[LAB 14 · 图像问答：视觉条件与问题先验](/course/notebooks/11d9f74871544a77b42024a9fe06b81a.html)

# 5. 学习曲线与可以被推翻的结论 · 实现代码
[独立阅读](/course/notebook-blocks/040689f0e3594ff988d47a70c68dbab6.html)
按问题继续

RGB+问题输入、question-only对照、同图换问与同问换图
展开完整实现代码
```
fig,axes=plt.subplots(1,3,figsize=(11,4))
for i,ax in enumerate(axes):
    ax.imshow(controls_x[i].permute(1,2,0));ax.axis('off')
    rows=['count: ','left color: ','right shape: ']
    text='\n'.join(rows[j]+ANSWERS[int(controls_pred[i,j])]+' [Q-only: '+ANSWERS[int(controls_qo[i,j])]+']' for j in range(3))
    ax.set_title(text,fontsize=10)
fig.suptitle('LAB14: same questions, changed image; three questions per image',fontsize=13)
fig.tight_layout();fig.savefig(OUT/'lab14-vqa-controls.png',dpi=160);plt.show();plt.close(fig)
fig,ax=plt.subplots(figsize=(10,3.6));labels=['image + Q','Q only','zero pixels','shuffled pixels'];keys=['test_image_question','test_question_only','test_zero_pixels','test_shuffled_pixels'];values=[metrics[k]['overall'] for k in keys]
ax.bar(labels,values,color=['#356b91','#ad8251','#9da6ad','#9da6ad']);ax.set(ylim=(0,1.1),ylabel='answer accuracy (540 held-out pairs)')
for i,v in enumerate(values):ax.text(i,v+.02,f'{v:.1%}',ha='center')
fig.tight_layout();fig.savefig(OUT/'lab14-vqa-ablation.png',dpi=160);plt.show();plt.close(fig)
fig,ax=plt.subplots(figsize=(8,3.2))
for label in ['image+question','question-only']:
    r=[v for v in curve if v['model']==label];ax.plot([v['epoch'] for v in r],[v['train_ce'] for v in r],label=label)
ax.set(xlabel='epoch',ylabel='answer cross entropy');ax.legend();fig.tight_layout();fig.savefig(OUT/'lab14-vqa-learning.png',dpi=160);plt.show();plt.close(fig)
result={'notebook_id':'11d9f74871544a77b42024a9fe06b81a','execution':'local CPU, fresh process; not a Deepnote cloud run','created_utc':datetime.now(timezone.utc).isoformat(),'elapsed_seconds':time.perf_counter()-START,'environment':{'python':sys.version,'torch':torch.__version__,'numpy':np.__version__,'platform':platform.platform()},'split':{'train_seed':14031,'test_seed':14032,'train_images':720,'test_images':180,'questions_per_image':3,'split_before_question_expansion':True,'train_red_prior':.75,'test_color_prior':'uniform by construction'},'metrics':metrics,'curve':curve,'controls':controls,'predictions':[{'truth':[ANSWERS[int(v)] for v in Yte[i]],'image_question':[ANSWERS[int(v)] for v in normal[i]],'question_only':[ANSWERS[int(v)] for v in qo[i]]} for i in range(len(Xte))],'files_sha256':{p.name:hashlib.sha256(p.read_bytes()).hexdigest() for p in sorted(OUT.glob('*.pt'))},'data_sha256':hashlib.sha256((OUT/'lab14-data.npz').read_bytes()).hexdigest()}
(OUT/'lab14-results.json').write_text(json.dumps(result,indent=2));print('saved',OUT,'seconds',round(result['elapsed_seconds'],2))
```