJ9lM2nC6EtIwPj6V12RRpjpDIyyv0%7EY9h32v2HDomyyNO6-Ry5UeIm3UjyESR9In3kmTKAqZX2zzlslTQMXDQghmLpIEQGgmh5-5RHrFgYPNxKmVICyQL1Vz9IgFQRMfdug6RBTGgmmXfLgksa9IU7TdvZcqvOb68HCdmv1hEt2U5vH4A9MF81ohMBqrvTb9389jzrlP1tZtNFb5wjNdZDmr57XIsvQRZB0ZDUIsMT1nc5QehNpWpX4jMLBSnkj1-oL9XN7%7EhAXDbB1mTH9kbrD3UUNKRm4%7ER-gVhegqsfirdSFi66sP3bg__&Key-Pair-Id=K24J24Z295AEI9 [following]\n",
+      "--2024-11-02 13:42:46--  https://cdn-lfs-us-1.hf.co/repos/19/39/19396cd98fe8b02e39b1be815db29f6b251fee34fc5d6550db0b478083fdda2f/f7af6f66802f4df86eda10fe9bbcfc75c39562bed48ef6ace719a251cf1c2fdb?response-content-disposition=attachment%3B+filename*%3DUTF-8%27%27nomic-embed-text-v1.5.Q8_0.gguf%3B+filename%3D%22nomic-embed-text-v1.5.Q8_0.gguf%22%3B&Expires=1730810566&Policy=eyJTdGF0ZW1lbnQiOlt7IkNvbmRpdGlvbiI6eyJEYXRlTGVzc1RoYW4iOnsiQVdTOkVwb2NoVGltZSI6MTczMDgxMDU2Nn19LCJSZXNvdXJjZSI6Imh0dHBzOi8vY2RuLWxmcy11cy0xLmhmLmNvL3JlcG9zLzE5LzM5LzE5Mzk2Y2Q5OGZlOGIwMmUzOWIxYmU4MTVkYjI5ZjZiMjUxZmVlMzRmYzVkNjU1MGRiMGI0NzgwODNmZGRhMmYvZjdhZjZmNjY4MDJmNGRmODZlZGExMGZlOWJiY2ZjNzVjMzk1NjJiZWQ0OGVmNmFjZTcxOWEyNTFjZjFjMmZkYj9yZXNwb25zZS1jb250ZW50LWRpc3Bvc2l0aW9uPSoifV19&Signature=R7WUe1icdziIE4kS%7EHMcsyiySLOHJpkJ9lM2nC6EtIwPj6V12RRpjpDIyyv0%7EY9h32v2HDomyyNO6-Ry5UeIm3UjyESR9In3kmTKAqZX2zzlslTQMXDQghmLpIEQGgmh5-5RHrFgYPNxKmVICyQL1Vz9IgFQRMfdug6RBTGgmmXfLgksa9IU7TdvZcqvOb68HCdmv1hEt2U5vH4A9MF81ohMBqrvTb9389jzrlP1tZtNFb5wjNdZDmr57XIsvQRZB0ZDUIsMT1nc5QehNpWpX4jMLBSnkj1-oL9XN7%7EhAXDbB1mTH9kbrD3UUNKRm4%7ER-gVhegqsfirdSFi66sP3bg__&Key-Pair-Id=K24J24Z295AEI9\n",
       "Resolving cdn-lfs-us-1.hf.co (cdn-lfs-us-1.hf.co)... 18.66.2.2, 18.66.2.116, 18.66.2.98, ...\n",
       "Connecting to cdn-lfs-us-1.hf.co (cdn-lfs-us-1.hf.co)|18.66.2.2|:443... connected.\n",
       "HTTP request sent, awaiting response... 200 OK\n",
       "Length: 274290560 (262M) [application/octet-stream]\n",
-      "Saving to: ‘nomic-embed-text-v1.5.f16.gguf’\n",
+      "Saving to: ‘nomic-embed-text-v1.5.Q8_0.gguf’\n",
       "\n",
       "nomic-embed-text-v1 100%[===================>] 261.58M  23.8MB/s    in 10s     \n",
       "\n",
-      "2024-11-02 13:42:56 (24.9 MB/s) - ‘nomic-embed-text-v1.5.f16.gguf’ saved [274290560/274290560]\n",
+      "2024-11-02 13:42:56 (24.9 MB/s) - ‘nomic-embed-text-v1.5.Q8_0.gguf’ saved [274290560/274290560]\n",
       "\n"
      ]
     }
    ],
    "source": [
-    "EXPORT_PATH = \"nomic-embed-text-v1.5.f16.gguf\"\n",
+    "EXPORT_PATH = \"nomic-embed-text-v1.5.Q8_0.gguf\"\n",
     "! wget \"https://huggingface.co/nomic-ai/nomic-embed-text-v1.5-GGUF/resolve/main/{EXPORT_PATH}?download=true\" -O  {EXPORT_PATH}"
    ]
   },
@@ -158,7 +158,7 @@
     }
    ],
    "source": [
-    "autoGGUFEmbeddings.write().overwrite().save(f\"nomic-embed-text-v1.5.f16.gguf_spark_nlp\")"
