Error: command must be one of "from", "license", "template", "system", "adapter", "parameter", or "message"

[fahdmirza]

Hi ,
Thanks for the tool. I was facing error while running it and fixed main.py as follows to get it working:

def create_meta_file(local_dir, file_path):
# Create the metafile for Ollama

meta_file_content = f"""
FROM {file_path}

TEMPLATE \"""{{ if .System }}<|start_header_id|>system<|end_header_id|>
{{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|>
{{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|>
{{ .Response }}<|eot_id|>\"""
    """

print("Metafile content:")
print(meta_file_content)

meta_file_path = os.path.join(local_dir, "metafile.txt")
with open(meta_file_path, "w") as meta_file:
    meta_file.write(meta_file_content)
return meta_file_path

===============

Following was the error:

(olguff) Ubuntu@0136-ict-prxmx50056:~/olguff$ python3 main.py
Enter the Hugging Face model ID: shenzhi-wang/Llama3.1-8B-Chinese-Chat
Available files in the repository:

1. .gitattributes
2. README.md
3. config.json
4. gguf/llama3.1_8b_chinese_chat_f16.gguf
5. gguf/llama3.1_8b_chinese_chat_q4_k_m.gguf
6. gguf/llama3.1_8b_chinese_chat_q8_0.gguf
7. model-00001-of-00004.safetensors
8. model-00002-of-00004.safetensors
9. model-00003-of-00004.safetensors
10. model-00004-of-00004.safetensors
11. model.safetensors.index.json
12. special_tokens_map.json
13. tokenizer.json
14. tokenizer_config.json
    Enter the number of the file you want to download: 5
    File 'gguf/llama3.1_8b_chinese_chat_q4_k_m.gguf' already exists. Do you want to redownload it? (yes/no): no
    Skipping download of 'gguf/llama3.1_8b_chinese_chat_q4_k_m.gguf'.
    Enter the Ollama name for the model (default: gguf/llama3.1_8b_chinese_chat_q4_k_m): myllama
    Do you want to proceed with the 'ollama create' command? (yes/no): yes

Error: command must be one of "from", "license", "template", "system", "adapter", "parameter", or "message"
Model imported successfully!

PS. I am also doing a video on it for my channel https://www.youtube.com/@fahdmirza and it will be published soon. Thanks.

[fahadshery]

were you able to resolve this?

[fahdmirza]

were you able to resolve this?

This tool doesn't seem ready yet, but I have covered various other similar tools at https://www.youtube.com/@fahdmirza

[fahadshery]

thank you Mirza sahb. I have a gpu and I am interested in safetensor files. have you covered those or are planning those files to convert to ollama?

[ZhangChi7]

Error: command must be one of "from", "license", "template", "system", "adapter", "parameter", or "message"

[ZhangChi7]

My situation is that the first line in the generated Modelfile is "failed to get console mode for stdout: The handle is invalid." Deleting this line and then executing the command to create a new model works.