Add files using upload-large-folder tool

lm-evaluation-harness/lm_eval/tasks/afrixnli/lai prompt/translate/afrixnli_manual_translate_yor.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: yor
+include: afrixnli_manual_translate_yaml
+task: afrixnli_manual_translate_yor

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_fra.yaml

@@ -0,0 +1,15 @@
+# Generated by utils.py
+dataset_name: fra
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_fra_prompt_1

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_ibo.yaml

@@ -0,0 +1,15 @@
+# Generated by utils.py
+dataset_name: ibo
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_ibo_prompt_1

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_sot.yaml

@@ -0,0 +1,15 @@
+# Generated by utils.py
+dataset_name: sot
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_sot_prompt_1

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_swa.yaml

@@ -0,0 +1,15 @@
+# Generated by utils.py
+dataset_name: swa
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_swa_prompt_1

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_twi.yaml

@@ -0,0 +1,15 @@
+# Generated by utils.py
+dataset_name: twi
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_twi_prompt_1

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_yaml

@@ -0,0 +1,27 @@
+tag: afrixnli_tt_tasks
+dataset_path: masakhane/afrixnli-translate-test
+dataset_name: null
+output_type: multiple_choice
+test_split: test
+fewshot_split: test
+doc_to_target: !function utils.doc_to_target
+doc_to_choice:
+  - "entailment"
+  - "neutral"
+  - "contradiction"
+should_decontaminate: true
+doc_to_decontamination_query: premise
+metric_list:
+  - metric: f1
+    aggregation: !function utils.weighted_f1_score
+    average: weighted
+    higher_is_better: True
+    ignore_case: true
+    ignore_punctuation: true
+  - metric: acc
+    aggregation: mean
+    higher_is_better: true
+    ignore_case: true
+    ignore_punctuation: true
+metadata:
+  version: 1.0

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_1/utils.py

@@ -0,0 +1,19 @@
+from lm_eval.utils import weighted_f1_score
+
+
+def doc_to_text(doc):
+    output = """Please identify whether the premise entails or contradicts the hypothesis in the following premise
+    and hypothesis. The answer should be exact entailment, contradiction, or neutral.
+
+    Premise: {premise}
+    Hypothesis: {hypothesis}
+
+    Is it entailment, contradiction, or neutral?"""
+
+    text = output.format(premise=doc["premise"], hypothesis=doc["hypothesis"])
+    return text
+
+
+def doc_to_target(doc):
+    replacements = {0: "entailment", 1: "neutral", 2: "contradiction"}
+    return replacements[doc["label"]]

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_ewe.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: ewe
+include: afrixnli_translate_yaml
+task: afrixnli_translate_ewe_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_ibo.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: ibo
+include: afrixnli_translate_yaml
+task: afrixnli_translate_ibo_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_kin.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: kin
+include: afrixnli_translate_yaml
+task: afrixnli_translate_kin_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_lin.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: lin
+include: afrixnli_translate_yaml
+task: afrixnli_translate_lin_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_lug.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: lug
+include: afrixnli_translate_yaml
+task: afrixnli_translate_lug_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_orm.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: orm
+include: afrixnli_translate_yaml
+task: afrixnli_translate_orm_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_sna.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: sna
+include: afrixnli_translate_yaml
+task: afrixnli_translate_sna_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_sot.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: sot
+include: afrixnli_translate_yaml
+task: afrixnli_translate_sot_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_swa.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: swa
+include: afrixnli_translate_yaml
+task: afrixnli_translate_swa_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_twi.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: twi
+include: afrixnli_translate_yaml
+task: afrixnli_translate_twi_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_wol.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: wol
+include: afrixnli_translate_yaml
+task: afrixnli_translate_wol_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_xho.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: xho
+include: afrixnli_translate_yaml
+task: afrixnli_translate_xho_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_yaml

@@ -0,0 +1,31 @@
+tag: afrixnli_tt_tasks
+dataset_path: masakhane/afrixnli-translate-test
+dataset_name: null
+output_type: multiple_choice
+test_split: test
+fewshot_split: test
+doc_to_text: "{{premise}}\nQuestion: {{hypothesis}} True, False, or Neither?\nAnswer:"
+# True = entailment
+# False = contradiction
+# Neither = neutral
+doc_to_target: !function utils.doc_to_target
+doc_to_choice:
+  - "True"
+  - "Neither"
+  - "False"
+should_decontaminate: true
+doc_to_decontamination_query: premise
+metric_list:
+  - metric: f1
+    aggregation: !function utils.weighted_f1_score
+    average: weighted
+    higher_is_better: True
+    ignore_case: true
+    ignore_punctuation: true
+  - metric: acc
+    aggregation: mean
+    higher_is_better: true
+    ignore_case: true
+    ignore_punctuation: true
+metadata:
+  version: 1.0

