speechless 0.5.0

LLM based agents with proactive interactions, long-term memory, external tool integration, and local deployment capabilities.

Speechless LLM based Agents

LLM based agents with proactive interactions, long-term memory, external tool integration, and local deployment capabilities.

Speechless.AI is committed to integrating the superior language processing and deep reasoning capabilities of large language models into practical business applications. By enhancing the model's language understanding, knowledge accumulation, and text creation abilities, and introducing long-term memory, external tool integration, and local deployment, our aim is to establish an intelligent collaborative partner that can independently interact, continuously evolve, and closely align with various business scenarios.

• Firstly, we focus on building a large model with enhanced reasoning capabilities, ensuring its outstanding performance in language processing and logical analysis.

• Next, we design and implement an efficient operational framework for the intelligent entity. This framework not only supports rapid deployment and invocation of the model but also boasts features like autonomous interaction, real-time feedback adjustment, context awareness, and long-term memory. For instance, in customer service scenarios, the intelligent entity can provide more precise and personalized responses based on a user's historical interactions and current context. In content recommendation scenarios, it can dynamically adjust its strategies by capturing real-time shifts in user interests.

• Ultimately, we integrate it with real business scenarios, ensuring that the intelligent entity seamlessly aligns with various business processes, delivering tangible value to enterprises.

Models

Models Repositry

• speechless-mistral-dolphin-orca-platypus-samantha-7b 2023.10.14 GPTQ GGUF AWQ

• speechless-tora-code-7b-v1.0 2023.10.10 GPTQ GGUF AWQ

• speechless-code-mistral-7b-v1.0 2023.10.10 GPTQ GGUF AWQ

• speechless-codellama-34b-v2.0 2023.10.04 GPTQ GGUF AWQ

• speechless-llama2-13b 2023.09.14 GPTQ GGUF GGML (deprecated)

• speechless-codellama-airoboros-orca-platypus-13b 2023.09.19

• speechless-codellama-dolphin-orca-platypus-34b 2023.09.14

• speechless-llama2-dolphin-orca-platypus-13b 2023.09.16

• speechless-codellama-34b-v1.0 2023.09.14

• speechless-codellama-platypus-13b 2023.09.13

• speechless-codellama-orca-13b 2023.09.13

• speechless-llama2-hermes-orca-platypus-wizardlm-13b 2023.09.10 GPTQ GGUF AWQ

• speechless-llama2-hermes-orca-platypus-13b 2023.09.02

• speechless-llama2-luban-orca-platypus-13b 2023.09.01

• speechless-hermes-coig-lite-13b 2023.08.22

CodeLlama based Models

speechless-codellama-34b-v2.0 2023.10.04 GPTQ GGUF AWQ

speechless-codellama-airoboros-orca-platypus-13b 2023.09.19

Mistral based Models

speechless-code-mistral-7b-v1.0 2023.10.10 GPTQ GGUF AWQ

Tora based Models

speechless-tora-code-7b-v1.0 2023.10.10 GPTQ GGUF AWQ

Llama2 based Models

speechless-llama2-hermes-orca-platypus-wizardlm-13b 2023.09.10 GPTQ GGUF AWQ

speechless-mistral-dolphin-orca-platypus-samantha-7b 2023.10.14 GPTQ GGUF AWQ

Datasets

jondurbin/airoboros-2.2.1

Open-Orca/OpenOrca

garage-bAInd/Open-Platypus

WizardLM/WizardLM_evol_instruct_V2_196k

ehartford/dolphin

ehartford/samantha-data

speechless.finetune

Install speechless

pip install speechless

Prepare train dataset

Focus on instruction following, currently not supporting multi-turn dialogue scenarios.

The training dataset is a jsonl file, with each line containing a JSON formatted instruction data. The data format is as follows:

{
    "instruction": "<instruction>", # Cann't be empty.
    "input': "<input>", # Can be an empty string.
    "response': "<response>", # Cann't be empty.
    "category": "mydata",
    "skip_prompt_formatting': False, # Keep this setting
    "system": "",
}

