OSError: Not found: "/mnt/data0/Garmin/LAVIS/lavis/weight/vicuna-7b-2/tokenizer.model"

[kristen-ding]

load_path = './EmoVIT/model_weights1.pth' model = torch.load(load_path, map_location=torch.device('cpu'))

When conducting my tests, I encountered an OSError. How should I modify the default path?

[loknini]

I've encountered the same issue. I've gone through all the configuration files but couldn't find where it's being set. I hope to get an answer.

[ditto66]

I have the same question. Can you pls help fix that? @aimmemotion

[dazzleeee]

I have the same question. Can you pls help fix that? @aimmemotion

I have the same question,do you fix that?

[dazzleeee]

I've encountered the same issue. I've gone through all the configuration files but couldn't find where it's being set. I hope to get an answer.

I find the dir in" ./model_weights1.pth",but I don't how to fix it.

[yutchina]

确实有问题 需要重新训练一个

[ditto66]

确实有问题 需要重新训练一个

Do you success running the training code?

I encountered the TypeError when I tried to retrain the model:

File "EmoVIT/LAVIS/lavis/datasets/datasets/base_dataset.py", line 47, in _add_instance_ids
    ann[ key] = str(idx)
TypeError: list indices must be integers or slices, not str

[yutchina]

是的 我成功复现了 但是修改了很多代码

---- Replied Message ---- | From | @.> | | Date | 09/25/2024 14:36 | | To | aimmemotion/EmoVIT @.> | | Cc | yutchina @.>, Comment @.> | | Subject | Re: [aimmemotion/EmoVIT] OSError: Not found: "/mnt/data0/Garmin/LAVIS/lavis/weight/vicuna-7b-2/tokenizer.model" (Issue #3) |

确实有问题 需要重新训练一个

Do you success running the training code?

I encountered the TypeError when I tried to retrain the model:

File "EmoVIT/LAVIS/lavis/datasets/datasets/base_dataset.py", line 47, in _add_instance_ids ann[ key] = str(idx) TypeError: list indices must be integers or slices, not str

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>

[dazzleeee]

可以参考下你的代码嘛~

[ditto66]

是的 我成功复现了 但是修改了很多代码 … ---- Replied Message ---- | From | @.> | | Date | 09/25/2024 14:36 | | To | aimmemotion/EmoVIT @.> | | Cc | yutchina @.>, Comment @.> | | Subject | Re: [aimmemotion/EmoVIT] OSError: Not found: "/mnt/data0/Garmin/LAVIS/lavis/weight/vicuna-7b-2/tokenizer.model" (Issue #3) | 确实有问题 需要重新训练一个 Do you success running the training code? I encountered the TypeError when I tried to retrain the model: File "EmoVIT/LAVIS/lavis/datasets/datasets/base_dataset.py", line 47, in _add_instance_ids ann[ key] = str(idx) TypeError: list indices must be integers or slices, not str — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>

If you don't mind, could you please share the code? We would really appreciate it.

[yutchina]

是的 我成功复现了 但是修改了很多代码 … ---- Replied Message ---- | From | @.**> | | Date | 09/25/2024 14:36 | | To | aimmemotion/EmoVIT @.**> | | Cc | yutchina @.**>, Comment @.**> | | Subject | Re: [aimmemotion/EmoVIT] OSError: Not found: "/mnt/data0/Garmin/LAVIS/lavis/weight/vicuna-7b-2/tokenizer.model" (Issue #3) | 确实有问题 需要重新训练一个 Do you success running the training code? I encountered the TypeError when I tried to retrain the model: File "EmoVIT/LAVIS/lavis/datasets/datasets/base_dataset.py", line 47, in _add_instanceids ann[ key] = str(idx) TypeError: list indices must be integers or slices, not str — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @_.***>

If you don't mind, could you please share the code? We would really appreciate it.

这花费了我很多精力和算力资源 所以我更倾向于合作

[kristen-ding]

是的 我成功复现了 但是修改了很多代码 … ---- Replied Message ---- | From | @.> | | Date | 09/25/2024 14:36 | | To | aimmemotion/EmoVIT @.> | | Cc | yutchina @.>, Comment @.> | | Subject | Re: [aimmemotion/EmoVIT] OSError: Not found: "/mnt/data0/Garmin/LAVIS/lavis/weight/vicuna-7b-2/tokenizer.model" (Issue #3) | 确实有问题 需要重新训练一个 Do you success running the training code? I encountered the TypeError when I tried to retrain the model: File "EmoVIT/LAVIS/lavis/datasets/datasets/base_dataset.py", line 47, in _add_instance_ids ann[ key] = str(idx) TypeError: list indices must be integers or slices, not str — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>

Can you share your code, thanks.

[ditto66]

是的 我成功复现了 但是修改了很多代码 … ---- Replied Message ---- | From | @.**> | | Date | 09/25/2024 14:36 | | To | aimmemotion/EmoVIT @.**> | | Cc | yutchina @.**>, Comment @.**> | | Subject | Re: [aimmemotion/EmoVIT] OSError: Not found: "/mnt/data0/Garmin/LAVIS/lavis/weight/vicuna-7b-2/tokenizer.model" (Issue #3) | 确实有问题 需要重新训练一个 Do you success running the training code? I encountered the TypeError when I tried to retrain the model: File "EmoVIT/LAVIS/lavis/datasets/datasets/base_dataset.py", line 47, in _add_instanceids ann[ key] = str(idx) TypeError: list indices must be integers or slices, not str — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @_.***>

If you don't mind, could you please share the code? We would really appreciate it.

这花费了我很多精力和算力资源 所以我更倾向于合作

Just have a question: How are multiple conversation data sets used in training? For example, are they concatenated into a single text input or treated as individual samples?"

[JoJoistheBestOne]

it's a pretty that the talented work is shared with open source but with some trivial mistakes..

[kristen-ding]

I solved it using this.