added splits to new cards and updated split_annotators to update furt…

…her cards to be added Signed-off-by: dafnapension <[email protected]>
IBM · Feb 6, 2025 · aa4eb0d · aa4eb0d
1 parent c747353
commit aa4eb0d
Show file tree

Hide file tree

Showing 37 changed files with 226 additions and 71 deletions.
diff --git a/prepare/cards/mlsum.py b/prepare/cards/mlsum.py
@@ -18,7 +18,12 @@
 
 for lang in langs:
     card = TaskCard(
-        loader=LoadHF(path="mlsum", name=lang),
+        loader=LoadHF(
+            path="mlsum",
+            name=lang,
+            splits=["train", "validation", "test"],
+            streaming=True,
+        ),
         preprocess_steps=[
             Rename(field_to_field={"text": "document"}),
             Wrap(field="summary", inside="list", to_field="summaries"),

diff --git a/prepare/cards/numeric_nlg.py b/prepare/cards/numeric_nlg.py
@@ -10,7 +10,7 @@
 from unitxt.test_utils.card import test_card
 
 card = TaskCard(
-    loader=LoadHF(path="kasnerz/numericnlg"),
+    loader=LoadHF(path="kasnerz/numericnlg", splits=["train", "validation", "test"]),
     preprocess_steps=[
         Set(
             fields={

diff --git a/prepare/cards/ragbench.py b/prepare/cards/ragbench.py
@@ -25,6 +25,7 @@
         loader=LoadHF(
             path="rungalileo/ragbench",
             name=subset,
+            splits=["train", "test", "validation"],
         ),
         preprocess_steps=[
             Copy(field="documents", to_field="contexts"),
@@ -51,6 +52,7 @@
         loader=LoadHF(
             path="rungalileo/ragbench",
             name=subset,
+            splits=["train", "test", "validation"],
         ),
         preprocess_steps=[
             Set({"context_type": "documents"}),

diff --git a/prepare/cards/trec.py b/prepare/cards/trec.py
@@ -81,7 +81,7 @@
 classes = [expand_label_text[label] for label in classlabels.names]
 
 card = TaskCard(
-    loader=LoadHF(path=dataset_name),
+    loader=LoadHF(path=dataset_name, splits=["train", "test"]),
     preprocess_steps=[
         Shuffle(page_size=sys.maxsize),
         SplitRandomMix(

diff --git a/src/unitxt/catalog/cards/mlsum/de.json b/src/unitxt/catalog/cards/mlsum/de.json
@@ -3,7 +3,13 @@
     "loader": {
         "__type__": "load_hf",
         "path": "mlsum",
-        "name": "de"
+        "name": "de",
+        "splits": [
+            "train",
+            "validation",
+            "test"
+        ],
+        "streaming": true
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/mlsum/es.json b/src/unitxt/catalog/cards/mlsum/es.json
@@ -3,7 +3,13 @@
     "loader": {
         "__type__": "load_hf",
         "path": "mlsum",
-        "name": "es"
+        "name": "es",
+        "splits": [
+            "train",
+            "validation",
+            "test"
+        ],
+        "streaming": true
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/mlsum/fr.json b/src/unitxt/catalog/cards/mlsum/fr.json
@@ -3,7 +3,13 @@
     "loader": {
         "__type__": "load_hf",
         "path": "mlsum",
-        "name": "fr"
+        "name": "fr",
+        "splits": [
+            "train",
+            "validation",
+            "test"
+        ],
+        "streaming": true
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/mlsum/ru.json b/src/unitxt/catalog/cards/mlsum/ru.json
@@ -3,7 +3,13 @@
     "loader": {
         "__type__": "load_hf",
         "path": "mlsum",
-        "name": "ru"
+        "name": "ru",
+        "splits": [
+            "train",
+            "validation",
+            "test"
+        ],
+        "streaming": true
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/mlsum/tu.json b/src/unitxt/catalog/cards/mlsum/tu.json
@@ -3,7 +3,13 @@
     "loader": {
         "__type__": "load_hf",
         "path": "mlsum",
-        "name": "tu"
+        "name": "tu",
+        "splits": [
+            "train",
+            "validation",
+            "test"
+        ],
+        "streaming": true
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/numeric_nlg.json b/src/unitxt/catalog/cards/numeric_nlg.json
@@ -2,7 +2,12 @@
     "__type__": "task_card",
     "loader": {
         "__type__": "load_hf",
-        "path": "kasnerz/numericnlg"
+        "path": "kasnerz/numericnlg",
+        "splits": [
+            "train",
+            "validation",
+            "test"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/covidqa.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/covidqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "covidqa"
+        "name": "covidqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/cuad.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/cuad.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "cuad"
+        "name": "cuad",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/delucionqa.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/delucionqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "delucionqa"
+        "name": "delucionqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/emanual.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/emanual.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "emanual"
+        "name": "emanual",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/expertqa.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/expertqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "expertqa"
+        "name": "expertqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/finqa.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/finqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "finqa"
+        "name": "finqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/hagrid.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/hagrid.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "hagrid"
+        "name": "hagrid",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/hotpotqa.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/hotpotqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "hotpotqa"
+        "name": "hotpotqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/msmarco.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/msmarco.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "msmarco"
+        "name": "msmarco",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/pubmedqa.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/pubmedqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "pubmedqa"
+        "name": "pubmedqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/tatqa.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/tatqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "tatqa"
+        "name": "tatqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/rag/response_generation/ragbench/techqa.json b/src/unitxt/catalog/cards/rag/response_generation/ragbench/techqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "techqa"
+        "name": "techqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/ragbench/covidqa.json b/src/unitxt/catalog/cards/ragbench/covidqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "covidqa"
+        "name": "covidqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/ragbench/cuad.json b/src/unitxt/catalog/cards/ragbench/cuad.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "cuad"
+        "name": "cuad",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/ragbench/delucionqa.json b/src/unitxt/catalog/cards/ragbench/delucionqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "delucionqa"
+        "name": "delucionqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/ragbench/emanual.json b/src/unitxt/catalog/cards/ragbench/emanual.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "emanual"
+        "name": "emanual",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/ragbench/expertqa.json b/src/unitxt/catalog/cards/ragbench/expertqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "expertqa"
+        "name": "expertqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {

diff --git a/src/unitxt/catalog/cards/ragbench/finqa.json b/src/unitxt/catalog/cards/ragbench/finqa.json
@@ -3,7 +3,12 @@
     "loader": {
         "__type__": "load_hf",
         "path": "rungalileo/ragbench",
-        "name": "finqa"
+        "name": "finqa",
+        "splits": [
+            "train",
+            "test",
+            "validation"
+        ]
     },
     "preprocess_steps": [
         {