Llm ppl workflow bug fix (#10128)

* add llm-ppl workflow

* update the DATASET_DIR

* test multiple precisions

* modify nightly test

* match the updated ppl code

* add matrix.include

* fix the include error

* update the include

* add more model

* update the precision of include

* update nightly time and add more models

* fix the workflow_dispatch description, change default model of pr and modify the env

* modify workflow_dispatch language options

* modify options

* modify language options

* modeify workflow_dispatch type

* modify type

* modify the type of language

* change seq_len type

.github/workflows/llm-ppl-evaluation.yml

@@ -17,9 +17,9 @@ on:
   workflow_dispatch:
     inputs:
       seq_len:
-        description: 'sequence length, input one number'
+        description: 'sequence length, input one number and must be quoted.'
         required: true
-        type: number      
+        type: string      
       model_name:
         description: 'Model names, seperated by comma and must be quoted.'
         required: true
@@ -29,13 +29,9 @@ on:
         required: true
         type: string
       language:
-        description: 'language, can be en, zh, or all'
+        description: 'language, can be en, zh, or all and must be quoted.'
         required: true
-        type: choice
-        options:
-          - "en"
-          - "zh"
-          - "all"
+        type: string
       runs-on:
         description: 'Labels to filter the runners, seperated by comma and must be quoted.'
         default: "accuracy"
@@ -202,4 +198,4 @@ jobs:
           --precisions ${{ matrix.precision }} \
           --device ${{ matrix.device }} \
           --dataset_path ${DATASET_DIR} \
-          --language ${{ matrix.language }}
+          --language ${{ matrix.language }}