Merge branch 'main' into feature/rename_more_udfs

Author: MarleneKress79789

doc/changes/changes_1.0.0.md

@@ -16,7 +16,7 @@ The 'token_conn' parameter was removed from the UDF calls. You can now call the
 as follows (Example case for the filling mask udf):
 
 ```sql
-SELECT TE_FILLING_MASK_UDF(
+SELECT AI_FILL_MASK_EXTENDED(
     device_id,
     bucketfs_conn,
     sub_dir,

doc/developer_guide/developer_guide.md

@@ -118,7 +118,7 @@ Users should implement the following methods in the UDF class
 that extends the `BaseModel UDF`:
  - `extract_unique_param_based_dataframes` : Even if the data in a given
 dataframe all have the same model, there might be differences within the given
-dataframe with different model parameters (e.g. _top_k_ parameter in [FillingMaskUDF](../../exasol_transformers_extension/udfs/models/filling_mask_udf.py)).
+dataframe with different model parameters (e.g. _top_k_ parameter in [AIFillMaskExtendedUDF](../../exasol_transformers_extension/udfs/models/ai_fill_mask_extended_udf.py)).
 This method is responsible for extracting unique dataframes which share both the
 same model and model parameters.
  - `execute_prediction` : Performs prediction on a given text list using

doc/user_guide/invoke_models.md

@@ -4,8 +4,9 @@ We provide 7 prediction UDFs in the Transformers Extension package. Each perform
 
 ### Table of Contents
 
-  * [Sequence Classification for Single Text UDF](#sequence-classification-for-single-text-udf)
-  * [Sequence Classification for Text Pair UDF](#sequence-classification-for-text-pair-udf)
+
+  * [AI Custom Classify Extended](#ai-custom-classify-extended)
+  * [AI Entailment Extended](#ai-entailment-extended)
   * [AI Answer Extended](#ai-answer-extended)
   * [Masked Language Modelling UDF](#masked-language-modelling-udf)
   * [AI Complete Extended](#ai-complete-extended)
@@ -14,14 +15,14 @@ We provide 7 prediction UDFs in the Transformers Extension package. Each perform
   * [AI Classify Extended](#ai-classify-extended)
 
 
-### Sequence Classification for Single Text UDF
+### AI Custom Classify Extended
 
 This UDF classifies the given text according to a given number of classes of the specified model.
 
 Example usage:
 
 ```sql
-SELECT TE_SEQUENCE_CLASSIFICATION_SINGLE_TEXT_UDF(
+SELECT AI_CUSTOM_CLASSIFY_EXTENDED(
     device_id,
     bucketfs_conn,
     sub_dir,
@@ -54,14 +55,14 @@ Example:
 | conn_name     | dir/    | model_name | text 1    | ALL          | label_2 | 0.23  | 2    | None          |
 | ...           | ...     | ...        | ...       | ...          | ...     | ...   | ...  | ...           |
 
-### Sequence Classification for Text Pair UDF
+### AI Entailment Extended
 
-This UDF takes two input sequences and compares them. Among other things, it can be used to determine if two sequences are paraphrases of each other.
+This UDF takes two input texts and compares them. Among other things, it can be used to determine if two texts are paraphrases of each other.
 
 Example usage:
 
 ```sql
-SELECT TE_SEQUENCE_CLASSIFICATION_TEXT_PAIR_UDF(
+SELECT AI_ENTAILMENT_EXTENDED(
     device_id,
     bucketfs_conn,
     sub_dir,
@@ -144,12 +145,16 @@ Example:
 
 ### Masked Language Modelling UDF
 
-This UDF is responsible for masking tokens in a given text with a masking token, and then filling that masks with appropriate tokens. The masking token of this UDF is ```<mask>```.
+This UDF needs to be given an input text containing the ```<mask>``` token. It can then 
+replace these masks with appropriate tokens. 
+I.E the input text could be "<mask> is the best database Software for Machine 
+Learning Enthusiasts.", resulting in an output like "Exasol is the best database 
+Software for Machine Learning Enthusiasts."
 
 Example usage:
 
 ```sql
-SELECT TE_FILLING_MASK_UDF(
+SELECT AI_FILL_MASK_EXTENDED(
     device_id,
     bucketfs_conn,
     sub_dir,

doc/user_guide/user_guide.md

@@ -8,8 +8,9 @@ The extension provides two types of UDFs:
 