+    "autoGGUFEmbeddings.write().overwrite().save(f\"nomic-embed-text-v1.5.Q8_0.gguf_spark_nlp\")"
    ]
   },
   {
@@ -197,12 +197,12 @@
      "text": [
       "total 267872\n",
       "drwxr-xr-x 2 root root      4096 Nov  2 13:48 metadata\n",
-      "-rwxrwxr-x 1 root root 274290560 Nov  2 13:48 nomic-embed-text-v1.5.f16.gguf\n"
+      "-rwxrwxr-x 1 root root 274290560 Nov  2 13:48 nomic-embed-text-v1.5.Q8_0.gguf\n"
      ]
     }
    ],
    "source": [
-    "! ls -l nomic-embed-text-v1.5.f16.gguf_spark_nlp/"
+    "! ls -l nomic-embed-text-v1.5.Q8_0.gguf_spark_nlp/"
    ]
   },
   {
@@ -221,11 +221,11 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "24/11/02 13:48:57 WARN SparkContext: The path /home/root/Workspace/scala/spark-nlp/examples/python/llama.cpp/nomic-embed-text-v1.5.f16.gguf_spark_nlp/nomic-embed-text-v1.5.f16.gguf has been added already. Overwriting of added paths is not supported in the current version.\n",
+      "24/11/02 13:48:57 WARN SparkContext: The path /home/root/Workspace/scala/spark-nlp/examples/python/llama.cpp/nomic-embed-text-v1.5.Q8_0.gguf_spark_nlp/nomic-embed-text-v1.5.Q8_0.gguf has been added already. Overwriting of added paths is not supported in the current version.\n",
       "24/11/02 13:48:57 WARN DAGScheduler: Broadcasting large task binary with size 1028.0 KiB\n",
       "24/11/02 13:48:57 WARN DAGScheduler: Broadcasting large task binary with size 1028.0 KiB\n",
       "24/11/02 13:48:57 WARN DAGScheduler: Broadcasting large task binary with size 1028.0 KiB\n",
-      "llama_model_loader: loaded meta data with 22 key-value pairs and 112 tensors from /tmp/spark-6de50aee-1059-4698-98e2-db9d68663467/userFiles-932de0e7-9a8f-41f5-9aaf-94bb7406df74/nomic-embed-text-v1.5.f16.gguf (version GGUF V3 (latest))\n",
+      "llama_model_loader: loaded meta data with 22 key-value pairs and 112 tensors from /tmp/spark-6de50aee-1059-4698-98e2-db9d68663467/userFiles-932de0e7-9a8f-41f5-9aaf-94bb7406df74/nomic-embed-text-v1.5.Q8_0.gguf (version GGUF V3 (latest))\n",
       "llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.\n",
       "llama_model_loader: - kv   0:                       general.architecture str              = nomic-bert\n",
       "llama_model_loader: - kv   1:                               general.name str              = nomic-embed-text-v1.5\n",
@@ -390,7 +390,7 @@
     "\n",
     "document_assembler = DocumentAssembler().setInputCol(\"text\").setOutputCol(\"document\")\n",
     "\n",
-    "autoGGUFEmbeddings = AutoGGUFEmbeddings.load(\"nomic-embed-text-v1.5.f16.gguf_spark_nlp\")\n",
+    "autoGGUFEmbeddings = AutoGGUFEmbeddings.load(\"nomic-embed-text-v1.5.Q8_0.gguf_spark_nlp\")\n",
     "\n",
     "pipeline = Pipeline().setStages([document_assembler, autoGGUFEmbeddings])\n",
     "\n",

