Add all cli examples

cli/foundation-models/system/distillation/conversation/README.md

@@ -2,4 +2,4 @@
 
 ### Run the Distillation CLI command pointing to the .YAML file in this folder plus the Azure ML IDs needed:
 
-az ml job create --file ./distillation_conversation.yaml --workspace-name [YOUR_AZURE_WORKSPACE] --resource-group [YOUR_AZURE_RESOURCE_GROUP] --subscription [YOUR_AZURE_SUBSCRIPTION]
+az ml job create --file distillation_conversation.yaml --workspace-name [YOUR_AZURE_WORKSPACE] --resource-group [YOUR_AZURE_RESOURCE_GROUP] --subscription [YOUR_AZURE_SUBSCRIPTION]

cli/foundation-models/system/distillation/conversation/distillation_conversation.yaml

@@ -9,14 +9,19 @@ data_generation_type: label_generation
 data_generation_task_type: conversation
 
 # Input data
-training_data: azureml:train_conversation:1
-validation_data: azureml:validation_conversation:1
+training_data:
+  type: uri_file
+  path: ./train_conversation.jsonl
+validation_data:
+  type: uri_file
+  path: ./validation_conversation.jsonl
 
 # Teacher model serverless endpoint information
+# REPLACE WITH YOUR ENDPOINT INFORMATION
 teacher_model_endpoint_connection: 
   type: serverless
-  name: Meta-Llama-3-1-405B-Instruct-vum
-  endpoint: https://Meta-Llama-3-1-405B-Instruct-vum.westus3.models.ai.azure.com/chat/completions
+  name: Meta-Llama-3-1-405B-Instruct-vkn
+  endpoint: https://Meta-Llama-3-1-405B-Instruct-vkn.westus3.models.ai.azure.com/chat/completions
   api_key: EXAMPLE_API_KEY
 
 # Model ID
@@ -36,13 +41,9 @@ teacher_model_settings:
     max_tokens: 100
     top_p: 0.95
   endpoint_request_settings:
-    request_batch_size: 5
+    request_batch_size: 10
     min_endpoint_success_ratio: 0.7
 
-# System prompt settings (OPTIONAL)
-prompt_settings:
-  enable_chain_of_thought: true
-
 # For finetuning (OPTIONAL)
 hyperparameters:
   learning_rate_multiplier: "0.2"
@@ -51,5 +52,4 @@ hyperparameters:
 
 # Resource for Data Generation Step (OPTIONAL)
 resources:
-  instance_types:
-    - Standard_D2_v2
+  instance_type: Standard_D2_v2

cli/foundation-models/system/distillation/math/README.md

@@ -2,4 +2,4 @@
 
 ### Run the Distillation CLI command pointing to the .YAML file in this folder plus the Azure ML IDs needed:
 
-az ml job create --file ./distillation_math.yaml --workspace-name [YOUR_AZURE_WORKSPACE] --resource-group [YOUR_AZURE_RESOURCE_GROUP] --subscription [YOUR_AZURE_SUBSCRIPTION]
+az ml job create --file distillation_math.yaml --workspace-name [YOUR_AZURE_WORKSPACE] --resource-group [YOUR_AZURE_RESOURCE_GROUP] --subscription [YOUR_AZURE_SUBSCRIPTION]

cli/foundation-models/system/distillation/math/distillation_math.yaml

@@ -9,14 +9,19 @@ data_generation_type: label_generation
 data_generation_task_type: math
 
 # Input data
-training_data: azureml:train_math:1
-validation_data: azureml:validation_math:1
+training_data: 
+  type: uri_file
+  path: ./train_math.jsonl
+validation_data: 
+  type: uri_file
+  path: ./validation_math.jsonl
 
 # Teacher model serverless endpoint information
+# REPLACE WITH YOUR ENDPOINT INFORMATION
 teacher_model_endpoint_connection: 
   type: serverless
-  name: Meta-Llama-3-1-405B-Instruct-vum
-  endpoint: https://Meta-Llama-3-1-405B-Instruct-vum.westus3.models.ai.azure.com/chat/completions
+  name: Meta-Llama-3-1-405B-Instruct-vkn
+  endpoint: https://Meta-Llama-3-1-405B-Instruct-vkn.westus3.models.ai.azure.com/chat/completions
   api_key: EXAMPLE_API_KEY
 
 # Model ID
@@ -33,10 +38,10 @@ outputs:
 teacher_model_settings:
   inference_parameters:
     temperature: 0.1
-    max_tokens: 100
+    max_tokens: 1024
     top_p: 0.95
   endpoint_request_settings:
-    request_batch_size: 5
+    request_batch_size: 10
     min_endpoint_success_ratio: 0.7
 
 # System prompt settings (OPTIONAL)
@@ -51,5 +56,4 @@ hyperparameters:
 
 # Resource for Data Generation Step (OPTIONAL)
 resources:
-  instance_types:
-    - Standard_D2_v2
+  instance_type: Standard_D2_v2

cli/foundation-models/system/distillation/nli/README.md

@@ -2,4 +2,4 @@
 
 ### Run the Distillation CLI command pointing to the .YAML file in this folder plus the Azure ML IDs needed:
 
-az ml job create --file ./distillation_nli.yaml --workspace-name [YOUR_AZURE_WORKSPACE] --resource-group [YOUR_AZURE_RESOURCE_GROUP] --subscription [YOUR_AZURE_SUBSCRIPTION]
+az ml job create --file distillation_nli.yaml --workspace-name [YOUR_AZURE_WORKSPACE] --resource-group [YOUR_AZURE_RESOURCE_GROUP] --subscription [YOUR_AZURE_SUBSCRIPTION]

cli/foundation-models/system/distillation/nli/distillation_nli.yaml

@@ -9,14 +9,19 @@ data_generation_type: label_generation
 data_generation_task_type: nli
 
 # Input data
-training_data: azureml:train_nli:1
-validation_data: azureml:validation_nli:1
+training_data:
+  type: uri_file
+  path: ./train_nli.jsonl
+validation_data:
+  type: uri_file
+  path: ./validation_nli.jsonl
 
 # Teacher model serverless endpoint information
+# REPLACE WITH YOUR ENDPOINT INFORMATION
 teacher_model_endpoint_connection: 
   type: serverless
-  name: Meta-Llama-3-1-405B-Instruct-vum
-  endpoint: https://Meta-Llama-3-1-405B-Instruct-vum.westus3.models.ai.azure.com/chat/completions
+  name: Meta-Llama-3-1-405B-Instruct-vkn
+  endpoint: https://Meta-Llama-3-1-405B-Instruct-vkn.westus3.models.ai.azure.com/chat/completions
   api_key: EXAMPLE_API_KEY
 
 # Model ID
@@ -33,10 +38,10 @@ outputs:
 teacher_model_settings:
   inference_parameters:
     temperature: 0.1
-    max_tokens: 100
+    max_tokens: 1024
     top_p: 0.95
   endpoint_request_settings:
-    request_batch_size: 5
+    request_batch_size: 10
     min_endpoint_success_ratio: 0.7
 
 # System prompt settings (OPTIONAL)
@@ -51,5 +56,4 @@ hyperparameters:
 
 # Resource for Data Generation Step (OPTIONAL)
 resources:
-  instance_types:
-    - Standard_D2_v2
+  instance_type: Standard_D2_v2