# Error when loading models trained with more than 999 samples

**URL:** <https://support.prodi.gy/t/error-when-loading-models-trained-with-more-than-999-samples/31>\
**Category:** Uncategorized\
**Tags:** textcat, done, spacy\
**Created:** [September 14, 2017, 5:53am UTC](https://support.prodi.gy/t/error-when-loading-models-trained-with-more-than-999-samples/31 "2017-09-14T05:53:34Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![plusepsilon](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/plusepsilon/32/49_2.png) [@plusepsilon](https://support.prodi.gy/u/plusepsilon)\
**Post date:** [September 14, 2017, 5:53am UTC](https://support.prodi.gy/t/error-when-loading-models-trained-with-more-than-999-samples/31/1 "2017-09-14T05:53:34Z")

</div>

Hi, sorry to spam all these questions.

I found a peculiar error where if my annotated dataset is at 1000 samples it gives a dimension error.

Let’s say I have an `annotated-data.json` with more than 1000 samples. When I have 999 samples I can load the model ok, but not at 1000 samples.

```python
prodigy drop test
prodigy dataset test "yoyo"
head -n 1000 annotated-data.jsonl > small-data.jsonl
prodigy db-in test small-data.jsonl
prodigy textcat.batch-train test en_core_web_sm --output test -n 1
python -c "import spacy; spacy.load('test')"

```

Here is the error.

```python
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/Users/apewu/writelab/prodigy/lib/python3.6/site-packages/spacy/ __init__.py", line 13, in load
    return util.load_model(name, **overrides)
  File "/Users/apewu/writelab/prodigy/lib/python3.6/site-packages/spacy/util.py", line 107, in load_model
    return load_model_from_path(Path(name), **overrides)
  File "/Users/apewu/writelab/prodigy/lib/python3.6/site-packages/spacy/util.py", line 138, in load_model_from_path
    return nlp.from_disk(model_path)
  File "/Users/apewu/writelab/prodigy/lib/python3.6/site-packages/spacy/language.py", line 541, in from_disk
    util.from_disk(path, deserializers, exclude)
  File "/Users/apewu/writelab/prodigy/lib/python3.6/site-packages/spacy/util.py", line 483, in from_disk
    reader(path / key)
  File "/Users/apewu/writelab/prodigy/lib/python3.6/site-packages/spacy/language.py", line 537, in <lambda>
    deserializers[proc.name] = lambda p, proc=proc: proc.from_disk(p, vocab=False)
  File "spacy/pipeline.pyx", line 170, in spacy.pipeline.BaseThincComponent.from_disk (spacy/pipeline.cpp:11298)
  File "/Users/apewu/writelab/prodigy/lib/python3.6/site-packages/spacy/util.py", line 483, in from_disk
    reader(path / key)
  File "spacy/pipeline.pyx", line 163, in spacy.pipeline.BaseThincComponent.from_disk.load_model (spacy/pipeline.cpp:10856)
  File "/Users/apewu/writelab/prodigy/lib/python3.6/site-packages/thinc/neural/_classes/model.py", line 352, in from_bytes
    copy_array(dest, param[b'value'])
  File "/Users/apewu/writelab/prodigy/lib/python3.6/site-packages/thinc/neural/util.py", line 48, in copy_array
    dst[:] = src
ValueError: could not broadcast input array from shape (128) into shape (64)

```

---

<div class="post-metadata">

**Author:** ![honnibal](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/honnibal/32/35_2.png) [@honnibal](https://support.prodi.gy/u/honnibal)\
**Post date:** [September 14, 2017, 10:31am UTC](https://support.prodi.gy/t/error-when-loading-models-trained-with-more-than-999-samples/31/2 "2017-09-14T10:31:42Z")

</div>

Thanks for the report.

As weird as this issue sounds, it does make sense: there’s a heuristic in spaCy’s textcat class that uses a smaller neural net architecture if very few examples are available. The cutoff is 1000 examples. It appears the larger-text model is broken. I thought I had a test for this, but apparently not!

---

<div class="post-metadata">

**Author:** ![honnibal](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/honnibal/32/35_2.png) [@honnibal](https://support.prodi.gy/u/honnibal)\
**Post date:** [September 19, 2017, 12:30pm UTC](https://support.prodi.gy/t/error-when-loading-models-trained-with-more-than-999-samples/31/3 "2017-09-19T12:30:00Z")

</div>

The issue only occurs when there are no pre-trained vectors loaded (as in the `en_core_web_sm` model). I’d been testing with a vectors model such as `en_vectors_web_lg`, which is why the problem didn’t show up before.

The root cause is actually a spaCy bug, which should be fixed in the next release of `spacy-nightly` – which is not _quite_ nightly, because the models take some time to retrain :p.

In the meantime, if you base your textcat stuff off `en_vectors_web_lg`, you’ll be able to take advantage of the GloVe vectors, and both the small and large models should work properly.

Thanks again for the report!

---

<div class="post-metadata">

**Author:** ![ines](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/ines/32/3_2.png) [@ines](https://support.prodi.gy/u/ines)\
**Post date:** [October 4, 2017, 2:32pm UTC](https://support.prodi.gy/t/error-when-loading-models-trained-with-more-than-999-samples/31/4 "2017-10-04T14:32:33Z")

</div>

Will be fixed in the upcoming Prodigy `v0.3.0`! 🎉

---

<div class="post-metadata">

**Author:** ![rodrigoccurvo](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/rodrigoccurvo/32/147_2.png) [@rodrigoccurvo](https://support.prodi.gy/u/rodrigoccurvo)\
**Post date:** [January 17, 2018, 1:40am UTC](https://support.prodi.gy/t/error-when-loading-models-trained-with-more-than-999-samples/31/5 "2018-01-17T01:40:30Z")

</div>

What is the underlying spacy bug? Is there a way around it?

I’ve stumbled on the same problem (exception stack), but without using Prodigy…

PS: I’m using pt\_core\_news\_sm model