Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
This is the tool set of we crawled in Apr 2404 in the ToolBench format and used to train the StableToolBench-Mirror API model. To use this tool set, download the .tar.gz file and unzip it with `tar xzvf toolenv2404_filtered.tar.gz` and copy the path of the tools folder in the running script of ToolBench. For example,
|
2 |
+
|
3 |
+
|
4 |
+
```bash
|
5 |
+
export TOOLBENCH_KEY=""
|
6 |
+
export OPENAI_KEY=""
|
7 |
+
export OPENAI_API_BASE=""
|
8 |
+
export PYTHONPATH=./
|
9 |
+
export GPT_MODEL="gpt-3.5-turbo-16k"
|
10 |
+
export SERVICE_URL="http://localhost:8080/virtual"
|
11 |
+
export OUTPUT_DIR="data/answer/virtual_chatgpt_cot"
|
12 |
+
group=G1_instruction
|
13 |
+
mkdir -p $OUTPUT_DIR; mkdir -p $OUTPUT_DIR/$group
|
14 |
+
|
15 |
+
python toolbench/inference/qa_pipeline_multithread.py \
|
16 |
+
--tool_root_dir toolenv/toolenv2404_filtered \ # This is the place where you fill the path to the tools folder.
|
17 |
+
--backbone_model chatgpt_function \
|
18 |
+
--openai_key $OPENAI_KEY \
|
19 |
+
--max_observation_length 1024 \
|
20 |
+
--method CoT@1 \
|
21 |
+
--input_query_file solvable_queries/test_instruction/${group}.json \
|
22 |
+
--output_answer_file $OUTPUT_DIR/$group \
|
23 |
+
--toolbench_key $TOOLBENCH_KEY \
|
24 |
+
--num_thread 1
|
25 |
+
|
26 |
+
```
|