### Start Qdrant Server

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Starts a Qdrant Docker container, exposing the default Qdrant port 6333 to the host.

```bash
docker run -p 6333:6333 qdrant/qdrant
```

--------------------------------

### Dockerfile Installation and Configuration

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Commands to install dependencies and configure GuardRails. Comment out GuardRails lines if not needed.

```bash
RUN python -m pip install -r requirements.txt
RUN python -m pip install guardrails-ai==0.5.13
RUN guardrails configure --disable-metrics --disable-remote-inferencing --token xxxxx
RUN guardrails hub install hub://guardrails/valid_choices
```

--------------------------------

### Install jq Package

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Installs the jq package on Debian-based systems using apt-get.

```bash
sudo apt-get update
sudo apt-get install jq
```

--------------------------------

### Start VLLM Server

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Starts the VLLM server using a provided bash script. Assumes the script is located at 'scripts/start_vllm.sh'.

```bash
bash scripts/start_vllm.sh
```

--------------------------------

### ETH Start & End Times

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Sets the warmup and testing periods for Ethereum (ETH).

```bash
warmup_start_time = "2023-02-13"
warmup_end_time = "2023-04-02"
test_start_time = "2023-04-03"
test_end_time = "2023-12-19"
```

--------------------------------

### Equities Start & End Times

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Sets the warmup and testing periods for equities like HON, JNJ, UVV, and MSFT.

```bash
warmup_start_time = "2020-07-01"
warmup_end_time = "2020-09-30"
test_start_time = "2020-10-01"
test_end_time = "2021-05-06"
```

--------------------------------

### BTC Start & End Times

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Sets the warmup and testing periods for Bitcoin (BTC).

```bash
warmup_start_time = "2023-02-11"
warmup_end_time = "2023-04-04"
test_start_time = "2023-04-05"
test_end_time = "2023-12-19"
```

--------------------------------

### ETF Start & End Times

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Sets the warmup and testing periods for Exchange-Traded Funds (ETF). Note the trailing commas.

```bash
warmup_start_time = "2019-07-29",
warmup_end_time = "2019-12-30",
test_start_time = "2020-01-02",
test_end_time = "2020-09-21",
```

--------------------------------

### Run Framework Warm-up

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Executes the 'warmup' command within the 'devon' Docker container, mounting the current directory and using host networking.

```bash
docker run -it -v .:/workspace --network host devon warmup
```

--------------------------------

### Run Framework Testing

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Executes the 'test' command within the 'devon' Docker container, mounting the current directory and using host networking.

```bash
docker run -it -v .:/workspace --network host devon test
```

--------------------------------

### Build Docker Container

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Builds a Docker image tagged as 'devon' using the provided Dockerfile in the current directory.

```bash
docker build -t devon -f Dockerfile .
```

--------------------------------

### Environment Variables for API Keys

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Configuration for API keys required for OpenAI and HuggingFace. Save these in a .env file.

```bash
OPENAI_API_KEY=XXXXXX-XXXXXX-XXXXXX-XXXXXX-XXXXXX
HUGGING_FACE_HUB_TOKEN=XXXXXX-XXXXXX-XXXXXX-XXXXXX-XXXXXX
```

--------------------------------

### Configure Meta Settings with Pkl

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Define the meta configuration for the framework in Pkl, including run names, agent names, trading symbols, time ranges, and model parameters.

```pkl
hidden config = new meta.MetaConfig {
    run_name = "exp"
    agent_name = "finmem_agent"
    trading_symbols = new Listing {
            "BTC-USD"
    }
    warmup_start_time = "2023-02-11"
    warmup_end_time = "2023-03-10"
    test_start_time = "2023-03-11"
    test_end_time = "2023-04-04"
    top_k = 5
    look_back_window_size = 3
    momentum_window_size = 3
    tensor_parallel_size = 2
    embedding_model = "text-embedding-3-large"
    chat_model = "catMemo"
    chat_vllm_endpoint = "http://0.0.0.0:8000"
    chat_parameters = new Mapping {
        ["temperature"] = 0.6
    }
}
```

--------------------------------

### Resume Framework Warm-up from Checkpoint

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Restarts the framework warm-up process from the last saved checkpoint using the 'devon warmup-checkpoint' command in Docker.

```bash
docker run -it -v .:/workspace --network host devon warmup-checkpoint
```

--------------------------------

### Resume Framework Testing from Checkpoint

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Restarts the framework testing process from the last saved checkpoint using the 'devon test-checkpoint' command in Docker.

```bash
docker run -it -v .:/workspace --network host devon test-checkpoint
```

--------------------------------

### Generate Configuration with Docker

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Runs the 'devon config' command within a Docker container, mounting the current directory to /workspace and using host networking.

```bash
docker run -it -v .:/workspace --network host devon config
```

--------------------------------

### Generate Metric Report

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Generates a metric report for the evaluation framework using the 'devon eval' command within a Docker container.

```bash
docker run -it -v .:/workspace --network host devon eval
```

--------------------------------

### Configure Chat Model with Pkl

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Define a chat model configuration in Pkl, specifying parameters like model path, type, and inference engine. This is used for deploying fine-tuned or merged LLM models.

```pkl
llama3_1_instruct_8b: ChatModelConfig = new {
    chat_model = "meta-llama/Meta-Llama-3.1-8B-Instruct"
    chat_model_type = "instruction"
    chat_model_inference_engine = "vllm"
    chat_endpoint = null
    chat_template_path = null
    chat_system_message = "You are a helpful assistant."
    chat_parameters = new Mapping {}
  }
```

--------------------------------

### Pull Qdrant Docker Image

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Pulls the latest official Qdrant Docker image from Docker Hub.

```bash
docker pull qdrant/qdrant
```

--------------------------------

### BibTeX Entry for InvestorBench

Source: https://github.com/felis33/investor-bench/blob/main/README.md

This is the BibTeX entry for the InvestorBench paper, useful for academic citations.

```bibtex
@inproceedings{li-etal-2025-investorbench,
    title = "{INVESTORBENCH}: A Benchmark for Financial Decision-Making Tasks with {LLM}-based Agent",
    author = "Li, Haohang and Cao, Yupeng and Yu, Yangyang and Javaji, Shashidhar Reddy and Deng, Zhiyang and He, Yueru and Jiang, Yuechen and Zhu, Zining and Subbalakshmi, Koduvayur and Xiong, Guojun and others",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.acl-long.126/",
    doi = "10.18653/v1/2025.acl-long.126",
    pages = "2509--2525",
    ISBN = "979-8-89176-251-0"
}
```

--------------------------------

### Pull VLLM Docker Image

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Pulls the latest official VLLM Docker image with OpenAI compatible server.

```bash
docker pull vllm/vllm-openai:latest
```

--------------------------------

### Register Chat Model in Pkl

Source: https://github.com/felis33/investor-bench/blob/main/README.md

Register the defined chat model identifier in a Pkl mapping for use within the framework. This associates a short name with the model identifier.

```pkl
chat_model_dict = new Mapping {
    ["llama-3.1-8b-instruct"] = llama3_1_instruct_8b
  }
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.