Time Series Forecasting
Transformers
Safetensors
Timer-S1
text-generation
time series
time-series
forecasting
foundation models
pretrained models
time series foundation models
quantized
4-bit precision
bitsandbytes
unofficial
custom_code
Instructions to use geetu040/Timer-S1-quantized-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use geetu040/Timer-S1-quantized-4bit with Transformers:
# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("geetu040/Timer-S1-quantized-4bit", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
File size: 1,274 Bytes
8bc63f3 f8acd96 8bc63f3 f8acd96 8bc63f3 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 | {
"architectures": [
"TimerS1ForPrediction"
],
"auto_map": {
"AutoConfig": "configuration_TimerS1.TimerS1Config",
"AutoModelForCausalLM": "modeling_TimerS1.TimerS1ForPrediction"
},
"dropout_rate": 0.1,
"dtype": "bfloat16",
"hidden_act": "silu",
"hidden_size": 1024,
"initializer_range": 0.02,
"input_token_len": 16,
"intermediate_size": 4096,
"max_position_embeddings": 12800,
"model_type": "Timer-S1",
"num_attention_heads": 16,
"num_experts": 32,
"num_experts_per_token": 2,
"num_hidden_layers": 24,
"num_mtp_tokens": 16,
"output_token_lens": [
16
],
"quantiles": [
0.1,
0.2,
0.3,
0.4,
0.5,
0.6,
0.7,
0.8,
0.9
],
"quantization_config": {
"_load_in_4bit": true,
"_load_in_8bit": false,
"bnb_4bit_compute_dtype": "bfloat16",
"bnb_4bit_quant_storage": "uint8",
"bnb_4bit_quant_type": "fp4",
"bnb_4bit_use_double_quant": false,
"llm_int8_enable_fp32_cpu_offload": false,
"llm_int8_has_fp16_weight": false,
"llm_int8_skip_modules": null,
"llm_int8_threshold": 6.0,
"load_in_4bit": true,
"load_in_8bit": false,
"quant_method": "bitsandbytes"
},
"rope_theta": 10000,
"transformers_version": "4.57.6",
"use_cache": true
}
|