deepseek-ai/Janus

Fork 0

mirror of https://github.com/deepseek-ai/Janus.git synced 2026-11-05 00:57:18 +00:00

[GH-ISSUE #77] JanusFlow示例代码修复建议 #58

New issue

Open

opened 2026-10-26 21:11:18 +00:00 by bsakaguchi · 3 comments

bsakaguchi commented

2026-10-26 21:11:18 +00:00

Owner

Originally created by @songafu on GitHub (Jan 28, 2025).
Original GitHub issue: https://github.com/deepseek-ai/Janus/issues/77

在使用JanusFlow的过程中，发现示例代码中的文生图部分在当前最新的Transformer版本（>=4.48.0）下无法正常运行且示例代码可能存在代码缺陷，建议修复如下：
（1）JanusFlow示例代码（文生图）中数据流的处理，存在变量引用逻辑错误
if step == 0:
outputs = vl_gpt.language_model.model(inputs_embeds=llm_emb,
use_cache=True,
attention_mask=attention_mask,
past_key_values=None)
past_key_values = []
for kv_cache in outputs.past_key_values: #should be outputs.past_key_values
k, v = kv_cache[0], kv_cache[1]
past_key_values.append((k[:, :, :inputs_embeds.shape[1], :], v[:, :, :inputs_embeds.shape[1], :]))
past_key_values = tuple(past_key_values)

（2）在当前最新的Transformer版本（>=4.48.0）下无法正常运行JanusFlow服务，报错如下，建议在Quick Start 提示用户选择使用较低版本的transformer（如4.38.2）或兼容最新的transformer版本修复。
llama/modeling_llama.py", line 551, in forward
past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'tuple' object has no attribute 'get_seq_length'

Originally created by @songafu on GitHub (Jan 28, 2025). Original GitHub issue: https://github.com/deepseek-ai/Janus/issues/77 在使用JanusFlow的过程中，发现示例代码中的文生图部分在当前最新的Transformer版本（>=4.48.0）下无法正常运行且示例代码可能存在代码缺陷，建议修复如下：（1）JanusFlow示例代码（文生图）中数据流的处理，存在变量引用逻辑错误 if step == 0: outputs = vl_gpt.language_model.model(inputs_embeds=llm_emb, use_cache=True, attention_mask=attention_mask, past_key_values=None) past_key_values = [] for kv_cache in outputs.past_key_values: #should be outputs.past_key_values k, v = kv_cache[0], kv_cache[1] past_key_values.append((k[:, :, :inputs_embeds.shape[1], :], v[:, :, :inputs_embeds.shape[1], :])) past_key_values = tuple(past_key_values) （2）在当前最新的Transformer版本（>=4.48.0）下无法正常运行JanusFlow服务，报错如下，建议在Quick Start 提示用户选择使用较低版本的transformer（如4.38.2）或兼容最新的transformer版本修复。 llama/modeling_llama.py", line 551, in forward past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: 'tuple' object has no attribute 'get_seq_length'

bsakaguchi commented

2026-10-26 21:11:19 +00:00

Author

Owner

@SimonYS001 commented on GitHub (Jan 29, 2025):

Good job!

@SimonYS001 commented on GitHub (Jan 29, 2025): Good job!

bsakaguchi commented

2026-10-26 21:11:20 +00:00

Author

Owner

@scifisatan commented on GitHub (Feb 4, 2025):

just created a fix for this https://github.com/deepseek-ai/Janus/pull/137

Replace the code on line 108 - 122

if step == 0:
            outputs = vl_gpt.language_model.model(inputs_embeds=llm_emb, 
                                             use_cache=True, 
                                             attention_mask=attention_mask,
                                             past_key_values=None)
            past_key_values = []
            for kv_cache in past_key_values:
                k, v = kv_cache[0], kv_cache[1]
                past_key_values.append((k[:, :, :inputs_embeds.shape[1], :], v[:, :, :inputs_embeds.shape[1], :]))
            past_key_values = tuple(past_key_values)
        else:
            outputs = vl_gpt.language_model.model(inputs_embeds=llm_emb, 
                                             use_cache=True, 
                                             attention_mask=attention_mask,
                                             past_key_values=past_key_values)

with this

if step == 0:
            past_key_values = None  # Ensure it starts as None
        else:
            past_key_values = tuple(past_key_values) if past_key_values else None  # Convert only if it's valid

        outputs = vl_gpt.language_model.model(
            inputs_embeds=llm_emb, 
            use_cache=True, 
            attention_mask=attention_mask,
            past_key_values=past_key_values  # Now correctly assigned
        )

Hope it helps :)

@scifisatan commented on GitHub (Feb 4, 2025): just created a fix for this https://github.com/deepseek-ai/Janus/pull/137 Replace the code on line 108 - 122 ``` if step == 0: outputs = vl_gpt.language_model.model(inputs_embeds=llm_emb, use_cache=True, attention_mask=attention_mask, past_key_values=None) past_key_values = [] for kv_cache in past_key_values: k, v = kv_cache[0], kv_cache[1] past_key_values.append((k[:, :, :inputs_embeds.shape[1], :], v[:, :, :inputs_embeds.shape[1], :])) past_key_values = tuple(past_key_values) else: outputs = vl_gpt.language_model.model(inputs_embeds=llm_emb, use_cache=True, attention_mask=attention_mask, past_key_values=past_key_values) ``` with this ``` if step == 0: past_key_values = None # Ensure it starts as None else: past_key_values = tuple(past_key_values) if past_key_values else None # Convert only if it's valid outputs = vl_gpt.language_model.model( inputs_embeds=llm_emb, use_cache=True, attention_mask=attention_mask, past_key_values=past_key_values # Now correctly assigned ) ``` Hope it helps :)

bsakaguchi commented

2026-10-26 21:11:20 +00:00

Author

Owner

@nv-samcheng commented on GitHub (Feb 15, 2025):

Just want to confirm if the kv cache bu used in Janus Flow?

@nv-samcheng commented on GitHub (Feb 15, 2025): Just want to confirm if the kv cache bu used in Janus Flow?

No labels

pull-request

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

deepseek-ai/Janus#58

No description provided.

Rows
Columns