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_yor.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: yor
+include: afrixnli_translate_yaml
+task: afrixnli_translate_yor_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/afrixnli_translate_zul.yaml

@@ -0,0 +1,4 @@
+# Generated by utils.py
+dataset_name: zul
+include: afrixnli_translate_yaml
+task: afrixnli_translate_zul_prompt_2

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_2/utils.py

@@ -0,0 +1,6 @@
+from lm_eval.utils import weighted_f1_score
+
+
+def doc_to_target(doc):
+    replacements = {0: "True", 1: "Neither", 2: "False"}
+    return replacements[doc["label"]]

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_3/afrixnli_translate_amh.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: amh
+doc_to_text: "Given the following premise and hypothesis in Amharic, identify if the\
+  \ premise entails, contradicts, or is neutral towards the hypothesis. Please respond\
+  \ with exact 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}}\
+  \ \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_amh_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_3/afrixnli_translate_hau.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: hau
+doc_to_text: "Given the following premise and hypothesis in Hausa, identify if the\
+  \ premise entails, contradicts, or is neutral towards the hypothesis. Please respond\
+  \ with exact 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}}\
+  \ \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_hau_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_3/afrixnli_translate_ibo.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: ibo
+doc_to_text: "Given the following premise and hypothesis in Igbo, identify if the\
+  \ premise entails, contradicts, or is neutral towards the hypothesis. Please respond\
+  \ with exact 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}}\
+  \ \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_ibo_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_3/afrixnli_translate_kin.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: kin
+doc_to_text: "Given the following premise and hypothesis in Kinyarwanda, identify\
+  \ if the premise entails, contradicts, or is neutral towards the hypothesis. Please\
+  \ respond with exact 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}}\
+  \ \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_kin_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_3/afrixnli_translate_lin.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: lin
+doc_to_text: "Given the following premise and hypothesis in Lingala, identify if the\
+  \ premise entails, contradicts, or is neutral towards the hypothesis. Please respond\
+  \ with exact 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}}\
+  \ \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_lin_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_3/afrixnli_translate_lug.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: lug
+doc_to_text: "Given the following premise and hypothesis in Luganda, identify if the\
+  \ premise entails, contradicts, or is neutral towards the hypothesis. Please respond\
+  \ with exact 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}}\
+  \ \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_lug_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_3/afrixnli_translate_orm.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: orm
+doc_to_text: "Given the following premise and hypothesis in Oromo, identify if the\
+  \ premise entails, contradicts, or is neutral towards the hypothesis. Please respond\
+  \ with exact 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}}\
+  \ \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_orm_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_3/afrixnli_translate_sot.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: sot
+doc_to_text: "Given the following premise and hypothesis in Sesotho, identify if the\
+  \ premise entails, contradicts, or is neutral towards the hypothesis. Please respond\
+  \ with exact 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}}\
+  \ \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_sot_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_3/afrixnli_translate_swa.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: swa
+doc_to_text: "Given the following premise and hypothesis in Swahili, identify if the\
+  \ premise entails, contradicts, or is neutral towards the hypothesis. Please respond\
+  \ with exact 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}}\
+  \ \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_swa_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_3/afrixnli_translate_wol.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: wol
+doc_to_text: "Given the following premise and hypothesis in Wolof, identify if the\
+  \ premise entails, contradicts, or is neutral towards the hypothesis. Please respond\
+  \ with exact 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}}\
+  \ \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_wol_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_4/afrixnli_translate_fra.yaml

@@ -0,0 +1,9 @@
+# Generated by utils.py
+dataset_name: fra
+doc_to_text: "You are an expert in Natural Language Inference (NLI) specializing in\
+  \ the French language.\nAnalyze the premise and hypothesis given in French, and\
+  \ determine the relationship between them.\n Respond with one of the following options:\
+  \ 'entailment', 'contradiction', or 'neutral'. \n\nPremise: {{premise}} \nHypothesis:\
+  \ {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_fra_prompt_4

lm-evaluation-harness/lm_eval/tasks/afrixnli/translate/prompt_4/afrixnli_translate_orm.yaml

@@ -0,0 +1,8 @@
+# Generated by utils.py
+dataset_name: orm
+doc_to_text: "You are an expert in Natural Language Inference (NLI) specializing in\
+  \ the Oromo language.\nAnalyze the premise and hypothesis given in Oromo, and determine\
+  \ the relationship between them.\n Respond with one of the following options: 'entailment',\
+  \ 'contradiction', or 'neutral'. \n\nPremise: {{premise}} \nHypothesis: {{hypothesis}}"
+include: afrixnli_translate_yaml
+task: afrixnli_translate_orm_prompt_4

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_2/xx.py

@@ -0,0 +1,5 @@
+from datasets import load_dataset
+
+
+data = load_dataset("HausaNLP/AfriSenti-Twitter", "yor", trust_remote_code=True)
+print(data)

