Merge pull request #5 from JessicaOjo/africanli

update afrixnli tasks

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_ewe.yaml

+# Generated by utils.py
+dataset_name: ewe
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_ewe

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_fra.yaml

+# Generated by utils.py
+dataset_name: fra
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_fra

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_hau.yaml

+# Generated by utils.py
+dataset_name: hau
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_hau

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_ibo.yaml

+# Generated by utils.py
+dataset_name: ibo
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_ibo

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_kin.yaml

+# Generated by utils.py
+dataset_name: kin
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_kin

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_lin.yaml

+# Generated by utils.py
+dataset_name: lin
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_lin

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_lug.yaml

+# Generated by utils.py
+dataset_name: lug
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_lug

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_orm.yaml

+# Generated by utils.py
+dataset_name: orm
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_orm

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_sna.yaml

+# Generated by utils.py
+dataset_name: sna
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_sna

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_sot.yaml

+# Generated by utils.py
+dataset_name: sot
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_sot

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_swa.yaml

+# Generated by utils.py
+dataset_name: swa
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_swa

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_twi.yaml

+# Generated by utils.py
+dataset_name: twi
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_twi

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_wol.yaml

+# Generated by utils.py
+dataset_name: wol
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_wol

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_xho.yaml

+# Generated by utils.py
+dataset_name: xho
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_xho

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_yaml

+group:
+    - xnli
+    - afrixnli
+dataset_path: masakhane/afrixnli
+dataset_name: null
+output_type: multiple_choice
+validation_split: validation
+test_split: test
+fewshot_split: validation
+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
+    hf_evaluate: true
+    higher_is_better: True
+    ignore_case: true
+    ignore_punctuation: true
+    regexes_to_ignore:
+      - ","
+      - "\\$"
+  - metric: acc
+    aggregation: mean
+    higher_is_better: true
+    ignore_case: true
+    ignore_punctuation: true
+    regexes_to_ignore:
+      - ","
+      - "\\$"
+metadata:
+  version: 1.0

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_yor.yaml

+# Generated by utils.py
+dataset_name: yor
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_yor

lm_eval/tasks/afrixnli/en-direct/afrixnli_en_direct_zul.yaml

+# Generated by utils.py
+dataset_name: zul
+include: afrixnli_en_direct_yaml
+task: afrixnli_en_direct_zul

lm_eval/tasks/afrixnli/en-direct/utils.py

+from sklearn.metrics import f1_score
+
+
+def doc_to_target(doc):
+    replacements = {
+        0: 'True',
+        1: 'Neither',
+        2: 'False'
+    }
+    return replacements[doc["label"]]
+
+
+def weighted_f1_score(items):
+    unzipped_list = list(zip(*items))
+    golds = unzipped_list[0]
+    preds = unzipped_list[1]
+    fscore = f1_score(golds, preds, average="weighted")
+    return fscore

lm_eval/tasks/afrixnli/en_direct.sh

+#!/bin/bash
+
+models=(
+  "masakhane/African-ultrachat-alpaca"
+  "masakhane/zephyr-7b-gemma-sft-african-alpaca"
+  "masakhane/zephyr-7b-gemma-sft-african-ultrachat-5k"
+  "google/flan-t5-xxl"
+  "bigscience/mt0-xxl-mt"
+  "CohereForAI/aya-101"
+  "bigscience/bloomz-7b1-mt"
+  "meta-llama/Llama-2-7b-chat-hf"
+  "meta-llama/Meta-Llama-3-8B-Instruct"
+  "meta-llama/Meta-Llama-3-70B-Instruct"
+  "google/gemma-1.1-7b-it"
+  "RWKV/v5-EagleX-v2-7B-HF"
+  "RWKV/rwkv-6-world-7b"
+)
+task=afrixnli_en_direct_amh,afrixnli_en_direct_eng,afrixnli_en_direct_ewe,afrixnli_en_direct_fra,afrixnli_en_direct_hau,afrixnli_en_direct_ibo,afrixnli_en_direct_kin,afrixnli_en_direct_lin,afrixnli_en_direct_lug,afrixnli_en_direct_orm,afrixnli_en_direct_sna,afrixnli_en_direct_sot,afrixnli_en_direct_swa,afrixnli_en_direct_twi,afrixnli_en_direct_wol,afrixnli_en_direct_xho,afrixnli_en_direct_yor,afrixnli_en_direct_zul
+
+for model in "${models[@]}"
+do
+  echo "Evaluating model: $model"
+  for fewshot in 0 2 4 6 8
+  do
+    export OUTPUT_DIR=results/${model##*/}/$fewshot
+
+    mkdir -p "$OUTPUT_DIR"
+
+    lm_eval --model hf \
+            --model_args "pretrained=${model}" \
+            --tasks $task\
+            --device cuda:0 \
+            --batch_size 16 \
+            --output_path "$OUTPUT_DIR" \
+            --num_fewshot $fewshot \
+            --limit 1 \
+            --verbosity DEBUG
+  done
+done
\ No newline at end of file

lm_eval/tasks/afrixnli/native-direct/afrixnli_native_direct_amh.yaml

+# Generated by utils.py
+dataset_name: amh
+doc_to_choice: '{{[premise+", ትክክል? አዎ, "+hypothesis,premise+", ትክክል? እንዲሁም, "+hypothesis,premise+",
+  ትክክል? አይ, "+hypothesis]}}'
+include: afrixnli_native_direct_yaml
+task: afrixnli_native_direct_amh