 * Utility UDFs: UDFs which deal with installation and deletion of pretrained Transformers models in the Exasol BucketFS.
 * Prediction UDFs: These are a group of UDFs for each supported task. Each of them uses the downloaded pre-trained model and performs prediction. These are the supported tasks:
-   1. Sequence Classification for Single Text
-   2. Sequence Classification for Text Pair
+
+   1. AI Custom Classify Extended
+   2. AI Entailment Extended
    3. AI Answer Extended
    4. Masked Language Modelling
    5. AI Complete Extended

exasol_transformers_extension/deployment/constants.py

@@ -9,10 +9,10 @@
 UDF_CALL_TEMPLATES = {
     "model_downloader_udf_call.py": "model_downloader_udf.jinja.sql",
     "ls_models_udf.py": "ls_models_udf.jinja.sql",
-    "sequence_classification_single_text_udf_call.py": "sequence_classification_single_text_udf.jinja.sql",
-    "sequence_classification_text_pair_udf_call.py": "sequence_classification_text_pair_udf.jinja.sql",
+    "ai_custom_classify_extended_udf_call.py": "ai_custom_classify_extended_udf.jinja.sql",
+    "ai_entailment_extended_udf_call.py": "ai_entailment_extended_udf.jinja.sql",
     "ai_answer_extended_udf_call.py": "ai_answer_extended_udf.jinja.sql",
-    "filling_mask_udf_call.py": "filling_mask_udf.jinja.sql",
+    "ai_fill_mask_extended_udf_call.py": "ai_fill_mask_extended_udf.sql",
     "ai_complete_extended_udf_call.py": "ai_complete_extended_udf.jinja.sql",
     "ai_translate_extended_udf_call.py": "ai_translate_extended_udf.jinja.sql",
     "delete_model_udf_call.py": "delete_model_udf.jinja.sql",

…classification_single_text_udf.jinja.sql …i_custom_classify_extended_udf.jinja.sqlexasol_transformers_extension/resources/templates/sequence_classification_single_text_udf.jinja.sql renamed to exasol_transformers_extension/resources/templates/ai_custom_classify_extended_udf.jinja.sql exasol_transformers_extension/resources/templates/sequence_classification_single_text_udf.jinja.sql renamed to exasol_transformers_extension/resources/templates/ai_custom_classify_extended_udf.jinja.sql

@@ -1,4 +1,4 @@
-CREATE OR REPLACE {{ language_alias }} SET SCRIPT "TE_SEQUENCE_CLASSIFICATION_SINGLE_TEXT_UDF"(
+CREATE OR REPLACE {{ language_alias }} SET SCRIPT "AI_CUSTOM_CLASSIFY_EXTENDED"(
     device_id INTEGER,
     bucketfs_conn VARCHAR(2000000),
     sub_dir VARCHAR(2000000),

…e_classification_text_pair_udf.jinja.sql …tes/ai_entailment_extended_udf.jinja.sqlexasol_transformers_extension/resources/templates/sequence_classification_text_pair_udf.jinja.sql renamed to exasol_transformers_extension/resources/templates/ai_entailment_extended_udf.jinja.sql exasol_transformers_extension/resources/templates/sequence_classification_text_pair_udf.jinja.sql renamed to exasol_transformers_extension/resources/templates/ai_entailment_extended_udf.jinja.sql

@@ -1,4 +1,4 @@
-CREATE OR REPLACE {{ language_alias }} SET SCRIPT "TE_SEQUENCE_CLASSIFICATION_TEXT_PAIR_UDF"(
+CREATE OR REPLACE {{ language_alias }} SET SCRIPT "AI_ENTAILMENT_EXTENDED"(
     device_id INTEGER,
     bucketfs_conn VARCHAR(2000000),
     sub_dir VARCHAR(2000000),

…ces/templates/filling_mask_udf.jinja.sql …/templates/ai_fill_mask_extended_udf.sqlexasol_transformers_extension/resources/templates/filling_mask_udf.jinja.sql renamed to exasol_transformers_extension/resources/templates/ai_fill_mask_extended_udf.sql exasol_transformers_extension/resources/templates/filling_mask_udf.jinja.sql renamed to exasol_transformers_extension/resources/templates/ai_fill_mask_extended_udf.sql

@@ -1,4 +1,4 @@
-CREATE OR REPLACE {{ language_alias }} SET SCRIPT "TE_FILLING_MASK_UDF"(
+CREATE OR REPLACE {{ language_alias }} SET SCRIPT "AI_FILL_MASK_EXTENDED"(
     device_id INTEGER,
     bucketfs_conn VARCHAR(2000000),
     sub_dir VARCHAR(2000000),

exasol_transformers_extension/udfs/callers/ai_custom_classify_extended_udf_call.py

@@ -0,0 +1,9 @@
+from exasol_transformers_extension.udfs.models.ai_custom_classify_extended_udf import (
+    AiCustomClassifyUDF,
+)
+
+udf = AiCustomClassifyUDF(exa)
+
+
+def run(ctx):
+    return udf.run(ctx)

exasol_transformers_extension/udfs/callers/ai_entailment_extended_udf_call.py

@@ -0,0 +1,9 @@
+from exasol_transformers_extension.udfs.models.ai_entailment_extended_udf import (
+    AiEntailmentExtendedUDF,
+)
+
+udf = AiEntailmentExtendedUDF(exa)
+
+
+def run(ctx):
+    return udf.run(ctx)