Merge pull request #2 from Saif-Shines/refactor/content

Refactor - Content

Author: roh26it

README.md

@@ -12,7 +12,7 @@ Portkey wants to make it easy to work with Large Models that expose an API by of
 
 Imagine portkey as a hop away from the LLMs enabling you with a lot of functionality out of the box.
 
-![Portkey diagram](https://github.com/roh26it/portkey-quick-start/blob/main/images/portkey-middleware.png?raw=true)
+![Portkey diagram](/images/portkey-middleware-v2.jpeg)
 
 #PromptOps #LLMOps #LMOps
 
@@ -40,4 +40,4 @@ We’re adding more features fast, but key areas of focus for Portkey will be hi
 
 The best part about Portkey is that you can get started with minimal effort. Just replace your OpenAI (or any other provider’s) endpoint to the portkey proxy and we’ll get you setup within minutes!
 
-Read the [Quickstart Guide](https://github.com/roh26it/portkey-quick-start/blob/main/quickstart.md) next
+Get Started with the [Quickstart Guide](https://github.com/roh26it/portkey-quick-start/blob/main/quickstart.md)

caching.md

@@ -8,7 +8,7 @@ This could be useful if you have fixed input prompts or are testing the app with
 
 To enable caching, pass the following headers in your requests.
 
-```
+```sh
     "x-portkey-cache": true
     "Cache-Control": "max-age:1000"
 ```
@@ -19,6 +19,6 @@ The `x-portkey-cache` enables or disables cache storage and retrieval. The `Cach
 
 You can choose to force refresh cache by using the `x-portkey-cache-force-refresh` header. Setting it to `true` ensures that the cache is invalidated, and a new value is stored in the cache.
 
-```
+```sh
     "x-portkey-cache-force-refresh": true
 ```

feedback.md

@@ -19,12 +19,10 @@ curl --location 'https://api.portkey.ai/v1/feedback/' \
 }'
 ```
 **URL**: `https://api.portkey.ai/v1/feedback/`
-<br>
 **Method**: `POST`
-
 **Request Body**<br>
 The request body should contain a `Feedback object` or an array of `Feedback objects`
-<br>
+
 
 | Key | Description | Options |
 |---|---|---|
@@ -37,13 +35,13 @@ The request body should contain a `Feedback object` or an array of `Feedback obj
 
 ### Examples
 
-#### 1. Implementing a 👍🏻 / 👎🏻 feedback
+#### Implementing a 👍🏻 / 👎🏻 feedback
 
 In many UI implementations, you would request a user to rate a generation with a simple thumbs up or thumbs down. You can use the `/feedback` endpoint to capture this user intent along with the generation's trace-id so you can aggregate the data at your end.
 
 You can make an API call as the user clicks a thumbs up button like this
 
-```
+```sh
 curl --location 'https://api.portkey.ai/v1/feedback/' \
 --header 'x-portkey-api-key: <YOUR PORTKEY API KEY>' \
 --header 'Content-Type: application/json' \
@@ -52,6 +50,6 @@ curl --location 'https://api.portkey.ai/v1/feedback/' \
     "value": 1
 }'
 ```
-Thumbs up can have a `value` of 1 and thumbs down can have a value of of -1. We're not adding any weights since both the values should be weighted equally.
+- Thumbs up can have a `value` of 1 and thumbs down can have a value of of -1. We're not adding any weights since both the values should be weighted equally.
 
-If the user also decides to leave a comment with their selection, you can add the `text` parameter to capture the user's comment.
+- If the user also decides to leave a comment with their selection, you can add the `text` parameter to capture the user's comment.

images/portkey-middleware-v2.jpeg

@@ -66,4 +66,4 @@ print(llm(text))
 
 ## Model completion metadata
 
-Coming soon
+_Coming soon_

metadata.md

@@ -8,7 +8,7 @@ This guide allows you to integrate with Portkey in a minute.
 You can create an account on Portkey with your work email address and get the authentication key needed later in the integration.
 
 You can get your API key from the profile dropdown
-![Portkey copy API key](https://github.com/Portkey-AI/quick-start/blob/main/images/copy-api-key.png)
+![Portkey copy API key](/images/copy-api-key.png)
 
 ## 2. Insert the Portkey endpoint
 Replace the OpenAI (or any other provider’s) endpoint to the portkey proxy endpoint and add the portkey authentication headers.
@@ -73,5 +73,5 @@ print(llm(text))
 ## 3. View real-time activity in your account
 As the API requests start flowing in, you’ll see the activity in your account instantly.
 
-## Metadata in Requests
+#### Metadata in Requests
 If you'd like to send additional metadata with your requests, you can add it as part of the headers. [Read more here](https://github.com/Portkey-AI/quick-start/blob/main/metadata.md).

quickstart.md

@@ -1,9 +1,9 @@
 # Automatic Retries with Portkey
 
-Portkey has an in-built retry mechanism that makes sure your requests get fulfilled. We implement a retry mechanism with exponential backoff. This can be managed by the `x-portkey-retry-count` header. 
+Portkey has an in-built retry mechanism that makes sure your requests get fulfilled. We implement a retry mechanism with exponential backoff. This can be managed by the `x-portkey-retry-count` header.
 
-```
-HEADER:
+```sh
+# HEADER:
 x-portkey-retry-count: "number of times you want the API to be retried (max 5)"
 ```
 
@@ -25,7 +25,7 @@ response = openai.ChatCompletion.create(
 ```
 
 For Javascript
-```javascript
+```js
 const configuration = new Configuration({
     organization: "YOUR_ORG_ID",
     apiKey: process.env.OPENAI_API_KEY,
@@ -66,4 +66,5 @@ agent = initialize_agent(tools, llm, agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION
 agent.run("What is 56 multiplied by 17654 to the 0.43 power?")
 ```
 
-Note: The max value of retries available is 5.
+> NOTE: The max value of retries available is 5
+

retry.md

@@ -2,8 +2,8 @@
 
 Portkey supports tracing that allows you to monitor your generative AI apps end-to-end. To enable tracing, you can pass a `trace-id` in the header of any Portkey request (proxy or otherwise) and it will be tracked through the journey of the request.
 
-```
-HEADER:
+```sh
+# HEADER:
 x-portkey-trace-id: "enter_your_trace_id_here"
 ```
 
@@ -43,9 +43,9 @@ const configuration = new Configuration({
 ### Using Tracing with Langchain Agents
 A common use case of tracing is to capture the flow of a langchain agent request. When using Portkey with langchain, you can pass your Trace ID in the headers and all the embeddings & completion requests will get tagged with the same Trace ID.
 
-Sample Langchain Agent Code with Tracing Enabled
+##### Sample Langchain Agent Code with Tracing Enabled
 ```python
-rom langchain.agents import load_tools
+from langchain.agents import load_tools
 from langchain.agents import initialize_agent
 from langchain.agents import AgentType
 from langchain.llms import OpenAI
@@ -68,7 +68,7 @@ agent.run("What is 56 multiplied by 17654 to the 0.43 power?")
 
 When we run this, we get view following logs on Portkey
 
-![Trace Logs in Portkey](https://github.com/Portkey-AI/quick-start/blob/main/images/trace-logs.gif)
+![Trace Logs in Portkey](/images/trace-logs.gif)
 
 ### Managing user feedback via Trace IDs
 Trace IDs can also be used to link user feedback to generations. You can implement a thumbs up / thumbs down user feedback system or something more complex with our feedback APIs. This feedback can be linked to traces which can span over 1 generation or multiple ones.