Update Workers docs

Author: mohamedalibarkailluin

docs/AristoteAPI_Workers_guide.md

@@ -1,6 +1,6 @@
 # Quiz generation worker
 
-First, make sure that you have an **llm-model** container :
+First, make sure that you have an **llm-model** container (not needed for quiz generation using Aristote Dispatcher):
 
 ```
 docker ps -a
@@ -12,26 +12,26 @@ If not, run this command
 docker run --runtime nvidia --gpus all —name llm-model -v ~/.cache/huggingface:/root/.cache/huggingface -p 8000:8000 --ipc=host vllm/vllm-openai:v0.2.4 --model teknium/OpenHermes-2.5-Mistral-7B --dtype float16 --tensor-parallel-size 1
 ```
 
-After a new Quiz Generation build with tag, these are the steps to follow on the Worker machine (**Note** : quiz-gen folder contains .env file necessary for running the docker container, you can make necessary changes if needed):
+After a new Quiz Generation build with tag, these are the steps to follow on the Worker machine for prod environment for example (**Note** : quiz-gen folder contains .env.prod file necessary for running the docker container, you can make necessary changes if needed):
 
 ```
 cd quiz-gen
-docker stop quiz-generator
-docker rm quiz-generator
-docker pull jq422pa7.gra7.container-registry.ovh.net/aristote/quizz-generation/quizgenerator:{tag}
-docker run --env-file .env --network="host" -p 3000:3000 --name quiz-generator jq422pa7.gra7.container-registry.ovh.net/aristote/quizz-generation/quizgenerator:{tag}
+docker stop quiz-generator-prod
+docker rm quiz-generator-prod
+docker pull jq422pa7.gra7.container-registry.ovh.net/aristote/opensource-quiz-generator:{tag}
+docker run --env-file .env --add-host host.docker.internal:host-gateway -p 3000:3000 -p 3000:3000 --name quiz-generator-prod jq422pa7.gra7.container-registry.ovh.net/aristote/opensource-quiz-generator:{tag}
 ```
 
 or simply (inside quiz-gen folder):
 
 ```
-./update.sh {tag}
+./update-prod.sh {tag}
 ```
 
 To start process of getting a job from AristoteAPI and generating the quiz (assuming quiz-generator container and the llm-model container are running):
 
 ```
-docker exec quiz-generator python server/generate_quizz.py
+docker exec quiz-generator-prod python server/generate_quizz.py
 ```
 
 This script can be used to handle switching on the llm-model container, waiting till it's responsive, then start the previous command, while taking into account a maximum number of parallel jobs :
@@ -51,15 +51,15 @@ docker save -o whisper.tar whisperapi-whisper
 scp whisper.tar ubuntu@ip_adress:/home/ubuntu/transcription
 ```
 
-These are the steps to follow on the Worker machine (**Note** : transcription folder contains .env file necessary for running the docker container, you can make necessary changes if needed):
+These are the steps to follow on the Worker machine (**Note** : transcription folder contains .env file necessary for running the docker container and the .env.preprod/.env.prod for the appropriate Aristote URL and credentials to be used, you can make necessary changes if needed):
 
 ```
 cd transcription
 docker stop whisper
 docker rm whisper
 docker image rm whisperapi-whisper
 docker load -i whisper.tar
-docker run --runtime nvidia --gpus all --env-file .env -p 3001:3000 --name whisper whisperapi-whisper
+docker run --runtime nvidia --gpus all --env-file .env -p 3001:3000 -v {MODEL_FOLDER_PATH}:/server_app/custom_model -v ./.env.prod:/server_app/.env.prod -v ./.env.preprod:/server_app/.env.preprod --name whisper whisperapi-whisper
 ```
 
 or simply (inside transcription folder):
@@ -71,7 +71,13 @@ or simply (inside transcription folder):
 To start the process of getting a job from AristoteAPI and transcribing the media (assuming whisper container and the whisper container are running):
 
 ```
-docker exec whisper python transcribe_aristote.py
+docker exec whisper python transcribe_aristote.py --env .env.prod
+```
+
+or
+
+```
+docker exec whisper python transcribe_aristote.py --env .env.preprod
 ```
 
 This script can be used to handle switching on the whisper container, waiting till it's responsive, then start the previous command, while taking into account a maximum number of parallel jobs :
@@ -90,7 +96,6 @@ To start the process of getting a job from AristoteAPI and evaluating a quiz
 docker exec quiz-generator python server/evaluate_quizz.py
 ```
 
-
 # Cron jobs
 
 To access cronjob list :
@@ -102,4 +107,9 @@ crontab -e
 There is **schedule.sh** that is run as a cron job and it handles the scheduling of quiz generation/transcription workers using **transcription-cron-job.sh** and 
 **quiz-generation-cron-job.sh** because there is not enough VRAM from them to run simultaneously.
 
+
+**evaluation-cron-job.sh** that is run as a separate cron job since it doesn't interfere with the other workers.
+
+**quiz-generation-dispatcher-cron-job.sh** handles the quiz generation using Aristote Dispatcher, so no local LLM is needed.
+
 There is also **evaluation-cron-job.sh** that is run as a separate cron job since it doesn't interfere with the other workers.