Daniel

Now, it can be compiled with CMake.

prompt learning, loss:0.4483, acc: 87.27%

1

未做数据增强，划分为7:3 7份为训练集，３份用作evaluate/test set，　evaluate时关闭train(其实就是测试状态了）

new record: 20220526

1

class Precision Recall F1 Support 民生,故事 0.890 0.856 0.873 912 文化,文化 0.920 0.918 0.919 4194 娱乐,娱乐 0.922 0.942 0.932 5940 体育,体育 0.967 0.962 0.964 5614 财经,财经 0.861 0.828 0.844 4102...

Can not infer quantized model, but fp32 works well.

9

run main.exe with a q4_0 quantized model. for q4_0, it complains assert in ggml.c in function static void ggml_compute_forward_soft_max_f32( line 9342: assert(sum > 0.0); sum is -nan (ind)

About the calculation of overhead.

4

https://github.com/ggerganov/ggml/issues/356

unsupported ops

4

There's 5 ops are not supported yet =========== top_k =========== =========== fake_quantize_moving_average_abs_max =========== =========== ctc_align =========== =========== lstm =========== =========== fake_channel_wise_dequantize_max_abs =========== There's 2 ops are not supported yet ===========...

where is directory splits ?

1

where? didn't find it in this repo.

建议在右上角位置加个带彩色 图标的按钮，方便找到。

Doesn't support input vector? There is no inputs_embeds

6

in the parameter list of forward, there is no inputs_embeds parameter, so it doesn't support inputing vector (embedding vector not token id)?

### Installation Method | 安装方法与平台 OneKeyInstall (一键安装脚本-windows) ### Version | 版本 Latest | 最新版 ### OS | 操作系统 Windows ### Describe the bug | 简述 Windows 10 latex编译时找不到字体，编译失败。 ### Screen...