Fix broken links Python BE (triton-inference-server#193)

Tabrizian · web-flow · commit 8466d33ad69b · 2022-10-31T16:55:11.000-04:00
diff --git a/examples/auto_complete/README.md b/examples/auto_complete/README.md
@@ -31,15 +31,15 @@
 This example shows how to implement
 [`auto_complete_config`](https://github.com/triton-inference-server/python_backend/#auto_complete_config)
 function in Python backend to provide
-[`max_batch_size`](https://github.com/triton-inference-server/server/blob/main/docs/model_configuration.md#maximum-batch-size),
-[`input`](https://github.com/triton-inference-server/server/blob/main/docs/model_configuration.md#inputs-and-outputs)
-and [`output`](https://github.com/triton-inference-server/server/blob/main/docs/model_configuration.md#inputs-and-outputs)
+[`max_batch_size`](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#maximum-batch-size),
+[`input`](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#inputs-and-outputs)
+and [`output`](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#inputs-and-outputs)
 properties. These properties will allow Triton to load the Python model with
-[Minimal Model Configuration](https://github.com/triton-inference-server/server/blob/main/docs/model_configuration.md#minimal-model-configuration)
+[Minimal Model Configuration](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#minimal-model-configuration)
 in absence of a configuration file.
 
 The
-[model repository](https://github.com/triton-inference-server/server/blob/main/docs/model_repository.md)
+[model repository](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_repository.md)
 should contain [nobatch_auto_complete](./nobatch_model.py), and
 [batch_auto_complete](./batch_model.py) models.
 The max_batch_size of [nobatch_auto_complete](./nobatch_model.py) model is set
diff --git a/examples/bls/README.md b/examples/bls/README.md
@@ -30,7 +30,7 @@
 
 In this section we demonstrate an end-to-end example for
 [BLS](../../README.md#business-logic-scripting) in Python backend. The
-[model repository](https://github.com/triton-inference-server/server/blob/main/docs/model_repository.md)
+[model repository](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_repository.md)
 should contain [pytorch](../pytorch), [addsub](../add_sub).  The
 [pytorch](../pytorch) and [addsub](../add_sub) models calculate the sum and
 difference of the `INPUT0` and `INPUT1` and put the results in `OUTPUT0` and
diff --git a/examples/decoupled/README.md b/examples/decoupled/README.md
@@ -36,7 +36,8 @@ how to write a decoupled model where each request can generate 0 to many respons
 These files are heavily commented to describe each function call.
 These example models are designed to show the flexibility available to decoupled models
 and in no way should be used in production. These examples circumvents
-the restriction placed by the [instance count](https://github.com/triton-inference-server/server/blob/main/docs/model_configuration.md#instance-groups)
+the restriction placed by the
+[instance count](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#instance-groups)
 and allows multiple requests to be in process even for single instance. In
 real deployment, the model should not allow the caller thread to return from
 `execute` until that instance is ready to handle another set of requests.
@@ -341,4 +342,4 @@ stream stopped...
 
 Look how responses were delivered out-of-order of requests.
 The generated responses can be tracked to their request using
-the `id` field.
+the `id` field.
diff --git a/inferentia/README.md b/inferentia/README.md
@@ -239,22 +239,23 @@ their need.
 
 To enable dynamic batching, `--enable_dynamic_batching`
 flag needs to be specified. `gen_triton_model.py` supports following three 
-options for configuring [Triton's dynamic batching](https://github.com/triton-inference-server/server/blob/main/docs/model_configuration.md):
+options for configuring [Triton's dynamic batching](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md):
 
-1. `--preferred_batch_size`: Please refer to [model configuration documentation](https://github.com/triton-inference-server/server/blob/main/docs/model_configuration.md#preferred-batch-sizes) for details on preferred batch size. To optimize
+1. `--preferred_batch_size`: Please refer to [model configuration documentation](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#preferred-batch-sizes) for details on preferred batch size. To optimize
    performance, this is recommended to be multiples of engaged neuron cores.
    For example, if each instance is using 2 neuron cores, `preferred_batch_size`
    could be 2, 4 or 6. 
 2. `--max_queue_delay_microseconds`: Please refer to
-   [model configuration documentation](https://github.com/triton-inference-server/server/blob/main/docs/model_configuration.md#delayed-batching) for details.
+   [model configuration documentation](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#delayed-batching) for details.
 3. `--disable_batch_requests_to_neuron`: Enable the non-default way for Triton to
    handle batched requests. Triton backend will send each request to neuron
    separately, irrespective of if the Triton server requests are batched.
    This flag is recommended when users want to optimize performance with models
    that do not perform well with batching without the flag.
 
 Additionally, `--max_batch_size` will affect the maximum batching limit. Please
-refer to the [model configuration documentation](https://github.com/triton-inference-server/server/blob/main/docs/model_configuration.md#maximum-batch-size)
+refer to the
+[model configuration documentation](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#maximum-batch-size)
 for details.
 
 ## Testing Inferentia Setup for Accuracy