python/sparknlp/annotator/embeddings/auto_gguf_embeddings.py

@@ -32,7 +32,7 @@ class AutoGGUFEmbeddings(AnnotatorModel, HasBatchedAnnotate):
     ...     .setInputCols(["document"]) \\
     ...     .setOutputCol("embeddings")
 
-    The default model is ``"nomic-embed-text-v1.5.f16.gguf"``, if no name is provided.
+    The default model is ``"nomic-embed-text-v1.5.Q8_0.gguf"``, if no name is provided.
 
     For extended examples of usage, see the
     `AutoGGUFEmbeddingsTest <https://github.com/JohnSnowLabs/spark-nlp/tree/master/src/test/scala/com/johnsnowlabs/nlp/embeddings/AutoGGUFEmbeddingsTest.scala>`__
@@ -513,13 +513,13 @@ def loadSavedModel(folder, spark_session):
         return AutoGGUFEmbeddings(java_model=jModel)
 
     @staticmethod
-    def pretrained(name="nomic-embed-text-v1.5.f16.gguf", lang="en", remote_loc=None):
+    def pretrained(name="nomic-embed-text-v1.5.Q8_0.gguf", lang="en", remote_loc=None):
         """Downloads and loads a pretrained model.
 
         Parameters
         ----------
         name : str, optional
-            Name of the pretrained model, by default "nomic-embed-text-v1.5.f16.gguf"
+            Name of the pretrained model, by default "nomic-embed-text-v1.5.Q8_0.gguf"
         lang : str, optional
             Language of the pretrained model, by default "en"
         remote_loc : str, optional

python/test/annotator/embeddings/auto_gguf_embeddings_test.py

@@ -42,10 +42,7 @@ def setUp(self):
 
     def runTest(self):
         model = (
-            # AutoGGUFEmbeddings.pretrained()
-            AutoGGUFEmbeddings.loadSavedModel(
-                "models/nomic-embed-text-v1.5.f16.gguf", SparkContextForTest.spark
-            )
+            AutoGGUFEmbeddings.pretrained()
             .setInputCols("document")
             .setOutputCol("embeddings")
             .setBatchSize(4)
@@ -88,7 +85,7 @@ def runTest(self):
         model = (
             # AutoGGUFEmbeddings.pretrained()
             AutoGGUFEmbeddings.loadSavedModel(
-                "models/nomic-embed-text-v1.5.f16.gguf", SparkContextForTest.spark
+                "models/nomic-embed-text-v1.5.Q8_0.gguf", SparkContextForTest.spark
             )
             .setInputCols("document")
             .setOutputCol("embeddings")

src/main/scala/com/johnsnowlabs/nlp/embeddings/AutoGGUFEmbeddings.scala

@@ -38,7 +38,7 @@ import org.apache.spark.sql.SparkSession
   *   .setInputCols("document")
   *   .setOutputCol("embeddings")
   * }}}
-  * The default model is `"nomic-embed-text-v1.5.f16.gguf"`, if no name is provided.
+  * The default model is `"nomic-embed-text-v1.5.Q8_0.gguf"`, if no name is provided.
   *
   * For available pretrained models please see the [[https://sparknlp.org/models Models Hub]].
   *
@@ -191,7 +191,7 @@ class AutoGGUFEmbeddings(override val uid: String)
 trait ReadablePretrainedAutoGGUFEmbeddings
     extends ParamsAndFeaturesReadable[AutoGGUFEmbeddings]
     with HasPretrained[AutoGGUFEmbeddings] {
-  override val defaultModelName: Some[String] = Some("nomic-embed-text-v1.5.f16.gguf")
+  override val defaultModelName: Some[String] = Some("nomic-embed-text-v1.5.Q8_0.gguf")
   override val defaultLang: String = "en"
 
   /** Java compliant-overrides */

src/test/scala/com/johnsnowlabs/nlp/embeddings/AutoGGUFEmbeddingsTestSpec.scala

@@ -23,9 +23,9 @@ class AutoGGUFEmbeddingsTestSpec extends AnyFlatSpec {
     "The sun is " //
   ).toDF("text").repartition(1)
 
-  // nomic-embed-text-v1.5.f16.gguf
+  // nomic-embed-text-v1.5.Q8_0.gguf
   def model(poolingType: String): AutoGGUFEmbeddings = AutoGGUFEmbeddings
-    .loadSavedModel("models/nomic-embed-text-v1.5.f16.gguf", ResourceHelper.spark)
+    .pretrained()
     .setInputCols("document")
     .setOutputCol("embeddings")
     .setBatchSize(4)