aboutsummaryrefslogtreecommitdiff
path: root/benchmark/reports/mini-agi/10_TestRememberMultipleWithNoise.json
diff options
context:
space:
mode:
Diffstat (limited to 'benchmark/reports/mini-agi/10_TestRememberMultipleWithNoise.json')
-rw-r--r--benchmark/reports/mini-agi/10_TestRememberMultipleWithNoise.json31
1 files changed, 31 insertions, 0 deletions
diff --git a/benchmark/reports/mini-agi/10_TestRememberMultipleWithNoise.json b/benchmark/reports/mini-agi/10_TestRememberMultipleWithNoise.json
new file mode 100644
index 000000000..08c2b7075
--- /dev/null
+++ b/benchmark/reports/mini-agi/10_TestRememberMultipleWithNoise.json
@@ -0,0 +1,31 @@
+{
+ "command": "agbenchmark start --test TestRememberMultipleWithNoise",
+ "completion_time": "2023-07-17-21:19",
+ "metrics": {
+ "run_time": "74.3 seconds",
+ "highest_difficulty": "No successful tests"
+ },
+ "tests": {
+ "TestRememberMultipleWithNoise": {
+ "data_path": "agbenchmark/challenges/memory/m3",
+ "is_regression": false,
+ "task": "Follow the instructions in the instructions_1.txt file",
+ "answer": "3145\n3791\n9317\n9471",
+ "description": "Tests ability for the agent to remember information between each action. Multiple ids are presented in different files and the last file instructs the agent to write these ids into another file. Some noise is also added to the files to test the agent's ability to filter out noise.",
+ "metrics": {
+ "difficulty": "intermediate",
+ "success": false,
+ "fail_reason": "[Errno 2] No such file or directory: 'C:\\\\Users\\\\silen\\\\miniagi\\\\result.txt'",
+ "success_%": 0.0,
+ "run_time": "74.059 seconds"
+ },
+ "reached_cutoff": true
+ }
+ },
+ "config": {
+ "workspace": "${os.path.join(Path.home(), 'miniagi')}"
+ },
+ "additional": {
+ "model": "gpt-4"
+ }
+}