Merge pull request #17 from JessicaOjo/afrixnli_translate_test

Afrixnli translate test

lm_eval/tasks/afrixnli/translate/afrixnli_translate_amh.yaml

+# Generated by utils.py
+dataset_name: amh
+include: afrixnli_translate_yaml
+task: afrixnli_translate_amh
+

lm_eval/tasks/afrixnli/translate/afrixnli_translate_ewe.yaml

+# Generated by utils.py
+dataset_name: ewe
+include: afrixnli_translate_yaml
+task: afrixnli_translate_ewe

lm_eval/tasks/afrixnli/translate/afrixnli_translate_fra.yaml

+# Generated by utils.py
+dataset_name: fra
+include: afrixnli_translate_yaml
+task: afrixnli_translate_fra

lm_eval/tasks/afrixnli/translate/afrixnli_translate_hau.yaml

+# Generated by utils.py
+dataset_name: hau
+include: afrixnli_translate_yaml
+task: afrixnli_translate_hau

lm_eval/tasks/afrixnli/translate/afrixnli_translate_ibo.yaml

+# Generated by utils.py
+dataset_name: ibo
+include: afrixnli_translate_yaml
+task: afrixnli_translate_ibo

lm_eval/tasks/afrixnli/translate/afrixnli_translate_kin.yaml

+# Generated by utils.py
+dataset_name: kin
+include: afrixnli_translate_yaml
+task: afrixnli_translate_kin

lm_eval/tasks/afrixnli/translate/afrixnli_translate_lin.yaml

+# Generated by utils.py
+dataset_name: lin
+include: afrixnli_translate_yaml
+task: afrixnli_translate_lin

lm_eval/tasks/afrixnli/translate/afrixnli_translate_lug.yaml

+# Generated by utils.py
+dataset_name: lug
+include: afrixnli_translate_yaml
+task: afrixnli_translate_lug

lm_eval/tasks/afrixnli/translate/afrixnli_translate_orm.yaml

+# Generated by utils.py
+dataset_name: orm
+include: afrixnli_translate_yaml
+task: afrixnli_translate_orm

lm_eval/tasks/afrixnli/translate/afrixnli_translate_sna.yaml

+# Generated by utils.py
+dataset_name: sna
+include: afrixnli_translate_yaml
+task: afrixnli_translate_sna

lm_eval/tasks/afrixnli/translate/afrixnli_translate_sot.yaml

+# Generated by utils.py
+dataset_name: sot
+include: afrixnli_translate_yaml
+task: afrixnli_translate_sot

lm_eval/tasks/afrixnli/translate/afrixnli_translate_swa.yaml

+# Generated by utils.py
+dataset_name: swa
+include: afrixnli_translate_yaml
+task: afrixnli_translate_swa

lm_eval/tasks/afrixnli/translate/afrixnli_translate_twi.yaml

+# Generated by utils.py
+dataset_name: twi
+include: afrixnli_translate_yaml
+task: afrixnli_translate_twi

lm_eval/tasks/afrixnli/translate/afrixnli_translate_wol.yaml

+# Generated by utils.py
+dataset_name: wol
+include: afrixnli_translate_yaml
+task: afrixnli_translate_wol

lm_eval/tasks/afrixnli/translate/afrixnli_translate_xho.yaml

+# Generated by utils.py
+dataset_name: xho
+include: afrixnli_translate_yaml
+task: afrixnli_translate_xho

lm_eval/tasks/afrixnli/translate/afrixnli_translate_yaml

+group:
+    - xnli
+    - afrixnli
+    - afrixnli-translate-test
+dataset_path: masakhane/afrixnli-translate-test
+dataset_name: null
+output_type: multiple_choice
+test_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_eval/tasks/afrixnli/translate/afrixnli_translate_yor.yaml

+# Generated by utils.py
+dataset_name: yor
+include: afrixnli_translate_yaml
+task: afrixnli_translate_yor

lm_eval/tasks/afrixnli/translate/afrixnli_translate_zul.yaml

+# Generated by utils.py
+dataset_name: zul
+include: afrixnli_translate_yaml
+task: afrixnli_translate_zul

lm_eval/tasks/afrixnli/translate/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/translate_test.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"
+)
+
+models=("masakhane/zephyr-7b-gemma-sft-african-ultrachat-5k")
+task=afrixnli_translate_amh,afrixnli_translate_ewe,afrixnli_translate_fra,afrixnli_translate_hau,afrixnli_translate_ibo,afrixnli_translate_kin,afrixnli_translate_lin,afrixnli_translate_lug,afrixnli_translate_orm,afrixnli_translate_sna,afrixnli_translate_sot,afrixnli_translate_swa,afrixnli_translate_twi,afrixnli_translate_wol,afrixnli_translate_xho,afrixnli_translate_yor,afrixnli_translate_zul
+
+for model in "${models[@]}"
+do
+  echo "Evaluating model: $model"
+  for fewshot in 0
+  do
+    export OUTPUT_DIR=results/${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 \
+            --verbosity DEBUG 
+  done
+done
\ No newline at end of file