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_3/afrisenti_amh.yaml

@@ -0,0 +1,7 @@
+# Generated by utils.py
+dataset_name: amh
+doc_to_text: "You are an assistant able to detect sentiments in tweets. \n\nGiven\
+  \ the sentiment labels Neutral, Positive or Negative; what is the sentiment of the\
+  \ Amharic statement below? Return only the labels. \n\ntext: {{tweet}} \nlabel:"
+include: afrisenti
+task: afrisenti_amh_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_3/afrisenti_ibo.yaml

@@ -0,0 +1,7 @@
+# Generated by utils.py
+dataset_name: ibo
+doc_to_text: "You are an assistant able to detect sentiments in tweets. \n\nGiven\
+  \ the sentiment labels Neutral, Positive or Negative; what is the sentiment of the\
+  \ Igbo statement below? Return only the labels. \n\ntext: {{tweet}} \nlabel:"
+include: afrisenti
+task: afrisenti_ibo_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_3/afrisenti_tir.yaml

@@ -0,0 +1,7 @@
+# Generated by utils.py
+dataset_name: tir
+doc_to_text: "You are an assistant able to detect sentiments in tweets. \n\nGiven\
+  \ the sentiment labels Neutral, Positive or Negative; what is the sentiment of the\
+  \ Tigrinya statement below? Return only the labels. \n\ntext: {{tweet}} \nlabel:"
+include: afrisenti
+task: afrisenti_tir_prompt_3

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_4/afrisenti_kin.yaml

@@ -0,0 +1,6 @@
+# Generated by utils.py
+dataset_name: kin
+doc_to_text: "Label the following text as Neutral, Positive, or Negative. Provide\
+  \ only the label as your response. \n\ntext: {{tweet}} \nlabel: "
+include: afrisenti
+task: afrisenti_kin_prompt_4

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_4/afrisenti_orm.yaml

@@ -0,0 +1,6 @@
+# Generated by utils.py
+dataset_name: orm
+doc_to_text: "Label the following text as Neutral, Positive, or Negative. Provide\
+  \ only the label as your response. \n\ntext: {{tweet}} \nlabel: "
+include: afrisenti
+task: afrisenti_orm_prompt_4

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_4/afrisenti_pcm.yaml

@@ -0,0 +1,6 @@
+# Generated by utils.py
+dataset_name: pcm
+doc_to_text: "Label the following text as Neutral, Positive, or Negative. Provide\
+  \ only the label as your response. \n\ntext: {{tweet}} \nlabel: "
+include: afrisenti
+task: afrisenti_pcm_prompt_4

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_4/afrisenti_swa.yaml

@@ -0,0 +1,6 @@
+# Generated by utils.py
+dataset_name: swa
+doc_to_text: "Label the following text as Neutral, Positive, or Negative. Provide\
+  \ only the label as your response. \n\ntext: {{tweet}} \nlabel: "
+include: afrisenti
+task: afrisenti_swa_prompt_4

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_4/afrisenti_tir.yaml

@@ -0,0 +1,6 @@
+# Generated by utils.py
+dataset_name: tir
+doc_to_text: "Label the following text as Neutral, Positive, or Negative. Provide\
+  \ only the label as your response. \n\ntext: {{tweet}} \nlabel: "
+include: afrisenti
+task: afrisenti_tir_prompt_4

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_4/afrisenti_tso.yaml

@@ -0,0 +1,6 @@
+# Generated by utils.py
+dataset_name: tso
+doc_to_text: "Label the following text as Neutral, Positive, or Negative. Provide\
+  \ only the label as your response. \n\ntext: {{tweet}} \nlabel: "
+include: afrisenti
+task: afrisenti_tso_prompt_4

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_4/afrisenti_twi.yaml

@@ -0,0 +1,6 @@
+# Generated by utils.py
+dataset_name: twi
+doc_to_text: "Label the following text as Neutral, Positive, or Negative. Provide\
+  \ only the label as your response. \n\ntext: {{tweet}} \nlabel: "
+include: afrisenti
+task: afrisenti_twi_prompt_4

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_4/afrisenti_yor.yaml

@@ -0,0 +1,6 @@
+# Generated by utils.py
+dataset_name: yor
+doc_to_text: "Label the following text as Neutral, Positive, or Negative. Provide\
+  \ only the label as your response. \n\ntext: {{tweet}} \nlabel: "
+include: afrisenti
+task: afrisenti_yor_prompt_4

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_4/utils.py

@@ -0,0 +1 @@
+from lm_eval.utils import weighted_f1_score

lm-evaluation-harness/lm_eval/tasks/afrobench/afrisenti/prompt_4/xx.py

@@ -0,0 +1,5 @@
+from datasets import load_dataset
+
+
+data = load_dataset("masakhane/afrisenti", "orm", trust_remote_code=True)
+print(data)