baalajimaestro/ollama
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

better readme

Browse source 
Signed-off-by: Matt Williams <m@technovangelist.com>
This commit is contained in:
Matt Williams 2023-10-31 12:40:46 -07:00
parent 5757925060
commit 80362fedce
3 changed files with 12 additions and 9 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
examples/bash-comparemodels/README.md
View file

@ -1,11 +1,5 @@
# Bash Shell examples # Bash Shell examples
![Example Gif](https://uce9aa94fbc06b088ca05a92fe37.previews.dropboxusercontent.com/p/thumb/ACF0g3yrdu5-tvuw59Wil8B5bwuLkvWFFQNrYJzEkqvJnv4WfyuqcTGfXhXDfqfbemi5jGr9-bccO8r5VZxXrAeU1l_Plq99HCqV6b10thwwlaQCNbkXkw4YSF0YlYu-wu5A6Vn2SlrdcfiwTl6et-m7CPYx8ad2jSZXcPEozDUqXqB-f_zZNskASYzWwQko9n6UjMKx6qt54FYvIiW6n3ZiNVlM0GGt91FAA2Y0zD23aBlOlIAN8wH7qLznS2rZsn1n_7ukJMwegcEVud_XNPbG8Hn_13NtwkVsf4uWThknUpslNRmxWisqlRCaxZY71Me9wz3puH3nlpxtNlwoNAvQcXf0S4u_r1WLx22KwWqmvYFU41X2j_1Kum8amUrAv_5WVnOL6ctWnrbV4fauYfT9ClwgmLAtLoHwaQSXo2R2Kut_QIAkFIDAyMj9Fe9Ifj0/p.gif) When you review the examples on this site, it is possible to think that making use of AI with Ollama will be hard. You need an orchestrator, and vector database, complicated infrastructure, and more. But that is not always the case. Ollama is designed to be easy to use, and to be used in any environment.
When calling `ollama`, you can pass it a file to run all the prompts in the file, one after the other. This concept is used in two examples The two examples here show how to list the models and query them from a simple bash script.
## Bulk Questions
`bulkquestions.sh` is a script that runs all the questions in `sourcequestions` using the llama2 model and outputs the answers.
## Compare Models
`comparemodels.sh` is a script that runs all the questions in `sourcequestions` using any 4 models you choose that you have already pulled from the Ollama library or have created locally.

1
examples/bash-comparemodels/bulkquestions.sh
View file

@ -1,3 +1,4 @@
#!/usr/bin/env bash #!/usr/bin/env bash
# Run a model and pass in a text file of questions
ollama run llama2 < sourcequestions ollama run llama2 < sourcequestions

10
examples/bash-comparemodels/comparemodels.sh
View file

@ -1,9 +1,14 @@
#! /usr/bin/env bash #! /usr/bin/env bash
# Compare multiple models by running them with the same questions
NUMBEROFCHOICES=4 NUMBEROFCHOICES=4
CHOICES=$(ollama list | awk '{print $1}')
SELECTIONS=() SELECTIONS=()
declare -a SUMS=() declare -a SUMS=()
# Get the list of models
CHOICES=$(ollama list | awk '{print $1}')
# Select which models to run as a comparison
echo "Select $NUMBEROFCHOICES models to compare:" echo "Select $NUMBEROFCHOICES models to compare:"
select ITEM in $CHOICES; do select ITEM in $CHOICES; do
if [[ -n $ITEM ]]; then if [[ -n $ITEM ]]; then
@ -18,6 +23,7 @@ select ITEM in $CHOICES; do
fi fi
done done
# Loop through each of the selected models
for ITEM in "${SELECTIONS[@]}"; do for ITEM in "${SELECTIONS[@]}"; do
echo "--------------------------------------------------------------" echo "--------------------------------------------------------------"
echo "Loading the model $ITEM into memory" echo "Loading the model $ITEM into memory"
@ -26,6 +32,8 @@ for ITEM in "${SELECTIONS[@]}"; do
echo "Running the questions through the model $ITEM" echo "Running the questions through the model $ITEM"
COMMAND_OUTPUT=$(ollama run "$ITEM" --verbose < sourcequestions 2>&1| tee /dev/stderr) COMMAND_OUTPUT=$(ollama run "$ITEM" --verbose < sourcequestions 2>&1| tee /dev/stderr)
# eval duration is sometimes listed in seconds and sometimes in milliseconds.
# Add up the values for each model
SUM=$(echo "$COMMAND_OUTPUT" | awk ' SUM=$(echo "$COMMAND_OUTPUT" | awk '
/eval duration:/ { /eval duration:/ {
value = $3 value = $3