Run run_finetune.sh

mkdir -p tasks/speechless-llama2-13b && cd tasks/speechless-llama2-13b
python -m speechless.finetune \
    --base_model meta-llama/Llama-2-13B-hf \
    --dataset <path_to_your_dataset_file> \
    --dataset_format instruction-input-response \
    --model_max_len 4096 \
    --bits 4 \
    --lora_r 16 \
    --min_batch_size 4 \
    --gradient_accumulation_steps 16 \
    --learning_rate 2e-4 \
    --num_train_epochs 2 \
    --num_gpus 2 \

python -m speechless.tools.merge_peft_adapters \
    --base_model meta-llama/Llama-2-13B-hf \
    --peft_model_path ${CHECKPOINT_DIR} \
    --merged_model_path ${TASK_MODEL_PATH}

speechless.completion

python -m speechless.completion \
    create \
    --model ${TASK_MODEL_PATH} \
    --prompt ${PROMPT} \
    --prompts_file ${PROMPTS_FILE_PATH} \
    --temperature 0.75 \
    --top_p 0.9 \
    --top_k 50 \

speechless.api.server

python -m speechless.api.server \
    start \
    --model ${TASK_MODEL_PATH} \
    --backbone vllm \
    --host 0.0.0.0 \
    --port 5001

speechless.eval

Speechless supports HumanEval, MultiPL-E, SQLEval, lm-evaluation-harness.

lm-evluation-harness

# ${SPEECHLESS_ROOT}/speechless/scripts/run_lmeval.sh
python -m speechless.eval.lmeval \
   genrate \
    --model ${TASK_MODEL_PATH} \
    --output_dir ${EVAL_OUTPUT_DIR} \

python -m speechless.eval.lmeval \
    eval \
    -eval_dir ${EVAL_OUTPUT_DIR}

# git clone https://github.com/EleutherAI/lm-evaluation-harness
# cd lm-evaluation-harness
# pip install -e .
make lm_eval

bigcode-evaluation-harness

docker pull ghcr.io/bigcode-project/evaluation-harness
docker tag ghcr.io/bigcode-project/evaluation-harness evaluation-harness

HumanEval

Execute the HumanEval geenrate command on the GPU server where the model is located.

python -m speechless.eval.humaneval \
    genrate \
    --model ${TASK_MODEL_PATH} \
    --output_dir ${EVAL_OUTPUT_DIR} \

python -m speechless.eval.humaneval \
    eval \
    --eval_dir ${EVAL_OUTPUT_DIR}

# make humaneval_gen
# call eval/humaneval_gen_vllm.py
bash ./eval/run_human_eval_gen.sh ${TEST_MODEL_PATH} ${HUMANEVAL_GEN_OUTPUT_FILE}

# make humaneval
python eval/run_humaneval.py \
    ${HUMANEVAL_GEN_OUTPUT_FILE} \
    --problem_file ${PWD}/eval/datasets/openai_humaneval/HumanEval.jsonl.gz

MultiPL-E

docker pull ghcr.io/bigcode-project/evaluation-harness-multiple
docker tag ghcr.io/bigcode-project/evaluation-harness-multiple evaluation-harness-multiple

python -m speechless.eval.multiple \
    genrate \
    --name ${TASK_MODEL_PATH} \
    --output_dir_prefix ${EVAL_OUTPUT_DIR} \

python -m speechless.eval.multiple \
    eval \
    --results_dir ${EVAL_OUTPUT_DIR}

MULTIPL_E_RESULTS_DIR=eval_results/multipl_e/${SERVED_MODEL_NAME}
SERVED_MODEL_NAME=$(shell basename ${TEST_MODEL_PATH})

#make multipl_e_gen
python eval/multiple.py \
    generate \
    --name ${TEST_MODEL_PATH} \
    --temperature 0.2 \
    --batch-size 20 \
    --completion-limit 20 

# docker pull multipl-e-eval
#make multipl_e_eval
# cd eval_results/multipl_e/${SERVED_MODEL_NAME} && bash ../../eval/run_multipl_e_eval.sh
python eval/multiple.py eval --results_dir eval_results/multipl_e/Mistral-7B-v0.1

# make multipl_e_results
# python ${PWD}/eval/MultiPL-E/pass_k.py -k 1 ${MULTIPL_E_RESULTS_DIR}/*
python eval/multiple.py results --results_dir eval_results/multipl_e/Mistral-7B-v0.1

SQLEval

python -m speechless.eval.sqleval \
    genrate \
    --model ${TASK_MODEL_PATH} \
    --output_dir ${EVAL_OUTPUT_DIR} \

python -m speechless.eval.sqleval \
    eval \
    --eval_dir ${EVAL_OUTPUT_DIR}

# Run docker postgres-sql-eval
cd nl2sql/sqleval && make codellama